The URL can be used to link to this page

Home My WebLink AboutNorthwest Conservation & Electric Power Plan, Vol 2 1986Meeting the region’s electrical energy needs at the lowest cost through regional cooperation. NORTHWEST POWER PLANNING COUNCIL ae ace 1986 Power Plan Volume II Contents Chapter Page i; UN RQEDUIGSTIONN 5 csess icic ste sapere apes nec res nes sere sa re ress 1-1 2 ECONOMIC, DEMOGRAPHIC AND FUEL PRICE ASSUMPTIONS . 2-1 WARDEN fare. < cristo eters ro oie sims eros Sins ven sams mis ewe ean kus 2-1 Forecast Overview wee 22 Overview of the Regional Economy .....................6065 2-2 IMG OOINNS s cor3 sisi stores ors sys e SS 5 sts Sys's Gis nine BAe aie FGIes 2-3 Description of the Scenarios 2-4 Employment and Production «625 The Forest Products Industries ...............0...ce eee eens 2-5 Aluminum Industry ............cccccececcccsceceerecsceeees 2-9 Chemicals ............ 0000s cee cece eee ... 2-10 Agriculture and Food Processing 2-11 The High Technology Industries 2-12 Growth in Nonmanufacturing Industries ..................0005 2-16 Changes in Productivity Growth .............0. cece eee eee 2-19 Population, Households and Housing Stock .................055 2-19 Real per Capita Income .......... 20... cece cece eee eee eee 2-20 Alternative Fuel Prices .......... 2.000. c cece cece cence eee eee 2-21 Forecasts for Utility Service Areas ............... 00 ccc eee e eee 2-22 Appendix 2-A: Detail on Economic Input Assumptions ........... 2-A-1 Appendix 2-B: SIC Code Listings ........... 00.00 cece eee e eee 2-B-1 3 FORECAST OF DEMAND FOR ELECTRICITY ..............--.+- 3-1 ITT a5: 6 2:05 05s! sere oars: s ayes njere ten y ater 31 OVGIVIOW 6 site < 2.0 cor coms same oes oar 5 ome 3-2 Residential Demand 3-4 Commercial Demand 3-10 WIGGLE OGINGING iirc «5c cris hiss sie esl cee sis 2s SS ele SESS 3-12 WTUSERION DENNING core te: 6 ative: « tess ores s wre 6 rete sais Sores Sere siete Siew 4 3-16 Retail Electric Prices 0.0.0.2... cece cece ete eee eens 3-17 The Role of Demand Forecasts in Planning ..............-.+..+- 3-19 NEOCRON ars oye estas sears 6 apes tate a ate 6 orate siete store 6 ates tere rain ets 3-19 Defining Range of Uncertainty ............ 0.00. cece eee eee 3-19 Effects of Resource Choices on Price . . 3-20 Conservation Analysis 3-20 Forecast Concepts .......... [= 3-20 Electric Loads for Resource Planning .............-.-..0.0005 3-22 4 FINANCIAL ASSUMPTIONS AND RESOURCE COST EFFECTIVENESS. ...............-.00005 41 Explanation of Terms ................ 4-1 Escalation Rates ...........c.00005 4-4 Cost of Capital .................020005 45 Discount Rates ....................055 4-6 Resource Cost Effectiveness 4-10 Cost Effectiveness of the Model Conservation Standards ....... 4-10 Cost Effectiveness of Discretionary Resources ...............- 4-13 Cost Effectiveness of Near-Term Acquisitions ................. 4-14 Conclusions ......... 0... e cece cece cece teen eee e eee eee 4-14 Contents Chapter Page 5 CONSERVATION RESOURCES ..........-.02-00ee eee eect eee 5-1 Estimating the Conservation Resource ...........-.-0.es0e0e00e 5-1 Supply CONV eis ie. ote int cies « cine wien oie 0 ine wine sioininevieisis 5-1 Conservation Programs for Portfolio Analysis . -. 52 Compatibility with the Power System ...............0..00.00- 5-3 Residential Sector... 20.0... c cece cece eee ee eee 5-4 Space Heating Conservation in Existing Residential Buildings ... 5-4 Space Heating Conservation in New Residential Buildings .... . . 5-11 Electric Water Heating Conservation ...............00.00e eee 5-27 Conservation in Other Residential Appliances ................ 5-30 The Interaction between Internal Gains and Electric Space Heat. 5-34 Primary Sources for the Residential Sector ................... 5-35 Commercial Sector .............0.0020005 5-36 Waste Water Treatment ................5 5-40 Primary Sources for the Commercial Sector 5-40 Industrial Sector ............. 000 e eee eee 5-40 Primary Sources for the Industrial Sector 5-42 Irrigation Sector ..............0.0.0e0 ee .. 5-42 Primary Sources for the Irrigation Sector .................0005 5-44 6—_ GENERATING RESOURCES ‘isiavencsinasmes ert cms sums amass 6-1 Selection of Available Resources . 6-1 Cost and Availability ................... -. 62 Bonneville System Efficiency Improvements .................- 6-2 Utility System Efficiency Improvements ...................005 6-2 Development of Detailed Planning Information ................ 6-2 Transmission and Distribution System Efficiency Improvements ... 6-2 CONN CU ISHN rm ose winte water otrie cg ww oper cle olen eis eee ere eats 6-2 Hydropower Efficiency Improvements .. 6-3 Efficiency Improvement Measures 6-3 FROSOUICO COSt xr cai s arse meen cei tjet sesame sere canie 6-5 Resource Availability . . .. 66 Conclusion ......... 0.0. c cece cece cece eee eects 6-7 Thermal Plant Efficiency Improvements ...................00055 6-7 Geothermal Electric Power ............ 6-7 Generation Technology ....... 6-8 Project Cost and Performance ... 6-8 Resource Availability ......... 6-9 Conclusion .......... 6-10 Hydroelectric Power. ... . 6-11 Generation Technology ...... . 6-11 Project Cost and Performance . 6-11 Resource Availability . . 6-11 Conclusion ...............0..0000e 6-12 Municipal Solid Waste Electric Generation . 6-12 Generation Technology ........ 6-12 Project Cost and Performance . 6-13 Resource Availability ......... 6-13 Conclusion .......... 6-13 Solar Electric Power ..... 6-13 Generation Technology ....... -. 6-13 Project Cost and Performance ............. 00. c cece cece eee 6-14 Resource Availability ........0..... 0020s eee e cece eee eee 6-14 Conclusion ......... 6.6. c cece cece eee eect e eect eee 6-15 Chapter Page 6-15 6-15 6-16 6-16 6-17 6-17 6-17 Cogeneration Technology ... .. . 6-17 Project Cost and Performance . . 6-18 Resource Availability 6-18 (Conchision) ft) 85-0 ein a otete eine 6-19 Coal-Fired Electric Generation . .. . 6-19 Generation Technologies ...... 6-19 Project Cost and Performance ................ cece cece eee ee 6-20 FResource Aaeieblily 1% ore cisfois5siaystcin estar tials exci e clore tichel fore teres 6-21 Conclusion 6-21 Gas-Fired Electric Generation 6-23 Generation Technology 6-23 Project Cost and Performance 6-23 Resource Availability 6-23 Conclusion 6-23 INUCIGGN, 2.Sy2isiors « ayela-aerats sen Nalete + arctstowtie Wels aiatars hin acloynsaelehies anes 6-24 VIN Scr tose: o pcte sectets un satote « ated tones statol tates erenyshyo S teleeiels syeet f 6-25 WINES eter siite sacred ait eain easier reste tale seers 6-25 WNP-1 and WNP-3 Cost and Performance ................005 6-25 WNP-1 and WNP-3 Availability .........0.00 000.0. cee eee eee 6-25 Continued Ability to Finance Preservation . . -.. 6-25 Availability and Cost of Construction Financing . . ... 6-26 Physical Preservation (25.5 2:4 <1 coe aarorseretsisinre s ateis are ste isio ects a 6-26 Maintenance of Site Certification Agreement ............... 6-26 Claims Against WNP-1 or WNP-3 Assets by WNP-4/5 BIER IOUIONS 75 ore) olases sistas re -hs eres aisyate svejavessin aile)o ae sfelats eiaiofs ata 6-26 NRC Construction Permit and Operating License ........... 6-26 More Stringent Seismic Design Criteria for WNP-3 .......... 6-27 Continued Availability of Nuclear Components . . --» 6-27 Technical Continuity ...................02. Bee) | Leer Litigation Regarding Shared Assets ....... 6-28 Operating Lifer rrics-- eee sere a ysb mace 6-28 GOTICHIBIONN «ts ercishers t suatels Gra ats late datars lera sats ve a'stale ar visrale ave steer 6-28 MATION OES <2 acter seo sha feted ctize e eats toze rotor nietare Beorste se ciate) i yo atalole suatate ace 6-28 Energy aneters crc sad ers robe aot eidee ere ate ere cree elcid 6-28 QOUEOF-Fragion imports | cycisla ara stsjere a ctelo avalats (ave siomw sears sie eaters aya 6-30 Contents Chapter Page Appendix 6-A: Existing and Assured Regional Generating FROSOUNCOS: fn 5 c1s < oe alas srste's ores oles axere erie seis 6-A-1 Appendix 6-B: Planning Assumptions — Generic Conventional Coal Project, Two 603-Megawatt Units ........... 6-B-1 Appendix 6-C: Planning Assumptions - Generic Conventional Coal Project, Two 250-Megawatt Units ........... 6-C-1 Appendix 6-D: Planning Assumptions — Generic AFBC Coal Project, Single 110-Megawatt Unit .............. 6-D-1 Appendix 6-E: Planning Assumptions — Representative Geothermal-Electric Area (Newberry Volcano, ON GON) ices aces sw ote ti sterw opel ima cistert availa (oheias omies 6-E-1 Appendix 6-F: Planning Assumptions — Representative Windpark (Columbia Hills East, Washington) ..... 6-F-1 Appendix 6-G: Planning Assumptions — Generic Combustion Turbine Project, Two 105-Megawatt (Nominal) MII aise sicee eos oes corse wie so cre sates tee eae Sos 6-G-1 Appendix 6-H: Planning Assumptions — Generic Combined Cycle Project, Two 286-Megawatt Units .......... 6-H-1 Appendix 6-I:_ Planning Assumptions — Washington Public Power Supply System Nuclear Project No.1...... 6-1-1 Appendix 6-J: Planning Assumptions — Washington Public Power Supply System Nuclear Project No.3...... 6-J-1 7 BETTERUSE OF THE HYDROPOWER SYSTEM ............-..- 7-1 WNOCHICHION cletene s rere tote o oes 2 ots 5 ted sroys sve slau 5 Stein se sleteis 2 chs 7-1 Critical Period Planning ...... 2.2.2... 6 sce c eee e cece ence eens 7-2 Nonfirm Strategies ........ 0... cece cece cence eee eee 7-3 System Reliability and Its Implications .................02..e eee 7-4 Summary of Results ........ 2.0.0.0 e ese e cece e eee eee 7-4 1S F-Tot Ce] (0) [0 7-4 AMVANYSISI ss icjac sini |t-cteus a rao siete ware aernie cheat sins arate setae ese 7-4 COMICIUISIONS) 5 sic!o)s 1015.4 res oor 's usa yee otsrs stares © 7-12 Institutional Issues 7-12 Strategies for the Increased Use of Nonfirm inthe Region ........ 7-12 Backup Generation: Combustion Turbines and Extra-Regional PRICES OS fae: cccele cig ioc = ants lorove & ahe'n fats opstem etal wnexe, olevore|a since: + = 7-12 Background ... 7-12 Analysis....... 7-13 Assumptions .. 7-14 End Effects .... 7-14 Results ......... 7-15 Sensitivity Analysis . 7-16 Fuel Use Act ..... 7-17 Using Nonfirm Energy without Backup Generation: Load Management s.<ct.:é.<11/ ois icre s sin-s stoie + ots oiomnse tole s seiniers 7-17 OPCS fs reyes! sree scssie teres fare. 6 scale lane wloroie ofarsls eters tiene aly ietetale « 7-21 6||| RESOURCE PORTFOLIO: 35 2 rele esis craye sme ars «sts sams clots 5 lelorora ss 8-1 Section A: Resource Portfolio Analysis ...............0.00ece eee 8-1 Introduction . 0.00... ccc ccc cece eee een eee eee 8-1 Analytical TOONS (sisi: ect = artes ores aise cic 2 eine cists seis s siete ta onora os 8-1 Portfolio Development Process .............0.0ee eee eee eens 8-2 Load Treatment .......... 8-3 Resource Requirements . . 8-4 Resource Availability and Cost-Effectiveness Studies .......... 8-5 Option and Build Decision Rules ................0..0.0eeee ee 8-7 Description of the Resource Portfolio .............6.20eeeeeee 8-8 iv Chapter Page Section B: Portfolio Uncertainty .......... 006.0. cece cece eee ee 8-16 Impact of Less Conservation Supply ...............2.2-020 00 8-18 Impact of Slower Conservation Ramp Rates .................. 8-18 Impact of Less Conservation Combined with Sever Famer a ns Sok oe cetera estiet potarespeps mrsretatersr eaters 8-19 Impact of Higher Conservation Supply 8-20 Impact of Delay in Implementation of the MCS ...............- 8-21 Impactial Losing MOS ieee enc siete titre serie srede arstotoin ecerete 8-22 Impact of Not Being Able to Option Generating Resources ..... . 8-22 Impact of Increased Direct Service Industry Uncertainty ........ 8-25 Lack of Regional Cooperation ............ 00. c cece cece eee 8-25 Section C: WNP-1 and WNP-3 Cost Effectiveness ............... 8-27 NON os bcs Saree store Projets atats|orau clare s ayelsfsing aiaiese weysiaiss atete 8-27 AMMEN 5). 2.5. a1bie oleh Slots piets sere stay ¢ ays] stove taka ations stele 8-27 Probability of Need for WNP-1 and WNP-3 ...............0565 8-28 Resa) ye Flo cay ceerererarctte eretctareiste tere stoke satel stems oka eden str ereU 8-29 Option Value of WNP-1 and WNP-3 8-29 Impact of the Future Status of the Direct Service Industries ... 8-31 Impact of Plant Operating Life .............. 0... c cece eee 8-32 Sensitivity to Cost Assumptions .................0202-0 eee 8-32 Impact of Equivalent Availability ...................022..0 ee 8-33 Malus ot Forced Feetarts tctcrticjs\ re ators t aiolels ais stall sielole ech otore 8-33 SUNTITIANY | ooh le Siseete ie riots estore hoteliers sfalo ed atstels Ore alersye atctate eyegoee 8-34 Section D: Decision Model ............ 0. cece cece cence ee eeeee 8-34 WRROGHICUON eteye: sisters tre stelorn states starsinere fess diefetectad tortie Gefote arch olate 8-34 Background: [20 25.5 tre ser aes 8-34 Decision Model Overview 8-36 Major Features .............. 8-37 Load Uncertainty 8-37 Two-stage Resource Decisions .................eeeeeeeeee 8-37 Conservation Program Management .................00005 8-37 Major Decision Variables ........... 8-37 A Typical Model Simulation ..... 8-37 Load Selection .............. 8-37 Option and Build Requirements . . 8-38 Resource Choice .............. 8-38 Capital Costing................ 8-39 Production Costing ............ 8-40 Treatment of End Effects .......... 8-40 Section E: Lost Opportunity Resources .......... 8-41 Availability of Potential Lost Opportunity Resources . . 8-42 Loss of Generation Potential ................. .. 842 OUt-OF FagiOn Sales 2.2... cece teen c ewes cesses ccinses 8-42 Loss of Development Rights ................. cee e cece ee eee 8-43 Loss of Development Incentives ..... . 843 Generation in Lieu of Transmission .. . 8-43 Additional Resource Information ....... 8-43 Resource Evaluation and Acquisition . . . 8-43 Appendix 8-A: Regional and Public Utility Resource Schedul 8-A-1 Contents Chapter Page 9 CONSIDERATION OF ENVIRONMENTAL QUALITY AND FISH AND VILLI tt areal atals ateyes 4 te + ashe) shedelolstore ls (ore © eke a fete} /stotel lanece enakahel «late 9-1 Environmental Quality ......... 0.0.0 cece cece eee eee eee 9-1 Due Consideration Process 9-1 Analysis and Resource Alternatives ...........-0..:e0eeeeee 9-1 GOMSETVEMIONT: eis fadere | ots teased dol obec ndste a cat) sidealo|wbole dean xan cre tana 9-2 Better Uses of the Hydropower System ............20.0e00eee 9-3 Hydropower Development ........... 0.0: e cece eee eee eee eee 9-4 Industrial Cogeneration... 2.0.0... .0 520 c eee e cece eee eee eee 9-4 Coal-Fired Power Plants’ 2) .)2(.J3)</c1s's/sto'ss-n2 + pse)s\ata)clanre bat a alates 9-5 Nuclear Power Plants ............c0secccescesevcccsccesres 9-6 OMT FROSOUTCOS [aah at: odie! al atele elcheleleiare exert: seueteletele pre cris) «el elelele 9-6 Geothermal Energy .........0.0 0c c cece cece eee eee eens 9-6 Wind Power 9-6 Solar Power ... 9-6 Additional Fish and Wildlife Concerns .............-0.000eeee eee 9-6 Due Consideration Process ........... 00.0 c cece eee cence eee 9-6 Analysis of the Fish and Wildlife Impacts of Hydropower Development ......... 0.6 ccc cece cece eee een eee ee eens 9-6 10:| (PUBLIC INVOLVEMENT | jo frcen sterol sietclclares exse 2 re olatels eters siexe 2 s|atehele 10-1 GEOSSARIY) scpesnledeleststcere dest claleteleltsteg data otehale otascHe eden stea eda! ajapelsieiats a aen9 Sickais GL-1 Appendix Page I-A © METHOD FOR DETERMINING QUANTIFIABLE ENVIRONMENTAL GOS TS, AND BENEGITS! |slajaie tess 4 ogo sacers sletelelare <b s elsbeloratots feseeenn ace II-A-1 II-B CONDITIONS FOR BONNEVILLE FINANCIAL ASSISTANCE TO HYDROPOWER DEVELOPMENT IN THE REGION.............-- lI-B-1 vi List of Illustrations Figure Title Page 2-1 Total Employment, Pacific Northwest Region and U.S. Assumptions .. 2-1 2-2 Percentage Change by Age Group, 1985-2005 .................... 2-3 2-3 Comparison of Pacific Northwest Lumber and Plywood Production with U.S. Housing Starts, 1960-1983 .............-. 0. sees eee eee 2-6 2-4 Forecasts of World Oil Prices, Comparison of 1986 and 1983 Plan WROUAMIIIOG 5 oe oe ee se sess: 2-22 31 Northwest Power Planning Council Demand Forecast System . peel 3-2 Sales of Electricity, Historical and Forecast ............... 3 3-3 Historical and Forecast Growth ................005 3-3 3-4 Electricity Use per Capita 3-5 3-5 1983 Firm Sales Shares 3-5 3-6 1983 Residential Use by Application = 736 3-7 Average Size of Electrically Heated Housing Units ......... .. $8 3-8 Thermal Efficiency of Electrically Heated Single Family Houses ...... 3-9 3-9 1983 Commercial Sector Use by Application 3-10 3-10 WAR sStry DGRVINS: og iirc 2 atch 2 oe, 6 cise 0 stuels yeas ele «002 31 -» 310 3-11 Assumed Aluminum Operating Rates ............... -. S15 3-12 Wrigetion Demian: ooo. cece eh sec ens 3-16 3-13 Average Retail Electric Rates 3-17 3-14 Relative Residential Energy Prices (Ratio of Electricity to NaturalGas). 3-18 3-15 Demand Uncortaitty <<< 6s— 536 etcetera 3-19 3-16 Comparison of High Forecasts ................-.-- .. 3-21 41 Actual Nominal Dollar Expenditures ............... -. 42 42 GCepiel COGtG ioe -. 42 43 Operating Costs |. talsece sons estes cee cians mes mes we. 43 4-4 Cost Effectiveness for Evaluating the MCS ..... eeaett 45 Estimates of the MCS Program Marginal Value 4-12 4-6 Cost-Effectiveness Method for Evaluating Discretionary Resources ... 4-13 47 Cost-Effectiveness Method for Evaluating Near-Term Acquisitions .... 4-14 48 Value of Lost Opportunities in the Resource Portfolio . .. 415 49 Comparison of Cost-Effectiveness Criteria .................-.00005 415 5-1 Technical Conservation Potential from Space Heating Measures in Existing RESIMOMCES | .).!.2 da)... tsc/alsiois oy since vale ave 9 aiesa sfelsse Ohs)S sfgia 31316 2 5-4 5-2 Comparison of Regional Thermal Integrity Curve, Estimated Cost and Savings Compared to Observed Bill Changes in Existing Utility Weatherization Progranrs) ti)! oi.) losc!a ast cia loiecssscale sso a'ele) 6 sieve) 6iasoia ayaye a 5-11 5-3 Technical Conservation from Space Heating Measures in New RNS tact ea g tezt aaha eae weno of opera lots enol onze a tote: tetra cree tae 5-11 5-4 Residential Heat Loss 5-14 5-5 Technical Conservation Potential from Residential Water Heating Measures 5-27 5-6 Technical Conservation Potential from the Commercial Sector ........ 5-36 5-7 Annual Energy Use of New Commercial Buildings in the Northwest as a Percentage of the Model Conservation Standards ................ 5-39 5-8 Technical Conservation Potential from the Industrial Sector .......... 5-41 5-9 Technical Conservation Potential from Irrigated Agriculture ........... 5-42 6-1 Pacific Northwest Geothermal Resource Areas .............. 6.00005 6-10 6-2 Pacific Northwest Wind Resource AreaS .............0.0 cece ee eee 6-15 7-1 Average Daily Columbia River Natural Flow at The Dalles, Oregon .... 7-1 7-2 Probability of Nonfirm Energy Availability ..................0.00000e 7-2 7-3 Firm Load Service, Current Rules 746 7-4 Firm Load Service, Shift for Deficit 7-6 7-5 Firm Load Service, Shift for Deficit, Full Adjustment to Annual ANGST Yee ee eee er eee tat dated Uae lated cele (oe dlalele Hla tad 7-7 vii List of Illustrations Figure Title Page 7-4 Firm Load Service, Current Rules, (1) Shift for Deficit + 500 MW 1st Year FELCC, (2) 500 MW Firm Deficit ............-.02..-00000e 7-7 7-7 Firm Load Service, Current Rules (Full Scale) ..........-........5- 7-8 7-8 lop Quartile Service) jays /aysersratays eters stererstorerctereyotsyerel stereyer stored e/atelols 3 etefere > 7-8 7-9 Ly Un JOSH AUOABHA BU APRA RUA SHA BHS O HU oer Ria a AHa Bes 7-9 7-10 System Refill, Shift for Deficit, Full Adjustment to Annual and 1st Year FELCC 7-9 7-11 Priest Rapids Flow . . 7-10 7-12 Lower, Granite FIOW FM sinetissrie nis srioseieceitacttseeteelas cite setdelt 7-10 7-13 Priest Rapids Flow, Shift for Deficit, Full Adjustment to Annual ‘and 1st Year, FELCC tise lisrtacelc cisteteterctorsistatelelstelelersinvelare stalereictatevete 7-11 7-14 Lower Granite Flow, Shift for Deficit, Full Adjustment to Annual land i stivear REE CC iris stra rare snictolate fare piateicteverstelerevorsstetals eyeta asta 7-11 7-15 Net Benefits of Combustion Turbines vs. Coal 7-15 7-16 Impacts to Southwest Sales and Top Quartile Service ............... 7-15 7-17 Average Capacity Factor of Coal and Combustion Turbine Plants ..... 7-16 7-18 Sensitivity to Fuel Escalation Rate 7-16 7-19 Sensitivity to Debt/Equity Ratio ......... 0... cece ccc eee cece eee es 7-18 8-1 Northwest Power Planning Council Resource Portfolio Development, Process) sis sera stcicis/olora stets ctoters/yareictoilotalstercietercts ciate a siete 8-2 8-2 Load Growth Probability Distribution ............6. 00 ee ee eeee eee ee 8-3 8-3 Distribution for, Uncertain DSI Load een false evereeteis/efelsisiosvetctele ovata le ete 8-3 8-4 Regional Resource Requirements .............seceeceeceeeeeeeee 8-4 8-5 Public Utility Resource Requirements .................0eeeeee ee eee 8-4 8-6 Option'and Build Level ee ete an ete t techie tasters ery eel eyaietala dete 8-7 8-7 Cost of Option/Build Level Combinations ...............---...-.--- 8-8 8-8 Regional Resource Schedules: High, Medium-High, Mediuin-Low: low) iret ite setdacttnebtanelcarel telnet aefeteisnetotsiet teres sats 8-9 8-9 Public Utility Resource Schedules: High, Medium-High, MeCN LOW, LOW) icselotcrns otaislolats state sstcrsl «| to elersveltoneletarers olarstctaioredojsis tele 8-10 8-10 Nondiscretionary Conservation Program Energy as a Funclon of load Pathe recs t rice citer err sree teeters 8-11 8-11 Discretionary Conservation Program Energy as a Function ofioad Patni itterte seiroccmecectreeminetrrettcieiicistrceetescicts 8-11 8-12 Hydropower Efficiency Energy as a Function of Load Path ........... 8-11 8-13 Combustion Turbine (Nonfirm) Energy as a Function of Load Path .... 8-11 8-14 Small Hydropower Energy as a Function of Load Path............... 8-12 8-15 Cogeneration Energy as a Function of Load Path ................055 8-12 8-16 ‘Coal| Energy as a\ Function of Load Path </i0 102 2) c)-)erslare leroeicls ce 8-12 8-17 Conservetion' Program| Start-Ups eis coscicsii sce sissies teieciasierls 8-13 8-18 Initial Decisions, Hydropower Efficiency Improvements .............. 8-13 8-19 Initial Decisions, Combustion Turbines (Nonfirm) ...............2555 8-14 8-20 Initial Decisions, Small Hydropower .............ccceceeeeeeeeeees 8-14 8-21 Initial Decisions, Cogeneration ........... 0. cece cece eee eee eee 8-15 8-22 ual Decisions Licensed Coalliirrcpiccite otiactitsyas/tereictsta ctetericte ciate 8-15 8-23 Initial Decisions, Unlicensed Coal ..............c cece e cece eee e eee 8-16 8-24 System Cost'Distrioudion tins siies tian cet neiettet te ceili 8-16 8-25 First Coal Options) Base) Portiolio he -tastiecticbiisrflcrcinces seettaci ae 8-17 8-26 Cost Impact of One-Third Less Conservation ..................0005 8-17 8-27 First Coal Options, One-Third Less Conservation ................005 8-18 8-28 Cost Impact of Slower Conservation Ramp Rates .................. 8-18 8-29 First Coal Options, Slower Conservation Ramp Rates ............... 8-19 8-30 Cost Impact of Less Conservation and Slower Ramp Rates .......... 8-19 8-31 First Coal Options, Less Conservation and Slower Ramp Rates ...... 8-20 8-32 Cost Impact of One-Third More Conservation ...............0000005 8-20 viii Figure Title Page 8-33 First Coal Options, One-Third More Conservation ...............+.. 8-21 8-34 ‘Coat inipactiof MCS Deby ass ieis cesses ie 5 ar 5 wierd winre swiss ere e swe cteis'e are 8-21 8-35 First Coal Options, MCS Delay ............. 00.0. c cece eee ee eee 8-22 8-36 Cost Impact of Losing the MCS ............. 0.2 cece eee eee e eee 8-22 8-37 FWSUGee ORMOIS RIN x x rei iers 5 is: sis s eras sie 6 iss wre esis 5 ares He 8-23 8-38 Coat tripact Gf tnaiility to Option « .. . 22... ce ee cee cence cence 8-23 8-39 Build Decision Impact of Inability to Option ....................005 8-24 8-40 Cost Impact of 100 Percent DSI Uncertainty .....................0. 8-24 8-41 First Coal Options, DSIs 100 Percent Uncertain ...................- 8-25 8-42 Expected Value of Regional Cooperation .................0.000e0ee 8-26 8-43 Distribution of Benefits Due to Regional Cooperation................ 8-26 8-44 First Coal Options, No Regional Cooperation ...................005 8-27 8-45 Arrival Distribution of First WNP Unit .............0 2000. e eee eee eee 8-28 8-46 Arrival Distribution of Second WNP Unit 8-28 8-47 WNP-1 and WNP-3 Option Value ........... 8-29 8-48 Value of WNP-1 and WNP-3 ..... 0.6622 8-29 8-49 VERIO OF FUSE UIE 0 0isle 05.0) 6 ccoe tie ease sieiesleia wis oatie 4 ies sale sare cre 8-30 8-50 VMI OF SECO IIE aie «eis 6 aie 5 201s wists viele s ams isis Sie6 0 vie arer v's 8-30 8-51 Impact of DSIs on WNP-1 and WNP-3 ...... 2.0.00 c eee eee eee eee 8-31 8-52 Impact of Plant Lives ....... 2.2.6.6. e cece cece cece ences 8-31 8-53 Impact of Changing Cost Assumptions .................02ce eee eee 8-32 8-54 Impact of Equivalent Availability .............0 00... c cece cece eees 8-32 8-55 Impact of Forced Restart ...........00 00. cc ccceceeeeeeseeeeeeeees 8-33 8-56 Decision Model Overview acs srs sere 5 sin > sans sjazes ozas stele scorers tos ass 3 0% 8-35 8-57 Decision Model Load Selection ............... 00sec eee cence eee 8-37 8-58 Example Decision Model Load Paths ............. 0c cece eee eee 8-38 8-59 Decision Model Option and Build Level ................002e eee euee 8-38 8-60 Decision Model Option and Build Requirements ...................- 8-39 8-61 Decision Model Build Decisions ................0.00eceeee eee eens 8-39 8-62 Decision Model Process of Calculating Capital Costs ............... 8-40 8-63 Decision Model End Effects Treatments ........... 00. 00ece eee eens 8-41 List of Illustrations ix List of Tables Table 2-1 2-2 2-3 2-4 2-5 2-6 2-7 3-10 3-11 3-12 3-13 3-14 41 4-2 4-3 Title Page Comparison of U.S. and Pacific Northwest Employment Trends ....... 2-3 Summary and Comparison of Forecasts, Pacific Northwest and U.S., Comparison of 1980 and 2008 «ic. 6:50 «mcs siya wee c sie sins snare 5 ares 2-4 Average Annual U.S. Housing Starts.............6. 00 eee eee ee eee 2-7 Forecasts of Production and Employment, Lumber and Wood Products, Pacific Northwest, 1980-2005 .......... 0... cece eee cence eee 2-8 Forecasts of Production and Employment, Pulp and Paper Products (SIC 26), Pacific Northwest, 1985-2005. .......... 0... ee eeee ee eee 2-9 Forecasts of Chemicals Industry Production, Pacific Northwest, 1985-2005 oo recs 5 ies alee osie ste 6 oso 6 os oie sees iw 6 slew sole eT «siete 2-11 Forecasts of Employment, Agriculture and Food Processing, Pacific Northwest, 1985-2005 .. 0.2.2... ccc cece cece eee eee eee ees 2-12 High Technology Industries. . -. 2-13 Employment in High Technology Industries, 1982................... 2-14 Factors That Influence Regional Location of High Technology Companies 2-15 Forecasts of Employment, High Technology Industries, Pacific Northwest; 1985-2005) cic: ova cress oie: « nary wiciars vars scape s:ncnle nroiere! aecore-6 2-15 Total Employment Shares, U.S. and the Pacific Northwest............ 2-16 Nonmanufacturing Employment Projections, Pacific Northwest ....... 2-17 Nonmanufacturing Shares of Total Employment in 2005 «2417 Real Output per Employee, U.S.......... 00.00 e eee 2-18 Total Population and Households... .......... 0000s eee ee eee eee eee 2-18 Forecast of Population and Households, Pacific Northwest, 1980-2005. 2-20 Housing Stock Projections, Pacific Northwest, 1980-2005............ 2-20 Ratio of State per Capita Income to National per Capita Income, 1980. 2-21 Growth Rates of Real Income per Capita .............0..0..2 ee eee 2-21 World Oil Prices... 0... e eee ence eee 2-23 Residential Sector Fuel Prices 2-23 Commercial Sector Fuel Prices 2-23 Industrial Sector Fuel Prices .. .. 2-24 Employment-Population Ratios . 2-A-1 Persons per Household. ...... 2. 2A Housing Additions by Type ..... 2.0.6... ccc cece eee eee eee ee eee 2-A-1 Production per Employee by Industry, Average Annual Rates of Change (%), 1985-2005. ... 0.0... cece cece cece cece ee eeeeeee ees 2-A-1 SIC Code Listings........ .. 2-B-1 Firm Sales of Electricity . 3-3 Per Capita Use of Electricity 3-4 Firm Sales Forecast for Public and Investor-Owned Utilities. .. 34 Residential Sector Electricity Demand ...............0 00 cece ence ee 3-6 Residential Sector Summary Indicators ........... 060.002 0eeee eee 3-7 Share of Housing Stock by Building Type, 1980-2005 (%) .. 38 Commercial Sector Electricity Demand. ...............0.02.0020 eee 3-8 Commercial Sector Summary Indicators ..............6.0-.00ee0ee 3-11 Industrial Sector = 12 Industrial Forecasting Methods................. 0c cece cece eee eee 3-13 Composition of Industry Growth, 1983-2005: Medium-High Forecast .. 3-15 Innigetlon SOCtON sais sis vcs cates ctc sce cams tnt soit se sis ems aris 3-16 Electric Price Forecasts (1985 Cents per Kilowatt-Hour).............. 3-18 Demand Growth by Forecast Concept, 1983-2005.................. 3-22 Cost Analysis Summary .......0.. 06.00 e cece cece eee eee . 43 Sample Calculation of Levelized Cost of Conservation Measure 4-4 Fuel Price Escalation Assumptions, Average Annual Real Rate of Growth (%)... 0. 0c cece eee cnet e eee e ee eee eee eee ene eeeeees 4-5 List of Tables Table Title Page 4-4 Discount Rates Used for Present Value by Source ....... 4-6 45 Comparative Values of Financial Variables.............. 4-7 46 Estimates of Average Implicit Discount Rates by Source .. .. - 48 47 Input Data for Consumers Implicit Discount Rates .................. 48 4-8 Summary — Financial Assumptions, 1983 Plan and 1986 Plan....... 49 5-1 Conservation Program Assumptions in the Decision Model. . . . 5-3 5-2 Costs of Weatherizing Single Family Houses 5-5 5-3 Costs and Savings of Single Family Weatherization Measures in Zone 3 — Missoula. ............ 20.0256 cece cece cence eee eeee 5-6 5-4 Costs and Savings of Single Family Weatherization Measures in (ZOO 2 — SPOKBNG x 5cc- aiie cas 2 ae sis si Foie sae ees om ee RS 5-6 5-5 Costs and Savings of Single Family Weatherization Measures in ZORG 1 SOM Fa 5a lore 2st ic Satein > nnclnbosaso elese sical « esetate ee Mesaie Ae =e 5-7 5-6 Costs and Savings of Multifamily Weatherization Measures i 5-7 Weights Used to Reflect Regional Weather for Existing Space Heating. 5-8 5-8 Regionally Weighted Costs and Savings of Single Family Weatherization Measures «2... ccc cece ccc cece sce vinecscsones 5-8 5-9 Regionally Weighted Costs and Savings of Multifamily Weatherization Measures ....... 2.0.0... 0 0 cece e cece eee eee e eee es 5-9 5-10 Regionally Weighted Single Family Weatherization Savings by COST COMOGIONY 2. Fae cle ccorniticres rinse» pefelnios ities Hine cies cress stators 5-9 5-11 Regionally Weighted Multifamily Weatherization Savings by COGLCAMBQONY fiers stats 5 afets + arctel ors seelsrsi< uetetareis alors Fes sisi Sie ee 5-10 5-12 Technical Conservation from Existing Space Heating................ 5-10 5-13 New Residential Construction Base Case Efficiency Levels and Annual Space Heating Use Assumptions ..................2..2-55- 5-13 5-14 Typical New Dwelling Characteristics. ..................0020eee eee 5-13 5-15 Costs and Savings from Conservation Measures in New Single Family Houses, Zone 1 — Seattle ...... 2.2.20... 02. e eee eee eee ee 5-15 5-16 Costs and Savings from Conservation Measures in New Single Family Homes, Zone 2— Spokane.............0..0e cece cece eens 5-16 5-17 Costs and Savings from Conservation Measures in New Single Family Houses, Zone 3— Missoula ................0..02 0 eee eee ee 5-17 5-18 Costs and Savings from Conservation Measures in New Multifamily ROGIAONCOG iets stats sistets sll tale seis e asttl-icisettre seine netele «cite cei teste = 5-18 5-19 Costs and Savings from Conservation Measures in New Manufactured Homes, Zone 1 — Seattle ..................02200005 5-19 5-20 Costs and Savings from Conservation Measures in New Manufactured Homes, Zone 2— Spokane.....................0055 5-20 5-21 Costs and Savings from Conservation Measures in New Manufactured Homes, Zone 3— Missoula.......................04 5-21 5-22 Weighting Factors Used to Aggregate Individual Building and Location Savings to Region. ........... 0... cece cece cece cece ec eececeeeees 5-22 5-23 Regionally Weighted Savings and Costs in New Single Family EARRINGS «re She ors stays o slate siecle austetaross Sinveys cic pleas ale alate.» oie nsenettove » 5-22 5-24 Regionally Weighted Savings and Costs in New Multifamily Dwellings . 5-22 5-25 Regionally Weighted Savings and Costs in New Manufactured Dwellings ........ 0... cece eee cece cece eee eee eeeeeeeees 5-22 5-26 Forecast Model vs. Engineering Estimate for Space Heating in New Dwellings, Regional Average Use ............... 0.0.0 eee ee eee eee 5-24 5-27 Forecasting Mode! Dwelling Size vs. Average New Dwellings ......... 5-24 5-28 Internal Gain Changes from More Efficient Appliances. .............. 5-25 5-29 Technical Savings per Unit and Megawatts for New Single Family Lo ee ee ee I Ee Coe ee A ee 5-26 xi List of Tables Table 5-30 5-31 5-32 5-33 5-34 5-35 5-36 5-37 ESUSTLEL LETTE £ Eg Leee 6-10 6-11 6-13 6-14 6-15 6-16 6-17 6A 6-A-1 6-A-2 6-A-3 6-A-4 6-A-5 6-A-6 6-B xii Title Technical Savings per Unit and Megawatts for New Multifamily Units. . . Technical Savings per Unit and Megawatts for New Manufactured EMONIVOS ose ores = rere ereie/s ote’ s1 0110's ntl on\e\ slate) slels otele olelei= Sele a1 -1a1-)~lor=l='elel Data on Standby Losses from Conventional Water Heater Tanks . Variable Demand Use for Hot Water ................5-5+ Savings from Water Heating Measures ..... 5 Measure Costs and Savings for Water Heaters .................0045 Sensitivity Analysis on the Cost Effectiveness of Heat Pump and Solar Water, HOatOrs x1. farts leierereerr scissile oss eletee eisle sella ilec eines Number of Eligible Units by 2005 and Achievable Conservation Percent for Water Heating Measures, High Demand Forecast . . Conservation Available from Water Heaters ......... Measure Cost and Savings for Prototype Refrigerators. . Measure Cost and Savings for Prototype Freezers Summary of Annual Energy Use for Existing Commercial Buildings coeentecd trite) FROCHONT crererctere opelorstetariciel etalon Yer e e cele lelt setts ae eiatelsloreta Summary of Annual Energy Use for New Commercial Buildings ocean in tne FROION Es crete: eres erctars tone elotelelotcte re oye] a) elete e overele weveleserereie Retrofit Savings from Existing Commercial Buildings: Puget Power's Technical Conservation from Commercial Buildings . . Technical Conservation from Waste Water Treatment Facilities . Industrial Sector Technical Conservation Potential ........ ; Technical Conservation Potential from the Irrigation Sector ........... Cost and Availability of Transmission and Distribution System Efficiency Improvements ............:eseeeeeeeeee Mee cuyess Generic Hydropower Efficiency Improvement Measures .. . Availability of Energy from Hydropower Efficiency Improvements . Cost and Availability of Hydropower Efficiency Improvements . Pacific Northwest Geothermal-Electric Resources . Planning Assumptions, New Hydropower ‘ Generic Solar Generating Projects, Cost and Performance Summary. . Representative Wind Turbine Cluster, Cost and Performance Summary Cost and Availability of Energy from Better Pacific Northwest Wind FASSOUNCS Ares re yerayereteyeye te fore total lore fared tater etetereietainre ere ET eletorel ee ePeto eieye Planning Assumptions, New Cogeneration se Generic Coal Projects, Cost and Performance Summary............. Generic Combustion Turbine and Combined-Cycle Projects, Cost and Performance Summary............0.cccsecesccceesceces Cost and Performance Characteristics of WNP-1 and WNP-3. Summary of Firm Energy Exports .............-0.00e0 ee Summary of Firm Energy Imports .. . aor Summary of Peaking Capacity Exports................55 Summary of Peaking Capacity Imports................-. Existing and Assured Regional Generating Resources .... Federal Hydropower Projects ..........-.0+:0esseeeeeee Investor-Owned Utility Hydropower Projects Publicly-Owned Utility Hydropower Projects Contracted Resources’... i.e: <iaicisicin cus sais) s ates cree orotate ersians s arsie s Large Thermal Units. . . Reserve Units Planning Assumptions, Generic Conventional Coal Project, Two 603-Megawatt Units .... 2.2.0.2... cece cece cece eee ences 5-26 5-26 5-28 5-28 5-28 5-29 5-30 5-31 5-31 5-32 5-33 5-37 5-37 6-12 6-14 6-16 6-17 6-18 6-20 6-22 6-24 6-29 6-29 6-30 6-30 6-A-1 6-A-2 6-A-3 6-A-6 6-A-7 6-A-10 6-A-11 List of Tables Table Title Page 6-C Planning Assumptions, Generic Conventional Coal Project, Two 250-Megawatt Units ....... 6... cece eee c eee eee ence eee 6-C-1 6-D Planning Assumptions, Generic AFBC Coal Project, Single 110-Megawatt Unit (January 1985 dollars) ..................0..008- 6-D-1 6-E Planning Assumptions, Representative Geothermal-Electric Area (Newberry Volcano, Oregon). ........... 2.6. e cece cece eee eee eee 6-E-1 6-F Planning Assumptions, Representative Windpark (Columbia Hills East, WH MAQIOR oo ois oa ons no cio cp tie sce nc ese vee cogs come cee 6-F-1 6G Planning Assumptions, Generic Combustion Turbine Project, Two 105-Megawatt (Nominal) Units (January 1985 dollars)........... 6-G-1 6-H Planning Assumptions, Generic Combined-Cycle Project, Two 286-Megawatt Units (January 1985 dollars) .................005 6-H-1 6&1 Planning Assumptions, Washington Public Power Supply System Nuclear Project No. 1 (January 1985 Dollars)..................0005 6-I-1 6-J Planning Assumptions, Washington Public Power Supply System Nuclear Project No. 3 (January 1985 Dollars). 6-J-1 7-1 Assumptions ... 2.2.2... 0.6 c cece eee eee ee 7-13 7-2 End Effect Corrections for Combustion Turbine Studies.............. 7-14 7-3 NQUEDQRG RIS 5555 tops tesas cra sate teesayn sori orate stores rete Fiano atatn vrata eco 7-14 7-4 End Effect Adjustments for Sensitivity Study ................0.00008 7-18 7-5 Curtailment: Firm, 6.2 Cents, and Top Quartile, 2.2 Cents. . 7-19 7-6 Curtailment: All Loads, 6.2 Cents...............0005 7-19 7-7 Variable Cost of Direct Service Industries . . . 7-20 7-8 Curtailment: Firm, 10.0 Cents, and Top Quartile, 5.7 Cents........... 7-20 8-1 Resource Availability 8-5 8-2 FROGOUNOR Frlortty Onder ores. ctoreo'or01- ores) oieiz/e oars Gis seer a since Ole ease 8-6 8-3 Priority Order Studies. ....... 00... ccc cece cece cece teen eee e eee eee 8-7 8-4 Inventory of Potential Lost Opportunity Resources .................- 8-42 8-A-1 Observed Loads and Resources, Regional High (1985-1995) ........ 8-A-1 8-A-2 Observed Loads and Resources, Regional Medium-High (1985-1995). 8-A-4 8-A-3 Observed Loads and Resources, Regional Medium-Low (1985-1995). 8-A-6 8-A-4 Observed Loads and Resources, Regional Low (1985-1995)......... 8-A-8 8-A-5 Observed Loads and Resources, Public High (1985-1995)........... 8-A-9 8-A-6 Observed Loads and Resources, Public Medium-High (1985-1995)... 8-A-11 8-A-7 Observed Loads and Resources, Public Medium-Low (1985-1995).... 8-A-13 8-A-8 Observed Loads and Resources, Public Low (1985-1995) ........... 8-A-14 10-1 Council and Advisory Committee Meetings, 1986 Power Plan ........ 10-3 10-2 Issue Paper List for Draft Power Plan... ..........0 20.0 eee ee eee eee 10-5 xiii The overall conclusions of this 1986 Power Plan are described in Volume I. It includes summaries of the basic planning strategies, the important regional power issues, the lowest cost mix and schedules for new resource acquisitions, and the priorities of the Action Plan the region needs to follow to ensure an adequate and reliable supply of power at the lowest cost. This volume (Il) contains supporting docu- mentation for the conclusions and positions in Volume I. It describes the analytical work and technical details leading to the policy decisions. Chapter 2, “Economic, Demographic and Fuel Price Assumptions,” describes the methods and results of an analysis on which the regional load forecasts were based. Pop- ulation and employment trends, and devel- opments in each of the economic sectors, will determine to a large extent the future need for electricity. Chapter 3, “Forecast of Demand for Elec- tricity,” explains how the load forecasts were derived, what they say and what role they play in the planning process. Chapter 4, “Financial Assumptions and Resource Cost Effectiveness,” examines the financial variables used to estimate quan- tities and costs of resources, to project future demand for electricity, and to simulate opera- tion of the regional power system with alter- native sets of resources. These values, used in the overall analysis, directly influence results and permit consistent comparison of components. Chapter 5, “Conservation Resources,” pres- ents the methods and results of studies that determined how much conservation could be secured in the region and at what cost. This chapter looks at conservation savings and techniques in the residential, commercial, industrial and irrigation sectors. Based on the work outlined in this chapter, the Council identified specific amounts of conservation savings available for ttre 20-year portfolio of resources that will meet the region's electrical energy needs. Chapter 6, “Generating Resources,” dis- cusses a variety of technologies that could potentially meet future electricity require- ments in the Pacific Northwest. This chapter describes the current status of development, estimated cost and availability of these possi- ble sources of electric energy. The most cost- effective and available resources were con- sidered in the development of the Council's resource portfolio. Resources considered promising but not yet fully reliable or cost effective are recommended in the Action Plan for further research, development or demonstration to better establish their role in future power plans. Tables in the chapter appendices describe resource costs and characteristics in detail. Chapter 7, “Better Use of the Hydropower System,” explains the methods and benefits of meeting more firm load in the region with hydropower that is presently nonfirm. Chapter 8, “Resource Portfolio,” describes in detail the Council's resource portfolio. Sec- tion A, “Resource Portfolio Analysis,” describes the analysis that led to the Coun- cils choice of the portfolio, and gives a brief overview of the decision rules employed. Section B, “Portfolio Uncertainty,” presents the results of sensitivity studies that examine the cost and scheduling impacts of assump- tions other than those the Council used in developing the resource portfolio. Section C gives details of the Council's analysis of the cost effectiveness of the two Washington Public Power Supply System's nuclear plants 1 and 3 under a variety of future circum- stances. Section D presents a more detailed description of the Decision Model. Finally, Section E discusses generating resource lost opportunities. An appendix to Chapter 8 presents the expected increments of resources required over the next 20 years by the region as a whole and by the public utility and direct service industry customers of Bon- neville only. Chapter | Introduction Chapter 9, “Consideration of Environmental Quality and Fish and Wildlife,” reviews the environmental effects of all resources con- sidered for use in this plan and summarizes their known likely impacts on fish and wildlife. In addition, this chapter examines the costs and effectiveness of ways to mitigate such impacts. (Appendices II-A and II-B deal fur- ther with methods for determining environ- mental costs and benefits, and with condi- tions hydropower projects must meet to gain financial assistance from the Bonneville Power Administration.) Chapter 10, “Public Involvement,” describes the public information and public involvement activities the Council has conducted, and will conduct, as part of its responsibility to ensure widespread participation in policy making and power planning by the region's rate- payers, customer groups, state and local governments, and users of the Columbia River system. Council publications and meetings are listed. 1-1 Chapter 2 Economic, Demographic and Fuel Price Assumptions Introduction Under the Pacific Northwest Electric Power Planning and Conservation Act, Congress charged the Council with forecasting electric power requirements as the basis for a plan for meeting regional electricity needs. The economic and demographic assump- tions are the dominant factors influencing the forecasts of demand for electricity. A good tule of thumb is that demand for electricity will parallel economic activity in the absence of other changes. This relationship is modified by shifts in relative energy prices, including the price of electricity and other fuels; by changes in the composition of economic activity; and by the gradual depreciation and replacement of the buildings and other cap- ital stock of the region. Future sales of elec- tricity will also be affected by conservation activities, although the Council treats conser- vation as a resource. Recognizing that the future is highly uncer- tain, the Council has adopted a planning strategy that incorporates flexibility and risk management. The economic and demo- graphic assumptions are both extremely important determinants of future electricity needs and, at the same time, highly uncer- tain. The objective of the range of planning assumptions discussed in this chapter is to help define the extent of uncertainty. The plan must address a range of future elec- tricity needs that reflects, among other fac- tors, this underlying economic uncertainty. In developing the range of forecasts for the plan, the Council adopts forecasts that bracket the highest and lowest plausible eco- nomic scenarios for the next 20 years. The purpose of this approach is to develop a flexi- ble resource strategy that provides an ade- quate supply of electricity at the lowest possi- ble cost. The risks are twofold: the risk of not having an adequate supply of electricity and the risk of being saddled with expensive investments in unnecessary resources. 1980=1.0 2.25 2.00 1.75 1.50 1.25 1.00 0.75 SITTTETL SIIMIDTTAISSEESL 4s wt f ple ae Low 1980 1985 1990 mmm §=Pacific Northwest vt77, United States 1995 2000 2005 Figure 2-1 Total Employment—Pacific Northwest Region and U.S. Assumptions The high and low forecasts are designed to describe scenarios that have a low proba- bility of occurrence. While actual levels for any single variable or industry may fall out- side the high and low ranges, it is assumed that the probability of the scenarios as a whole is very low. Equally important to the planning process are the medium-high and medium-low sce- narios. These scenarios are assumed to bound an area of most likely load growth. The higher probability of load growth falling within this range will have an impact on the analysis of resource decisions. The total employment forecasts presented in this report are similar in many respects to the forecasts for the 1983 Power Plan. The fore- casts encompass a range of employment growth between 1985 and 2005 comparable to the range in the 1983 plan between 1980 and 2000. The high forecast assures that the Council's plan will accommodate record regional economic growth should it occur. In the high forecast, total regional employment grows 130 percent faster than a high national forecast of employment. The high forecast represents a case where the region grows faster relative to the nation than in any histor- ical five-year period. The low forecast assumes that the Pacific Northwest grows at a rate 40 percent lower than a low growth national forecast. The low case implies a rela- tive performance well below that which has characterized the region in the long term. Figure 2-1 illustrates the forecast range rela- tive to a national forecast range from Whar- ton Econometric Forecasting Associates. 21 Chapter 2 In spite of the general similarity of the fore- cast range to that in the 1983 Power Plan, there are several important changes in the details of the economic and demographic forecasts and in the fuel price assumptions. In general, the forecasts show lower levels of employment and population growth in all scenarios. For example, in the high case, employment was forecast to increase at a rate of 3.7 per- cent per year from 1980 to 2000 in the 1983 Power Plan forecasts. In the forecasts pre- sented in this chapter, employment is pro- jected to increase at a rate of 3.2 percent per year from 1985 to 2005 for the high case. Asa result of the combination of a lower growth rate and a different base year, total employ- ment in the 1986 Power Plan high case is approximately 20 percent lower in the year 2000 than in the 1983 Power Plan high case. This pattern is similar across the range of forecasts. Other significant changes include: ™@ Lower fuel price assumptions. @ Lower aluminum smelter operating rates in most forecasts. @ Lower heavy manufacturing employment forecasts. @ Increased relative importance of the non- manufacturing sector. The forecasts for oil and natural gas prices are generally lower than those in the 1983 Power Plan, reflecting recent history and an improved understanding of the world oil mar- ket. The ability of oil producers to achieve ever higher prices for their oil is severely lim- ited by market responses, both on the demand side and on the supply side. Recent changes in the structure of the world aluminum market and rapid increases in regional electricity rates have raised ques- tions about the long-term viability of some of the aluminum plants in the region. To encom- pass this potential uncertainty, lower alumi- num smelter operating rates were assumed in all forecasts, except the high case. The high case assumes that plants are operating at 100 percent in the long run. 2-2 Forecasts of employment growth in a number of heavy manufacturing industries are lower in these forecasts than in the 1983 Power Plan forecasts. The industries include forest products, transportation equipment, food and kindred products, machinery, and pri- mary and fabricated metals. The impact of increased foreign competition, rising relative costs of production (such as electricity and transportation), and the length and severity of the recent recession have taken a toll on the region's manufacturing industries. In the lumber and wood products industry, higher productivity growth in the last few years has decreased the need for labor rela- tive to a given level of production. For ex- ample, employment was estimated to be approximately 20 percent lower in 1984 than in 1979, while production was estimated to be similar to 1979 levels. Other industries have achieved productivity gains, although they may not have been as dramatic. The nonmanufacturing industries accounted for 82 percent of total employment in the region in 1980. Nonmanufacturing industries are projected to increase employment faster than manufacturing industries in all sce- narios. These and other aspects of the fore- casts are discussed in more detail in this chapter. Forecast Overview Overview of the Regional Economy The Pacific Northwest is blessed with rich natural resources in its minerals, agricultural lands, fisheries and extensive forests. The abundance of natural resources has pro- vided the region's inhabitants with sources of jobs and income, as well as providing a desir- able environment for recreation and main- taining a high quality of life. The development of the vast Columbia/ Snake River system for navigation, electricity production, irrigation and recreation has con- tributed to economic growth in the region. Low electricity rates, relative to those found elsewhere in the nation, have attracted elec- tricity-intensive industries, such as the alumi- num industry, to the Pacific Northwest. More recently, industries such as electronics have grown in the region, attracted primarily by the quality of the labor force and quality of life. The development of port facilities and growing trade with Alaska and the Pacific Rim countries have provided a source of new jobs for the region. Growth in the non- manufacturing sectors, in general, has occurred at a rapid rate. These develop- ments have lent diversity to a region depen- dent on resource-based industries. During the 1960s and 1970s, total employ- ment grew faster in the region than in the nation. Table 2-1 shows a comparison of growth patterns between the region and the nation for the last two decades. Since 1979, the region has experienced slower growth than the nation. From 1979 to 1984, it is esti- mated that total employment increased 5.9 percent nationally, while total employment in the region decreased 0.2 percent. This can be explained, in part, by the composition of the region's industrial sector. In 1979, manufacturing employment accounted for 19 percent of total employ- ment. Within manufacturing, lumber and wood products was the largest employer, accounting for 27 percent of regional man- ufacturing employment. The lumber industry has suffered from depressed U.S. housing markets induced by high interest rates, com- petition from other producing areas and new product lines. It is estimated that, in 1984, employment in lumber and wood products in the Northwest was more than 20 percent lower than in 1979. The lumber and wood products category includes logging activities, some of which are related to pulp and paper production. In addi- tion, many companies have production facili- ties producing both wood and paper prod- ucts. Including pulp and paper products, the forest products industry accounted for 31 per- cent of manufacturing employment. The second largest regional manufacturing industry is transportation equipment, which is composed primarily of aerospace. It accounted for 17.5 percent of manufacturing employment in 1979. The aerospace industry began to turn around during 1984. Even so, from 1979 to 1984, it is estimated that employment in transportation equipment declined 15 percent. Primary metals is the largest industrial con- sumer of electricity in the region, accounting for nearly half of all industrial electricity con- sumption. Most of the electricity consump- tion is concentrated in the primary aluminum industry, which operates ten plants in the Northwest. This industry has suffered from dramatic swings in prices of aluminum, increasing electricity prices, and increasing competition from lower-cost producing areas. Pulp and paper is the second largest indus- trial consumer of electricity, followed by chemicals and lumber and wood products. In 1977, the top four industrial consumers of electricity accounted for almost 90 percent of the electricity used by industrial customers in the region. Major Trends There are a number of basic trends common to the range of forecasts. While the extent of change resulting from these trends varies somewhat in each forecast, it nevertheless forms a context for the future. Many of the trends relate to demographic patterns in the existing population. One of the primary demographic changes that will occur is the aging of the population. From 1985 to 2005, the national population aged 45-54 is projected to increase almost 60 percent, while the population aged 20-29 is projected to decline by 10 percent. The population over the age of 55 is projected to increase by 35 percent during this period. Figure 2-2 shows the percentage change in population by age group for the nation from 1985 to 2005. Although the age composition of the population in the region will vary among scenarios because of migration, the general patterns of demographic change will persist. Table 2-1 Comparison of U.S. and Pacific Northwest Employment Trends Average Annual Rate of Growth (%) Chapter 2 1960-1979 1979-1984 PNW US. PNW U.S. Total Employment 3.0 2.2 0.0 11 Manufacturing Employment 2.2 1.2 -2.0 -1.4 SIC@ 20—Food & Kindred Products 1.3 -0.2 -1.6 “1.41 SIC 24—Lumber & Wood Products 1.0 0.8 -4.5 -1.6 SIC 26—Pulp & Paper Products 0.3 0.9 0.1 -0.7 SIC 28—Chemicals & Allied Products> -0.1 1.6 3.6 -0.9 SIC 33—Primary Metals 2.9 0.3 5.9 -7.0 SIC 35—Non-Electric Machinery 6.3 2.8 0.1 -2.4 SIC 36,38—Electrical Equipment and Instruments 9.0 2.2 3.9 1.0 SIC 37—Transportation Equipment 2.3 141 -3.1 -1.1 Other Manufacturing 3.4 1.0 -1.6 -1.5 Nonmanufacturing Employment 3.2 25 0.4 1.9 @ Standard Industrial Classification (SIC) code. > Change in classification of a facility in the region to chemicals has artificially raised the rate of growth from 1979-1984. Excluding this facility in the 1984 data would yield a growth rate of 2.0 percent. Year -20 019 20-29 30-44 45-54 55-64 65 Age TOTAL Figure 2-2 Percentage Change by Age Group— 1985-2005 2-3 Chapter 2 This aging of the population is expected to affect consumption patterns, the labor force and labor productivity. Productivity growth should be enhanced by the dramatic slow- down in the growth of the labor force during the forecast period. Slower growth in the labor force will result in upward pressure on wages. Producers will seek to substitute cap- ital for labor, which tends to increase produc- tivity and stimulate technological change. Consumption patterns are expected to emphasize personal services, clothing, travel, and health services. A second major trend is the increase in the proportion of women in the labor force. From 1960 to 1980, the female labor force par- ticipation rate increased from 37 percent to 52 percent. This trend is expected to con- tinue to varying extents in all forecasts. Growth in the importance of nonmanufactur- ing industries is projected in each of the fore- casts. Traditionally, studies of regional economic growth have focused on the man- ufacturing industries. Recently, the non- manufacturing industries have attracted more attention because of their size and rapid growth. In 1980, nonmanufacturing industries accounted for 81 percent of total employment in the region. Nonmanufactur- ing employment increased at a rate nearly 70 percent higher than manufacturing employ- ment from 1960 to 1979. The outlook is strong for industries, such as communications and machinery, that will play a key role in growing technological changes and productivity-enhancing invest- ments. The foreign trade sector is expected to continue to increase in importance. The Pacific Northwest is well positioned to partici- pate in trade to the Pacific Rim countries, and that possibility is assumed to be an important component of the higher growth forecasts. The continued stagnation of the region's large resource-based industries charac- terizes all of the forecast range. Lumber, alu- minum, and basic chemicals are not expected to be important sources of eco- nomic growth for the region even in the high forecasts. 2-4 Table 2-2 Summary and Comparison of Forecasts Pacific Northwest and U.S. Comparison of 1980 and 2005 AVERAGE ANNUAL RATE OF GROWTH (%) 1960-1980 1985-2005 Medium- Medium- U.S. PNW U.S.* High High Low Low Total Employment 2.1 3.1 1.2 3.2 2.4 1.5 0.5 Manufacturing 1.0 2.0 -0.5 1.6 14 0.5 -0.4 Nonmanufacturing 2.4 3.4 1.5 3.4 a7 47, 0.7 Population 141 1.9 0.8 2.0 1.5 0.9 0.2 Households 21 28 1.4 28 2.0 1.3 0.3 haa Medium- a Medium- High High Low Low Persons per Household 2.7 2.2 2.4 2.4 2.6 Employment/Population Ratio 0.41 0.50 0.47 0.44 0.42 Percent of Total Employment 100.0 100.0 100.0 100.0 100.0 Manufacturing 18.2 12.5 13.0 13.7 14.0 Nonmanufacturing 81.8 87.5 87.0 86.3 86.0 Percent of Manufacturing 100.0 100.0 100.0 100.0 100.0 Lumber & Wood Products 23.7 13.8 14.2 14.1 16.9 Transportation Equipment 18.5 17.5 17.0 17.3 15.6 Food & Kindred Products 12.6 11.2 10.8 11.1 11.8 Electronics (SIC 35,36,38) 14.6 26.4 26.0 24.0 20.7 Other 30.6 31.1 32.0 33.5 35.0 *The U.S. forecast is Wharton's medium case projection. Description of the Scenarios The economic assumptions presented in this chapter rely on basic policy assumptions, many of which operate at the national level. Each of the four regional economic forecasts was made within the context of a correspond- ing view of the national economy. However, the linkages between the national forecast and the regional forecast are indirect. Certain results of the national forecasts are included directly in the regional forecasts. These include inflation rates, interest rates, industry specific productivity growth, and basic demographic patterns. Other assump- tions create a greater variation in the regional forecasts than in the national forecasts, how- ever. These include wider fuel price ranges, regional shares of national employment growth by industry, and specific assumptions about the viability of the regional aluminum industry. Forecasts developed by Wharton Econometric Forecasting Associates (Whar- ton)‘ were the primary source for forecasts of national economic variables used in develop- ing regional projections. Chapter 2 In developing the range, the primary objec- tive was internal consistency for each fore- cast. That is, incompatible assumptions were not combined in any one forecast just to achieve a wide forecast range. In some cases, there are three forecasts for each industry projection or other assumption. These were combined into four scenarios. For example, there are three forecasts for production and employment in the lumber and wood products industry. These were combined with other industries into four sce- narios. In the case of lumber and wood prod- ucts, the high case forecast was included in the high economic growth scenario, the medium case forecast was included in the medium-high economic growth scenario and the low case forecast was included in the low and medium-low scenarios. This combina- tion of assumptions is intended to reflect the downside risk assumed for the lumber and wood products industry. In developing the scenarios, it is important to recognize the wide range of possible out- comes for the regional economy. A short- term view of the future was rejected in favor of developing scenarios that would encompass wide range of uncertainty about the region's economy in the long run. The high case pre- sents quite a different view of the regional economy in the year 2005 than the low case. For example, there are 40 percent more peo- ple living in the region in the high case than in the low case by the year 2005. In addition to an underlying high growth sce- nario on the national level, the regional out- look for the high growth case implies that the region's economy fares better, relative to the nation, than it has in the past. The large resource-based industries, such as forest products, aluminum, agriculture and basic chemicals, maintain a vital presence in the region's economy, but are not expected to contribute to new jobs. In the high case, employment in lumber and wood products is projected to decline 20 percent from 1984 to 2005. Other resource-based industries show no increase in jobs. On the other hand, indus- tries such as electronics, trade and services expand rapidly, more than doubling their employment in 20 years. As shown in Table 2-2, total employment is projected to increase at a rate of 3.2 percent per year, whichis slightly higher than the rate of growth sustained by the region from 1960-1980. Population is projected to grow at 2.0 percent per year, while households grow at 2.8 per- cent per year. It is assumed in these projec- tions that the region will continue to be a favorable location for growth, because of the richness and diversity of its natural resources, the quality of the environment and labor force, the quality of the educational sys- tem, relatively lower electricity prices, and proximity to expanding markets in Japan and other Pacific Rim nations. In the medium-high scenario, rapid growth in high technology and commercial industries is coupled with moderate levels of activity in forest products, agriculture, and basic chem- icals. Employment in lumber and wood prod- ucts is projected to decline 25 percent from 1984 to 2005. This is accompanied by slight declines in other resource-based industries. The operating level of the region's aluminum plants is assumed to average 85 percent. Employment in electronics and nonmanufac- turing increases by nearly 80 percent. These changes result in employment growth of 2.4 percent per year, and population and house- hold growth of 1.5 and 2.0 percent per year, respectively. Although the overall level of employment growth in the medium-high sce- nario is slower than the region experienced in the 1960s and 1970s, it still represents a case where employment growth is 100 percent faster than national growth in the medium case. In the medium-low growth forecast, tradi- tional industries experience low levels of eco- nomic activity while other manufacturing and commercial industries experience moderate growth levels. Employment in lumber and wood products is projected to decrease by more than a third of its 1984 level. The operat- ing level of the region's aluminum plants is assumed to average 70 percent. The region continues to increase its share of employ- ment in electronics and nonmanufacturing industries, however. Total employment is pro- jected to increase at a rate of 1.5 percent per year, with population and households increasing at rates of 0.9 and 1.3 percent per year, as shown in Table 2-2. In the medium- low scenario, employment growth is 25 per- cent faster than national growth in the medium case, which is lower than the relative rate of growth experienced by the region from 1960 to 1980. The regional outlook for the low case shows total employment increasing at a rate of 0.5 percent per year, indicating a rate of growth 40 percent lower than the national rate of employment growth in the low case. The dis- proportionate impact of the recent recession on major regional industries leads to more severe long-term problems than in the other scenarios. Growth in nonmanufacturing is offset by declines in many of the larger tradi- tional industries. In the low case, the operat- ing level of the region's aluminum plants is assumed to average only 50 percent. In addi- tion, employment in aerospace is projected to decline by more than a third. Total popu- lation and households are projected to increase at rates of 0.2 and 0.3 percent per year, respectively. This slow level of growth implies net out-migration of population throughout the forecast period. Employment and Production The Forest Products Industries The long-term outlook for the region's forest products industry is clouded by the roller coaster housing markets of the last few years. New housing accounts for 40 percent of the market for lumber and wood products. Figure 2-3 is a graph showing U.S. housing starts, Pacific Northwest lumber production and plywood production for 1960 to 1983. The graph shows that regional lumber and plywood production follows a cyclical pattern similar to U.S. housing starts. In 1979, the regional wood products industry accounted for 38 percent of U.S. lumber pro- duction and 55 percent of U.S. softwood plywood production. The bulk of production in the region— almost half of lumber produc- tion and over 70 percent of the softwood plywood production—occurred in Oregon. Furthermore, a large proportion of production in both Oregon and Washington is west of the Cascades. In recent years, the regional lumber industry has been threatened by a poor housing industry, a loss of market share to other com- peting regions and Canada, and competition to plywood from lower-cost substitutes such as waferboard and oriented strandboard. 2-5 Chapter 2 Production Starts 16,000 13,500 11,000 8,500 6,000 2.5 2.0 1.5 1.0 0.5 1960 62 64 66 68 70 72 74 76 78 80 82 meee (umber (Million Board Feet) “4/74 Plywood (Million Square Feet ) == Starts (Million Units) Figure 2-3 Comparison of Pacific Northwest Lumber and Plywood Production with U.S. Housing Starts— 1960-1983 The housing industry in the U.S. has under- gone a number of fundamental changes through the last recession. Real mortgage rates are projected to remain higher in the future than during the 1970s. The housing boom that led to speculation and high-return investment opportunities for households has fizzled as housing values failed to keep up with the general rate of inflation. The deregulation of the financial industry has opened additional avenues for investment by households. These factors raise the cost and lower the demand for housing. One important characteristic of the housing market with consequences for lumber and plywood demand is the percentage of total housing units that are single family units. A single family unit uses approximately three times as much lumber and wood products as a multifamily unit. From 1970 to 1974, the average share of single family units to total units was 58 percent. This share increased to 73 percent for the years 1975 to 1979. Whar- 26 ton projects that the share of single family units will average between 64 and 69 percent over the next 20 years. The share of single family units is affected by the cost of housing and demographic factors. Another important factor is the average size of new housing units. The average size of new single family units increased from 1,355 square feet in 1962 to 1,760 square feet in 1979. During this period, the average number of persons per household declined from 3.3 to 2.7. Since 1979, the average size of new single family housing units has decreased by approximately 10 percent. The Real Estate Research Corporation projects that average unit size will decrease to 1,200 square feet by the end of the 1980s. This is in contrast to recent U.S. Forest Service (USFS) forecasts, which assume that the average size of new single family housing units will increase grad- ually, reaching 1,850 square feet by 1990 and 1,950 square feet by 2000. This would have an important impact on the demand for lumber. Another important area of concernis the fore- cast of housing starts. The USFS and Whar- ton forecasts of U.S. housing starts are shown in Table 2-3. The numbers shown are housing starts, excluding manufactured homes. As shown in Table 2-3, the forecasts differ considerably in the 1985-1990 period, but are similar during the 1990s. The region's lumber industry has experi- enced increasing competition from lumber- producing areas in the Southeastern United States. Higher transportation, labor and stumpage costs have made it difficult for the Northwest to retain its historical market shares. For example, wage rates are as much as 40 percent lower in the Southeast. In the Southeast region, timber resources are owned primarily by the lumber industry and other private parties. The timber harvest can respond to fluctuations in demand, relieving pressure on stumpage prices. In the North- west, the federal government owns more than half of the commercial timberlands. Tim- ber resources under the management of the U.S. Forest Service are governed by laws limiting the level of cuttings to an even flow, nondeclining yield. Stumpage prices have been bid competitively, raising costs dramat- ically for some mills that rely extensively on timber from National Forest lands. In addi- tion, the tree growth cycle is faster in the Southeast, approximately 35 years com- pared to 50 years in the Northwest. One area of uncertainty is in the estimates of future timber resources. Recent studies show that more privately-held timberlands in the Southeast are being lost to other uses, such as agriculture or urban development, than previously thought. In addition, the intensity of management science applied by nonindustry private timber owners is subject to uncertainty. Other factors that add to the uncertainty of future timber resources include natural disasters, improvement of timber management techniques, and changes in wilderness or recreational desig- nations, to name a few. Canadian producers have increased their share of the U.S. market to one-third, from 28 percent in 1979. Further competitive inroads into U.S. markets may continue. This is a subject of controversy in U.S./Canada trade relations currently. Some U.S. producers claim that the Canadian government is using unfair trade practices by selling public timber at subsidized prices. Competition to the region's plywood industry is provided by the introduction of low-cost substitute products. The substitutes include products such as waferboard and oriented strandboard. These products are fabricated from faster-growing trees and waste chips. Their main cost advantage is the use of lower cost materials. Estimates of the impact of these new products range from capturing 25 percent of the plywood market by 1985 to 20 percent by 1989. The potential exists for expanding markets for lumber and wood products in other coun- tries, but represents an area of uncertain magnitude. One factor interfering with the growth of exports has been trade restrictions in other countries, particularly in Japan, against finished wood products. In some potential export markets, wood housing is viewed as inferior or lower quality. Industry and state organizations have carried out mar- keting programs to increase export markets, but little information is available to assess the impacts on the Northwest lumber industry. The production forecasts presented in this plan are based on earlier studies adopted by the Council, comparisons with recent Forest Service forecasts, and comments received in the review of the proposed draft assumptions.2 The high case combines the assumptions of high levels of housing activity, intensive man- agement of forest industry lands, and increased export of finished wood products from the region. Chapter 2 Table 2-3 Average Annual U.S. Housing Starts (Million Units) YEAR High Vessel) Low USFS 1985-1990 1.78 1.65 1.50 2.07 1992-1995 2.04 1.85 1.65 1.89 1996-2000 2.09 1.82 1.54 1.82 2002-2005 1.82 1.57 1.33 1.71 The medium case assumes there are no major changes in management of timber resources. Housing demand is high in the 1980s, then declines because of demo- graphic shifts in the population. Exports of finished wood products increase slowly through the forecast period. The region's share of national consumption decreases because of competition from the Southeast and Canada. The low case combines the assumptions of low demand from housing and further loss of market share to other producing regions. Plywood production declines dramatically because of competition from alternative products. The projections presented in this plan differ from the Council's 1983 studies in that the forecast for production in the plywood indus- try has been reduced substantially in all sce- narios. The lower plywood forecast is in response to the increased competition from substitute products such as oriented strandboard and waferboard. The proposed draft lumber forecasts were reduced for the state of Idaho based on comments from Idaho state agencies and utilities. These changes were consistent with the com- parison of recent Forest Service forecasts, which showed a lower forecast for lumber in Idaho. In general, the Forest Service fore- casts are slightly lower than the Council's high range for the remainder of the 1980s, and fall halfway between the high and medium range from 1990 to 2005. The fore- casts for the lumber and wood products industry are shown in Table 2-4. The high case for lumber and plywood was used in the high scenario. The medium case was used in the medium-high scenario, and the low case was used in the medium-low and low scenarios. The pulp and paper industry is the second largest industrial consumer of electricity in the region. In 1977, firms in pulp and paper products accounted for 19 percent of the electricity consumed by industry. The indus- try employed 30,100 people in 1980. The region's pulp and paper industry sup- plied an average of 14 percent of national pulp production and an average of 10 percent of national paper and paperboard production in the 1970s. The region's share of pulp pro- duction was down from an average of 17 percent during the 1960s. Most of the raw material used in the pulp- making process is wood chips, byproducts from lumber and plywood plants. Availability and cost of wood chips in the future will oper- ate as a constraint on capacity expansion in this region. Competition for portions of the timber resource has increased because of improvements in yield from each log by sawmills and plywood plants, and timber management practices that produce more uniform logs. Another factor has been the growth of the export market for chips during the 1970s. The long-term outlook for the Pacific North- west industry is favorable with regard to prox- imity to markets in the West. Other factors, however, including fiber availability and com- parative production costs (the costs of labor and environmental regulation, for example) compare less favorably to the Southeastern producing areas. The region's advantage in electricity costs has decreased as a result of large increases in electricity rates since 1979. Not only are electricity costs a major portion of operating costs, but the costs of chemicals used in the bleaching process are important as well. Chlorine and caustic soda are pro- duced through an electrolytic process, which is highly electricity intensive. 27 Chapter 2 Lumber (SIC 2421) (Billion board feet) High Medium Low Plywood (SIC 2436) (Billion square feet) High Medium Low High Lumber (SIC 2421) Plywood (SIC 2436) Other SIC 24 Total SIC 24 Medium Lumber (SIC 2421) Plywood (SIC 2436) Other SIC 24 Total SIC 24 Low Lumber (SIC 2421) Plywood (SIC 2436) Other SIC 24 Total SIC 24 Table 2-4 Forecasts of Production and Employment Lumber and Wood Products Pacific Northwest 1980-2005 AVERAGE ANNUAL PRODUCTION RATE OF GROWTH (%) 1980 1985 2005 1985-2005 12.3 -0.5 11.2 13.5 11.2 -0.9 8.9 -2.1 9.4 0.0 8.6 9.3 7.0 1.4 5.8 -2.3 EMPLOYMENT AVERAGE ANNUAL (in thousands) RATE OF GROWTH (%) 1980 1985 2005 1985-2005 31.6 -2.0 16.3 “1.4 58.5 0.3 106.4 -0.8 52.0 47.3 29.3 -2.4 26.7 22.3 12.5 -2.9 61.2 54.6 56.7 0.2 139.9 124.2 98.5 1.2 23.7 3.4 10.6 -3.7 52.0 0.2 86.3 1.8 2-8 Nationally the demand for paper products is expected to be strong, with paper holding its own against petroleum-based plastic prod- ucts. In addition, the Northwest has the largest inventory of preferred long-fiber soft- woods, and access to ports to serve world markets. The production forecasts for pulp (SIC 2611), paper (SIC 2621) and paperboard (SIC 2631) were based on work performed by Ekono, Inc., for the Brookhaven National Laborato- ries contract with the Bonneville Power Administration. Ekono, Inc., supplied Brookhaven with a range of projections by industry for the region, based on surveys col- lected from most of the region's companies and their own analysis of fiber availability and cost.3 The Northwest Pulp and Paper Association conducted a survey of regional pulp and paper producers in early 1982,4 requesting information on raw material use in 1980, pulp and paper production and capacity in 1980, and projections of production increases for the next 20 years. Ekono, Inc., estimated that participating companies represented approximately 75 percent of the installed capacity of pulp, paper and paperboard prod- ucts in the region. The survey was compiled through Arthur Andersen Company to ensure the privacy of individual companies. In developing the projections, Ekono, Inc., relied on the survey results, as well as esti- mates of capacity and production for 1980 and 1981 by product, and trends in fiber avail- ability, production costs, and regional market share in domestic and foreign markets. These projections were updated to reflect data on capacity and production provided by a 1985 Northwest Pulp and Paper Associa- tion survey for the years 1982 through 1984. In addition, changes to the relative rates of growth by end-product for Oregon, Idaho and Western Montana were incorporated to reflect differences in historical growth rates by state. Forecasts for regional production and employment in the pulp and paper industry are shown in Table 2-5. The changes result in a slightly higher regional rate of growth in paperboard production and a slightly lower regional rate of growth in paper production. Chapter 2 The residual category consists of mis- cellaneous converted paper products (SIC 264), paperboard containers and boxes (SIC 265), and building paper and board mills (SIC 266). These categories include the manufacture of bags, boxes and containers, writing paper, tissue paper and building board at sites where primary products are not produced. Industries within these categories locate close to population centers, and the forecast of production is dependent on regional population growth. Aluminum Industry The Pacific Northwest is an important world center for aluminum production. Almost 9 percent of the world’s aluminum production capacity is located in this region. The histor- ically low electricity rates and large supplies of hydropower originally attracted aluminum production to the region. The aluminum industry is far more significant as a consumer of electricity than as an employer. At full operation, the Northwest aluminum plants can consume up to 3,000 megawatts of electricity. This represents about 20 percent of 1982 regional electricity sales. In contrast, the aluminum companies employ about 9,500 persons, or about 0.3 percent of the region's employment. The aluminum industry is an important eco- nomic presence in the region. It supplies intermediate products to a number of indus- tries, including fabricated metals, machinery, transportation equipment, and electronics. Proximity to primary aluminum reduction is an important locational advantage to these industries. Thus, additional numbers of jobs may be indirectly related to the presence of aluminum smelters in the region. All of the primary aluminum plants are direct service industrial (DSI) customers of the Bonneville Power Administration (Bon- neville). As such, they are entitled to contrac- tually-specified maximum amounts of power, of which 25 percent is interruptible under certain conditions. Because of the existence of these long-term contracts, there is an upper limit to the demand for electricity by the aluminum industry. The question that seems to be more current is, what is the lower limit on aluminum industry demand for electricity, Table 2-5 Forecasts of Production and Employment Pulp and Paper Products (SIC 26) Pacific Northwest 1985-2005 PRODUCTION Average Annual Rate of Growth (%) 1985-2005 INDUSTRY High Medium Low Pulp (SIC 2611) 1.9 1.6 1.4 Paper (SIC 2621) 26 2.2 1.8 Paperboard (SIC 2631) 2.0 1.3 0.5 Other Paper 3.9 3.3 1.1 EMPLOYMENT (in thousands) 2005 1985 High Medium Low Pulp (SIC 2611) 2 1.5 1.5 ee Paper (SIC 2621) 12.5 10.7 10.5 11.1 Paperboard (SIC 2631) 48 3.6 3.3 3.3 Other Paper 8.0 86 82 62 Total SIC 26 27.4 24.4 23.5 22.3 and under what conditions is lower demand likely to occur? Since 1978, the low price of electricity that the region's aluminum companies had enjoyed has increased dramatically. The DSI rates that aluminum companies pay for electricity increased from about .3 cents per kilowatt hour in 1978 to 2.7 cents per kilowatt hour in 1984. These increased costs have made it more difficult for the Northwest's smelters to operate profitably in the recent cyclical depressions of the aluminum market. Bon- neville has offered some rate relief to alumi- num companies in order to help keep them operating at higher levels, but the long-term viability of the region's smelters is being questioned. Because of the present condition of the region's aluminum plants, and the doubts being raised about their long-term viability, the size of aluminum industry demands for electricity has become a major uncertainty facing the region's electricity planners. Not only is the long-term viability of the aluminum industry uncertain, but the increased cyclical sensitivity of the industry has important implications for the electricity system. These are planning questions that fall outside the scope of this chapter, however. The assumptions proposed here should be viewed as assumptions about the pos- sibilities facing the region's aluminum plants in the absence of other policy decisions that may affect their rates for electricity or change the conditions of electric service they receive. The basic assumption is that reasonably strong aluminum markets will make the Northwest aluminum smelters competitive in the world aluminum market. It is further assumed that strong economic growth is compatible with growing aluminum demand and prices. Therefore, the highest operating rates for Northwest aluminum plants are assumed to occur in the high forecast. The specific assumptions for operating rates of the region's smelters are shown below. The direct service industry loads are treated dif- ferently, however, in the analysis of electrical loads faced by the region for resource plan- ning purposes. Further explanation of this treatment is in Chapter 3, Volume II. 2-9 Chapter 2 Assumed Aluminum Operating Rates High Forecast: 100 percent Medium-high Forecast: 85 percent Medium-low Forecast: 70 percent Low Forecast: 50 percent The high forecast assumes strong growth in the world economy and strong growth in alu- minum demand. Higher world oil prices keep plastics less competitive with aluminum than in the lower cases. As new capacity is added throughout the world, the average cost of producing aluminum increases, making the Northwest plants relatively more profitable on average. Higher electric rates in the high case may cause some of these plants to close during cyclical weaknesses in the alu- minum market. It is assumed that DSI power contracts are transferable to purchasers of Northwest aluminum plants and that the con- tracts are renewed in 2001. In the high fore- cast, any efficiency improvements are accomplished along with capacity increases so that electricity demand is unchanged. The medium cases would be consistent with reasonably good world aluminum markets, but reflect the considerable risk that some of the region's aluminum plants may not survive through the current recession. Such out- comes could result from corporate strategic decisions or pessimistic views of the future aluminum market. These declines in oper- ating rates could also reflect efficiency improvements made in the absence of capacity increases. It is also expected that the real price of electricity for the DSI plants will be stable in the medium cases. The low case reflects a world in which alumi- num markets remain highly cyclical and on average weak. There is only very slow capac- ity growth in the world, and areas with extremely low or subsidized electric rates are able to attract smelting capacity more suit- able to operating through price cycles. In this situation, a larger number of the Northwest plants find it advantageous to close. How- ever, this would take place over a longer period because of the large share of the world capacity that resides in the Northwest, and because there are many less profitable plants in the world that are likely to close before the Northwest plants. 2-10 Chemicals The manufacture of chemicals consumes approximately 12 percent of electricity pur- chased by the industrial sector in the region. Elemental phosphorus production accounts for approximately half of the electricity con- sumed by the chemicals industry, followed by chlorine and caustic soda, which accounts for approximately 20 percent. In the Council's forecasting models, the consumption of elec- tricity by these industries is modeled on a plant-by-plant basis. Two of the chlorine and caustic soda plants are direct services indus- tries (DSIs) of Bonneville. The remainder of the chemicals industry in the region is dominated by nuclear fuels pro- cessing and agricultural chemicals (such as fertilizers). The nuclear fuels processing component has exhibited large swings in employment, as policies of the federal gov- ernment have changed over the last 20 years. The agricultural chemicals compo- nent has increased at a steady rate in the last decade, but it is not likely to increase rapidly in the future. The manufacture of chlorine and caustic soda involves the electrolytic separation of salt into two co-products: chlorine and sodium as sodium hydroxide (caustic soda). Approximately 1.12 pounds of caustic soda are produced per pound of chlorine. The market outlook for the two products dif- fers substantially. In the past, chlorine has held the stronger market and higher price. Expansion plans were based on growth in chlorine demand. As little as ten years ago, caustic soda was considered an undesirable “byproduct,” and for years producers sought to develop a commercial process to produce chlorine without producing caustic soda. In the last few years, the price of caustic soda has risen and supplies have tightened, while chlorine demand has dropped and prices have remained stable. Industry experts have predicted growth rates for national chlorine demand in the 1980s to range from an average of 1 to 3 percent per year, whereas demand for caustic soda could increase at rates ranging from 2.5 to 5 per- cent. This is slower than the rate of growth in production from 1960 to 1980, which aver- aged 4.1 percent per year. From 1970 to 1980, however, production increased at an annual rate of only 1.6 percent. The outlook for chlorine has been affected by environ- mental regulations on effluent standards. Pulp and paper producers may substitute other chemicals in pulp bleaching to reduce emissions. The outlook for caustic soda is much more favorable because it has a broader base of end-uses. One of the fastest growing end-uses is in the neutralization of waste acids. Tougher environmental stan- dards would enhance the outlook for caustic soda. Soda ash can be substituted for caustic soda, and although the initial invest- ments required to handle soda ash are high, projections of relative price increases for caustic soda and soda ash favor some con- version to soda ash. Production of chlorine and caustic soda is likely to be constrained by the price of chlorine, since chlorine is more difficult to store. Chlorine and caustic soda are produced in five plants in the region, with four located in Washington and one in Oregon. Nationally, over half of the chlorine produced is used within the chemicals industry in the manufac- ture of a variety of organic and inorganic chemicals. An additional 13 percent is used by the pulp and paper industry as a bleaching agent in the production of paper. In the Pacific Northwest, a much larger portion of produc- tion goes to the pulp and paper industry. In fact, two of the five plants in the region are owned by pulp and paper companies. The proportion of product going to the pulp and paper industry in the Northwest varies from 32 percent to 80 percent, depending on the plant and temporary shifts in market con- ditions. This is a much larger proportion than nationally, although the pattern is similar in the Southeastern U.S. Although not all of the chlorine produced in the region is sold to pulp and paper producers, growth in the produc- tion of paper (SIC 2621) was chosen as a reasonable indicator of growth in the produc- tion of chlorine and caustic soda. The projec- tions presented here are within the range of projections for national production cited in the preceding paragraphs. Comparison of the production growth rates for chlorine and caustic soda and paper (SIC 2621) shows that the projection for chlorine and caustic soda is 0.4 percent per year higher in the high case to allow for higher rates of growth in other end-uses. The medium case growth rate is similar to the medium rate of growth in Paper, and the low case is 0.5 percent per year lower than the low case paper projection to reflect lower rates of growth in other end- uses or market penetration by British Colum- bia producers. Table 2-6 shows projections of production for SIC 2812, chlorine and caustic soda. Elemental phosphorus production is located in only four states (Idaho, Florida, Montana and Tennessee), near deposits of phosphate rock. Elemental phosphorus is extracted from phosphate rock in electric furnaces, and frequently converted nearby to phosphoric acid and other compounds. Elemental phosphorus plants are classified under industrial inorganic chemicals, not elsewhere classified (SIC 2819). In the North- west, firms producing elemental phos- phorus, nuclear fuel, corn starch, chemical catalysts and a variety of other products are classified under SIC 2819. About half of total U.S. elemental phosphorus production capacity is located in the Northwest. Of this, 85 percent of capacity is located in Idaho, with the remainder in Montana. The major end-use markets for elemental phosphorus are cleansers and detergents (45 percent), food and beverages (15 per- cent), metal treating (10 percent) and other chemicals and cleansers (30 percent). The outlook for elemental phosphorus production in the Northwest depends, in part, on the demand for these products. The detergent market has been projected to remain stable or increase slightly over the forecast period, with growth rates ranging from 0 percent to 1 percent per year. Non- detergent uses, such as food and beverage products and other uses, have been forecast to increase at rates of 1.4 percent to 2.4 per- cent per year. The problems facing elemental phosphorus producers in the region include the cost and availability of electricity and mature markets for their products. The costs of additional electricity beyond current contracted Chapter 2 Table 2-6 Forecasts of Chemicals Industry Production Pacific Northwest 1985-2005 Average Annual Rate of Growth (%) MEDIUM- MEDIUM- sic HIGH HIGH LOW Low Chiorine/Caustic Soda (SIC 2812) 3.0 22 1.3 1.3 Elemental Phosphorus (SIC 2819) 1.6 0.7 0.0 0.0 Other Chemicals (SIC 28XX) 44 3.6 25 1.4 amounts may lead to no expansion in capac- ity over the forecast period. This was assumed to be the case for the low scenario. The high case projection is a weighted aver- age of the higher ranges of forecasts for detergent and nondetergent uses of elemen- tal phosphorus. Projections of production are shown in Table 2-6. The residual category for chemicals (SIC 28XX) includes a wide variety of products manufactured in the region. The larger groups in employment and energy use are the nuclear engineering, fuels and waste pro- cessing segments, and agricultural chem- icals (primarily fertilizers and pesticides). There are also many other types of chemical products manufactured in the region. The forecasts for the other chemicals cate- gory are shown in Table 2-6. The forecast range for the region was based on selecting ranges around national forecasts for chem- icals, with the exception of the forecasts for Idaho. Comments were received that indi- cated that the industry in Idaho is dominated by agricultural chemicals. The demand for agricultural chemicals is expected to increase at a slower rate of growth than the demand for other chemicals products. In the high case, production increases at a 30 per- cent higher rate than Wharton's high case forecast for the nation, while the low case increases at a 30 percent slower rate than Wharton's low case forecast for the nation. Agriculture and Food Processing Over the past decade, agriculture has found itself increasingly at the mercy of circum- stances beyond its control. These circum- stances run the gamut from changing foreign markets for farm products to federal farm policy and state decisions on groundwater pumping. Northwest agriculture markets were primarily regional and national. How- ever, increasing production and sales of farm products from the Midwest and Northeast for large eastern markets has put increasing pressure on Northwest producers to sell over- seas. The Orient has been an important des- tination for many of these sales. A recent comprehensive study of Northwest agri- culture concluded that if Northwest agri- culture is to maintain its share of national production and make reasonable growth, it must continue to develop foreign markets. Regional agriculture has been fairly suc- cessful in doing so. However, farm production and marketing efforts are often offset by a lack of clear agricultural policy from the U.S. government. There are mixed policy signals on the level and structure of price supports, overseas marketing assistance, environ- mental enforcement, taxes and water policy, just to name a few. 2-11 Chapter 2 Table 2-7 Forecasts of Employment Agriculture and Food Processing Pacific Northwest 1985-2005 EMPLOYMENT AVERAGE ANNUAL (in thousands) RATE OF GROWTH (%) 1985 2005 1985-2005 Agriculture High 157.9 0.0 Medium-high 157.4 150.9 -0.2 Medium-low 144.9 -0.4 Low 129.5 -1.0 Food Processing High 86.0 0.9 Medium-high 72.5 74.8 0.2 Medium-low 68.0 -0.3 Low 60.3 -0.9 The region's agriculture has found itself in increasingly difficult times during the past few years. Crop prices have been low, production costs have continued to increase, and export markets have been shrinking. Furthermore, there are several issues on the horizon which further cloud development of new irrigated land. An unsettled lawsuit in Idaho over pri- ority of use of Snake River water— irrigation or hydrogeneration—has halted most irri- gated land development there. In Oregon, groundwater pumping has been restricted in some irrigated areas. The restriction may be expanded to additional areas. In Washington, half the Columbia Basin irrigation project is not yet developed. Federal funding for devel- opment is restricted, and there is not yet a decision on partial state funding nor even on the advisability of developing the remaining land. During the past few years, there are indica- tions that irrigated land development may have leveled off. During this same period, irrigation pumping loads have become more erratic and appear to be leveling off. Many irrigators are installing water and electricity conservation equipment and measures. 2-12 Agriculture employed 157,100 persons directly in the region in 1980, accounting for almost 5 percent of total employment. Direct employment in agriculture decreased at a rate of 1.5 percent per year from 1960 to 1980, even though agricultural production increased throughout this period. This resulted because of large increases in mech- anization. Agricultural employment is pro- jected to decrease at a slower rate than in the past to reflect increasing costs of capital and fuels. The projections for agricultural employ- ment are shown in Table 2-7. Agricultural production supports a large food processing industry. In 1980, 74,200 persons were employed in food and kindred products (SIC 20), which represented 13 percent of manufacturing jobs. Activity in this industry is concentrated in preserved fruits and vegeta- bles (SIC 203), which accounted for nearly half of the employment in food and kindred products and over half of the electricity con- sumption. Processed potatoes are the major products in this category, accounting for over half of the value added in the regional food processing industry. Another portion of the industry important to coastal areas is the seafood canning and freezing industry. Poor commercial fishing conditions have forced closure of a number of these plants. The outlook for plants in preserved fruits and vegetables relies on future demand for pro- cessed foods domestically and in Pacific Rim countries, a recently expanding market. Changes in lifestyle and consumer prefer- ences have had an impact on the market for food products. Comments received during the planning process indicated that high transportation costs and rising electricity rates were leading to more plants locating in the midwestern U.S., closer to markets. In addition, it was pointed out that Wharton's forecast of national employment growth in this sector led to employment decreasing at a rate of 1.4 percent per year from 1985-2005. The projections of employment in food pro- cessing for the region are shown in Table 2-7. The High Technology Industries A great deal of attention has been focused of late on the so-called high technology indus- tries. State and local governments in the U.S. and national governments around the world have initiated studies and programs designed to understand and attract eco- nomic development through the encourage- ment of growth in high technology industries. In the region, the recent growth of electronics and software firms has been heralded by some as a panacea for stagnation in some of the region's resource-based industries. The first step in a discussion of high tech- nology industries is to define the group of industries to be discussed. Several methods of defining high technology have been pro- posed, but general agreement does not exist on which definition is the most appropriate. To a certain extent, the nature of technology intensive activity makes definition difficult, because the industries are changing so rapidly. New industries are created and oth- ers become obsolete, thus causing any defi- nition of high technology industries to be tied to a particular point in time. Most definitions have looked at one or a com- bination of three factors: research and devel- opment expenditures as a proportion of value added, the percentage of scientific and tech- nical personnel in industry employment, and product sophistication. The definition described in this chapter was adopted from a Battelle study® for the state of Washington and reflects a combination of all three factors. The Battelle study included a number of chemical industries in its definition of high technology industries. These industries were excluded from the definition of high tech- nology industries used in this chapter, for reasons described below. Even at the level of industry detail shown in Table 2-8, it is difficult to categorize industries as high technology industries. At more detailed levels of categorization, however, data are not available to analyze the indus- tries because of disclosure laws that protect companies rights to proprietary information. Comments were received during the plan- ning process that it may be inappropriate to apply definitions developed to describe high technology industries in the state of Wash- ington to the same industries in other states. In particular, concerns were raised about the inclusion of some chemical industries, partic- ularly industrial inorganic chemicals (SIC 281) and agricultural chemicals (SIC 287) in the high technology group. The chemical industry forecasts have been discussed in a previous section. The list of industries included in the high technology group and their SIC codes are shown in Table 2-8. In the U.S., the industries listed in Table 2-8 comprised approximately 5.3 percent of total wage and salary employment in 1982, com- pared to 6.0 percent for the region. The high technology share of total employment was 7.9 percent in Washington, 5.0 percent in Oregon, 3.6 percent in Idaho, and 0.4 per- cent in the state of Montana. In 1982, high technology industries em- ployed 145,700 persons in the region, with almost half of the employment concentrated in the aerospace category. The second largest category was professional instru- ments, with 16.2 percent, followed by elec- trical equipment, with 16.0 percent of high technology employment. Table 2-9 shows employment in 1982 by state for the major high technology groupings. Chapter 2 Table 2-8 High Technology Industries SIC CODE INDUSTRY NAME Machinery 351 Engine and Turbines 357 Office, Computing and Accounting Machines Electrical Equipment 361 Electric Transmission and Distribution Equipment 362 Electrical Industrial Apparatus 365 Radio and Television Receiving Equipment 366 Communication Equipment 367 Electronic Components and Accessories 369 Miscellaneous Electrical Machinery Transportation Equipment 372 Aircraft and Parts 376 Guided Missiles and Space Vehicles and Parts Professional Instruments 381 Scientific Instruments 382 Measuring and Controlling Instruments 383 Optical Instruments 384 Medical and Dental Instruments 386 Photographic Equipment and Supplies Business Services 737 Computer and Data Processing Services 7391 Research and Development Laboratories 2-13 Chapter 2 Machinery (SIC 351, 357) percent of high tech Electrical Equipment (SIC 361, 362, 365, 366, 367, 369) percent of high tech Transportation Equipment (SIC 372, 376) percent of high tech Professional Instruments (SIC 381, 382, 383, 384, 386) percent of high tech Business Services (SIC 737, 7391) percent of high tech Total High Tech Percent of Total Employment TOTAL EMPLOYMENT Table 2-9 Employment in High Technology Industries, 1982 PACIFIC UNITED STATES NORTHWEST WASHINGTON OREGON IDAHO MONTANA 521,380 13,110 4,580 6,510 2,000 20 13.2% 9.0% 4.7% 16.9% 23.9% 2.4% 1,678,140 23,300 11,160 10,225 1,575 340 42.6% 16.0% 11.4% 26.5% 18.8% 41.0% 705,820 69,265 67,800 1,450 15 0 17.9% 47.5% 69.2% 3.8% 0.2% 0.0% 581,740 23,570 6,230 17,000 200 140 14.8% 16.2% 6.4% 44.1% 2.4% 16.9% 456,160 16,500 8,210 3,380 4,580 330 11.6% 11.3% 8.4% 8.8% 54.7% 39.8% 3,943,240 145,745 97,980 38,565 8,370 830 5.3% 6.0% 7.9% 5.0% 3.6% 0.4% 74,297,300 2,434,545 1,239,700 763,975 232,400 198,470 SOURCES: U.S. Census Bureau County Business Patterns, 1982. The employment figures shown in this table are based on a survey of employment during the pay period including March 12. As such, they are not comparable to annual average data used in other segments of this report. They are used for illustration purposes here because they are available at the level of industry detail needed. The aerospace industry in the region is domi- nated by the Boeing Company, which has a number of production facilities in the state of Washington. Employment in aerospace in the state of Washington has been extremely cyclical, dropping from 104,000 in 1968 to 40,000 by 1971. In 1980, it reached a level of 79,600, only to drop to 64,400 by 1983. From 1970 to 1982, the high technology industries increased employment at an aver- age annual rate of 4.2 percent. This com- pares to a national growth rate of 2.2 percent over the same period. Removing aero- space from the calculation shows that non- aerospace high technology employment increased at an average annual rate of 11.8 percent in the region, compared to a national rate of 2.8 percent. 2-14 The factors often cited as favorable for the region's growth in high technology include the quality of the region's labor force, avail- able land, good educational facilities and an environment suitable for maintaining a high quality of life. A survey of high technology companies regarding location factors was completed by the Congressional Joint Eco- nomic Committee. The results are shown in Table 2-10. The existing concentration of firms in the region also testifies to the impor- tance of spin-off activity from Pacific North- west firms and California firms. The factors often cited as unfavorable for the region's growth in high technology industries include high labor costs, unfavorable tax pol- icies, and complex regulatory practices that make it difficult to expand or locate facilities. There is also some question as to the region's commitment to improving or maintaining the quality of the educational systems in light of tax revolts and state and local budget crises. Many states and cities in the U.S. are com- peting aggressively to attract high technology industries. Some areas of the country, such as New England and North Carolina's Research Triangle Park, enjoy advantages in their traditions of high quality academic institutions. While the region will most assuredly continue to see growth in its high technology industries, the question is whether or not the region will be able to increase or maintain its share of national growth. National forecasts of employment prepared by the U.S. Bureau of Labor Statistics show employment in high technology industries increasing at an annual rate of 2.4 to 2.5 percent between 1982 and 1995. Although this rate of growth is a third to a fifth faster than that projected for totalemployment, high technology would nevertheless account for only 8 to 9 percent of new jobs. The impact could be greater in particular states and regions. Forecasts of employment for high technology industries are shown in Table 2- 11. The table shows forecasts for industries at the two-digit SIC level, which includes some businesses that are not classified as high technology industries. Electrical equipment and profes- sional instruments are the only categories where nearly all of the employment is in the high technology category. In machinery and business services, only 34 and 20 percent, respectively, of the employment are in the high technology industries. Approximately 73 percent of the employment in transporta- tion equipment is in the high technology category. A rapidly growing sector of the machinery industry in the region has been the computer machinery category. Much of the remainder of the machinery industry, however, is farm, construction, logging and other heavy machinery. These categories are not forecast to grow rapidly. Aerospace employment, which is dominated by the Boeing Corporation, accounts for nearly 85 percent of employment in the trans- portation equipment industry in the region. Commercial aircraft production represents the largest portion of production in the region. During the recent recession, annual average employment in aerospace declined almost 20 percent. Commercial aircraft orders had dropped substantially because of low profits in the airline industry and declines in pas- senger miles. Since then, Boeing has started to increase employment as orders increased, in response to improvements in economic conditions and the financial condition of air- lines. Boeing is well positioned for the next few years because of its fuel-efficient 757 and 767 model aircraft. Its primary competi- tion is Airbus, a European aircraft consor- tium. The market for commercial aircraft is projected to improve, although it will probably continue to be highly cyclical. Because employment in this category is dominated so much by one company, the forecasts encom- pass a wide range of uncertainty. Table 2-10 Factors That Influence Regional Location of High Technology Coinpanies Chapter 2 FACTOR PERCENTAGE OF FIRMS CITING FACTORS AS SIGNIFICANT OR VERY SIGNIFICANT Labor Skills and Availability Labor Costs Tax Climate Academic Institutions Cost of Living Transportation Access to Markets Regulatory Practices Energy Costs and Availability Cultural Amenities Climate Access to Raw Materials NOTE: cant, or not significant. 89.3 72.2 67.2 58.7 58.5 58.4 58.1 49.0 41.4 36.8 35.8 27.6 Firms were asked to rate each factor as very significant, significant, somewhat signifi- SOURCE: U.S., Congress, Joint Economic Committee. Location of High Technology Firms and Regional Economic Development, 1 June 1982, p. 23.; and from Battelle Seattle Research Center, High Technology Employment, Education and Training in Washington State, June 1984. Table 2-11 Forecasts of Employment High Technology Industries Pacific Northwest Average Annual Rate of Growth (%) 1985-2005 MEDIUM- MEDIUM- HIGH HIGH Low Low Machinery (SIC 35) 3.8 3.3 2.5 1.4 Electrical Equipment (SIC 36) 4.3 3.5 25 0.4 Transportation Equipment (SIC 37) 1.6 1.0 0.4 -1.0 Aerospace (SIC 372)* 1.4 0.6 -0.2 -2.1 Professional Instruments (SIC 38) 44 3.5 2.0 0.0 Business Services (SIC 73) 47 3.8 2.8 eZ, *Washington only. 2-15 Chapter 2 Table 2-12 Total Employment Shares U.S. and the Pacific Northwest Percent of Total (%) PACIFIC NORTHWEST U.S. 1970 1980 1970 1980 Total Employment 100.0 100.0 100.0 100.0 Manufacturing 20.7 18.2 25.1 21.6 Nonmanufacturing 79.3 81.8 74.9 78.4 Mining 0.5 0.4 0.8 11 Agriculture 7.5 49 43 3.6 Construction 4.4 49 5.1 4.6 Transportation & Public Utilities 6.3 5.5 5.8 5.5 Wholesale and Retail Trade 20.9 23.0 20.7 21.7 Finance, Insurance & Real Estate 47 5.7 5.0 5.5 Services 14.6 17.9 16.0 19.1 Government 20.5 19.4 17.1 17.3 Growth in Nonmanufacturing Industries Employment in nonmanufacturing has grown faster in the last two decades than employ- ment in manufacturing. Table 2-12 shows the shares of total employment by industry for the region and the U.S. Nonmanufacturing employment accounted for 81.7 percent of total employment in the region in 1980. The largest category of nonmanufacturing employment in the region is wholesale and retail trade, followed by government. The third largest nonmanufacturing industry is services, which includes such industries as health care, business services, and personal services. The growth in the nonmanufacturing sectors has occurred on a national level, as well as at the regional level. A larger proportion of man- ufactured goods are produced in other coun- tries, which has had a negative impact on the proportion of employment in manufacturing. Productivity gains in the past have occurred to a greater extent in manufacturing indus- tries, and this has lowered employment rela- tive to output. Computerization of some activities could lead to higher productivity gains in nonmanufacturing, however. 2-16 A closer look at specific industries may add some insight into the growth in the non- manufacturing sectors. The services indus- try was the fastest growing industry in the region from 1970-1981, increasing employ- mentat a rate of 6.1 percent per year. In 1981, health services accounted for 33 percent of the region’s employment in services. Employment in health services increased at an annual rate of 6.2 percent from 1970-1981. Growth in this sector resulted from the expan- sion of health care benefits for workers and elderly people and growing public interest in personal health. The second largest service category, busi- ness services, accounted for 16 percent of the region's employment in services. This category was among the fastest growing sec- tors in services, increasing employment at an annual rate of 8.6 percent. This category includes a diverse group of industries, such as computer and data processing services, advertising agencies, building services com- panies, and personnel agencies. Although it only accounted for 3 percent of services employment in 1981, the legal ser- vices industry was the fastest growing among services industries. Employment increased at an annual rate of 10.1 percent from 1970-1981. Employment in construction increased at a rate of 5.0 percent per year from 1970-1981. Since 1979, however, construction employ- ment has decreased, as a result of slower population growth and the cancellation or delay of construction on nuclear power plants. The finance, insurance and real estate sector increased employment at an average annual rate of 4.8 percent between 1970 and 1981. The most rapidly growing sectors in this industry were credit agencies (other than banks) and investment offices. Deregulation of the financial industry has led to the crea- tion of a wide range of services by a diverse group of businesses. The combination of deregulation, high interest rates, and loan defaults has put a great deal of strain on financial institutions. This may result in an industry shakeout in the next few years, accompanied by slower employment growth. Wholesale and retail trade accounted for the largest share of total employment in 1981, as shown in Table 2-12. Wholesale trade accounted for approximately one-fourth of employment in trade and increased at an annual rate of 3.7 percent from 1970-1981. Employment in retail trade increased at a rate of 4.4 percent per year during the same time period. Eating and drinking establishments ac- counted for 35 percent of employment in retail trade. This was also the fastest growing category of employment in retail trade, increasing at an annual rate of 7.9 percent from 1970-1981. The increase in household consumption of food away from home reflects the increase in household income and the increase in the participation of women in the labor force. A larger proportion of household budgets for persons aged 25-44 is spent on food away from home than for other groups. As the “baby boom” gener- ation continues to move into this age cate- gory, growth in the restaurant industry is expected to continue. Other fast growing retail trade categories included apparel and accessory stores and miscellaneous retail stores, which includes sporting goods stores and mail order houses. Employment in both categories increased at an average annual rate of 4.9 percent. Chapter 2 The public sector was the second largest Table 2-13 employment category in the region in 1980, Nonmanufacturing Employment Projections as shown in Table 2-12. State and local gov- Pacific Northwest ernment accounted for over 80 percent of Average Annual Rate of Growth (%) employment in government. Between 1970 1906-2005 and 1981, employment in the federal govern- 1970-1981 High Low ment increased 1.4 percent per year, while state and local government increased em- Construction 5.0 3.6 0.7 ployment at a rate of 3.2 percent per year. Transportation, Communications and Public Utilities 28 2.7 -0.2 Education accounts for the largest proportion of state and local government employment. Trade 4.2 3.8 0.9 Since 1981, cutbacks in federal, state, and Wholesale Trade o7 3.9 1.0 local budgets have led to decreases in public ; tail Tr : H 7 sector employment. The outlook for future erat tres 38 ed employment changes in this sector is depen- Food Stores 42 3.0 0.2 dent on the level of population growth and Eating & Drinking Places 79 5.0 2.0 Policy decisions. Finance, Insurance & Real Estate 48 3.6 0.7 Employment in transportation, communica- Services® 6.1 4.2 1.3 tions and public utilities increased at an Hotels and Lodging Places 3.9 36 07 annual rate of 2.8 percent from 1970 to 1981. | | The fastest growing category was transpor- Business Services 8.9 4.7 1.7 tation services, which includes travel agen- Amusement & Recreational Services 48 3.7 0.8 Gies, freight forwarding services, and ship- Health Services 6.2 45 16 ping agents and brokers. Employment in transportation services increased at an aver- Government 28 2.9 0.3 age annual rate of 9.5 percent from 1970 to Federal Government 1.4 1.6 0.3 1981. The two largest categories of transpor- i ; te k i : tation and public utilities employment in 1981 State & Local Governmen $2 $2 pa were motor freight transportation and ware- | | housing with 29 percent, and communication @ Excludes Educational Services, SIC 82. services with 32 percent. Motor freight trans- » Includes Educational Services, SIC 82. portation and warehousing employment increased at an average annual rate of 3.6 Table 2-14 percent. Employment in communications Nonmanufacturing Shares of Total Employment increased at an average annual rate of 3.5 in 2005 percent. SHARE OF EMPLOYMENT (%) The discussion of nonmanufacturing indus- Pacific Northwest tries presented thus far has centered on . industries as defined by the Standard Indus- High 87.5 trial Classification (SIC) system. Industries Medium-high 87.0 such as the travel industry and port activity 7 -| 7 are not separated from other economic data Medtunnctow ees to allow historical analysis of their impor- Low 85.9 tance to the regional economy. U.S. (Wharton) The travel industry, which includes tourism High ot and business travel, has impacts on retail Medium 85.2 trade sectors, such as eating and drinking Low 84.5 places, retail stores, and service stations. It has an impact on transportation industries, such as transportation services, and air or rail transportation. It has an impact on the services industry, which includes hotels and lodging places, personal services, and amusement and recreation services. It also 2-17 Chapter 2 Table 2-15 Real Output per Employee, U.S. Average Annual Rate of Growth (%) YEARS ALL INDUSTRIES MANUFACTURING 1953-1963 1.9 2.6 1963-1973 1.8 3.1 1973-1983 0.3 1.8 ALL INDUSTRIES MANUFACTURING FORECAST High Base Low High Base Low 1983-1993 1.4 1.2 el 3.2 2.8 2.4 1993-2003 1:6) 1.3 0.9 3.3 3.0 7431) Table 2-16 Total Population and Households AVERAGE ANNUAL TOTAL POPULATION (Thousands) RATE OF GROWTH (%) 1960 1970 1980 1960-70 1970-80 Washington 2,853.2 3,409.2 4,132.2 1.80 1.94 Oregon 1,768.7 2,091.4 2,633.1 1.69 2.33 Idaho 667.2 712.6 944.0 0.67 2.85 W. Montana 231.7 253.5 294.5 0.90 1.51 PNW 5,520.8 6,466.7 8,003.8 1.59 2.16 US. 180,671.0 204.878.0 227.020.0 Wer, 1.03 HOUSEHOLDS AVERAGE ANNUAL TOTAL POPULATION (Thousands) RATE OF GROWTH (%) 1960 1970 1980 1960-70 1970-80 Washington 894 1,106 1,540.5 2.15 3.37 Oregon 558 692 991.6 2.18 3.66 Idaho 194 219 324.1 1.22 4.00 W. Montana 70 79 106.4 1.25 3.47 PNW 1,716 2,096 2,962.6 2.02 3.52 U.S. 53,021 63,450 80,377 1.81 2.39 PERSONS PER HOUSEHOLD (Total Population/Total Households) 1960 1970 1980 PNW 3.22 3.09 2.70 U.S. 3.41 3.23 2.82 2-18 has an impact on the government sector, through parks and recreation, national parks, national and state forests, and the highway system. Because all of these services are consumed by the local population as well as out-of-state travelers, it is difficult to measure the impact of the travel industry on the economy. Nevertheless, the travel industry is an impor- tant activity in the region. The beauty and diversity of the region's natural environment provide opportunities for a variety of recrea- tional activities. Factors that will aid the growth of the travel industry in the future include increases in real income and changes in the age composition of the popu- lation. State and local governments in the region have developed programs to promote tourism and conventions, which will add to its growth. Another economic activity that appears to have increased in importance is port activity related to trade with Alaska and other coun- tries. The expansion of the economies of the Pacific Rim countries and the region's prox- imity to these countries point to increased trade and transportation activity. The employ- ment impacts are difficult to measure because they are spread across a number of SIC categories. Port activity has an impact on the transportation, wholesale trade, ser- vices, and financial industries. It has an impact on manufacturing industries as well, by providing markets for goods produced in the region. A study by the Port of Seattles showed a direct impact of 55,800 jobs result- ing from the harbor and airport facilities. This estimate was for 1982, which was a year of worldwide economic slowdown. In addition, the estimate included jobs in King County only, which would underestimate the impact of the port on the state of Washington and the region. In recent years, more attention has focussed on the nonmanufacturing industries as an increasing source of jobs to the economy. The traditional approach to understanding regional economic development empha- sized manufacturing, agriculture, and extrac- tive industries as the providers of the basis for economic growth. Other industries were treated as secondary, providing support ser- vices to these industries and to the local pop- ulation. A recent study of the services sector in the central Puget Sound region’ disputes this approach. The study interviewed firms from selected industries in the services sec- tor and estimated that approximately one third of the employment in these industries is linked to export markets. The study points out many areas where the dynamics of loca- tion and growth of nonmanufacturing indus- tries have remained largely unexplored. In developing its range of forecasts of employment growth in the nonmanufacturing industries, the Council has relied on national forecasts developed by Wharton and the Bureau of Labor Statistics, and comparison to historical regional growth rates by industry. Table 2-13 shows a comparison of the Coun- cils forecasts of nonmanufacturing employ- ment by industry with historical growth rates. Table 2-14 shows the Council's forecasts of the share of nonmanufacturing employment to total employment for the year 2005. The shares for the nation forecast by Wharton are shown as well. Changes in Productivity Growth The early phases of an economic recovery period often show large gains in productivity. The conditions may exist at this time, how- ever, for a more sustained growth in labor productivity in the U.S. that could last well beyond the cyclical impacts of recession and recovery. Some of the factors encouraging higher productivity growth were brought about by the recession. Intense foreign com- petition and a high value of the U.S. dollar against foreign currencies has put down- ward pressure on prices. Efforts to increase profitability have focused on improving productivity. Recent changes in federal government policies have led to increased financial in- centives for investment. These include decreased capital gains tax rates, generous tax credits for research and development expenditures, and the Accelerated Cost Recovery program, which allows faster depreciation of capital investments. Over the long-term, demographic factors will have an impact on productivity growth. With the maturation of the baby boom generation, there will be fewer young, inexperienced workers in the labor force. The impact of developments in high tech- nology are just beginning to be observed in office automation, robotics, electronic tech- nology, and telecommunications. Spurred by foreign competition and tempted by numer- ous success stories, U.S. companies are turning to new technology to remain com- petitive in world markets. Two factors that may have dampened pro- ductivity growth in the 1970s may contribute to productivity growth in the 1980s by their absence. These are energy price shocks and new federal regulations. The costs of adjust- ment to higher prices and higher environ- mental standards diverted funds from in- vestments that contribute more directly to measures of productivity. These factors are not likely to be as prominent in the near future. Table 2-15 shows rates of growth in real output per employee for all industries and for manufacturing. As shown, productivity growth in the 1970s was slow compared to previous decades. Wharton's forecasts for the next twenty years show a continuation of the rapid trends established in the 1950s and 1960s. Table 2-A-4 of Appendix 2-A shows productivity forecasts by industry for man- ufacturing industries. Chapter 2 Population, Households and Housing Stock Total population in the region was 8.0 million in 1980. Regional population increased at an average annual rate of 2.2 percent from 1970 to 1980, more than twice the rate of U.S. population growth (1.0 percent) in the same period. Population growth in the region was more than one-third faster in the 1970s than during the 1950s and 1960s. Idaho was the fastest growing state in the region during the 1970s, although it was the slowest growing in the 1960s. Table 2-16 summarizes historical data on population and households. The number of households in the region and the nation grew at a higher rate than popula- tion. Although population growth was slower nationally in the 1970s than in the 1960s, because of lower birth rates, growth in the number of households was considerabty higher in the later decade. During the 1970s, the ‘baby-boom’ generation reached the 20-29 year age group, where household for- mation rates are high. Decreasing fertility rates also lowered the average household size. Householder rates, or the proportion of the population in an age group designated to represent a household, increased rapidly with the rise in divorce rates and single per- son households. In the 1970s, householder rates have increased dramatically for females over the age of 65, as more women in this group have maintained their own household, rather than move in with family or to group quarters. In addition, women in the 20-29 age group have maintained house- holds at a higher rate. The combination of shifts in age composition and of changes in householder rates has lowered average household size in the region from 3.1 in 1970 to 2.7 in 1980. There were 2.963 million occupied housing units in the region in 1980. Results from the 1980 U.S. Census indicated that approx- imately 78 percent of the occupied housing stock were single family units (1-4 units per building). An additional 14 percent were multi- family units and 7 percent were manufac- tured homes. 2-19 Chapter 2 Table 2-17 Forecast of Population and Households Pacific Northwest 1980-2005 AVERAGE ANNUAL SCENARIO 1980 2005 RATE OF GROWTH (%) Total Population (in thousands) High 12,449.4 1.8 Medium-high 8,005.1 11,383.6 1.4 Medium-low 10,007.4 0.9 Low 8,785.7 0.4 Total Households (in thousands) High 5,605.6 2.6 Medium-high 2,963.7 4,761.1 1.9 Medium-low 4,184.8 1.4 Low 3,381.3 0.5 Table 2-18 Housing Stock Projections Pacific Northwest 1980-2005 Share of Occupied Housing Units (%) 2005 Medium- Medium- 1980 High High Low Low Single Family (2-4 units) 78.3 77.4 72.0 68.4 72.4 Multifamily (5 and more) 14.2 14.3 Wl 19.9 17.9 Manufactured Homes 7.5 8.3 10.9 11.7 9.7 The forecast for population is derived from the forecast of total employment through an average employment-population ratio. Changes in the employment-population ratio reflect changes in labor force participation, unemployment rates, and age composition of the population. The proportion of women in the labor force increased rapidly in the 1960s and 1970s. From 1960 to 1980, the percent- age of women in the labor force increased from 37 percent to 52 percent. The employ- ment-population ratios in this forecast incor- porate the impacts of continued increase in female labor force participation, although at slower rates than in the past. The range of projections was based on national trends as forecast by Wharton and the U.S. Bureau of Labor Statistics. Changes in employment- population ratios implied in the national fore- casts were tracked in the state-level fore- 2-20 casts, maintaining historical differences between the state and national ratios. Table 2-A-1 in Appendix 2-A shows employment- population ratios for each state. The forecast for total households is obtained from the forecast of population after dividing by average household size. Changes in aver- age household size reflect changes in the age composition and householder rates. The projections are based on national trends as forecast by the U.S. Bureau of the Census. The high and medium cases assume that householder rates will continue to increase, but at much slower rates than in the 1970s. This results in part because of increases in the relative cost of housing and in a slowing of increases in the divorce rate. The low case assumes that householder rates do not increase, but average household size decreases slightly because of changes in age composition. Average household size projections by state are shown in Table 2-A-2 of Appendix 2-A. Table 2-17 shows the forecasts of population and households which result from the assumptions described. Change in the hous- ing stock is the result of change in total households plus replacement of existing units. The proportion of new housing units by type is projected for each state. Table 2-A-3 in Appendix 2-A shows the proportion of housing additions by type for each state and scenario. In all scenarios, it is assumed that the affordability of new single family housing will lead to a smaller proportion of single family units than in the current stock of hous- ing in the region. As noted in Table 2-A-3 in Appendix 2-A, the same proportions of housing additions by type were used in the low and medium-low scenarios. In the low growth scenario, the existing stock is a much larger proportion of total stock in 2005 because there are fewer additions than in the other scenarios. Changes in the stock of housing shown in Table 2-18 for each scenario are the result, therefore, of assumptions regarding the rela- tive proportion of each housing type in addi- tions, and the rate at which additions are added to the stock. Real per Capita Income Real per capita income is an important input to many econometric models of energy demand. It plays a far less critical role in the more structural end-use models used by the Council. The only sector it affects directly is the residential sector, where it influences the penetration rate of certain types of appliances, and the long-run expected use of appliances. In 1980, the personal income per capita of the Pacific Northwest was $9,600, only 1.1 percent above the U.S. average of $9,494. The size of the Northwest states per capita incomes relative to the U.S. average in 1980 is shown in Table 2-19 along with extreme values for other states in the U.S. Chapter 2 Table 2-19 Ratio of State per Capita Income to National per Capita Income, 1980 Oregon 0.98 Idaho 0.85 Montana 0.88 Highest State 1.37 Lowest State 0.69 Alternative Fuel Prices This section describes assumptions about world oil prices and the retail prices of natural gas, oil, and coal. These fuel price assump- tions are important for two reasons. First, since these fuels are alternatives to electricity in end-use energy consumption, their prices will affect the forecasts of demand for elec- tricity. This is particularly true for the residen- tial and commercial sectors, where elec- tricity, natural gas, and oil compete for space heating, water heating, air conditioning, and cooking. Sensitivity tests on the demand models show that a doubling of natural gas prices would increase residential demand for electricity by about 4 percent by the year 2000. For the commercial sector the com- parable increase in demand for electricity would be about 20 percent. Industrial de- mand for electricity is not very sensitive to fuel prices, so that roughly weighing the three sectors responses would indicate that a dou- bling of gas prices could increase electricity use by about 5 percent. The second reason that fuel prices are impor- tantis that they are highly uncertain. Reason- able assumptions could support a variety of forecasts, ranging from a collapse of fuel prices over the next ten years to another doubling of prices within 20 years. Thus, even though the impacts of a small change in fuel prices on demand for electricity may not be large, the possibility of large variations is great, making fuel prices an important ele- ment of uncertainty about future demand for During development of the 1983 Power Plan, there was considerable controversy over forecasts of fuel prices. Therefore, the Coun- Table 2-20 Growth Rates of Real Income per Capita Average Annual Percent PACIFIC UNITED NORTHWEST STATES Historical 1960-70 29 3.2 1970-80 27 22 Forecast 1980-2005 High 27 25 Medium Ae? 2.0 Low 0.9 1.6 cil contracted with Energy Analysis and Plan- ning, Inc., to develop a method of relating world oil market conditions to the retail prices of fuels in the Pacific Northwest. In addition, Energy Analysis and Planning, Inc., provided an evaluation of conditions in the world oil market and developed three illustrative world oil market scenarios. The insights into the world oil market, and into the regional fuel markets, provided by the Energy Analysis and Planning, Inc., report appear to be very good. Although the Council did not directly adopt the world oil price forecasts in the report, its ranges were informed by the contractor's report. The Council adopted the simple model devel- oped by Energy Analysis and Planning, Inc., to determine regional retail oil and natural gas prices from world oil price forecasts, and continued to use this model in developing the retail price projections for the 1986 Power Plan. The process for developing world oil price assumptions included four phases. The first was to read recent studies of the world oil market and evaluated current world oil prices. Second, an informal survey of world oil price forecasts by various groups was made. Third, the Council sent, with its work- ing paper on economic and demographic issues, a questionnaire that contained, among others, a question about world oil prices. These steps did not, of course, elimi- nate uncertainty about oil prices, but they did help clarify the extent of uncertainty and what would be reasonable forecast ranges. The fourth step was to revise the forecast in response to comments on the draft plan. The final forecasts of fuel price reflect a change in the pattern of the lower cases, but not the 2005 levels, to reflect an increased liklihood of short-term price reductions. Figure 2-4 shows the range of world oil price forecasts. Figure 2-4 also shows, for com- parison, the assumptions used for the 1983 Power Plan and actual world oil prices for 1980-83. The forecasts are all in 1985 dollars. The forecasts are generally lower than those in the 1983 Power Plan, reflecting recent his- tory and a changing understanding of the world oil market. The ability of oil producers to achieve ever growing prices for their oil is severely limited by market responses, both on the demand side and on the supply side. The questionnaire responses indicated that somewhat lower forecasts would be appro- priate, although the support for lower fore- casts was far from unanimous. Table 2-20 shows historical and forecast growth of real personal income per capita in the Pacific Northwest and for the U.S. During the 1960s, income per capita increased at a slightly slower rate in the region than in the U.S. In fact, the region's real income per cap- ita dipped below the U.S. in 1970. Income per capita increased faster in the region than in the U.S. during the 1970s. Over the entire 20- year period from 1960 to 1980, the region's per capita income increased at almost the identical rate as the U.S. average. From 1980 to 1983, real income per capita declined to $9,455 (in 1980 dollars) in the region, while it increased to $9,764 in the U.S. The forecasts for 1980 to 2005 are shown in Table 2-20 as well. 2-21 Chapter 2 1985 $ Per Barrel 90 75 60 45 30 15 0 1980 1985 1990 Year 1995 2000 2005 ==" 1986 Plan "**** 1983 Plan eeee Actual Figure 2-4 Forecasts of World Oil Prices Comparison of 1986 and 1983 Plan Assumptions Table 2-21 shows the world oil price assump- tions for selected years and selected growth rates. All of the forecasts assume continued weakness in the world oil market through 1986. The cases reflect varying degrees of price increase after 1986. In the lower part of the Table 2-21 are some forecasts by other organizations for com- parison. Of particular interest is the forecast range from the Canadian National Energy Board (CNEB). This range was derived from a survey of independent forecasts of 30 differ- ent organizations, predominantly energy companies. With the exception of the U.S. Department of Energy forecasts, all of the forecasts fall well within the Council's pro- posed range of assumptions. The Depart- ment of Energy forecasts seem to lie outside most views of the world oil market, and it is difficult to imagine sustaining such large price increases over a 20-year period. 2-22 The world oil price assumptions described above were used to forecast the retail prices of fuel oil and natural gas. The relationship between world oil prices and retail prices is embodied in the Energy Analysis and Plan- ning model. The model is very simple. Wholesale prices of residual oil and distillate oil are related to world oil price through a simple model of refinery economics. Retail price markups, based on historical data, are added to obtain retail oil prices for the resi- dential, commercial, and industrial sectors. Natural gas prices are determined in com- petition with residual oil in the industrial boiler market. It is assumed that interruptible natu- ral gas prices will equate to residual oil prices in this market. Markups are added to the industrial interruptible gas price to obtain retail prices for the other sectors. The assumption that natural gas prices will be determined by competition with residual oil in the industrial market was essentially imple- mented in the new Canadian natural gas export policy. The retail fuel price forecasts are shown in Tables 2-22 through 2-24. Table 2-24 includes forecasts for industrial coal prices. These forecasts reflect the fact that weaker oil prices tend to cause weak coal demands and prices. There was no formal model involved in forecasting the coal prices. It was assumed that in the absence of oil price increases, coal prices would maintain their current real levels. As oil prices rise in the various forecasts, it was assumed that coal prices would follow with a delay of several years. The larger the oil price increase, the larger the proportion of the oil price increase received by coal. The proportional increases ranged from 20 percent to 75 percent as the amount and duration of the oil price growth increased. Forecasts for Utility Service Areas The economic and demographic assump- tions are divided into public and investor- owned utility service areas to provide inputs to the demand forecasting system, which forecasts electricity consumption by utility type. Industrial production at the detailed industry level, employment in the commer- cial sector, and housing units are divided into public and investor-owned utility areas for each state. The splits between public and investor-owned utility areas are provided by Bonneville. In the case of major manufactur- ing industries, the shares of production allo- cated to public or investor-owned utilities were developed by detailed industry analysis of plant location or county employment pat- terns. The shares of commercial employ- ment and housing stock were allocated on the basis of customer counts in the residen- tial sector at the utility and county level. The split into investor-owned and public utility ser- vice areas is based on historical data. It is assumed that the shares for each individual component do not change over time. These allocations were updated from the 1983 plan assumptions to reflect more recent Bon- neville analysis. Although there were a number of small changes to specific industry shares, the overall impact was negligible. Chapter 2 Table 2-21 Table 2-23 World Oil Prices Commercial Sector Fuel Prices HIGH “atch “iow” Low HIGH Mon MiGwh Low Prices (1985 $ per barrel) Natural Gas (1985 $ per thousand cubic feet) 1980 40 40 40 40 Prices 1981 4 41 41 41 1980 6.27 6.27 6.27 6.27 1982 40 40 40 40 1985 671 6.61 6.39 5.95 1983 33 33 33 33 1990 6.29 5.69 469 3.10 1984 30 30 30 30 2000 10.07 7.48 6.09 4.69 1985 29 28 28 27 2005 13.05 8.48 6.49 5.29 1990 4 30 24 13 Growth Rates (% per year) 1995, 50 38 28 17 1980-1985 1.4 Ww 04 1.0 2000 60 42 33 24 1985-2005 3.4 1.3 0.1 0.6 2005 80 49 36 28 Oil (1985 $ per Galion) Growth Rates (% per year) Prices 1980-1985 62 69 69 76 1980 1.13 1.13, 1.13 1.13, 1983-2005 41 18 04 07 1985 97 96 93 87 1985-2005 5.4 28 1.3 0.0 1990 1.04 94 78 52 2005 Forecasts (1985 $ per barrel) 2000 1.65 1.23 1.00 78 CNEB* 54 42 30 2005 2.13 1.39 1.07 88 Bonneville 66 49 36 Growth Rates (% per year) Wharton, 40 1980-1985 -3.0 3.2 3.8 51 Oregon DOE 45 1985-2005 40 19 07 01 U.S. DOE 118 82 53 *Canadian National Energy Board, Canadian Energy: Supply and Demand 1983-2005, Sept. 1984, Table A2-1, page A-30. The range of CNEB forecasts came from a survey of 30 independent forecasts of various energy companies and other organizations. Table 2-22 Residential Sector Fuel Prices won Maa” MEW Low Natural Gas (1985 $ per thousand cubic feet) Prices 1980 6.77 6.77 677 6.77 1985 7.39 7.28 7.07 6.65 1990 6.97 6.37 5.38 3.79 2000 10.75 8.16 677 5.38 2005 13.73 9.15 747 5.97 Growth Rates (% per year) 1980-1985 18 15 09 0.4 1985-2005 3.4 14 0.0 21 Oil (1985 $ per Gallon) Prices 1980 1.31 1.31 1.31 1.31 1985 1.05 1.04 1.01 1990 1.11 1.01 86 60 2000 1.71 1.30 1.08 86 2005 2.18 1.46 1.14 95 Growth Rates (% per year) 1980-1985 43 45 51 62 1985-2005 37 17 06 0.0 2-23 Chapter 2 Table 2-24 Industrial Sector Fuel Prices nich = “Ha TWh Low Natural Gas (1985 $ per thousand cubic feet) Prices 1980 5.48 5.48 5.48 5.48 1985 5.65 5.54 5.34 4.91 1990 5.64 5.04 4.05 2.46 2000 9.42 6.83 5.44 4.05 2005 12.40 7.83 5.84 4.64 Growth Rates (% per year) 1980-1985 0.6 0.2 -0.5 “2.1 1985-2005 4.0 1.7 0.4 -0.3 Oil (1985 $ per Barrel) Prices 1980 38.84 38.84 38.84 38.84 1985 35.31 34.70 33.55, 31.14 1990 37.89 34.16 27.93 17.87 2000 61.41 45.33 36.65 27.98 2005 79.92 51.51 39.13 31.73 Growth Rates (% per year) 1980-1985 1.9 2.2 -2.9 -4.3 1985-2005 4.2 2.0 0.8 -0.1 Coal (1985 $ per ton) Prices 1980 50.58 50.58 50.58 50.58 1985 50.58 50.58 50.58 50.58 1990 50.58 50.58 50.58 50.58 2000 55.85 53.16 51.86 48.10 2005 63.19 55.88 54.51 49.32 Growth Rates (% per year) 1980-1985 0.0 0.0 0.0 0.0 1985-2005 ie 0.5 0.4 -0.1 1./ Wharton Econometric Forecasting Associ- ates, Long-term Alternative Scenarios and 20-year Extension, July 1984, Volume 2, Number 4. 2./ Northwest Power Planning Council, “Eco- nomic, Demographic and Fuel Price Assump- tions,” various drafts, December 6, 1984; March 26, 1985; and July 15, 1985. 3./ Aho, William O., letter to Ms. Barbara Pierce, Brookhaven National Laboratory, March 17, 1982. 2-24 4./ Northwest Pulp and Paper Association, Results of NWPPA/Ekono Survey, Heidi Schultz, April 2, 1982. 5./ Battelle Seattle Research Center, High Tech- nology Employment, Education and Training in Washington State, June 1984. 6./ Port of Seattle, 1982 Economic Impact Study, October 1984. 7./ Beyers, William B., Alvine, Michael J., and Johnsen, Erik G., The Service Economy: Export of Services in the Central Puget Sound Region, Central Puget Sound Eco- nomic Development District, April 1985. 8./ Energy Analysis and Planning, Inc., “Fuel Prices in the Northwest,” August 1982. APPENDIX 2-A DETAIL ON ECONOMIC INPUT ASSUMPTIONS Table 2-A-1 Table 2-A-3 Employment-Population Ratios Housing Additions by Type 1985 1990 «1995 2000 «= 2005 ‘STATE HIGH MEDIUM" LOW" WASHINGTON WASHINGTON High 397 433 465 481.498 Single Family (1-4 units) 75 62 49 Medium-high 397 416 440 460 470 Multifamily (5 and more) 16 22 Hv Medium-low 397 413 430 445.445 Manufactured Homes 9 16 20 Low 397 410 415 416 416 OREGON 1-4 OREGON Single Family (1-4 units) 76 65 54 High 392 434 468 484.496 Multifamily (5 and more) 13 19 27 Medium-high 392 416 440 460 475 Manufactured Homes an 16 19 Medium-low 392 413 428 450.455 IDAHO Low 392 410 416 418 418 Stagle Femnly (1-4 ue) eo a bed Dato Multifamily (5 and more) 8 13 18 High 366 410 442 467 500 Manufactured Homes W 16 22 Medium-high 366 395 422 445 460 WESTERN MONTANA Medium-low 366 384 408 425.435 Singh Famiy (1-4 uta) =" 2 60 Low 366 381 400 410 420 Mubiternily (6 and more) 02 10 15 WESTERN MONTANA Manufactured Homes 14 20 25 High 341 365 390 410.425 er z ium case shown here was used in the medium-high io. The low case was used i Medium-high 341 360 375 390 .400 ae, pen iiten @eneners te yeees) Medium-low 341 355 365 370 375 Low 341 350 355 360 360 Table 2-4-4 Production per Employee by Industry” PACIFIC NORTHWEST. Average Annual Rates of Change (%) High 391 429 462 479.496 7965-2006 Medium-high 391 411 436 456.469 a aan a oe Medium-low 391 408 425 443.445 = a or ) Low 391 405 412 415 416 22 34 27 20 23 33 3.0 23 Table 2-A-2 25 23 19 1 Persons per Household 27 29 25 18 1980 «1985 = 1990 1995 2000 2005 29 25 22 14 30 3.2 28 24 WASHINGTON High 260 250 240 230 2.20 32 27 24 15 Medium’ 268 260 252 244 236 236 33xx 32 29 24 Low 260 260 258 258 258 34 25 22 14 OREGON 35 37 34 26 High 259 250 240 230 220 38 a7 as - Medium 266 259 252 245 238 2.38 x 28 aa 17 Low 259 258 257 256 256 28 32 “ es (AHO 39 29 26 18 High 283 266 252 245 240 : 2421 16 15 14 Medium 291 283 275 267 259 259 2436 16 15 14 Low 2.83 282 281 280 280 24xx 16 15 14 WESTERN MONTANA High 267 248 234 224 2.24 2611 35 32 24 Medium 275 267 258 248 239 239 2621 35 32 24 Low 267 266 264 262 262 2631 35 32 24 PACIFIC NORTHWEST 26XX 35 32 24 High 262 252 241 231 222 2812 34 - 22 Medium 270 262 255 247 239 239 2819 20 ‘0 0 Low 2.62 260 260 260 260 28xx aa 0 be 3334 25 25 25 * The medium case shown here is used in the medium-high and medium-low. * Please refer to Table 2-B-1 in Appendix 2-B for a listing of SIC Codes. 2-A-1 APPENDIX 2-B SIC CODE LISTINGS Table 2-B-1 Code Listings CODE INDUSTRY CODE INDUSTRY 20 Food and kindred products 3334 Primary aluminum 22 Textiles 40-49 Transportation & public utilities 23 Apparel 50-51 Wholesale trade 25 Furniture 52-53 Retail trade except food stores (54) 55-57, 59 and eating places (58) 27 Printing and publishing 54 Food stores 29 Petroleum refining 58 Eating and drinking places 30 Rubber and plastics 60-67 Finance, insurance & real estate 31 Leather and leather products 70 Hotels and lodging 32 Stone, clay, glass & concrete 72 Personal services 33XX Primary metals except aluminum 73 Business services 34 Fabricated metals 75 Automotive repair & garages 35 Machinery except electrical 76 Miscellaneous repair services 36 Electrical machinery 78 Motion pictures 37 Transportation equipment 79 Amusement and recreation services 38 Professional instruments 80 Health services 39 Miscellaneous manufacturing 81 Legal services 2421 Sawmills & planing mills 82, 941 Educational services 2436 Softwood veneer & plywood 83 Social services 24XX Other lumber & wood products 84 Museums, art galleries 2611 Pulp mills 86 Membership organizations 2621 Paper mills 89 Miscellaneous services 2631 Paperboard mills 90-99 Government except education (941) 26XX Other paper products 2812 Alkalies and chlorine 2819 Elemental phosphorus 28XX Other chemicals 2-B-1 B. Chapter 3 Forecast of Demand for Electricity Introduction Forecasts of the demand for electricity in the Pacific Northwest region are required by the Northwest Power Planning and Conservation Act. Demand forecasts play three important roles in the Council's power planning pro- cess. The first is the traditional role; they are the basis for deciding how much electricity is needed to support a healthy and growing economy. The second role is to explore and define the uncertainty surrounding future electrical resource needs. Finally, the demand forecasts are an essential compo- nent of conservation assessment. Conserva- tion, identified as the priority resource in the Act, is directly related to the demand for elec- tricity. Demand forecasts are needed to esti- mate conservation potential, but, in addition, the forecasting models help determine the effects of conservation actions taken as part of the Council's power plan. The Council has developed the best avail- able forecasting tools in its demand forecast- ing system. This system helps the Council determine how assumptions about the growth of the region's economy and energy prices affect the demand for electricity. Figure 3-1 illustrates the general structure of the forecasting system. The growth of the regional economy and changes in its com- position are the key factors affecting growth in demand for electricity. Assumptions about the prices of fossil fuels and electricity, how- ever, modify the effects of economic condi- tions. The Council's forecasting system cap- tures these relationships in considerable detail. The Council developed three preliminary forecasts of demand for electricity, which underwent public review and revision before draft forecasts were adopted by the Council. The first preliminary forecast was described in a staff issue paper dated February 13, 1985, and was presented to the Council in Boise on February 21, 1985. The second pre- liminary forecast was presented to the Coun- cil in Missoula on April 4, 1985, and was described in a staff report dated March 7, 1985. On April 24, 1985, in Seattle, the Coun- cil adopted the third preliminary forecast for purposes of resource portfolio analysis. Economic and Demographic Forecasts Fuel Price Forecasts Conservation Programs and Costs Electric Generating Resources and Costs Residential Commercial Demand Demand Resource Portfolio Figure 3-1 Northwest Power Planning Council Demand Forecast System Demand Determinants Industrial Demand Total Demand Supply Demand Balance Irrigation Demand Electric Price 3-1 Chapter 3 Average Megawatts 1950 1960 1970 1980 Figure 3-2 Sales of Electricity—Historical and Forecast These demand forecasts were revised slightly to reflect the effects of the proposed resource portfolio on electricity prices, and were presented in the draft plan adopted August 7, 1985. The forecasts described here reflect com- ments received on the draft plan, the adop- tion of new building codes by the states of Washington and Oregon, and correction of minor errors discovered in the draft plan forecasts. This chapter expands the discussion of demand forecasts that is included in Volume |, Chapter 4, of the 1986 Power Plan. Some of the material in that chapter is repeated here so this chapter can be read without referring back to Volume |. Following a summary of the forecast results and methods, each consum- ing sector is discussed. Each of these sec- tions describes the forecasting methods and results in detail. Following the sections on the consuming sectors is a discussion of fore- casts of retail electric rates. The last section 3-2 describes the role of the forecasts in resource planning. Additional detail about the forecast- ing methods and results is available from the Council. Overview The Council's forecast of demand for elec- tricity consists of a range of four forecasts; a low, medium-low, medium-high, and high forecast. The Council's high demand forecast is designed to ensure that power supplies never constrain the regional economy's growth potential. The high forecast reflects the effects of record high regional economic growth relative to the nation combined with less competitive prices for alternative fuels. The likelihood that such a rapid regional growth would occur is considered to be very small. The Council’s forecast range is bounded on the low side by a forecast whose pessimism about the regional economy is roughly proportional to the optimism of the high case. Inside the bounds of the low and high fore- casts is a smaller, most probable range of demands bounded by the medium-low and medium-high forecasts. These two medium forecasts will carry a greater weight in the planning of resources than will the high and low extremes. Nevertheless, the possibilities posed by the high growth forecasts must be addressed by appropriate resource options. Similarly, conditions that are implied by the low demand forecast will be considered within a flexible planning strategy designed to minimize regional electricity costs and risks. The demand forecast ranges are constructed by combining economic assumptions, fuel price assumptions, and some modeling assumptions. The combination of assump- tions is designed to explore a wide range of possible demands without combining assumptions unrealistically. That is, mutually inconsistent assumptions are not combined just to obtain extreme forecasts. In the high forecast, for example, the high economic assumptions are combined with high fuel price assumptions. In addition, it was assumed in the industrial sector and irriga- tion sectors that consumers have relatively low price response, and in the residential sector it was assumed that consumers were less likely to invest in energy efficiency improvements. Electricity prices, which have a significant effect on demand, are not assumed but are determined by the elec- tricity pricing model based on the amount and cost of resources needed to meet demand. Generally, electric prices will be higher for higher demand growth unless dif- ferent policy assumptions are used in the forecast scenarios. This was only done in the low forecast, where it was assumed that costs associated with Washington Public Power Supply System's Nuclear Projects 4 and 5 (WNP-4 and WNP-5) would have to be paid by electric consumers. In the other cases, it was assumed that those costs would not be reflected in electric rates. Chapter 3 In 1983, firm sales of electricity to the final consumer in the Pacific Northwest totaled 14,593 average megawatts, or 127.8 billion kilowatt-hours. The high forecast shows this demand could grow to 26,101 average mega- watts by 2005, an increase in electricity requirements equivalent to the power from 15 nuclear plants the size of WNP-2 at Hanford, Washington. Under the set of assumptions leading to the low forecast, demand only increases to 15,121 average megawatts, an amount little changed from current require- ments. Figure 3-2 illustrates the forecast range in the context of historical sales of electricity. This large uncertainty about future needs for electricity resources represents an important challenge for energy planning. The region needs to deal with this uncertainty in a manner that will neither prevent the region from attaining rapid growth, nor impose large and unnecessary costs should slower growth occur. Table 3-1 summarizes the demand forecasts. Before these forecasts are addressed further, however, their nature should be clarified. The basic concept presented in the Council's demand forecast is a “price effects’ forecast. The forecast indicates what demand would be if consumers responded to prices but if no new conservation programs were imple- mented. Two alternative concepts will be dis- cussed in the final section of this chapter. Table 3-1 shows that the rate of growth of demand could be as high as 2.7 percent per year, if the high case materialized, or as low as 0.2 percent. A more likely outcome, how- ever, is between the medium-low growth rate of 1.2 percent and the medium-high rate of 1.8 percent. Figure 3-3 compares the projected growth rates of demand to growth rates experienced in the region since 1950. Between 1950 and 1970, demand for electricity grew by an aver- age of 7.4 percent each year. During the 1970s, demand grew much more slowly, at about 3.7 percent per year. The forecasts show a continued decline in the rate of growth, even in the high forecast, over the next 20 years. Table 3-1 Firm Sales of Electricity (Average Megawatts) GROWTH RATE ACTUAL FORECASTS (% per year) 1983 1990 2000 2005 1983-2005 High 14,593 8,044 23,026 26,101 ae Medium-high 16,701 20,022 21,687 1.8 Medium-low 15,351 17,538 18,950 1.2 Low 13,697 14,370 15,121 0.2 Growth Rate 1950-60 1960-70 1970-80 High Medium Medium Low High Low Figure 3-3 Historical and Forecast Growth Decreasing growth rates of demand for elec- tricity, historically and in the forecasts, are a result of many factors. These factors include the rate of growth of the economy, changing standards of living, the price of energy rela- tive to other goods and services, and the changing mix of economic activity, both in the nation and in the region. However, the use of electricity is much different in the Pacific Northwest than in the rest of the nation. This difference is illustrated with use of electricity per person in Table 3-2. Although the historical patterns of growth in use of electricity are similar in the region and the nation, there is a striking difference in the amount of electricity used. The Pacific North- west uses nearly twice as much electricity per person as the nation as a whole. This pattern is due primarily to large supplies of low-cost hydroelectric power in this region. Recent large increases in the Northwest price of electricity, however, have changed the outlook for electricity demand. The fore- casts show that, while per capita use will Chapter 3 Table 3-2 Per Capita Use of Electricity (Kilowatt-Hours per Person) NORTHWEST STATES History 1960 8,930 3,810 1970 14,790 6,800 1980 17,230 9,230 1982 16,330 9,000 Forecast, 2005 High 18,366 12,310 Medium-high 16,689 12,310 Medium-low 16,588 12,310 Low 15,077 12,310 Table 3-3 Firm Sales Forecast for Public and Investor-owned Utilities (Average Megawatts) INVESTOR- SALES SALES Non BSE OS total Actual 1983 14,593 6,854 5,843, 1,896 7,739 Forecast 2005 High 26,101 13,300 10,324 2,477 12,801 Medium-high 21,687 10,896 8,645 2,146 10,791 Medium-low 18,950 9,516 7,692 1,742 9,434 Low 15,121 7,574 6,323 1,224 7,547 Growth rates, 1983-2005 High 2.7 3.1 2.6 1.2 2.3 Medium-high 1.8 2.1 1.8 0.6 1.5 Medium-low 1.2 1.5 1.3 -0.4 0.9 Low 0.2 0.5 0.4 -2.0 -0.1 remain well above national levels, growth in use per person will be slower than historically and could actually decline in a low forecast. Figure 3-4 illustrates historical and forecast patterns of electricity use per person. The summary of forecast results that is usu- ally presented in Council issue papers and reports hides the fact that the forecasts are done in great detail. This chapter presents more of that detail than has been presented in Council issue papers or in the 1983 Power 3-4 Plan. A major dimension of the demand anal- ysis system is the separate forecasting of residential, commercial, industrial, and irriga- tion uses of electricity. A second major dimension is the separate treatment of demand by customers of public utilities and customers of investor-owned utilities. Each component of demand, e.g., residential use of electricity in investor-owned utility service areas, is analyzed in many more dimensions within the sector forecasting models. Those additional levels of detail are discussed in the sections of this chapter dealing with each sector. The sectoral and utility ownership dimensions are characterized briefly below. In 1983, total regional firm sales of electricity were 14,593 average megawatts. Investor- owned utilities marketed 6,854 average megawatts or 47 percent of the total. Public utilities and the Bonneville Power Administra- tion marketed 53 percent of the firm sales. Table 3-3 shows the 1983 composition of firm sales and the four forecasts for 2005. In all of the forecasts, the investor-owned utility share of firm sales increases slightly. Separate forecasts are done for investor- owned and public utility service areas by run- ning the demand forecasting models inde- pendently for those groups of consumers. The economic assumptions driving the fore- casts are also done separately, as described in Chapter 2 of this volume. These economic assumptions, combined with differences in electric rates and existing conditions, lead to differences in the forecasts for the two cus- tomer groups. Table 3-3 shows the public utility and Bon- neville Power Administration sales sepa- rately for direct service industries (mostly alu- minum companies) and all other customer components. Direct service industries accounted for a third of Bonneville/public sales in 1983, but are forecast to increase only moderately in the high cases, and decrease in the low forecasts. Thus, the direct service industry forecast is an impor- tant reason for lower growth in the Bonneville/ public sales than in private sales. However, the other Bonneville/public sales are also shown growing somewhat more slowly than investor-owned utility sales. Figure 3-5 shows the composition by sector of the 1983 electricity sales in the region. The industrial, residential, and commercial sec- tors account for most of the region's electricity demand. Each of the demand sectors is dis- cussed in some detail in the sections that follow. Residential Demand The residential sector accounted for 36 per- cent of regional firm sales of electricity in 1983. Residential sector demand is influ- Chapter 3 enced by many social and economic factors, including fuel prices, per capita income, and the choices in efficiency of energy-consum- ing equipment available to consumers (avail- able technology). The most important factor, however, is the number of households. The structure of the residential sector demand model reflects this importance by using the individual household as the basic unit. The model simulates future demand for electricity by projecting future growth in households; their choice of housing type; the amount of electricity-using equipment the average household owns; choices of fuel for space heating, water heating, and cooking; the level of energy efficiency chosen; and the energy- using behavior of the household. These choices are influenced in the model by energy prices, equipment costs, per capita incomes, and available technology. The use of electricity is simulated for each of eight use classifications. Figure 3-6 shows estimated historical shares of these uses in total resi- dential use of electricity for 1983. The projections of residential demand for electricity cover a wide range. This range results partly from variation in projections of the number of households, per capita income and fuel prices from the economic and demographic growth scenarios. Pro- jected demand also varies because of differ- ent assumptions regarding consumers effi- ciency choice behavior (implicit discount rates; see Table 4-7 in this volume). In the absence of new conservation pro- grams, projected residential demand increases from 5,216 average megawatts in 1983 to a range which spans from 9,920 average megawatts in the high growth fore- cast to 5,825 average megawatts in the low growth forecast in 2005. As shown in Table 3-4, the average demand growth rate ranges from a low of 0.5 percent per year to a high of 3.0 percent. The Council's residential model of energy demand is the descendant of the computer model originally developed at Oak Ridge National Laboratory in 1978. Since that time, the model has been used in a wide variety of applications for the U.S. Department of Energy, state agencies and utilities. It has also incorporated improvements in logic and data to the extent that the current model is several generations removed from the original. KWH Per Person 20,000 15,000 10,000 5,000 | _ | | | : | | 1960 1970 1980 1982 2005 2005 2005 2005 High Medium Medium Low High Low Pacific Northwest @l™ United States . Figure 3-4 Electricity Use per Capita Irrigation Other 0.8% Commercial Industry Figure 3-5 1983 Firm Sales Shares 3-5 Chapter 3 Table 3-4 Residential Sector Electricity Demand (Average Megawatts) GROWTH RATE ACTUAL FORECASTS (% per year) 1983 1990 2000 2005 1983-2005 High 5,216 6,628 8,613 9,920 3.0 Medium-high 6,273 7,549 8,128 2.0 Medium-low 5,769 6,726 7,720 1.5 Low 5,206 5,535 5,825, 0.5 Refrigeration Space Heat Air Conditioning 0.8% Cooking Freezers Water Heat Figure 3-6 1983 Residential Use by Application The model is best described as a hybrid of engineering and econometric approaches. It is based on the fundamental idea that resi- dential energy is used by equipment such as furnaces, refrigerators and water heaters to provide amenities to the occupants of resi- dences. Residential energy use, as simu- lated by the model, is a function of: 1. The total number of residences. The projections for future years are taken from the economic and demographics projections. 3-6 2. The number of energy-using appli- ances in the average residence. Each year's appliance penetrations, or pur- chases of appliances per household, are simulated based on econometric analysis of historic sales patterns. Penetrations are influenced by equipment and energy costs and by per capita incomes. 3. The efficiencies of these appliances. Efficiency choice by consumers is simu- lated based on engineering analysis of costs of appliances of varying efficiencies and on econometric analysis of observed efficiency choices in the past. Efficiency choices are influenced by energy prices, the cost of more efficient appliances and the inclination of consumers to invest in conservation (represented by their implicit discount rates, described in Chapter 4 of this volume). Efficiency choices can also be constrained (e.g., thermal integrity choices will be no worse than some spec- ified level), which provides the means of representing conservation programs, such as the model conservation stan- dards, whose objectives are to modify consumers choices of efficiency. 4. The fuels used by these appliances. While some appliances such as air condi- tioners use electricity exclusively, others such as water heaters can use any of sev- eral fuels. Fuel choice is simulated based on the model’s simulated efficiency choices and econometric analysis of fuel choice behavior that has been observed in the past. Fuel choices are influenced by relative fuel prices, equipment prices, and relative efficiencies of the appliances using the various fuels. 5. The intensity of use of these appli- ances. Intensity of use is varied by such means as thermostat settings, reduced use of hot water for washing clothes, and the like. Variation in intensity of use is based on econometric analysis of observed short run response to fuel prices. Intensity of use is determined in the model by fuel costs, appliance efficien- cies, and per capita incomes. Since the adoption of the Council's 1983 plan, Council staff have worked on several projects to improve the performance and credibility of the residential demand model. These pro- jects fall into three categories: 1) comparison of the model's performance with that of other models that might be used for the Council's forecasting work; 2) development of the logic and structure of the model to eliminate recog- nized shortcomings; and 3) incorporation of improved data. Households (millions) Electricity Prices (1985 cents/kWh) Efficiency Measures: Thermal Integrity (New electrically heated single family, efficiency relative to regional 1979 stock) Refrigerators (New, efficiency relative to 1979 stock) Saturations: Electric Space Heat (% of homes with electric heat) Electric Hot Water (% of homes with electric hot water) Utilization Intensity (Relative to 1979) (Electrically space heated homes) kWh per Household (All homes) Space Heat kWh per Household (Electrically heated homes) Non-space-heat kWh per Household (All homes) Space Heat Sales (MW) Total Sales (MW) In the first category, the Council's model was compared to two alternatives. The first of these was the Residential End-use Energy Planning System (REEPS), developed by Cambridge Systematics, Inc. with financial support by the Electric Power Research Institute (EPRI). The objective in comparing the projections of the two models was not only to see by how many average megawatts they differed in some future year, but also to become more familiar with REEPS, its strengths and weaknesses, its ease of use, and its suitability for analysis of policy ques- tions important to the Council. The comparison required the adaptation of REEPS base-year data to Bonneville's ser- vice area, which Cambridge Systematics accomplished under contract from the Coun- cil. It also required the translation of eco- nomic and demographic projections used in Chapter 3 Table 3-5 Residential Sector Summary Indicators 2005 HIGH MEDIUM- MEDIUM- LOW 1980 HIGH LOW 2.964 5.606 4.761 4.185 3.381 Public 2.0 43 3.6 3.4 27 10U 3.7 49 43 3.6 3.3 Public 1.17 1.64 1.60 1.58 1.61 lou 1.26 1.63 1.59 1.57 1.60 Public 1.00 1.20 1.50 1.26 1OU 1.17 1.26 1.61 1.33 Public 59 57 55 54 49 ou 37 46 a4 38 33 Public 90 80 80 79 76 10U 81 83 78 77 75 Public 98 85 87 93 .93 lou 98 1.00 1.02 1.08 1.09 16,181 15,501 14,953 15,281 15,090 10,283 8,316 8,912 9,258 9,506 11,441 11,287 10,773 11,094 11,356 1,604 2,697 2,272 1,970 1,441 5,475 9,920 8,127 7,270 5,824 the 1983 plan into the appropriate form for use by REEPS, which was done by Council staff with assistance from Cambridge Systematics. The projections of REEPS and the Council's model were compared over a substantial range of assumptions. High and medium-low economic and demographic assumptions from the 1983 plan were used, and for each of these economic scenarios, energy use, with and without the model conservation stan- dards, was projected. As a result of these comparisons} some logical flaws in REEPS’ structure were discovered. Some of these flaws were remedied quickly, but one in par- ticular, the inability of REEPS to simulate the effect of model conservation standards on space heating fuel choice, was added to the longer run development agenda. The model- ing of model conservation standards is very important to the Council; REEPS’ problem in this area, in the absence of compelling advantages in other areas, left the Council's existing model preferred for the 1986 Power Plan. The Council's model was also compared to a version of the Oak Ridge National Laboratory model used by Bonneville (the Residential Reference House Energy Demand Model, or RRHED) in their long run forecasting work. The RRHED model was developed from the same ancestor as the Council's model and differs most importantly in the area of fuel choice. The RRHED model simulates fuel choice based on data gathered in the Pacific Northwest specifically for this purpose. These data were analyzed by researchers at the Massachusetts Institute of Technology and at Oak Ridge National Laboratory under contract from Bonneville. The resulting fuel 3-7 Chapter 3 Multifamil 2005 High 2005 Medium High 2005 Medium Low ‘005 ow 1980 2005 High 2005 Medium High 2005 Medium Low 2005 Low 1980 0.0 0.5 1.0 15 Relative To 1980 Stock Figure 3-7 Average Size of Electrically Heated Housing Units Table 3-6 Share of Housing Stock by Building Type 1980-2005 (%) 2005 MEDIUM- MEDIUN- 1980 HIGH HIGH LOW Low Single Family 78.3 77.4 72.0 68.4 72.4 Multifamily 14.2 14.3 t7at 19.9 17.9 Manufactured Homes 75 8.3 10.9 ted, 9.7 Table 3-7 Commercial Sector Electricity Demand (Average Megawatts) GROWTH RATE ACTUAL FORECASTS (% per year) 1983 1990 2000 2005 1983-2005 High 2,936 3,654 5,108 5,946 3.3 Medium-high 3,267 4,192 4,651 21 Medium-low 2,958 3,483 3,848 2 Low 2,727 2,579 2,773 -0.3 3-8 choice submodel should be the most credi- ble available for the Pacific Northwest. The adoption of the RRHED model for use by the Council would also have the advantage of making it easier to understand differences which might appear between Bonneville and Council forecasts. The comparison between the Council's model and Bonneville's followed the general procedure described above for the REEPS comparison. Input data from the Council's 1983 plan high and medium-low forecasts were translated into equivalent inputs for the RRHED model. Projections were compared for both these economic scenarios, with model conservation standards and without them. The process and results are described in detail in a Council staff working paper? Interestingly, the fuel choices simulated by the two models were very close. The dif- ferences observed in other components of the models did not indicate any clear superi- ority of the RRHED model. It became clear that the adoption of the RRHED model would have some disadvantages, however. For example, the RRHED model required about five times as much computer time to run, which would not be a prohibitive disadvan- tage but which would roughly double the run time of the entire forecasting system. Also, the adoption of the RRHED model would require that it be modified to include a number of important features already in the Council's model. Since the Council's model mimicked the most desirable component of the RRHED model quite closely, there seemed to be no compelling reason to accept the costs of changing models. In addition to comparisons between the Council residential energy demand model and others, considerable effort has been spent since 1983 on further development of the model. Perhaps the most significant example of this work is the modification of the model to include the interaction of appli- ance efficiencies and space conditioning requirements. Chapter 3 Since much of the energy used by ap- pliances such as refrigerators and stoves is converted to heat in the living space, that energy decreases space heating require- ments and increases air conditioning require- ments. Current space heating and air condi- tioning energy use implicitly reflects this fact. It follows, in the absence of other changes, that future space heating and air conditioning use will change if appliance efficiencies improve, since less appliance energy use will result in less heat in the living space. The Council's model did not take this interaction into account in 1983 (nor do REEPS or the RRHED model currently). As a result, the model's projections of total demand for energy and the estimated effects of model conservation standards were both biased downward, and the estimated effects of pro- grams to improve appliance efficiency were biased upward. Based on work by Palmiter and Kennedy? Council staff enhanced the Council's model to include future appliance efficiencies in the simulation of future space conditioning demands. This change had modest effects on total demand for electricity4 but estimates of effects of conservation programs were affected more significantly. Generally, the modified model makes appliance efficiency programs appear less attractive and space heating conservation programs look more attractive. The final category of testing and develop- ment of the residential model was the devel- opment of better input data for the model. These data include updated cost and per- formance data for thermal integrity, updated cost and performance data on efficient refrigerators and freezers, more recent data on the incidence of wood heating, data on retirement rates for appliances, and others. The effects of these individual data changes on projections of demand are varied; the net effect of all of them together is modest. Table 3-5 provides a summary of historical and projected values of some of the compo- nents that determine total demand for elec- tricity in both public and investor-owned utility (lIOU) areas. Although total residential use of electricity varies widely across the four growth forecasts, use per household shows much less variation. The table shows use per Year 2005 High 2005 Medium-high 2005 Medium-low 2005 Low 1980 0.0 0.5 Relative to 1980 Stock 1.0 1.5 Figure 3-8 Thermal Efficiency of Electrically Heated Single Family Houses household for 2005 for the four growth fore- casts, as well as historical use in 1980. Use per household decreases in all forecasts. The fairly narrow range of per household use projections for 2005 means that the variation in total residential demand projections is pri- marily due to variation in the projections of numbers of households. Use per household is the net result of changes in variables such as efficiency, housing type, housing size, and fuel choice. The changes in some of these individual components are substantial, but there is a tendency for them to offset one another in their effects on use per household. For exam- ple, efficiencies generally improve, tending to reduce use per household, while the sizes of multifamily units and mobile homes are pro- jected to increase, increasing the per house- hold energy requirements for space condi- tioning. These patterns are illustrated in Figures 3-7 and 3-8. Figure 3-7 shows the projected increases in the average size of manufactured homes and multifamily hous- ing units, ranging from 11 to 21 percent. Fig- ure 3-8 shows that the average thermal effi- ciency of electrically heated single family houses improves by between 15 and 30 per- cent in the various growth forecasts. The thermal integrity of new houses (shown in Table 3-5) improves significantly from 1980 practices. This is mainly due to more strin- gent building codes adopted in Washington and Oregon in 1985 taking effect in 1986. The greater thermal integrity of houses built after 1985 raises the average thermal integrity in 2005; the higher growth scenarios have more new houses, and higher average thermal integrity, as shown in Figure 3-8. Housing type and fuel choice also influence per household energy use. The general trend in housing type projections is a reduc- tion in the total share of homes which are single family houses and an increase in the shares of multifamily units and manufactured homes. Table 3-6 shows the 1980 historical shares of the three building types, along with the projected 2005 shares for each of the forecasts. The effect of this trend is to decrease average use per household, since multifamily units and manufactured homes are smaller and require less energy to heat and cool. 3-9 Chapter 3 Ventilation Water Heat Lighting Air Conditioning 0. Figure 3-9 1983 Commercial Sector Use by Application Refrigerator Cooking 1% Space Heat 3-10 Key Industries + Paper + Lumber + Chemicals + Food + Non-DSI + Primary Metals Other 8% Figure 3-10 Industry Demands Direct Service Industry Fuel choice projections have mixed effects on per household energy use. As shown in Table 3-5, the share of households with elec- tric water heating is expected to decrease in all forecasts, but the share with electric space heating shows no clear trend. Space and water heating saturations are influenced by electricity prices, per capita incomes and the share of recently constructed houses in the stock. In addition, they are influenced heavily by the relationship of. electricity prices to those of competing fuels such as natural gas and oil. As will be described in the section of electric prices, the higher growth scenarios have higher electricity prices, but relatively lower prices of electricity compared to com- peting fuels. This pattern helps to explain the higher saturation of electrical space and water heating in the higher growth scenarios. When all the conflicting influences just described are combined, the net effect is the observed pattern of relatively small changes in per household use. This projection of electrical equipment use is based on demand for electricity before taking into account the Council's proposed conser- vation programs. The effects of these pro- grams cause sales of electricity to grow at slower rates. In addition, the use of electricity per household would decline because of the increased thermal efficiency of buildings and improved appliance efficiencies. The effects of these efficiency increases would be some- what diminished, however, by the greater use of energy services due to cost savings from improved efficiency in space and water heating. Commercial Demand Commercial demand for electricity ac- counted for 20 percent of firm sales of electricity in 1983. Commercial sector elec- tricity demand, like that of the residential sec- tor, is influenced by many factors, such as fuel prices and available technology. In par- ticular, one fundamentally important factor used as a basis for energy use projections is the total floorspace of the buildings in the commercial sector. The commercial sector demand model projects the amount of com- mercial floorspace and then predicts fuel choice, efficiency choice, and the use of the energy-consuming equipment necessary to service this floorspace. These choices are based on investment factors, fuel prices, and Chapter 3 Commercial Sector Summary Indicators Table 3-8 2005 MEDIUM- MEDIUM- HIGH LOw 1980 HIGH LOW Floorspace (million sq ft.) 1,283 3,032 2,322 1,903 1,544 Electricity Prices (1985 cents/kWh) Public 2.0 41 3.4 29 2.5 10U 3.9 6.6 5.8 47 46 Efficiency Measures Electric Space Heat (Average office building, efficiency Public 1.03 1.67 1.59 1.47 1.32 relative to 1979) 10U 1.04 2.22 1.89 1.67 1.49 Lighting (Average office building, efficiency relative to 1979) Public 1.00 1.09 1.08 1.05 1.06 1OoU 1.01 1.11 peal 1.10 1.11 Saturation of Electric Space Heat (%) Public 46 94 89 80 60 10U 42 75 64 50 23 Utilization Intensity Electric Space Heat (Average office building, relative to 1979) Public 0.97 0.94 0.96 0.98 0.98 10U 0.97 1.00 1.00 1.02 1.00 Lighting (Average office building, relative to 1979) Public 0.96 0.87 0.90 0.93 0.95 lou 0.96 0.90 0.92 0.96 0.96 Sales—per Square Foot Floorspace (of those heated by electricity) Space Heat (kWh/year) 14.7 8.5 9.5 10.7 12.2 Lighting (kWh/year) 63 53 5.5 5.7 5.8 Total (kWh/year) 18.9 17.2 17.5 lez: 15.7 Space Heat Sales (MW) 947 2,439 1,873 1,454 819 Lighting Sales (MW) 923 1,841 1,451 1,249 1,021 Total Sales (MW) 2,768 5,946 4,651 3,848 2,773 available technology. Energy use projections are made separately for different building types, applications, and fuel types. Shares of historical commercial sector demand for electricity for various applications are shown in Figure 3-9. Since 1983, development of the Council's commercial sector energy demand model has been extensive. During that period, the commercial model used by Bonneville for their long range forecasting was modified by Synergic Resources, Inc. (SRC). Because the Bonneville mode! shares many compo- nents with the Council's model, the Council hired Jerry Jackson and Associates (JJA) to assist the Council staff in evaluating the desirability of adopting all or part of the SRC modifications for the Council model. JJA also incorporated new information on fuel and effi- ciency choice into the Council's model. The results of this work are described in detail in JJA's contractor's report To summarize, the basic structure of the Council model was retained, but much of the data developed by SRC from Pacific Northwest sources on com- mercial floorspace and energy use was adopted. In addition, information on space heating fuel choices in recent construction was used to recalibrate the fuel choice com- ponent of the model; this change significantly increased the amount of electric space heat- ing projected by the model. Finally, the model's characterization of consumer choices for lighting was completely restruc- tured to reflect new lighting technologies available and an improved simulation of con- sumer decision making. Projections of commercial demand for elec- tricity vary widely. In the low growth forecast, commercial demand for electricity decreases from 2,936 megawatts in 1983 to 2,773 megawatts by 2005. In the high growth fore- cast, it reaches 5,946 megawatts. As shown in Table 3-7, the average rate of growth of demand ranges from — 0.3 to 3.3 percent per year. Table 3-8 shows some of the components underlying these totals. Floorspace increases in all forecasts, as a result of increased employment in the commercial sector, and is the major driver of growth in demand for electricity. Use of electricity per square foot of floorspace decreases in all growth forecasts. The decrease in use per square foot from 1980 to 2005 is modest for all forecasts, ranging from 6 percent in the medium-low growth forecast to 17 percent in the low growth forecast. Chapter 3 Table 3-9 Industrial Sector (Average Megawatts) GROWTH RATE ACTUAL FORECASTS (% per year) 1983 1990 2000 2005 1983-2005 High 5,659 6,907 8,348 9,219 2.2 Medium-high 6,342 7,392 7,992 1.6 Medium-low 5,828 6,470 6,956 0.9 Low 5,006 5,417 5,655 0.0 The relatively small projected changes in energy use per square foot are the net result of changes in various components of the forecast that are significant but that tend to offset one another. For example, the fraction of commercial floorspace heated by elec- tricity is projected to increase in nearly all forecasts, with greater increases occurring in the higher-growth forecasts. This would tend to increase the use of electricity per square foot except for the offsetting changes in build- ing and equipment efficiency. The changes in equipment efficiency are also demonstrated by Table 3-8. Compared to 1980, the 2005 efficiency of space heating in offices improves by percentages that range from 28 percent in public utility service areas for the low forecast, to 113 percent in investor-owned utility service areas in the high forecast. These efficiency improve- ments are equivalent to reductions in use of 22 percent and 53 percent, respectively. These improvements are substantial but not unreasonable; estimates of space heating use in offices designed according to stan- dard 90-80E of the American Society of Heating, Refrigerating, and Air-conditioning Engineers (ASHRAE) are 59 percent lower than average 1980 levels. Smaller improve- ments in lighting efficiency are projected. These projections do not take into account the conservation programs included in this plan, but are based on existing building codes and market response to increased energy prices. The Council's programs will reduce overall demand for electricity, reduce demand per square foot, and improve equip- ment efficiency. Conservation savings esti- mated in the Council's conservation analysis will be reduced by increases in the intensity 3-12 of electricity use, because the programs will decrease operating costs, making the use of electricity more attractive. Industrial Demand The industrial sector is the largest of the four consuming sectors. In 1983 the industrial sector consumed 5,659 average megawatts of firm power, accounting for 39 percent of total firm demand in the region. In addition, the direct service industrial (DSI) customers of Bonneville consume varying amounts of nonfirm electrical energy, depending on eco- nomic and hydroelectric conditions. Unlike the residential and commercial sec- tors, where the general uses of electricity are similar in different houses or buildings, the industrial uses of electricity are extremely diverse. It is very difficult to generalize about the end-uses of energy or the amounts of energy used in a “typical” industrial plant. For example, the primary metals industry uses about 80 times as much electricity per dollar of output as the apparel industry. The industrial use of electricity in the North- westis highly concentrated in a few industrial sectors. Figure 3-10 illustrates the composi- tion of total industrial demand for electricity. The data for Figure 3-10 are based on 1977, the most recent year for which a comprehen- sive accounting of industrial energy use by detailed industry sector in the Northwest was attempted. Direct service industry customers accounted for 45 percent of total industrial demand for electricity, or about one-fifth of total regional sales to all sectors. The direct service industry sales are dominated by ten aluminum plants that consume about 90 per- cent of the direct service industry electricity. One-fourth of the direct service industry demand is considered nonfirm demand, or interruptible demand. Only the firm portion of direct service industry demands are included in the the Council's forecasts of energy requirements. However, the interruptible por- tion of direct service industry demand is con- sidered in system operation and electricity pricing analyses. A more current look at the composition of industrial demand would likely indicate some significant changes. The aluminum com- panies are currently operating at about 70 percent of capacity. In addition, the trends away from energy intensive industries, which will be discussed in the forecast, have already had some effect since 1977. For example, the medium-high forecast for 1985 shows the direct service industry share of total sales at 33 percent, key industries at 50 percent, and the minor industries share up to 17 percent. Five industries account for about 85 percent of the non-DSI industrial demand for elec- tricity. These industries are lumber and wood products, pulp and paper, chemicals, food processing, and primary metals. These five industries combined with the direct service industries account for over 90 percent of the region's industrial demand for electricity. Forecasts of industrial demand for electricity reflect production forecasts for the various industrial sectors, the amount of energy used per unit of output, and the effects of prices on their use of energy. Table 3-9 shows total industrial firm demand forecasts for selected years for all four forecasts. In the high fore- cast, consumption of electricity by the indus- trial sector grows to 9,219 average mega- watts by 2005—an average annual growth rate of 2.2 percent per year. In the low fore- cast there is no growth in industrial demand. The more likely range of industrial demand growth is from 0.9 to 1.6 percent per year. Chapter 3 Table 3-10 Industrial Forecasting Methods SIC CODE TITLE oe ‘SHARE FORECASTING METHOD VERSION 20 Food and Kindred Products 3.7 Econometric Model ODOE 203 Canned Fruits and Vegetables 1.9 Not Forecast 22 Textiles A Econometric Model AEA 23 Apparel A Simple 24 Lumber and Wood Products 8.4 Summed 2421 Sawmills and Planing Mills 3.7 Key Industry Model 2436 Softwood Veneer and Plywood 2.6 Key Industry Model 24XX Rest of SIC 24 241 Simple 25 Furniture at Simple 26 Pulp and Paper 18.8 Summed 2611 Pulp Mills 2.4 Key Industry Model 2621 Paper Mills 9.8 Key Industry Model 2621 Paper Mills — DSI 3 Assumption 2631 Paperboard Mills 5.3 Key Industry Model 26XX Rest of SIC 26 1.0 Simple 27 Printing and Publishing 4 Econometric Model ODOE 28 Chemicals 11.3 Summed 2812 Chlorine and Alkalies 2.3 Key Industry Model 2812 Chlorine and Alkalies — DSI 1.4 Assumption 2819 Elemental Phosphorous 5.6 Key Industry Model 2819 Elemental Phosphorous — DSI aA Assumption 28XX Rest of SIC 28 1.9 Econometric Model ODOE 29 Petroleum Refining 1.6 Simple 30 Rubber and Plastics 4 Econometric Model AEA 31 Leather and Leather Goods 0.0 Not Forecast 32 Stone, Clay, Glass and Concrete 1.2 Summed 3291 Abrasive Products — DSI 3 Assumption 32XX Rest of SIC 32 9 Econometric Model ODOE 33 Primary Metals 49.8 Summed 3334 Aluminum — DSI 41.6 Assumption 3313 Electrometallurgical — DSI 1.7 Assumption 3339 Non-ferrous N.E.C.— DSI a Assumption 33XX Rest of SIC 33 6.5 Simple 34 Fabricated Metals 8 Econometric Model AEA 35 Machinery Except Electrical 7 Simple 36 Electrical Machinery 2 Econometric Model ODOE 37 Transportation Equipment 1.6 Simple 38 Professional Instruments at Simple 39 Miscellaneous Manufacturing A Simple XX Residual Categories 8 Simple 3-13 Chapter 3 Methods of forecasting the industrial demand for electricity vary substantially among differ- ent components of the industrial sector. In general, the forecasting methods are most detailed for the activities that consume the greatest amounts of electricity. It is neces- sary to forecast industrial activity and demand for electricity individually for up to 40 industry components in order to obtain reli- able forecasts of total industry demands. The composition of the industrial forecasting system is shown in Table 3-10. The compo- nents of the industrial sector are defined using the Standard Industrial Classification (SIC) code. Table 3-10 shows the share of total industrial consumption of electricity estimated to have been consumed by each subsector in 1977. The concentration of demands for electricity that was described in Figure 3-10 is apparent in Table 3-10. There are four different forecasting methods used for the industrial sector. The methods are referred to as 1) key industry model, 2) econometric model, 3) simple relationships, and 4) assumptions. The method applied to each industry component is abbreviated in Table 3-10. All of the forecasting methods, except assumptions, are primarily driven by forecasts of industrial production for each industry component. In addition, each of those methods modifies the relationship between production and electricity use to reflect the effects of changing energy prices. Direct service industrial customers of Bon- neville are treated separately from other industrial components. All of their demands are forecast by assumption rather than being explicitly related to causative influences. This approach is used because direct service industry demands are limited on the high side by Bonneville contracts. There is sub- stantial uncertainty, however, whether direct service industry demands will be as large as their contracts allow. The three largest non-direct service indus- tries are forecast using the Key Industry Mod- els. The Key Industry Models are very detailed approaches to forecasting demand for electricity. The three so-called key indus- tries are lumber and wood products, pulp and paper, and chemicals. First, the industry is further divided into the most energy intensive 3-14 activities. For those activities, the uses of electricity are divided into several types of uses, such as motors for specific processes, electrolysis, or lighting. The fraction of elec- tricity use attributable to each of these end- uses is estimated for an average plant. In the case of the chemical production of phos- phorous and chlorine, the model is specified separately for each of the relatively few plants in the region. The forecast requires a specification of how the types of end-uses may change their shares over time. In addition, the degree must be specified to which electricity for each type of end-use could be conserved in response to price changes. The degree of price response was varied across forecast scenarios, being largest in the low forecast and smallest in the high forecast. Given these specifications, the demand for elec- tricity per unit of production will change from its base year value as production and elec- tricity prices change. The Key Industry Models require a great deal of data and judgment. This information goes beyond readily available sources of data. For this reason, specification of the Key Industry Models relied heavily on the participation and advice of industry representatives and trade organizations. The Council's industrial forecasting system includes a variety of econometric equations for non-direct service industry demand for electricity for all but the key industries. Econ- ometric models consist of equations esti- mated from historical data. The equations attempt to measure the effect of industry pro- duction and energy prices on the demands for different types of energy, including elec- tricity. Because historical data are generally of poor quality at the industrial subsector level, it is often difficult to obtain plausible relationships for econometric equations. Where econometric results appeared implausible, simple relationships between output and electricity use were used to obtain forecasts. Alternative econometric estimates are avail- able in the demand forecasting system for most industry components. In Table 3-10, the alternative equation used is specified under model version. The Oregon Department of Energy equations are noted as ODOE. Equations used by Bonneville are labeled AEA for the consulting firm that estimated the equations, Applied Economic Associates® The Oregon Department of Energy equa- tions were updated since the 1983 plan fore- casts. The new estimates were provided to the Council by ODOE. The sectors whose forecasting methods are listed as “simple” are those for which econo- metric results were unsatisfactory. The econ- ometric models that were used in the 1983 plan analysis for these industries were aban- doned in response to public comment criticiz- ing the behavior of those equations. In these simple forecasts, demand for electricity is assumed to grow at the same rate as produc- tion, but is modified by an assumed trend in electricity use per unit of production. There is substantial agreement in econometric mod- els and other research on industrial energy demand, that in the absence of other influ- ences, energy demand will grow with produc- tion. There is much less agreement about the degree of influence price changes will have on demand. To reflect this uncertainty, assumptions about changes in demand per unit of production were varied across forecast scenarios. Electricity use per unit of produc- tion was assumed constant in the high fore- cast for industry components that were fore- cast using the simple method. In the medium-high forecast, the electric intensity was assumed to decrease by 0.3 percent per year; in the medium-low forecast, by 0.7 per- cent per year; and in the low forecast by 1.0 percent per year. These assumptions are representative of the range of results from econometric equations that are more accept- able theoretically and behaviorally. The forecast growth rates for industrial demand for electricity are considerably smaller than the projected rates of growth in total industrial production. Production by Northwest manufacturing industries is expected to grow by 4.7 percent per year in the high forecast, 3.9 and 3.3 percent per year in the medium-high and medium-low forecasts, respectively, and by 1.8 percent per year in the low forecast. The relative growth rates of electricity demand and output imply an overall reduction in the electricity intensity of the Northwest industrial sector. The ratios of electricity use to production decline over the forecast period in all four forecasts. The rates of decline vary from 2.4 percent per year in the high case to 1.8 per- cent per year in the low case. Although these rates of decrease are significant, they are lower than recent regional history. Between 1977 and 1983, regional industrial electricity intensity is estimated to have declined by about 2.8 percent per year. Such decreases in energy intensity are not unprecedented. At the national level, for example, total energy use per unit of production in the industrial sector has been estimated to have decreased by 3.3 percent per year between 1970 and 1982. There are several factors operating to reduce industrial rates of electricity growth relative to production growth. The most important is a change in the mix of industry. Many of the large users of electricity are not expected to grow as fast as industry does on average. This is most notable in the case of the direct service industries, a very large portion of the industrial demand that is not expected to increase and may decline. The assumptions regarding direct service industry demand for electricity are shown in this chapter as a range of demand levels associated with specific forecast scenarios. The direct service industry loads are treated differently, however, for resource planning purposes. In the resource portfolio analysis, direct service industry load uncertainty is modeled by including 50 percent of alumi- num direct service industry load in all load cases and randomly adding portions of the remaining 50 percent of aluminum direct ser- vice industry loads. This is based on the conclusion that half of the aluminum produc- tion capacity in the region appears to be eco- nomically viable in the long run, while more uncertainty exists about the remaining capacity. The direct service industry assumptions described in this chapter are incorporated in the four forecast scenarios for purposes of defining the full range of electrical resource needs. Figure 3-11 shows the percent of cur- rent aluminum plant demand that is assumed to remain in the region by the end of the forecast period for each of the four forecasts. Chapter 3 20 30 40 50 60 70 80 90 Percent of Demand Figure 3-11 Assumed Aluminum Operating Rates Table 3-11 Composition of Industry Growth, 1983-2005: Medium-High Forecast HISTORICAL PRODUCTION DEMAND SHARE OF GROWTH RATE GROWTH RATE CONSUMPTION (%) (% per year) (% per year) DSIs 45 N.A. 0.6 Key industries 47 ont 1.6 Minor industries _8 51 48 TOTAL 100 3.9 1.8 Since Bonneville currently has contractual Obligations to serve all direct service industry capacity, 100 percent of DSI demands are included in the high forecast. It is assumed that 15 percent of DSI capacity will cease to operate in the medium-high forecast. The reductions in DSI demand in the medium-low and low forecast are 30 and 50 percent, respectively. The forecast of industrial electricity use is further dampened by the fact that some of the large non-direct service industrial users such as lumber and wood products, food process- ing, and pulp and paper are not growing as fast as less energy intensive industries. As shown in Table 3-11, output growth for the key non-direct service industries combined is expected to be 2.1 percent per year in the medium-high forecast, compared to 3.9 per- cent per year for all industrial production. Thus, the two components of the industrial sector that accounted for over 90 percent of the sector's electricity demand historically will show relatively weak growth over the next 20 years. 3-15 Chapter 3 Average Megawatts 1975 1980 1985 1990 2000 2005 Figure 3-12 Irrigation Demand Table 3-12 Irrigation Sector (Average Megawatts) GROWTH RATE ACTUAL FORECASTS (% per year) 1983 1990 2000 2005 1983-2005 High 615 735 838 896 1.7 Medium-high 699 768 796 1.2 Medium-low 676 739 756 0.9 Low 638 718 748 0.9 The second major reason for lower electricity . growth relative to production is the effects of Irrigation Demand the large changes in the relative price of elec- tricity in the region over the last several years. The effects of price on industrial demand can not be separated into components as they can for the residential and commercial sec- tors, but conceptually they include efficiency improvements, fuel switching, and product mix changes within individual industrial sectors. 3-16 Irrigation use of electricity is less than 5 per- cent of total regional firm electricity sales. In 1983, 615 average megawatts of electricity were used for irrigation. Until 1977, use of electricity for irrigation was increasing. As shown in Figure 3-12, irrigation sales since 1977 have become erratic and have not grown. In 1981, there were 8.6 million acres of irri- gated land in the region. Most electricity use in irrigation is associated with sprinkler irriga- tion. Currently, about half of the irrigated land in the region is irrigated with sprinkler systems. Table 3-12 shows the forecasts of use of elec- tricity for irrigation. The forecasts show some growth in electricity used for irrigation from its 1983 levels, but the growth is small relative to historical growth, which averaged nearly 4 percent a year from 1967 to 1983. Current use of electricity for irrigation, under normal weather conditions, was assumed to be 700 average megawatts, the average annual use from 1976 to 1983. The forecasts of demand for electricity by the irrigation sec- tor began with assumptions about growth in irrigated acres. The assumptions about growth in irrigated acres were judgmentally made, based on various studies in the region? There is sufficient growth to allow for the possible completion of the proposed Columbia Basin East High Project in the higher forecasts. The development of new irrigation, such as the East High Project, would be accompanied be reduced elec- tricity generating capability of the region's hydroelectric system and could impose addi- tional costs on Bonneville Power Administra- tion. These effects have not been included in the Councils analysis of the higher cases. The growth in demand for electricity implied by the range of assumptions about increases in irrigated acres is modified by assumptions about the level of price response by irrigators. A range of price responsiveness was assumed based on more detailed models of irrigation sector behavior. The lower forecasts were assumed to have more price response. Long-term price elasticity for the low forecast was assumed to be — 0.6; for both medium forecasts it was assumed to be — 0.4; and for the high forecast, —0.2. Since real electric rates decline the most in the lowest forecasts, the price response tended to raise those fore- casts the most. This results in a more narrow range of forecasts. Chapter 3 Retail Electric Prices The Council's forecasts of electrical rates in the Pacific Northwest show relatively stable prices over the next several years. The exact price outlook varies substantially in the differ- ent forecasts, however, due to differences in the amount of new resources to be acquired. Because nearly all new resources are more costly than the existing resource base, adding new resources will raise electrical rates. Retail prices are forecast using an electricity pricing model that is part of the demand fore- casting system. The pricing model develops forecasts of retail prices for each sector for investor-owned utilities and publicly-owned utilities. These rates are forecast through a detailed consideration of power system costs, secondary power sales, and the provi- sions of the Act. The model contains capacity and cost infor- mation on both generating and conservation resources. Cost and capacity of the federal base hydroelectric resources are included as a total. However, most other resources are treated on an individual basis. Capacity of each resource is specified for critical water conditions and for peak capacity. Capital cost and operating costs are specified for each generation resource. For conservation resources, only those costs that are to be paid through electric rates are included. The capacity of conservation resources are gen- erally predicted directly in the various demand models, although in some cases the savings are included within the pricing model and subtracted from demand there. The costs of generation and conservation are added up and allocated to the various owners (Bonneville, investor-owned utilities, and public utilities). The costs of resources used to provide power to customers of Bon- neville, public utilities, and investor-owned utilities are combined to reflect contractual agreements among utilities and the ex- change and other provisions of the Act. The model develops forecasts of wholesale power costs for three Bonneville rate pools— priority firm, direct service industries, and new resources. Similarly, wholesale rates are developed for investor-owned and public util- ities. Retail markups are added to these 1985 Mills/kWh 1980 1985 1990 Figure 3-13 Average Retail Electric Rates wholesale costs to obtain retail rates for each consuming sector of each type of utility. As demands grow, resources are added to meet demand and the new resource costs are melded with existing resource costs. The pricing model balances resources and demand based on critical water capacities. However, electricity is priced based on expected water conditions. The effects of different water conditions on secondary energy and electric rates is simu- lated by the pricing model. The operation of the hydroelectric system on a monthly basis for over 40 historical water years is the basis of this simulation. When there is surplus hydroelectric power in any month for a spe- cific water year, the model allocates that sec- ondary power to various uses according to a set of priorities specified in the model assumptions. These uses, in the assumed order of priority, are 1) serve the top quartile of direct service industry demand, 2) shut down combustion turbines, 3) sell outside the tegion, and 4) shut down other thermal generation. For purposes of the pricing model, firm sur- pluses are added to secondary power and allocated using the same priorities. If the region is in a deficit situation, instead of sur- plus, the model will import power at a pre- specified price until additional resources are added to meet demand. The revenues from sales of secondary power and firm surplus power, or the costs of import- ing to cover deficits, are averaged over months and water years to obtain estimates of expected prices of power given uncertain water conditions. Figure 3-13 shows real average retail rates in 1985 dollars for the four forecasts. As can be seen from Figure 3-13, real retail rates are projected to begin to decline in real terms after 1985. The exception to this is the low case, where it was assumed that the region would lose half of the aluminum companies by 1987. This loss of electrical sales during the surplus increases the rates that other consumers would have to pay and delays the downturn in real prices. In addition to the direct service industry assumption, the low 3-17 Chapter 3 Table 3-13 Electric Price Forecasts (1985 Cents per Kilowatt-Hour) om MEE MEMS AUeMAGE WHOLESALE CONSUMERS UTILITIES UTILITIES Estimated 1984 (1985 cents per kWh) 2.3 3.6 28 4.2 Forecast 2005 (1985 cents per kWh) High 3.0 45 4.0 5.3 Medium-high 2.4 3.8 3.3 46 Medium-low Vv 3.1 a 3.7 Low 1.3 2.8 2.4 3.4 Growth rates (% per year) (1984-2005) High 1.3 et 1.7 at Medium-high 0.2 0.3 0.8 0.4 Medium-low -1.4 -0.7 -0.2 -0.6 Low -2.7 1.2 -0.7 -1.0 1990 Figure 3-14 Relative Residential Energy Prices (Ratio of Electricity to Natural Gas) 3-18 case also assumes that the debts from the Washington Public Power Supply System Nuclear Projects 4 and 5 (WNP-4 and WNP-5) fall on the region's ratepayers some- how. This is to reflect the fact that there still remains some doubt about the final settle- ment of the WNP-4 and 5 debts. If those debts did fall on ratepayers, it would contrib- ute to a low case demand. That WNP-4 and 5 assumption accounts for most of the dif- ference in the beginning price level for the low forecast. Table 3-13 shows 1984 estimated average electric rates, forecasts for 2005, and aver- age annual rates of change for four different kinds of rates. The rates shown include Bon- neville wholesale rates for preference cus- tomers, average retail rates paid by all con- sumers combined, average retail rates paid by customers of public utilities, and average retail rates paid by customers of investor- owned utilities. Bonneville preference customer rates increase faster than inflation in the high and medium-high forecast. In the other forecasts real rates decline. Similar results are shown for retail rates of both public and investor- owned utilities. These results depend on the assumptions used in the pricing model. One important assumption is that the Council's resource portfolio is implemented, including the pro- posal that the region option to 90 percent of the load forecast range and build to the expected loads. Another important assump- tion is that no dramatically revised repayment requirement will be imposed for the federal debt on the region's hydroelectric system. Some of the more extreme versions of the revised repayment costs would have a signifi- cant effect on electric rates. For most of the demand sectors, the relative price of electricity compared to oil or natural gas is important. It is the relative price that is most relevant for consumers choice of fuel type. Figure 3-14 shows forecast prices of electricity relative to natural gas for residen- tial customers. Natural gas prices have been divided by 0.7 to adjust for differences in the end-use efficiency of gas and electricity. Thus, the relative prices shown in Figure 3-14 are more appropriate comparisons of the cost of heating than of the cost of buying fuel. Chapter 3 Although electric rates are highest in the high forecast, it is in the high forecast that relative electric rates are lowest. This stimulates the demand for electricity in the high forecast. When the ratio in Figure 3-14 is above 1.0, it means electricity is relatively more expensive than natural gas. During most of the 1970s, electricity in the Pacific Northwest was inex- pensive relative to natural gas, its main com- petitor. However, recent large increases in electric rates combined with decreases in natural gas prices have increased the com- petitiveness of natural gas. This result is only a general tendency, because the relative prices of electricity vary significantly for differ- ent utility areas. Further, the attractiveness of electricity or natural gas also can depend on consumer tastes and the relative cost of equipment used to convert energy to a useful service, such as heat. The general conclu- sion to be drawn from Figure 3-14 is that natural gas and electricity prices could remain competitive within a fairly broad range. The Role of Demand Forecasts in Planning Introduction The role of demand forecasts in the Council's resource planning is significantly different from the traditional role of demand forecasts. The traditional role of demand forecasts could be characterized as deterministic. That is, a best-guess demand forecast deter- mined the amount of new electricity genera- tion capacity needed. Before the early 1970s, it was generally assumed that demand for electricity would grow at close to its historical growth rates. That growth had been rapid and relatively steady. It was assumed that economies of scale in power generation could be relied on to keep prices for electricity from increasing as new capacity was added, so planners saw little reason for demand growth to slow down. In fact, it was widely assumed that there would be little or no response to price changes if they did occur. The dramatic reduction in demand growth that occurred in response to increases in electricity prices in the early 1970s caught Average Megawatts Probability Figure 3-15 Demand Uncertainty most planners by surprise. The initial reac- tion of planners seems to have been to develop much more sophisticated forecast- ing tools. The forecasting models adopted by the Council are representative of the results of those efforts. However, the Council has recognized that, even with the best available forecasting tools, the forecasts of future demands remain highly uncertain. This rec- cognition is moving forecasts away from their deterministic role in planning, to what may be described as an integral role. The integral planning role of demand fore- casts has three major components. First, forecasts of demand define the extent and nature of uncertainty that planners must face. Second, the level of demand is not indepen- dent of resource choices, but will respond to the costs of resource choices to meet future demands. Finally, sophisticated demand models are needed to assess the potential impacts of choosing conservation programs as alternatives to building new generating resources. Defining Range of Uncertainty The Council defines the uncertainty in future demands for electricity by developing a range of forecasts. The range of demands are based primarily on variations in the key assumptions that determine the demand forecasts. The forecast range has been described above in terms of four forecasts. A subjective probability distribution of future demands is developed based on the four forecasts. The probability distribution describes the liklihood that any given level of electricity demand within the range will occur. Figure 3-15 illustrates probability distribu- tions around the demand forecasts. The Council has adopted the trapezoidal distribu- tion. The implications of the trapezoidal dis- tribution are: (1) that demands outside the high and low forecasts are judged to be of sufficiently low probability that they are not formally considered in resource planning, and (2) that demands between the medium- high and medium-low forecasts are most likely and considered equally probable. 3-19 Chapter 3 Resource portfolio analysis is based on the entire probability distribution of future loads. This is a major change from the 1983 plan and is made possible by the new Decision Model. The Decision Model analysis utilizes hundreds of possible load paths that are dis- tributed according to the trapezoidal proba- bility distribution defined by the original four demand forecasts, as illustrated in Figure 3-15. Effects of Resource Choices on Price As was shown in Figure 3-1 and discussed in the previous section, there is an electricity pricing model in the demand forecasting sys- tem. This model translates resource deci- sions made by the Council into retail prices that various consumers will face in the fore- cast period. The price model ensures that the implications of future resource decisions, including conservation programs, are reflected in future prices and demands. Conservation Analysis In addition to defining uncertainty, the demand forecasting models play an impor- tant role in defining and evaluating conserva- tion opportunities. This is particularly true for the residential and commercial sectors, where the demand models are most detailed and conservation opportunities are best defined. There are two key roles for the demand mod- els in conservation analysis. The first is in helping define the size of the potential con- servation. The second is to predict the effec- tiveness of programs designed to achieve some portion of the potential conservation available. The stock of energy-using buildings and equipment, including its fuel type and effi- ciency characteristics, essentially deter- mines how much additional efficiency can be achieved to offset the need for new electricity generation. The building energy demand models provide the necessary stock fore- casts for analyzing conservation potential. Obviously, the demand models will show dif- ferent amounts of conservation potential for different forecasts. 3-20 The effects of conservation programs can be quite complicated and the demand models are designed to help assess those effects. For example, the effects of an energy effi- cient building code can affect all three com- ponents of building owner choice: efficiency, fuel type, and use. Of course, the direct impact is on efficiency choice, since a build- ing code constrains that choice directly. How- ever, there are also likely to be unintended effects on fuel choice and intensity of use. A more stringent code for residential elec- trical efficiency will tend to increase the con- struction cost of electrical homes. This rela- tive increase in the initial cost of electrical homes, if borne by homebuyers, may cause some increase in the number of homes heated by natural gas or oil, even though the cost of operating the more efficient elec- trically heated homes would be reduced. For cost-effective conservation actions, the cost of providing an end-use service, such as space heating, will decrease. With the decrease in cost, the consumers intensity of use may increase. Another important com- plication is that appliances give off waste heat affecting the heating and cooling require- ments in buildings. More efficient appliances give off less waste heat and, therefore, more heating and less cooling will be needed than with less efficient appliances. These second- ary effects can be assessed in the detailed building models to give a more accurate assessment of the actual effects of conserva- tion programs on demand for electricity. Forecast Concepts For any given forecast case (i.e., high, medium-high, medium-low, or low), there are three different demand forecast concepts used in the Council's planning activities. Most Council presentations and publications, including the preceding sections of this chap- ter, describe “price effects’ forecasts. Price effects forecasts show what the demand for electricity would be if customers were allowed to respond to price, but no new con- servation programs were implemented. Price effects forecasts also include no adoption of the proposed model conservation standards, but do include the more stringent building codes adopted in Washington and Oregon in 1985. An important factor affecting price effects forecasts is the resource mix that is assumed in the electricity price provided to the demand models. A ‘sales’ forecast is a forecast of the demand for electricity after the effects of the model conservation standards and other conserva- tion programs have been taken into account. This is the amount of electricity that would actually be sold by utilities and flow through power lines to consumers. The third demand concept, the “frozen effi- ciency” forecast, is somewhat more compli- cated to explain. Its purpose is to help avoid double counting of conservation—once as part of the response to price increases, and once as programmatic conservation poten- tial. Essentially, the frozen efficiency forecast attempts to eliminate from the demand fore- cast the effects of actions that are taken in response to price, but could also be achieved through the Council's proposed conservation programs. The method of developing frozen efficiency forecasts varies by sector. The first step in developing the three fore- casting concepts is to do a sales forecast. In the sales forecast, preliminary resource port- folios are assumed, including conservation resources. The effects of conservation pro- grams for the residential and commercial sector are estimated directly in the demand models. Industrial and irrigation programs are treated as resources that offset those sectors demands. The sales forecast results in a forecast of electricity prices that is based on the costs of the resources used to meet the demand forecasts. The price effects and frozen efficiency forecasts are done using these electricity prices. Using the electricity prices from the sales forecast, the price effects forecast answers the following question: What would demand for electricity be if consumers faced forecast prices but there were no new conservation programs? Clearly, electricity prices would be somewhat different than the prices from the sales forecasts if no conservation pro- grams were implemented. This is because the portion of demand served by conserva- tion program effects (beyond what would happen as price response) would have to be met by alternative generating resources whose costs and rate impacts might differ from those of conservation programs. Frozen efficiency forecasts are also done assuming sales forecast electricity prices. The term “frozen efficiency” comes from the residential and commercial sector demand models, which simulate three components of consumer decision making. The three com- ponents are energy efficiency levels, type of fuel, and intensity of use. It is the efficiency choice component of consumer behavior that could potentially duplicate the estimated effects of conservation programs. Therefore, the frozen efficiency forecasts add back the efficiency choice component of price response to the price effects forecast, but leave the other components of price response in the forecast. The frozen efficiency forecast is accom- plished by “freezing” the level of efficiency choice at the levels being simulated by the models for choices made in the years when conservation programs are assumed to go into effect. Thus, for example, thermal integ- rity choice for new buildings is kept at 1983 choice levels. Residential efficiency of refrigerators, freezers, and water heaters are frozen at 1992 levels, the year currently assumed for appliance code adoption. The industrial and irrigation models are not sufficiently detailed to use a similar approach to the frozen efficiency forecast. In the 1983 Power Plan analysis, the frozen efficiency forecasts for the irrigation sector assumed there was no price elasticity at all. The indus- trial frozen efficiency forecast assumed no elasticity with respect to electricity price in the pulp and paper industry, the industry where all of the identified conservation was to occur. The approach of assuming no price elasticity was weak in two respects. First, assuming no price elasticity eliminated all price effects, including those that resulted from price increases in the late 1970s and early 1980s. Second, the industrial and irrigation models do not separate out the components of price response—so not only was efficiency choice being held constant, but fuel choice and use responses were also being limited. Average Megawatts Chapter 3 30,000 25,000 20,000 15,000 10,000 1985 1990 1995 2000 2005 Figure 3-16 Comparison of High Forecasts The frozen efficiency forecasts for the indus- trial and irrigation sectors in the 1986 Power Plan have assumed that double counting would not occur. That is, the frozen efficiency and price effects forecasts are the same. The maximum double counting that might occur can be examined by running the models with electricity prices held constant at their 1985 levels and comparing the results to the price effects forecasts. In all but the high case, there is no double counting because elec- tricity prices do not increase during the 1985 to 2005 period. In the high forecast, prices reduce the industrial and irrigation demands by 206 megawatts. Comparison of the price effects on an industry-by-industry basis and for irrigation showed that the maximum dou- ble counting in the high case for 2005 could be 170 megawatts (140 megawatts in industry and 30 megawatts in irrigation). This double counting would occur if the conservation actions identified in the Council's industry survey of conservation potential were the same actions that would be taken in response to high case price increases from 1985 to 2005. Differences between the three forecast con- cepts have particular meanings. This section discusses those meanings and summarizes the differences. The three forecasts for the high scenario are shown in Figure 3-16 to help visualize the following discussion. Table 3-14 shows the growth rates for the three forecast concepts for each of the fore- cast scenarios. The price effects growth rates are the same as those shown in Table 3-1 and Figures 3-2 and 3-3. The frozen efficiency growth rates are slightly higher because part of the demand decreases due to price response have been eliminated. The differences between price effects and frozen efficiency forecasts are relatively small because prices are not forecast to increase much in most forecast scenarios. Demand growth rates for the sales forecasts are signif- icantly lower than the price effects and frozen efficiency forecasts, reflecting potential con- servation savings from the Council's pro- grams. Only in the low forecast are the dif- ferences among the three forecast concepts small, reflecting the fact that only new build- ing standards savings are acquired. 3-21 Chapter 3 Table 3-14 Demand Growth by Forecast Concept, 1983-2005 Average Annual Rate of Change (%) MEDIUM- MEDIUM- HIGH HIGH LOW LOW Price Effects 27 1.8 1.2 0.2 Frozen Efficiency 2.9 1.9 1.3 0.2 Sales 2.2 1.3 0.7 0.0 The difference between the highest forecast (the frozen efficiency forecast) and the lowest (the sales forecast) is the total effect on elec- tricity demand of conservation resources and cogeneration. The price effects forecast divides that total effect into two parts, that which would result from price response and the incremental effect of conservation pro- grams. The difference between the frozen efficiency and price effects forecasts repre- sents the price response portion. The dif- ference between the price effects and the sales forecasts represents the incremental program impacts. Electric Loads for Resource Planning Demand forecasts serve as the basis for the Council's resource portfolio analysis. The actual loads for resource planning are based on the various demand forecast concepts, but must be modified to the appropriate defi- nition for resource planning analysis. This section describes the forecast concepts used and their modifications. In the 1983 plan, resource loads were based on frozen efficiency forecasts of demand. The 1986 Power Plan loads are also based on frozen efficiency forecasts. However, sev- eral adjustments are made to these forecasts before they are used for resource planning. The assumptions regarding direct service industry demand for electricity are shown in this chapter as a range of operating levels associated with specific forecast scenarios. The direct service industry loads are treated differently, however, in the analysis of elec- trical loads faced by the region for resource planning purposes. In the resource portfolio analysis, direct service industry load uncer- 3-22 tainty is modeled by including 50 percent of aluminum direct service industry load in all load cases and randomly adding portions of the remaining 50 percent of aluminum indus- try loads. Thus, for resource analysis, the risk associated with the upper half of the alumi- num loads has been disassociated with any particular load scenario. This facilitates a bet- ter assessment of the uncertainty, because it is not clear that the health of the aluminum industry in this region will be related directly to the general economy; the positive influences of ahealthy economy may be offset for alumi- num producers by the higher electric rates that would come with a faster growing region. Several adjustments are made to the demand forecasts to create the load fore- casts for resource planning. First, demand forecasts are converted to load forecasts by adding transmission and distribution losses. The demand forecasts are for consumption of electricity at the point of use, while loads are the amount of electricity that needs to be generated. More electricity has to be gener- ated than is actually consumed by utility cus- tomers, because some electricity is used or lost in the transmission and distribution of power. The demand forecasts are converted to loads by adding 2.4 percent to direct ser- vice industry demand, and 7.5 percent to other demand. Most resource analysis is done on an operat- ing year basis. Since the demand forecasts are done on a calendar year basis, the demands must be converted from a year that begins in January, to a year that begins the previous September. This is done by cal- culating a weighted average of the previous and current calendar years. The previous year receives a one-third weight, and the cur- rent year a two-thirds weight. In addition, for resource planning, the 1985 and 1986 calen- dar year forecasts are set to be the same across forecast scenarios. This was done by averaging the four forecasts. The resulting 1986 forecast (a proxy for actual loads) is then interpolated to each scenario’s respec- tive 1990 level. Finally, it is important to restate that the resource portfolio analysis is based on the entire probability distribution of future loads. This major change from the 1983 plan is made possible by the new Decision Model. The Decision Model and resource portfolio analysis are described in Volume II, Chapter 8. 1./ The results of these comparisons are pre- sented in detail in Kenton R. Corum, “REEPS in the Pacific Northwest: Preliminary Results,” presented at the EPRI Annual Review of Demand and Conservation Research, Seat- tle WA, July 10-12, 1984. 2./ “Evaluation of the BPA RRHED Model,” North- west Power Planning Council, Portland OR 97205, November 1984. 3./ Larry Palmiter and Mike Kennedy, Annual Thermal Utility of Internal Gains, 8th National Passive Solar Conference, American Solar Energy Society, Santa Fe NM, September 4./ For a detailed description of methodology and results, see Kenton R. Corum, “Interaction of Appliance Efficiency and Space Conditioning Loads: Application to Residential Energy Demand Projections,” proceedings of the American Council for an Energy Efficient Economy Summer Study on nergy Effi- ciency in Buildings, Santa Cruz CA, August 1984; or “Linking Efficiency and Space Condi- tioning Loads in Residential Energy Demand Projections,” forthcoming in Energy — The International Journal. 5./ Jerry Jackson, “Northwest Power Planning Council Commercial Model Update,” contrac- tor's report, September 1985. 6./ Applied Economic Associates, Inc., Update and Re-estimation of the Northwest Energy Policy Project Energy Demand Forecasting Model, report to Bonneville Power Admin- istration, December 1981. 7./ Water Today and Tomorrow, Vol. Il, The Region, Pacific Northwest River Basins Com- mission, June 1979; Northwest Agricultural Development Project: Final Report, June 1981; Demand Response to Increasing Elec- tricity Prices by Pacific Northwest Irrigated Agriculture, College of Agricultural Research Center, Washington State University, 1981; and unpublished studies and computer runs from Bonneville Power Administration. Chapter 4 Financial Assumptions and Resource Cost Effectiveness The Councils planning process involves a number of analytical steps, including estima- tion of quantities and costs of resources, pro- jection of future demand for electricity under a variety of assumptions, and simulation of the operation of the regional power system to meet demands with alternative sets of resources. All of these analytical steps require that values for a number of financial variables be assumed. Consideration of these assumptions is important for two rea- sons: first, the values used directly influence the outcome of the analysis; second, the val- ues used in the various components of analy- sis must be consistent. Anumber of financial variables influence the Council's planning process. Like many com- ponents of the Council's analysis, the values of these variables cannot be known with absolute certainty. This chapter of Volume II describes the major issues and the reason- ing behind the values adopted by the Coun- cil. It also provides an explanation of terms used throughout this chapter: nominal dol- lars, real dollars, present value, levelized cost, and discount rate. Following this expla- nation, three categories of variables are examined: 1) escalation rates, including those of fuel prices, construction costs, and the general level of prices; 2) cost of capital, including home mortgage rates and the cost of capital for regional resource acquisition; and 3) discount rates, including the rate used for converting streams of regional costs to present values and rates used in projecting consumers efficiency and fuel choices in the future. The costs of conservation resources avail- able to the region vary widely. Choosing which of these resources is to be used requires the specification of a cost-effective- ness limit, based on the cost of alternatives available to the region. This specification, given uncertainty about future demand for electricity, is not entirely straightforward. The rationale for the cost-effectiveness limit, for conservation used in this 1986 plan, is described in the last section of this chapter. Explanation of Terms Nominal Dollars and Real Dollars. inflation distorts the apparent costs of any energy resource, making it appear to cost more if it is purchased at a later time. To control for this distortion, three concepts are used. Nominal dollars are the actual expenditure of dollars over time and include the effects of inflation. Nominal dollars are therefore dollars that, at the time they are spent, have no adjustments made for the amount of inflation that has affected their value over time. Real dollars adjust nominal expenditures to account for the effects of inflation. By correcting for the impact of inflation on a dollars purchasing power, a real dollar represents constant pur- chasing power or “real” value. That is, a real dollar has the same value in 1985 that it has in 1995. To convert nominal dollar costs to real dollar costs, a base year is chosen, and all costs are converted to that year's dollars; i.e., the inflation that occurs between years is accounted for. Real dollars can be compared across the board, regardless of the year, because they represent equal purchasing power. The Council used a 1985 base year and a forecast inflation rate of 5 percent per year. Present Value and Levelized Cost. Even after costs are converted to real 1985 dollars, it is difficult to compare the costs of different resources because costs occur in different years. For instance, a hydropower project involves a large outlay at the beginning for construction, but the fuel (water) is essen- tially free after completion. An oil or gas-fired combustion turbine has a low construction cost, but the fuel cost is high and may even escalate in real terms (thats, it may get more expensive to run even after removing the effect of inflation). Because of the various resources available in the region and the dif- ferent capital and operating cost structures associated with each, two methods may be used to place them on even footing for cost comparison. Present value and levelized cost are the methods used. Present value implies that money has a time value. That is, when money is spent is as important as the amount of money spent. A dollar is worth more today than it is a year from now because it could be invested during the year to earn a financial return. A dollar a year from now is converted back to its present value by calculating, over the year, the interest or return foregone. Present value then allows the equal comparison of costs of energy resources by using a standard discount rate to convert all costs back to the start of the plan. The uniform series of costs that has the same present value is called a resource'’s levelized costs. For instance, the amount borrowed from a bank is the present value cost of buying a house; the mortgage pay- ment is the levelized cost. Discount Rate. The value of money over time to the Northwest ratepayer is used in calculating present values and levelized costs and is called the discount rate. The discount rate used for the Council's analyses was an inflation-free real rate of 3 percent. Interest rates consist of a real rate and an inflation premium. To convert nominal costs to present values, a nominal discount rate of 8.15 percent that combines the real discount rate of 3 percent with a 5 percent rate of inflation is used. The application of all the concepts to a generic coal plants illustrated in Figures 4-1, 4-2, and 4-3. This is only a numerical exam- ple, and the actual costs for this hypothetical coal plant do not necessarily agree with the coal plants used in the resource portfolio. The plant produces 452 average megawatts and comes on line in 2001. The concepts are the same for all resources; only the actual costs would differ. Figure 4-1 shows the nom- inal (actual) expenditures for the plant through construction and during its opera- tion. The line labeled “construction” repre- sents the cumulative construction costs from the start of the project in 1995 to the time it comes on-line in 2001. The total capital cost is $2.3 billion, which includes labor and mate- tials of $1.5 billion and interest of $0.8 billion. For the purposes of this example, the assumption has been made that those costs are repaid to lenders at a uniform rate of $395 million a year beginning in 2001. Those annual payments are represented by the “debt service line.” The line labeled “O&M” (operations and maintenance) rises faster than the rate of inflation due to increased costs of fuel. O&M starts at $235 million a year and rises to $1.3 billion per year by the end of the plant's 30-year life. Again, all costs in this chart include the effects of inflation over time. Chi apter 4 DOLLARS (MILLIONS) 2500 2262 Construction 2000 1500 1000 500 Debt Service 1985 90 95 2000 05 10 15 20 25 30 Figure 4-1 Actual Nominal Dollar Expenditures DOLLARS (MILLIONS) Nominal $ 400 395 Real 2001 $ 300 224 Levelized-2001 $ 200 100 Levelized 1985 $ 1985 90 95 2000 05 10 15 20 25 30 4-2 Figure 4-2 Capital Costs Figure 4-2 takes the debt service line from Figure 4-1 and demonstrates the conversion of nominal dollars to real dollars applying the present value and levelized cost concepts. The line labeled “nominal” represents the repayment of the construction costs from 2001 forward. Those costs include inflation. By converting to real costs, hence adjusting for inflation (line labeled “real 2001”), the effect of inflation upon the nominal repay- ment costs is illustrated. Starting in 2001, debt service commences at a fixed payment of $395 million per year. Over the years, repayment is subject to general inflation, but cannot rise to reflect it. Therefore, by the end of the repayment period, the nominal repay- ment amount of $395 million is worth $91 million in actual 2001 dollars. Inflation has decreased the value of a fixed payment because other wages and costs have risen with inflation. The declining real costs are then annualized to levelized real costs (line labeled “levelized 2001$”). This line repre- sents the constant debt service payments restated to control for inflation. Finally, using the line labeled “levelized 1985$,” the debt service payments are restated to the base year 1985 dollars by removing inflation from 1985 to 2001. This process allows the com- parison of capital costs of different resource projects by controlling for timing, inflation, and interest rates. Figure 4-3 goes through the same process, but uses the O&M line from Figure 4-1 to analyze operating costs. Operating costs start at $235 million a year in 2001, and rise in nominal terms (line labeled “nominal”) to $1.3 billion by the end of the plant's life. The assumption is made that these costs rise faster than general inflation due to the costs of fuel. Those nominal costs are controlled for inflation, and are represented by the line labeled “real 2001$,” which reflects the slightly higher (than inflation) cost increases of fuel over time. Levelizing those costs yields the “levelized 2001$” line. This re- states the stream of real dollar costs as an annualized amount. “Levelized 1985$,” then, takes the levelized 2001 costs back to 1985 levelized costs by controlling for inflation for those years and using present value. Chapter 4 The various numbers that can describe the same plant are summarized in Table 4-1. The capital cost in nominal dollars is $2.3 billion. The first-year cost, as it would actually affect rates in 2001, the first year of operation, is 16.0 cents per kilowatt-hour. Levelized in 2001 dollars for comparison with other resources that come on-line in 2001, the cost is 12.4 cents per kilowatt-hour. Finally, con- verted to the base year used in the Council analysis, the levelized cost is 5.7 cents per kilowatt-hour. Table 4-2 gives a sample cal- culation of the levelized costs of a conserva- tion measure. It is important to remember that the process described above is used to put resource cost estimates on a consistent basis. It is not a prediction of the impact of any given resource on consumer rates in a given year. In fact, the two example resources mentioned (the hydropower plant and the combustion tur- bine) could have quite different effects on rates in any given year. The hydropower plant is the most expensive in the first year. Because the capital cost is fixed, its real cost declines through time as other costs and wages rise with inflation. Grand Coulee Dam, for example, was a very expensive project when it was finished in the early 1940s. It is only the succeeding 40 years of inflation that have made the cost of about 0.2 cent per kilowatt-hour relatively cheap compared to the cost of new power plants. A combustion turbine, on the other hand, has a large percentage of its total cost in its fuel cost. If operated at reasonable levels of annual output, its total cost (capital plus fuel) could be lower in the first years of its opera- tion than the hydropower plant. However, its fuel cost will continue to rise with inflation, if not faster, and its relative rate impact will be much higher 20 years from now than would that of a hydropower plant built now. A resource such as the hydropower plant could have the lowest present value and levelized cost even though it has the highest first-year cost. The Council's resource choice was not based on the rate impacts in any given year but was based on the present-value cost, taking into account the costs and their timing over the life of the resources. 1000 750 500 250 1250 Nominal $ 1270 266 Levelized1985$ Levelized 2001$ ee a 1985 90 95 2000 05 10 15 20 25 30 Figure 4-3 Operating Costs Table 4-1 Cost Analysis Summary Total Capital Cost Direct Construction First Year Cost Levelized 2001 dollars (first year of operation) Levelized 1985 dollars $2.3 billion $1.5 billion 16.0 cents per kilowatt hour 12.4 cents per kilowatt hour 5.7 cents per kilowatt hour 43 Chapter 4 Table 4-2 Sample Calculation of Levelized Cost of Conservation Measure Levelized Unit Cost = Measure Cost 1980 dollars x Annual Capital Recovery Factor MEASURE LIFE (Years) 5 10 15 20 25 30 Formula for Annual Capital Recovery Factor i(1+i)N Where N = Measure Life Annual Savings in kWh ANNUAL CAPITAL (ah Real Discount Rate) 0.218 0.117 0.084 0.067 0.056 0.051 (1+i)N-1 i = Real Discount Rate Measure Cost = $ 32 Annual Savings = 435 kWh Annual Capital Recovery Factor Levelized Unit Cost = -117 (10 years) $32 x 0.117 435 kWh $0.0086/kWh or 8.6 mills/kWh Levelized cost numbers are appropriate for rough comparison of resources. For the final analysis, the resources operating charac- teristics were simulated in the System Analy- sis Model, and the costs from that simulation converted to present values. This is a very important distinction, because levelized costs do not take into account the changes in system operations that will result when resources with different operating charac- teristics are added. The system models that the Council uses for evaluating the present value system cost of each resource added to the Northwest's existing system provide the best test of the cost effectiveness of each resource. Escalation Rates The regional economic and demographic projections reported in the Council staff report, “Economic, Demographic, and Fuel Price Assumptions’ (July 15, 1985), were con- structed to be consistent with Wharton Eco- nomic Forecasting Associates national eco- nomic forecasts. The Council subscribes to 4-4 the Wharton national forecasting service and uses Wharton forecasts as the basis for the regional economic scenarios. The specific Wharton forecast used for this 1986 plan is cited in Volume II, Chapter 2. In order to maintain maximum internal consistency among the economic and demographic pro- jections and the financial assumptions devel- oped in this 1986 plan, Wharton projections for the various escalation rates are used wherever feasible. Fuel Prices. Fuel price escalation rates are used to cost resource alternatives, to project demand for electricity, and to simulate the operation of the regional electrical power sys- tem. The 1983 Power Plan used variation in fuel price projections to help generate the range in projections of demand for electricity. Each of the four economic growth scenarios included a unique projection of fuel prices. In contrast, the fuel prices used in costing resource alternatives and in simulating the operation of the power system did not vary with economic growth assumptions. The rationale for this apparent inconsistency was that fuel prices were used to reflect uncertainty about future demand for elec- tricity. The method did not assume that a given demand for electricity could only result from a given set of fuel prices, but that a plausible set of conditions, including fuel prices, could lead to each of the Council's four demand scenarios. There was no implication that these were the only condi- tions that could lead to the demand scenario (e.g., that a medium-low demand scenario could come about only if the medium-low natural gas price assumptions were realized). As a result, fuel cost assumptions for the analysis of resource costs and system oper- ation were not required to be identical to those of each demand scenario. The region faces considerable uncertainty regarding future prices of fuels, as well as other determinants of costs and performance of generating resources. However, cost- effectiveness analysis would be impossibly complex if the whole spectrum of costs and performance of alternative resources was treated explicitly. The Council's approach was to select the “most likely” values for testing cost effectiveness. Sensitivity studies were done where resource alternatives cost approximately the same. In the 1986 plan, the Council decided to treat fuel prices as it did in the 1983 plan, using four sets of fuel price assumptions for demand forecasting and a single “best-esti- mate” set of assumptions for the resource costs and system operation work. Fuel price assumptions for each of the demand sce- narios were developed by Council staff as part of the economic-demographic projec- tions. The “best-estimate” set of price esca- lation assumptions for the resource cost and system operation analysis was obtained for oil and gas by averaging the escalations in the medium-high and medium-low sce- narios. Coal price escalations are based on Wharton projections of mine-mouth coal costs and Wharton projections of compo- nents of coal transportation costs. Price escalation assumptions used in resource evaluation and system analysis, along with the assumptions used in the 1983 Power Plan, are shown in Table 4-3. All assumptions were presented for extensive public review prior to their adoption. Chapter 4 Table 4-3 Fuel Price Escalation Assumptions Average Annual Real Rate of Growth (%) 1983 PLAN 1986 PLAN (1980-2002) (1985-2005) Natural Gas 2.4 1.8 Oil 1.8 1.6 Coal 1.5 1.0 Construction Costs. Escalation rates for construction costs are needed to estimate future resource costs, which influence both system operation cost and demand projec- tions (the latter through the effect of construc- tion costs on electricity prices). Escalation rates for construction costs of different resources influence the choice of new resources. The overall level of construction costs for all resources influences the number of new resources needed. Based on Whar- ton projections and Portland General Electric comments, the Council used a 0.4 percent real escalation rate for all resource construc- tion. This compares with the 0.8 percent real escalation rate for construction costs of coal plants and zero percent real escalation in residential conservation costs, used in the 1983 plan. Inflation. The rate of inflation affects all com- ponents of the Council's analytical process. It is impossible to project the effect of changes in costs without considering the changes from both the real and nominal perspective. For example, prices of electricity are deter- mined in part by historical (nominal) con- struction costs, but projection of demand is usually based on the inflation-corrected (real) path of electricity prices. The necessary translation between real and nominal values requires a set of assumptions regarding the rate of inflation. Economic forecasters gener- ally have lowered their forecasts of long-term inflation, compared to forecasts in 1983. This 1986 plan uses the average inflation rate of 5 percent from the Wharton forecast (in com- parison to 6 percent in the 1983 plan). Cost of Capital Home Mortgages. One of the most inten- sively analyzed resources for future elec- tricity conservation is improved thermal efficiency of new homes. The cost of this improved efficiency, both to the individual homeowner and to the region, is influenced by the extra construction cost due to energy efficient measures. These increased costs are mortgaged, and therefore the present value cost is a function of the interest rate charged on the mortgage. Mortgage rates, as projected by Wharton, change over time as the overall state of the national economy changes. Because these rates influence the costs of thermal efficiency, the use of varying mortgage rates would result in varying levels of optimal thermal efficiency. From a practical perspective, this would complicate the plan- ning process prohibitively, so the choice of a single mortgage rate assumption that is a reasonable long-run average seems more appropriate. The Council used a 6.2 percent real after tax rate or 11.5 percent nominal before tax for the mortgage rate assumption, which is the approximate 1994 rate projected by Wharton. This rate compares with the 6 percent real assumption used in the 1983 plan. Resource Acquisitions by Bonneville. The cost of capital for resources acquired by Bonneville for the region should reflect the actual regional cost of capital for the com- panies or organizations expected to develop the resources. The region's cost of capital is reduced by any federal tax benefits accruing to the owner of the resource, but includes any risk premium which the financial markets can be expected to attach to the investment. The assumptions for the real cost of capital in the 1983 plan, based on suggestions by the region's utilities, were 4 percent for debt financed by investor-owned utilities, 7.5 per- cent for equity of investor-owned utilities, 3 percent for debt financed by publicly-owned utilities, and 3 percent for Bonneville borrow- ing. Based on Wharton projections of the cost of capital and comments from Portland General Electric, the Pacific Northwest Util- ities Conference Committee, and Bonneville, these assumptions now appear low. There- fore, the Council adopted higher values of 7 percent, 8.5 percent, 4 percent, and 5 per- cent, respectively, for these real costs of cap- ital. Details of the comments and analysis leading to these assumed values are avail- able from the Council's staff and the Council's issue paper of January 4, 1985, “Assump- tions for Financial Variables.” Ownership and Capital Structure. The net financial cost of resources is a function both of who owns them and what capital structure is used. The generic coal plants and com- bustion turbines used in this plan were assumed to be owned 100 percent by the investor-owned utilities. The Council assumed that, with Bonneville acquisition available under the Act, generating projects would be financed using a capital structure of 80 percent debt and 20 percent equity. Conservation, including the model conserva- tion standards, was evaluated using utility financing. This assumption may be a conser- vatism in the case of the standards, which the Council ultimately expects to be embodied in building codes and financed directly by the consumer at lower rates. The capital struc- ture for conservation assumed regional coop- eration. Forty percent of the conservation was assumed to occur in public utility service territories and to be financed by Bonneville. The remaining 60 percent, in the investor- owned utility service territories, was assumed to be financed 75 percent by Bon- neville, under its cost-sharing principles, and 25 percent by the investor-owned utilities at their normal ratio of 50 percent debt and 50 percent equity. The small renewables, small hydropower and cogeneration were assumed to be PURPA resource purchases at 4.0 cents per kilowatt-hour. There was no explicit finance cost attached to these resources. 4-5 Chapter 4 Table 4-4 Discount Rates Used for Present Value by Source ORGANIZATION Northwest Power Planning 3% real Council (1983 20-year Plan) Bonneville Power Admin. 3% real U.S. Office of Management 10% real and Budget Pacific Northwest Utilities 3% real Conference Committee (PNUCC) Intercompany Pool 7.1% real Natural Resources Defense 1% real Council (NRDC) 2% real 3.5% real Robert C. Lind, et. al., 1% real Discounting for Time and Risk in Energy Policy 2% real 4.6% real DISCOUNT RATE TYPE OF PROJECT Power system analysis Power system analysis Federal government projects (water projects use lower discount rate) Power system analysis Transmission system investment (investor-owned utility perspective) Zero-risk social discount rate Costing of conservation Evaluation of investments of risk comparable to average common stock Evaluation of investments of risk comparable to U.S. Treasury bills Evaluation of investments of risk comparable to long-term U.S. government bonds Evaluation of investments of risk comparable to “market portfolio” Discount Rates The Social Discount Rate. A central feature of the Council's consideration of alternative strategies for providing adequate electricity to the region is the comparison of the strat- egies costs. Each strategy's stream of costs must be translated into a present value that can be compared to the present value of each of the other strategies. To accomplish this translation, it is necessary to use a discount rate that represents society's willingness to exchange consumption now for consumption in the future. For example, if the region puts the same value on $1.00 of consumption now as $1.08 a year from now, the region's rate of time preference, or its “nominal social dis- count rate,” is 8 percent. This would then be the appropriate nominal discount rate to use in converting regional power system costs to present values. While the concept of the social discount rate is fairly straightforward, its application is more complicated. The principal difficulty is that it is not possible to observe the social discount 4-6 rate directly; it must be imputed from rates of return on investments that are observable. In a perfectly competitive economy, the social discount rate would be equal to the market rate of interest, but in the real world things are less simple. For example: 1. Both corporations and individuals pay income taxes. Income taxes mean that when a consumer postpones current con- sumption to invest, the future consump- tion that investment makes possible is less than that implied by the (pre-tax) return to that investment. As a result, indi- viduals investing in a project with a 10 percent rate of return are not demonstrat- ing a rate of time preference of 10 percent, but rather a somewhat lower rate. 2. Allinvestments are risky, and this riskiness varies from one investment to another. This is reflected in varying costs of capital from one investment to another. Ordinarily, the rate of time preference is understood to be the willingness to trade (certain) con- sumption now for (certain) consumption in the future. For regional planning, however, the Council would like to use a rate that reflects the uncertainty the region con- fronts as it evaluates resource costs dec- ades from now. The Council is faced, then, with the task of determining how much of observed rates of return are risk pre- miums, and how much risk premium should be included in the regional social discount rate. 3. Inflation complicates the interpretation of observed costs of capital in terms of the social discount rate. Investors can be expected to insist on a rate of return that, in addition to covering their rate of time pref- erence, tax obligation and risk premium, will also cover the expected rate of infla- tion. Thus, observable (nominal) costs of capital, even after income taxes and risk premiums are taken into account, will be greater than investors rates of time prefer- ence by the amount of inflation they expect. Attempts to estimate the magni- tude of inflation's effect on the cost of cap- ital are complicated by the fact that although the inflation rate that the econ- omy actually experiences can be mea- sured, the inflation rate that investors expect cannot. For reasons such as these, the estimation of an appropriate social discount rate is fairly complicated. A typical estimate might begin with some measure of cost of capital for low- risk investments, translated to an after-tax return based on some assumed tax rate for the representative investor. This rate of return would be translated to real terms by some estimate of expected inflation, and the risk premium judged appropriate for regional power resource investments added. Each step in this process requires judgments (e.g., which investments are low-risk, should any year's data be excluded, what is a represen- tative investor, how is expected inflation related to historical inflation, etc.) that affect the results of the process. Table 4-4 includes a sample of discount rates suggested or used by various organizations. While it demonstrates a lack of perfect agree- ment among the sources represented, Table 4-4 also indicates a rough range of uncer- tainty for the social discount rate. Two of the sources, the Natural Resources Defense Council and the book Discounting for Time Chapter 4 and Risk in Energy Policy, describe an estimation process much like the one out- lined above. Both analyze data on long-run (1920s to 1970s) average returns to invest- ments of various levels of risk and both esti- mate real after-tax returns for the lowest-risk class of investment. They both conclude that these yields have varied from — 2 percent to +2 percent, depending on the historical period. Further, they conclude that 1 percent real is a reasonable estimate for a long-run average return to low- or no-risk investments. Given these estimates, the discount rate of 3 percent, which has been used by the Coun- cil, Bonneville, and PNUCC for power sys- tem analysis in the past, implies that the riskiness of power system investments justi- fies a 2 percent risk premium. Values used for the Intercompany Pool and for the U.S. Office of Management and Budget reflect the perspectives of the Pacific Northwest's inves- tor-owned utilities and the federal govern- ment. These are different from the perspec- tive of the region's consumers. Therefore, it is reasonable for discount rates for these orga- nizations to differ from those used for the Council's power system analysis. The 3 percent assumption for the social dis- count rate appears to be reasonable for the 1986 plan. Table 4-5 provides a comparison of the adopted real values for cost of capital and discount rate, their nominal equivalents and several reference values. Consumers’ Implicit Discount Rates; Demand Forecasting. The concept of con- sumers implicit discount rates appeared when analysts looked at consumers fuel and efficiency choices from the perspective of investment behavior. Viewed as an invest- ment, for example, an extra layer of insulation in a homeowners attic has a cost incurred either as a lump sum (if paid for out of current income) or a stream of payments (if money is borrowed to pay for it). The insulation pro- duces a stream of benefits in the form of energy savings. If the rate of return of the least attractive conservation measure adopted by consumers were known, given the energy costs they face, that rate of return could be used to predict the conservation measures consumers will adopt if their energy costs change. Table 4-5 Comparative Values of Financial Variables CURRENT® CURRENT ADOPTED ADOPTED NOMINAL REAL REAL NOMINAL VARIABLE VALUE VALUE VALUE VALUE Mortgage 12.9% 7.5% 6.2%° 11.5% OU Debt 13.54 8.1 7.0 12.4 Public Debt 10.1¢ 49 4.0 9.2 Treasury 11.3 6.0 5.0 10.3 IOU Equity 15! 9.5 8.5 13.9 Consumer Discount Rate _ _ 10.09 19.4 Social Discount Rate a _ 3.09 10.2" FURTHER COMPARISONS Wharton Forecasts DECEMBER 1984 FORECAST AUGUST 1984 FORECAST 1994 1994 1995 1995 VARIABLE NOMINAL REAL' NOMINAL REAL’ Mortgage 11.7 66 10.6 51 1OU Debt 13.4 8.2 12.4 6.8 Public Debt 8.8! 3.6 8.0! 2.6 Treasury 10.5 5.5 9.5 43 @ Current values from Wall Street Journal, week of February 4, 1985, unless noted. > Real/nominal conversions made using inflation rate of 5 percent. © This is a before tax real rate. Assuming a 20 percent marginal tax bracket, the 11.5 percent nominal rate is equivalent to 9.2 percent after tax. Correcting for 5 percent inflation, the after tax real rate would be 4.0 percent. ¢ Current value for A-rated utility debt 12.8 percent; Northwest investor- owned utilities are lower rated, PNUCC suggested current values at 12-14 percent, Portland General Electric (PGE) suggested current value of 14.5 percent. © Merrill Lynch retail electric 20-year revenue bond index in The Wall Street Journal; PNUCC suggests current values of 10-12 percent. ' PNUCC suggested current values of 15-15.5 percent, PGE suggested current value of 14.5 percent. 9 After-tax rates (all others pre-tax) + Nominal discount rate calculated from real after-tax rate using a 20 percent marginal tax rate. ' Real value calculated with 1994 forecast inflation of 4.74 percent. i Real value calculated with 1995 forecast inflation of 5.23 percent. « Baa utility yield in 1995 calculated using Aaa yield + 2.0 percent, which is approximate difference in 1994 forecast (Baa series not in 1995 data). ' Retail electric municipal yield calculated using average general obligation municipal yield + 0.5 percent, which is the approximate current difference between general obligation municipals and retail electrics. 4-7 Chapter 4 Table 4-6 Estimates of Average Implicit Discount Rates by Source SOURCE IMPLICIT RATE DECISIONS ANALYZED Berkovic, Hausman and Rust 25% Fuel choice (heating system, PNW)? 33% Fuel choice (water heat, PNW) Cole and Fuller 26% Thermal integrity (national) 12% Thermal integrity (PNW) 61-108% Efficiency of refrigerators Corum and O'Neal 10-21% Thermal integrity (national, 3 fuels) 7-20% Thermal integrity (Seattle, 3 fuels) Goett Fuel choice & heating system, 4% national (w/central AC) 21% national (w/o central AC) 3% PNW (w/ central AC) 27% PNW (w/o central AC) 7% Fuel choice for water heat, national 3% Choice of central AC, national Hausman 24-26% Efficiency of room air conditioners Johnson 4% Influence of total utility bills on resale price of existing house Arthur D. Little, Inc. 10% Windows and doors 32% Other thermal integrity Meier and Whittier 34% Efficiency of refrigerators (Pacific region of U.S.) a PNW = Pacific Northwest. > AC = Air conditioning. Table 4-7 Input Data for Consumers’ Implicit Discount Rates INITIAL LOWER LEVEL BOUND Thermal Integrity High scenario 50% real 7% real Medium-high scenario 40% real 7% real Medium-low scenario 30% real 7% real Low scenario 20% real 7% real Appliance Efficiency High scenario 85% real 85% real Medium-high scenario 65% real 20% real Medium-low scenario 65% real 7% real Low scenario 55% real 7% real A number of estimates have been made of this rate of return, or of what is commonly called the consumers “implicit discount rate.” As Table 4-6 demonstrates, the estimates vary widely (from as low as 3 percent real to over 100 percent real). Most estimates of the implicit discount rate, however, are signifi- cantly higher than the usual range of esti- mates of the social discount rate, which com- monly falls between zero percent real and 10 percent real. In a perfectly competitive econ- omy consumers might undertake all conser- vation investments with rates of return greater than the social discount rate. How- ever, this evidence indicates that in the real world consumers pass up many conserva- tion investments that have expected rates of return higher than any estimate of the social discount rate. The gap between estimated implicit discount rates and the estimates of the social discount rate has been attributed to various forms of market imperfections (e.g., lack of information about the performance of efficient equipment, uncertainty about resale value of more efficient houses, and limited access to loans for conservation investments). The demand projections in the Council's 1983 plan used a single set of implicit dis- count rate assumptions to simulate consum- ers efficiency choices. In the residential sec- tor, the initial value of this discount rate for thermal integrity, space conditioning effi- ciency, and water heating efficiency deci- sions was 30 percent real. For efficiency decisions for appliances such as refrigerators and freezers, the initial value of this discount rate was set at 65 percent real. All implicit discount rates were simulated to drop significantly (by one-half or more) as fuel prices increased, and consumers became more concerned and informed about their efficiency choices. These assumptions were based on research results and judgment and were the most reasonable single set of assumptions the Council could make at the time. In view of the range of estimates demonstrated in Table 4-6, the Council, in this 1986 plan, used implicit dis- count rate assumptions that varied by fore- cast scenario. For each scenario, two pairs of inputs are required, one pair for thermal integrity deci- sions and another pair for appliance effi- ciency decisions. The first value in each pair is the initial level of the implicit discount rate. The discount rate is adjusted during opera- tion of the model as fuel prices change through time and the second value is a lower bound on the range of possible adjustments. Given the projected fuel prices in this plan, implicit discount rates generally stay closer to their initial levels than to the lower bounds. The assumptions are listed in Table 4-7. Consumers’ Discount Rates; Evaluation of Model Conservation Standards. Another important part of the Council's analy- sis is to examine the effects of the model conservation standards from the consumer's point of view. In the 1983 plan, this analysis used a discount rate of 10 percent real. This value is in the lower part of the range of estimates shown in Table 4-6, and it is lower than the values used in the projections of demand for electricity. It should be pointed out, however, that the appropriate discount rate for evaluation of the standards from the consumer's perspective is only roughly com- parable to the discount rates estimated in Table 4-6 and to those used in the demand First, most of the discount rates in Table 4-6 were estimated using simplifying assump- tions which ignored the effects of mortgage financing. This is inconsistent with the detailed examination of cash flow—an important part of the Council analysis of the standards. Perhaps more importantly, the standards may themselves change the appropriate discount rate to use in evaluating effects on consumers. The relatively high implicit discount rates demonstrated in Table 4-6 are commonly attributed in part to con- sumers perceived risk. This perceived risk is due to the consumers lack of reliable infor- mation regarding the performance of more efficient equipment and structures. In the case of the standards, there is evidence that the region has learned a great deal and is in the process of learning even more about the costs and performance of thermal integrity measures in Pacific Northwest climates. Chapter 4 Table 4-8 Summary—Financial Assumptions, 1983 Plan and 1986 Plan VARIABLE 1983 PLAN 1986 PLAN Escalation Rates Natural gas 2.4% real 1.8% real Oil 1.8% real 1.6% real Coal 1.5% real 1.0% real Construction Residential conservation 0.0% real 0.4% real Other resources 0.8% real 0.4% real Inflation 6% 5% Cost of Capital Home mortgages 6% real 6.2% real Resource acquisition Debt (investor-owned utilities) 4% real 7% real Equity (investor-owned utilities) 7.5% real 8.5% real Debt (public utilities) 3% real 4% real Debt (Bonneville borrowing) 3% real 5% real Discount Rates Social discount rate 3% real 3% real Consumers implicit rates Thermal integrity Declining from Varying by 30% real economic scenario Appliance efficiency Declining from Varying by 65% real economic scenario Evaluation of model conservation 10% real 10% real standards from consumers’ perspective This new knowledge could be expected to increase consumers confidence in the per- formance they expect from the measures incorporated into their homes. To the degree this is so, the implicit discount rate for home- buyers would be lower than the level of two or three years ago. This is another reason not to rely on the values in Table 4-6, which reflect historical situations in which consumers had significantly less information regarding energy conservation investments. There are good reasons to consider a dis- count rate lower than 10 percent for the eval- uation of the standards from the perspective of the consumer. If the appropriate social dis- count rate for investments of risk comparable to common stock is in the 4-5 percent real range, as Table 4-4 suggests, there is the question of whether consumers view houses meeting the standards as so much riskier than common stock as to justify a 10 percent real discount rate. Even though 10 percent was appropriate for the 1983 plan, the sub- stantial improvement in information about thermal integrity investments of the last two years would suggest a reduced discount rate now. The other side of the risk from the con- sumers’ perspective, of course, is that owners of houses meeting the standards are largely insulated from unexpected increases in energy costs. In the climate of legal and political uncertainty affecting the region's expected electricity prices, this reduction in risk will be valued by many consumers. 4-9 Chapter 4 With these arguments in mind, the Council will continue for the present to use the 10 percent rate. As shown in Table 4-5, a dis- count rate of 10 percent real is the after-tax equivalent of a rate of return to the consumer of 19.4 percent in nominal terms before tax. This is a higher rate of return than is available to most consumers. The use of the 10 per- cent real rate thus requires that homebuyers receive a rate of return on their investment in the improved efficiency of homes built to the Council's standards, which is quite attractive compared to other investment opportunities available to them. The financial variable assumptions described in this chapter are summarized in Table 4-8. These assumptions have been influenced in many cases by public comment during the development of the 1986 plan. Resource Cost Effectiveness All resources included in the Council's resource portfolio are selected based on their relative cost effectiveness. Cost effective- ness is a measure of the relative cost of the contribution of a resource to the region's elec- trical power system, and is most frequently measured in cents per kilowatt-hour. The Council has chosen, as the appropriate mea- sure of cost effectiveness, the net present value of each resource in the resource port- folio. The Council uses the levelized life cycle cost of each resource only as a preliminary screening tool to select resources for detailed study in the resource portfolio analysis. In selecting the amount of resources to be included in the Council's plan, the Council uses cost-effectiveness criteria. These cost- effectiveness criteria have three primary roles in the development of the Council's resource portfolio. The first role is in deter- mining which measures to include in the model conservation standards. The North- west Power Act directed the Council to develop model conservation standards that include all cost-effective conservation mea- sures. In evaluating individual conservation measures for their potential inclusion in the MCS, the Council uses a cost-effectiveness criterion that selects those measures that are lower cost than the expected cost of other 4-10 resources that would be included in the Council's Plan, if the conservation measures are not included in the MCS. Because this criterion applies to actions taken today and over the next few years, the evaluation was done using the Council's Decision Model, which accounts for the present value of actions in the near-term given the uncertain- ties in the resource portfolio. MCS measures are lost opportunity resources and, with their seasonal and load-tracking characteristics, were evaluated in the Decision Model to determine the expected present value of incremental MCS savings. The results were adjusted by the transmission system losses and other costs, described below. The second role of cost-effectiveness criteria is in sizing the amount of resource that may be available from discretionary conservation and generating resources in the future. Dis- cretionary conservation and generating resources are resources that can be acquired when the region's power system has a need for additional energy capability. This criterion is not used to determine actions to be taken today, but rather is used only to size the total cost-effective non-coal resource. Here the evaluation was made using the System Anal- ysis Model to calculate the cost of a coal plant (the Council's marginal resource) put in place in the year 2000. For this purpose, the mar- ginal coal cost was discounted only to the in- service date of the plant, rather than to the present. The third role of cost-effectiveness criteria is in selecting among near-term acquisitions those opportunities that are cost effective for the region to secure now. These near-term acquisitions are difficult to predict in advance; however, a specific cost-effectiveness crite- rion will allow the region to select only those that contribute value to the region's power system. This criterion, like the first, is used to evaluate actions to be taken today or in the near future. In calculating the value of near- term acquisition, the Decision Model was used by assuming resources of different lives are acquired in either 1985 or 1990. These resources, of course, did not have the sea- sonal or load-tracking characteristics or the administrative and other cost adjustments mandated in the Act that apply to marginal MCS measures. The present value of these resources was calculated assuming acquisi- tion in the 1985-1990 time period. In the following sections, each of these roles of cost effectiveness will be discussed: 1) the cost-effectiveness criteria used in evaluating the model conservation standards; 2) the cost-effectiveness criteria used by the Coun- cil in sizing the amount of discretionary resources that will be available in the Coun- cil's resource portfolio; and 3) the role of cost- effectiveness criteria in selecting among near-term resource acquisition opportunities. Cost Effectiveness of the Model Conservation Standards Figure 4-4 illustrates how the Council sys- tematically approaches the evaluation of the cost effectiveness of the model conservation standards. This figure shows three basic analytical efforts that the Council conducts in evaluating the MCS. The first of these is an evaluation of the cost and energy savings expected from each measure that might be included in the MCS. The second analytical effort involves the evaluation of the value of lost opportunity resources in the Council's resource portfolio. The third analytical effort evaluates the value of marginal MCS invest- ments in the region's resource portfolio. Each measure that potentially may be included in the MCS is evaluated for its expected cost. Considerable uncertainty exists with respect to financial assumptions, the accuracy and validity of cost data, the amount of administrative and overhead costs needed to secure the measure, and the defi- nition of exactly what a measure includes. Given each of these aspects, the Council must choose the most likely assumption on which to base further analysis. The financial assumptions involved in evaluating the MCS and other resources were discussed earlier in this chapter. The financial characteristics are different for each individual consumer; however, the Council must evaluate the MCS based on assumed typical characteristics for an average regional consumer. In evaluating the cost of each individual mea- sure, the Council has accumulated a variety of cost information. In reviewing this cost data, it is clear that significant uncertainty exists with respect to the range of cost that is likely to be experienced on each measure. In Chapter 4 spite of this significant uncertainty, the Coun- cil has selected the median Residential Stan- dards Demonstration Program (RSDP) cost for most measures that are potential candi- dates for inclusion in the MCS. For the mea- sures dealing with infiltration control and management of indoor air quality, the Coun- cil used the lower quartile of RSDP costs, exceptin climate zone IIl where median costs were used to reflect the higher cost of install- ing heat recovery ventilators that can operate in the severe climate. This was done to account more accurately for the substantial cost reductions that have been seen in com- munities that have adopted the MCS. These are being achieved as builders learn to install the measures more cost effectively and the market infrastructure for heat recovery ven- tilators is developed. In reviewing administrative and overhead costs for conservation programs, the Council has decided these costs do not change as the marginal measure changes, but are instead more a function of the administrative characteristics of offering a conservation pro- gram. Because the administrative and over- head costs contribute to the average cost of each conservation program, and not incre- mentally to the marginal measure, the Coun- cil includes administrative and overhead costs in the average cost of the MCS program. Cost oF MEASURES LosT OPPORTUNITY ACQUISITIONS AVERAGE VALUE OF MARGINAL MCS. Cost-Effectiveness Method for Evaluating the MCS In defining measures to be included in the MCS, the Council has tried to select incre- mental actions that builders might take to improve the efficiency of new buildings. This process is difficult in that measures are not homogenous actions that are simply evalu- ated individually. Previously, the Council has evaluated as a single measure the combina- tion of two individual actions: the inclusion of a vapor barrier and a heat recovery ventilator (HRV) to compensate for indoor air quality problems. Although it is likely that these two actions are independent and should be eval- uated separately, they are evaluated as a single measure in this plan. Because this issue was not fully discussed during the development of this plan, the Council will review, through a public process, the appro- priateness of separating these measures in the future. The expected energy savings must be evalu- ated for each measure that potentially will be included in the MCS. In evaluating residential measures, a heat loss model is used to esti- mate the energy savings of each measure. Considerable uncertainty exists over the actual savings that each measure might include when installed in a building. Here again, the Council must deal with a typical structure operated in a typical way and located at a typical site within each climate LEVELIZED COST OF MARGINAL MEASURES (| = L, WOOD/APPLI./CLOSURES i LEVELIZED COST OF mcs PACKAGE VALUE OF ALL RESOURCES PORTFOLIO Zone. The actual performance that is likely from each measure in an actual building could be substantially different than the amount forecast by these models. The Council has validated these models on a sample of actual buildings, and the average savings estimates appear to be accurate. However, the specific performance in each building could differ substantially from the Council's predictions due to a number of fac- tors: differences in building type; location; how the owner chooses to operate the build- ing; the application of other heating sources, such as wood heat; different appliance effi- ciencies; and possible room closures. Each of these changes from the typical conditions assumed by the Council will result in a differ- ent set of expected energy savings from the measures that might be included in the MCS. While there are a number of factors in this analysis that cannot be known with certainty, the challenge of power planning is to make informed decisions in the face of substantial uncertainty. For this reason, the Council selects a set of typical assumptions using the best cost information available and the best heat loss models available to estimate the levelized life cycle cost of each measure that could be included in the MCS. In addition, when a package of measures has been SELECTION OF ‘ALL REGIONALLY" COST-EFFECTIVE FOR MCS MEASURES AS MCS ACQUISITION Se eT Ue COST EFFECTIVE UTILITY PROGRAM Figure 4-4 4-11 Chapter 4 4.5 4.0 3.5 3.0 MCS year by year MCS one time 2.5 2.0 1.5 1.0 -— CENTS/KILOWATT-HOUR 0.5 0.0 1985 1990 1995 2000 2005 Figure 4-5 Estimates of the MCS Program Marginal Value selected to be included in the MCS, the Council evaluates the levelized life cycle average cost of the MCS, including admin- istrative and overhead costs. In evaluating the cost effectiveness of the MCS, the Council first evaluates the marginal value to the Council's resource portfolio of an incremental investment in MCS savings. Fig- ure 4-5 illustrates the results of studies of the marginal value of MCS savings using the Decision Model. A curve and a straight line are shown in this figure. The curve shows the expected value of a marginal MCS invest- ment in each year over the 20-year planning horizon. This figure shows that the marginal value of the MCS in 1986 is approximately 3.0 cents per kilowatt-hour, and that value increases to nearly 4.0 cents per kilowatt- hour in 1995. The value of marginal MCS investments beyond 1995 is approximately flat at 4.0 cents per kilowatt-hour. In other words, if the MCS could be changed year-by- year, the region should begin with an MCS defined to secure all measures costing less than 3.0 cents per kilowatt-hour in 1986. As shown by this curve, the region would expect to change the MCS each year until 1995 when the MCS would be defined to secure all measures costing less than 4 cents per kilo- watt-hour. 4-12 The horizontal line in Figure 4-5 is an esti- mate of the marginal value of increased MCS savings. This estimate is based on the dis- counted present value of all future resources in the plan that are displaced by greater MCS investments. The horizontal line illustrates that the expected present value of a marginal MCS measure is approximately 3.5 cents per kilowatt-hour. This is the Council's best esti- mate of the marginal value of MCS measures that must be selected today and cannot easily be changed year-by-year. Therefore, if the Council and the region must select a definition of the MCS that is not easily changed year-by-year, but must achieve all cost-effective energy savings in the Council's resource portfolio over the next 20 years, the MCS, based on this analysis, should be set at securing all measures costing less than 3.5 cents per kilowatt-hour. Because these values do not include the 10 percent cost advantage for conservation in the Act, and the appropriate adjustments for transmission and distribution system cost and losses, the appropriate criteria for eval- uating the cost effectiveness of the marginal MCS measures is 3.5 cents increased by the 10 percent advantage in the Act, 7.5 percent for transmission and distribution system losses, and 2.5 percent for transmission and distribution system costs. With these adjust- ments, the appropriate cut-off for the mar- ginal MCS measure is 4.2 cents per kilowatt- hour in levelized life cycle cost. Because of the substantial range of uncer- tainty surrounding each of the individual cal- culations involved in measuring the cost effectiveness of the MCS, the appropriate range for careful inspection of measures selected to be part of the MCS is established by the Council to be between 4.0 and 4.5 cents per kilowatt-hour. The Council carefully reviews measures in this range of cost and exercises judgment on which of these mea- sures to include. Marginal MCS investments that cost more than 4.5 cents should not be considered further until the costs or perform- ance improve. Measures less than 4.0 cents per kilowatt-hour are clearly cost effective and should be included in the MCS. As a practical matter, only the HRV and infiltration control package in climate zone 1 was a close call. This package in a typical 1,850 square foot house was estimated to cost 4.1 cents per kilowatt-hour. The Council decided to include these measures for health and safety reasons and to monitor their cost effective- ness closely. Once the Council has selected all of the mea- sures to be included in the MCS, the Council also evaluates the average cost effective- ness of any MCS program designed to secure MCS-level construction over the next several years. This is particularly important at this time because, with this plan, the Council is initiating a new MCS proposal that focuses on a Bonneville/Utility MCS Program. This MCS program is designed to market MCS- level construction and to provide financial assistance to builders. The goal is to secure the MCS as a lost opportunity resource over the next several years, and also to assist the region's building industry in making the tran- sition to more efficient construction. In evaluating the Bonneville/Utility MCS Pro- gram, the Council evaluated the cost effec- tiveness of acquiring MCS savings during the period from 1986 until 1990, again using the Decision Model. Figure 4-5 showed that the expected value to the Council's resource portfolio of MCS-level savings during 1986 was approximately 3.0 cents per kilowatt- hour. Adjusting this for the 10 percent advan- tage in the Act, transmission and distribution system costs and losses would mean that an Chapter 4 MCS utility program operated during 1986 should secure MCS-level savings at no greater than 3.6 cents per kilowatt-hour. This same calculation would escalate to a level of 4.4 cents per kilowatt-hour in 1990. For the MCS utility program, the Council estimates show that MCS-level construction can be achieved at an average cost of about 3.0 cents per kilowatt-hour. This is much less than the expected value of MCS savings over the next five years. The Council therefore uses two independent tests of the cost effectiveness of the MCS. The first is the cost effectiveness of the mar- ginal measure in the MCS. In making this evaluation, the Council uses the expected value in the Council's resource portfolio of the marginal measure in the MCS. The second test is to evaluate the cost effectiveness of any utility program used over the next several years to secure MCS-level construction. These programs will be evaluated to ensure that the level of financial assistance offered to secure MCS-level construction does not exceed the average value of MCS savings during the next several years. The Council has found that the MCS, as they are currently formulated, successfully meet both cost- effectiveness evaluations. MAINTAIN CAPABILITY RESIDENTIAL WEATHERIZATION SECURE ALL REGIONALLY DISCRETIONARY coer RESOURCES ALL OTHER RESOURCES DON'T ACQUIRE UNTIL NEEDED POLICY FOR SUPPLY FUNCTION CUTOFF Cost Effectiveness of Discretionary Resources The process for determining the cost effec- tiveness of discretionary conservation and generating resources is shown in Figure 4-6. This analysis basically falls into two catego- ties. The first is the resources the region is currently acquiring (at this time only the resi- dential weatherization program) and all other discretionary resources included in the Council's plan. The primary function of this analysis in the Council's planning is to size the amount of each resource that the Council expects to have available in the future in order to meet regional load growth. The cost-effec- tiveness criteria for discretionary resources are used to cut off the resource supply func- tions for resources included in the Council's portfolio. Because most of these resource acquisitions will be made when the region is assumed to need resources, it is important that the amount of discretionary resources that are estimated to be available is consis- tent with other resources that will be acquired at the same time. Since many of the discre- tionary conservation programs and generat- ing resource programs are begun during the time when the Council's resource portfolio also calls for securing options on new coal plants, the Council sizes the amount of con- CONSTRAIN BUDGET TO MINIMUM VIABLE é HEAT LOSS 1 VALUE TO REGION OF MARGINAL MEASURES DECISION MODEL VALUE OF R/WX NOW POLICY FOR R/WX IN PORTFOLIO SIZE OTHER RESOURCES IN THE PORTFOLIO Figure 4-6 servation and generating resources included in the plan based on the estimated costs of a new generic coal plant in the region's power system. The residential weatherization program is a discretionary resource that is currently being acquired. The Council believes that the capability to secure this resource should be maintained by continuing to operate the pro- gram at a minimum viable budget. While the program is operating primarily to maintain capability, it should continue to secure all measures that would be required when the program is needed. For this reason, the Council believes that even under minimum viable operations, the residential weatheriza- tion program should be securing all mea- sures up to the cost of a new coal plant. Current estimates of the cost of a new generic coal plant in the region's power sys- tem, using the System Analysis Model, are between 4.0 and 4.5 cents per kilowatt-hour discounted to the in-service date of the coal plant. The Council therefore truncates supply functions for all discretionary generating resources at 4.5 cents per kilowatt-hour. This figure needs to be increased for new conser- vation programs in order to take into account the 10 percent advantage in the Act and DETERMINE p] CURRENT PROGRAM DESIGN SIZE RES/WX RESOURCE FOR PORTFOLIO REGIONAL REVIEW BEFORE STARTING ACQUISITION PROGRAMS Cost-Effectiveness Method for Evaluating Discretionary Resources 4-13 Chapter 4 transmission and distribution system losses and costs. Increasing the 4.0 to 4.5 cents per kilowatt-hour for a coal plant by 20 percent results in a range for new conservation sup- ply curves of 4.8 to 5.4 cents per kilowatt- hour. The Council has truncated all discre- tionary conservation supply functions included in the resource portfolio at 5.0 cents per kilowatt-hour. There are very few conser- vation opportunities between 5.0 and 5.5 cents per kilowatt-hour. Cost Effectiveness of Near-Term Acquisitions The process of analyzing the cost effective- ness of near-term acquisitions over the next five years is shown in Figure 4-7. The evalua- tion begins with an analysis, using the Deci- sion Model, of the value of lost opportunity resources that must be acquired over the next several years if they are not to be lost. For purposes of this plan, the Council has looked at resources that may need to be acquired over the next five years. Figure 4-8 illustrates the value of resources with life- times from zero to 70 years if they are acquired in 1986 or alternatively in 1990. These resources are assumed to have a flat seasonal shape over the year and do not adjust their output based on load growth. For these reasons, they are less valuable than the MCS, which have their peak savings in the winter and save considerably more in high load paths than in low loads. LOST OPPORTUNITY RESOURCES OVER NEXT 5 YEARS NEAR-TERM RESOURCE ACQUISITIONS NON-LOST OPPORTUNITY RESOURCES The curves in Figure 4-8 show that a resource with an expected lifetime of 30 years acquired in 1986 has an expected value to the region’s power system of approx- imately 2.5 cents per kilowatt-hour. If this same resource is not acquired until 1990, its expected value will increase to approximately 3.0 cents per kilowatt-hour. The Council will use these avoided cost estimates to deter- mine the value of potential lost opportunity resources that may be acquired during the next five years. If the lost opportunity resources are conservation resources whose cost and savings are measured at the load, these estimates would need to be increased by 20 percent and adjusted for seasonal shape and load following ability. Significant lost opportunities will be evaluated on a case- by-case basis. Non-lost opportunity resources (a resource that could be acquired any time in the future, such as adding additional generating capability to an existing dam) must be evalu- ated separately. Because the region currently has a surplus, the value of such a resource would be substantially less than the value of the resource if it were acquired when needed. The relative value of each resource depends on the cost effectiveness and priority of that resource in the Council's resource portfolio. For this reason, it is not possible to develop a uniform policy for all non-lost opportunity resources. An evaluation of each resource acquired before it is needed in the resource EXPECTED VALUE TO REGION CURRENT VALUE BASED ON SURPLUS portfolio is necessary on a case-by-case basis to determine the resource's value to the region when acquired early. Conclusions A comparison of the cost-effectiveness crite- ria used by the Council is shown in Figure 4-9. This figure illustrates the Council's cur- rent estimates of the cost of a new coal-fired plant to be between 4.0 and 4.5 cents per kilowatt-hour. This estimate is used by the Council to truncate supply functions for both conservation and generating resources. Based on these estimates of the cost of coal, the Council has selected a supply function cut-off for new generating resources of 4.5 cents per kilowatt-hour and a supply function cut-off for discretionary conservation pro- grams of 5.0 cents per kilowatt-hour to account for the advantages in the Act and transmission system costs and losses. These two values were used in developing this plan to estimate the amount of each resource expected to be available to the region in the future. The Council has sized the conservation resources that are included in the Council's resource portfolio based on a selection of all measures that have an expected cost less than 5 cents per kilowatt-hour. Figure 4-9 shows this as the estimated cost-effective- ness criterion for discretionary conservation POLICY FOR LOST OPPORTUNITIES POLICY FOR NON-LOST OPPORTUNITIES CASE BY CASE ANALYSIS OF VALUE Figure 4-7 Cost-Effectiveness Method for Evaluating Near-Term Acquisitions 4-14 programs. The purpose of this estimate in the Councils Plan is to select all conservation measures that would be cost effective in the future as the region needs to acquire new resources. Because of conservation's greater cost effec- tiveness and higher priority, discretionary conservation programs are initiated in the Council's resource portfolio as the region begins to experience electrical power defi- cits. Simultaneously, in most load paths, the region is beginning to secure options on new generating resources, particularly new coal- fired power plants. For this reason, the Coun- cil has decided to size the conservation resource potential based on the expected cost to the region of a new generic coal plant. Although there is substantial uncertainty concerning the cost of this new coal plant, the Council estimates that the cost of a new coal plant will be between 4.0 and 4.5 cents per kilowatt-hour. When these costs are adjusted to take account of the advantages in the Act and transmission system costs and losses, the appropriate range of cost-effectiveness criteria for discretionary conservation pro- grams is between 4.8 and 5.4 cents per kilo- watt-hour. There is very little conservation capability in the range of 5.0 to 5.5 cents per kilowatt-hour, and therefore the Council has selected 5.0 cents per kilowatt-hour as the appropriate cost-effectiveness criterion to size discretionary conservation programs. The Council has also estimated the value to the region of a lost opportunity resource with an expected lifetime of 30 years (as well as other lifetimes). If this resource is acquired in 1986, it has an expected value of 2.5 cents and, alternatively, if it is acquired in 1990, it has an expected value of 3.0 cents per kilo- watt-hour. These values will either increase of decrease, depending on how actual loads develop in the future. For the purposes of acquiring lost opportunity and generating resources over the next several years, the Council recommends that these resources should not be acquired if they cost more than 2.5 to 3.0 cents per kilowatt-hour. In evaluating the MCS, the Council has used two cost-effectiveness measures. The first is the value of marginal MCS investments to the region. The Council has estimated that cur- rently the value of marginal MCS investments is between 4.0 and 4.5 cents per kilowatt- hour. For purposes of defining the MCS at Mills/kWh a 20 30 40 50 60 Resource Life Figure 4-8 70 Value of Lost Opportunities in the Resource Portfolio Chapter 4 CENTS/KWH Heavier shaded area represents range of uncertainty for cost effectiveness. Resources within this range should be treated on a case- 5 by-case basis. 4 3 2 1 0 ST LOST MARGINAL MCS _ DISCRE. DISCRE. 8 OPP. MCS PRO- CONSV. RESOUR. COAL 30YR GRAM AVE Figure 4-9 Comparison of Cost-Effectiveness Criteria 4-15 Chapter 4 this time, the Council has evaluated the indi- vidual measures that are in the range of 4.0 to 4.5 cents per kilowatt-hour and, based on judgment, selected those measures that the Council believes are cost effective and should be included in the MCS. The second test of the cost effectiveness of the MCS was the cost effectiveness of the average energy savings that are acquired in each individual building through a utility MCS program over the next several years. In evaluating the cost effectiveness of average MCS savings in 1986, the Council found that the region could afford to pay, on average, 3.6 cents per kilo- watt-hour for MCS savings. By 1990, this fig- ure will increase to 4.4 cents per kilowatt-hour 4-16 if expected load growth is experienced. For this reason, in evaluating the average cost of the utility MCS program, the Council used a value of 3.6 cents per kilowatt-hour during 1986, and expects this to escalate to approx- imately 4.4 cents per kilowatt-hour by 1990. Finally, in evaluating the cost effectiveness of each individual resource, there are significant non-quantified attributes that must be included in the Council's judgment concern- ing the cost effectiveness and appropri- ateness of each resource included in the plan. In deciding on the cost effectiveness of individual actions, the Council included environmental concerns such as indoor air quality, acid rain, mining impacts, transporta- tion, employment, fish and wildlife, etc. Some of the resources included in the Council's plan will help reduce future load growth uncertainty. Also, some resources are partic- ularly flexible and assist the region in adapt- ing to the wide range of uncertainty it is fac- ing. Finally, due to the significant uncertainty that exists with respect to the cost and avail- ability of each resource included in the Coun- cil’s portfolio, the Council must decide whether sufficient, valid cost and perform- ance information is available on which to make an informed judgment. Chapter 5 Conservation Resources Conservation is a key ingredient in the Coun- cil’s resource portfolio for meeting future elec- trical energy needs. Each megawatt of elec- tricity conserved is one less megawatt that needs to be generated. The Council has identified close to 3,700' average megawatts of conservation in the high demand forecast available at an average cost of 2.4 cents per kilowatt-hour. This is enough energy to replace more than eight coal plants, at about half the cost. Conservation remains an extra- ordinarily cost-effective resource for the region to acquire. This chapter provides an overview of the procedures and major assumptions used to derive the Coun- cil’s estimates of regional conservation resources. In the Council's plan, conservation is the more efficient use of electricity. This means that less electricity is used to produce an amenity level comparable to the one existing before the implementation of the conserva- tion measure. Conservation resources are measures? that ensure new and existing residential buildings, household appliances, new and existing commercial buildings, and industrial and irrigation processes use energy efficiently. For example, buildings that cut down heat loss through insulation and tightening require less electricity for heating. These “savings’ of electricity mean that fewer power plants must be built to meet growing demand. Conservation also includes mea- sures to reduce electrical losses in the region's generation, transmission and dis- tribution system. These latter conservation resources are discussed in Volume Il, Chapter 6. Estimating the Conservation Resource The evaluation of conservation resources involves three major steps. The first step is to develop conservation supply curves. This step entails evaluating the levelized life cycle cost of all conservation measures and rank ordering them with the least-cost measure first. The second step is to group into programs all measures with levelized costs less than a given avoided cost. The avoided cost is the cost of the resource that would be used in the electrical system should conservation not be developed. Avoided cost varies somewhat, depending on the specific characteristics of the conservation program, such as whether the savings from the program can be devel- oped as need occurs or whether it is devel- oped today, during the current surplus. In general, the avoided cost in this plan is the cost of a new coal pliant. The third step involves using the cost and savings characteristics of each program to evaluate the conservation resource's cost effectiveness and compatibility with the exist- ing power system. Cost effectiveness of each conservation program is determined by com- paring the program against other resources to develop a least-cost resource portfolio. The bulk of this chapter deals with steps one and two, which are preliminary cost-effective- ness screens to size the conservation resource that is used in the resource port- folio. Step three is described primarily in the resource portfolio, Volume II, Chapter 8. Supply Curves Conservation supply curves are used to eval- uate the amount of conservation available at given costs. A supply curve is an economic tool used to depict the amount of a product available across a range of prices. Inthe case of conservation, this translates into the number of average megawatts that can be conserved (and made available for others to use) at various costs. For example, an indus- trial customer may be able to recover waste- heat from a process load and conserve 3 average megawatts at a cost of 2 cents per kilowatt-hour. This same customer may con- serve 5, 7 and 8 average megawatts of elec- tricity for the respective costs of 3, 4 and 5 cents per kilowatt-hour. These figures repre- sent the conservation supply curve for this particular customer. Individual conservation estimates for end-uses in each sector are merged to arrive at the regional supply curve for that sector. The supply curves used in this plan do not distinguish between conservation resulting from specific programs and conservation motivated by rising prices of electricity. This is a regional perspective; whether the con- sumer or the utility invests in a conservation measure, the region is purchasing those sav- ings at a particular price. Conservation supply curves are primarily a function of the conservation measure's sav- ings and cost. Each measure's savings and cost are used to derive a levelized cost, in terms of cents per kilowatt-hour, for that mea- sure. The absolute value (in terms of kilowatt- hours per year) of the savings produced by adding a conservation measure is a function of the existing level of insulation. The less efficient the existing structure or equipment, the greater the savings obtained from install- ing the measure. Consequently, the amount of conservation available is directly related to the amount of energy currently used. In order to minimize the costs of efficiency improve- ments, conservation measures are applied with the least costly measure first+ until all measures are evaluated. The levelized costs used to generate the supply curves are based on the capital, operation and maintenance expenditures incurred over the lifetime of the conservation measure. To ensure consistency between the conservation supply curves and the sys- tem models,§ capital recovery factors used in the levelized cost calculation (see Volume II, Chapter 4, for calculation procedure) are the same ones used in the system models. This means that the tax benefits, treatments, rate requirements and other financial considera- tions specific to the developer of the resource are accounted for in the levelized cost of the conservation resource. 5-1 Chapter 5 Conservation was assumed to be financed for 20 years by Bonneville and for the aver- age lifetime of the program by the investor- owned utilities. It was assumed that Bonneville would sponsor 40 percent of the conservation acquisition costs and the investor-owned utilities would sponsor 60 percent based on their share of total loads. Twenty-five percent of the investor-owned utilities share is financed equally between debt and equity, while 75 percent of the investor-owned utilities share is financed by Bonneville. Conservation rams for Portfolio Analysis After the supply curves are generated for each end-use or sector, the amount of con- servation to be used in the portfolio analysis is first sized by cutting off the supply curve at the point where the levelized cost of the last measure included is equal to or just slightly less than the avoided cost. This is called the “technical” conservation potential. The tech- nical potential is then reduced to reflect the portion of the conservation resource that is considered achievable. Achievable conser- vation is the net savings the Council antici- pates after taking into account factors such as changes in consumer behavior, consumer resistance, quality control, and unforeseen technical problems. The Council believes that the wide assortment of incentives and regulatory measures the Act makes available can persuade the region's electric consum- ers to install a large percentage of the tech- nically available conservation. As a conse- quence, the proportion of technical potential considered achievable in this plan varies from 50 percent to 90 percent depending on the sector and the conservation measures. As described in Volume II, Chapter 4, the avoided cost is 5.0 cents per kilowatt-hour for conservation resources that can be sched- uled to meet load. These are called “discre- tionary resources’ because they don't need to be developed during the current surplus. Conservation resources that fit into this cate- gory are based on existing end-uses—for example, commercial retrofit programs and residential weatherization. Residential weatherization is a special case within the discretionary resource category, because this resource is being secured today, even 5-2 though a surplus exists. The avoided cost for residential weatherization measures pur- chased in 1986 is approximately 3.5 cents per kilowatt-hour and increases over time up to 5.0 cents per kilowatt-hour as the surplus nears an end. The residential weatherization program is expected to be reduced to a mini- mum viable level in the near term, and the majority of savings should not be developed until near the end of the surplus. In addition, any weatherization that does occur should be aimed at developing the capability to deliver the full amount of savings when the program is required to ramp-up. Over the next few years, the weatherization program should be aimed primarily at the low income and rental sub-sectors, because capability needs to be developed here. As a consequence of these factors, the Council used the 5.0 cents per kilowatt-hour cutoff to size the weatherization resource in the portfolio. Even so, the vast majority of measures included in the residen- tial weatherization program cost less than 3.5 cents per kilowatt-hour. The 5.0 cents per kilowatt-hour avoided cost also applies to conservation resources that grow automatically with economic develop- ment, but are not expected to be developed until the later years of the forecast, when the region is no longer in a surplus condition. Savings from refrigerators and freezers, not anticipated to be developed until 1992, fall into this category. Resources that fall into this category have lifetimes that are shorter than expected building lifetimes. The avoided cost for conservation resources that grow with loads, have lifetimes longer than the duration of the surplus, and must be acquired today or their savings are lost for- ever is between 4.0 and 4.5 cents per kilo- watt-hour. However, the avoided costs for these resources will increase over time. Sav- ings from the model conservation standards in new residential and commercial buildings epitomize this type of conservation resource. Each conservation program is comprised of the package of measures that cost less than the avoided cost. The present value costs of the achievable savings for each program are adjusted in the following manner before they are used in the system models to determine compatibility with the existing power system and to derive a least-cost resource portfolio. First, since the system models use conserva- tion programs instead of measures in the resource portfolio, capital replacement costs have to be added to those measures with lifetimes shorter than the lifetime of the major measure in the program. For example, caulk- ing and weatherstripping have shorter life- times than insulation; therefore, replacement costs are incurred over the expected lifetime of the insulation to maintain the benefits of caulking and weatherstripping. Consistent with generating resources, these capital replacement costs were escalated at 0.4 per- cent per year for the first 20 years after net- ting out the effects of inflation. Second, in addition to the direct capital and replacement costs of the conservation mea- sures, administrative costs to run the pro- gram must be included in the overall cost. The Council believes that the administrative cost of a given program is generally indepen- dent from the level of measures that the pro- gram installs. For example, the admin- istrative expense of requiring an insulation contractor to install full levels of cost-effective ceiling insulation is no more than if the con- tractor were only required to install half the cost-effective amount. Processing of con- tracts, quality checks, and other admin- istrative actions still need to be taken. The Council reviewed current utility conservation programs and those operated by other agen- cies. This review indicated that conservation program administrative costs range from 10 to 30 percent of the direct cost of measures. As a consequence, the Council has as- sumed a 20 percent administrative cost in its calculations of cost-effectiveness evalua- tions for conservation. This means that the average cost of the conservation programs are increased 20 percent before the conser- vation is compared to other generating resources to determine which is cheaper. As more data becomes available on fully opera- tional conservation programs, the Council will move toward an estimate based on dol- lars per application instead of percent. Chapter 5 A third factor that must be accounted for when comparing conservation programs with other generating resources is the 10 per- cent credit given to conservation in the North- west Power Act. This credit means that con- servation can cost 10 percent more than the next lowest cost resource and still be consid- ered cost effective under the Act. This 10 percent benefit is assessed to all conserva- tion measures. Finally, to ensure that conservation and gen- erating resources are being compared fairly, the costs and savings of both types of resources must be evaluated at the same point of distribution in the electrical grid. Con- servation savings and costs are evaluated at the point of use—in the house, for example. In contrast, the costs and generation from a power plant are evaluated at the generator (busbar) itself. Thus, to make conservation and the traditional forms of generation com- parable, the costs of the generation plant must be adjusted to include transmission system losses (7.5 percent) and transmis- sion costs (2.5 percent). The net effect of all these adjustments is different for the marginal conservation mea- sure than for the average program, because administrative costs are assessed to the average program and not the marginal mea- sure. The cost threshold for investment in the marginal conservation measure is the busbar cost of coal plants, the resource that gener- ally establishes the avoided cost, plus 20 percent—10 percent for the Act's credit, 7.5 percent for transmission system losses and 2.5 percent for transmission costs. The effect on the average cost of conserva- tion programs that are compared to generat- ing resources is to increase the average cost of the conservation programs by 7.5 per- cent—20 percent added for administrative costs minus 10 percent for the Act's con- servation credit and 2.5 percent saved in transmission and distribution costs—and to increase the average savings from the pro- gram by 7.5 percent to account for line loss credits. The adjustments to the average costs and savings from conservation programs were made for purposes of comparing conserva- tion resources with generating resources, as Table 5-1 Conservation Program Assumptions in the Decision Model "VINBLE’ ACCELERATION DECELERATION RATE Transmission + Distribution 0 6% 15% 15% Efficiency Existing Residential 2% 5% 5% 11% Existing Commercial 2% 6% 6% 15% Existing Industrial 0 9% 22% 22% Agricultural 0 5% 10% 10% is done in the models used by the Council to simulate system responses. However, in this chapter, the 10 percent benefit from the Act is not included in the average cost calculations, in order to portray the true cost of conserva- tion programs. As a consequence, the level- ized program costs in this chapter are 10 percent higher than those used in the system models. In addition, this chapter is based on conservation savings at the end-use, so the savings presented are 7.5 percent lower than those used in the resource portfolio. Compatibility with the Power System After these adjustments are made, each con- servation program is evaluated in terms of its compatibility with the existing power system and is compared to the cost and savings characteristics of other electricity resources. To assess compatibility, and ultimately the cost effectiveness of the conservation pro- grams, the Council used two complex com- puter programs, called the Decision Model and the System Analysis Model. These served as a final screen to judge whether a conservation program was regionally cost effective. Both the Decision Model and the System Analysis Model are described fully in Volume II, Chapter 8. The Decision Model determines how much conservation is needed in each of the Coun- cils forecasts. The conservation that the model secures in any one year to meet energy needs depends on how fast a pro- gram can become operational, and on the ultimate amount of cost-effective conserva- tion available. If the region is surplus for a long time, but a conservation program is already operating, the speed at which the program can slow down and the minimum viable level of that program are also impor- tant. The minimum viable level of the pro- gram, if above zero, determines the amount of savings that would accrue even though the region would prefer to delay purchase of the resource during the surplus period. Table 5-1 displays the current conservation assumptions used by the Council in the Deci- sion Model. Maximum acceleration and deceleration represent how fast a program can start up or slow down, while the max- imum rate indicates how fast the program can run once it is up to top speed. Sensitivity analysis, described in Volume II, Chapter 8, considered the impact on the resource port- folio if these assumptions were altered. In general, the base case assumptions lead to the conclusion that it takes about ten years to secure the total average megawatts available from any given program. Residential weath- erization is the conservation program with which the region has the most experience. The values used in the Decision Model for this program reflect levels of weatherization that have been attained in programs operat- ing in the Northwest. If full payment for con- servation were offered, instead of incentives, it is likely that these values would be exceeded. Programs aimed at the model conservation standards, refrigerators and freezers, water heaters and manufactured homes are not included in the table, because their savings are driven by demographic assumptions instead of program operation. For example, once incorporated into building codes, the level of savings achieved from the standard would be driven primarily by the number of electrically heated building starts. 5-3 Chapter 5 Cumulative Average Megawatts 600 500 400 300 200 100 0 a 2 3 4 Conservation Cost (Cents/kWh) 5 6 7 8 Figure 5-1 Technical Conservation Potential from Space Heating Measures in Existing Residences The technical discussion that follows describes the evaluation of conservation resources conducted by the Council. The narrative is illustrated with calculations from the high demand forecast, but similar cal- culations were conducted for all of the Coun- cils forecasts. All costs are in 1985 dollars. This discussion, and the technical exhibits listed at the end of each sector, provide the capital cost data, energy savings, and mea- sure life used by the Council. Bonneville is expected to use comparable assumptions and procedures in any calculation of cost effectiveness. Residential Sector In 1983, the region's residential sector con- sumed 5,216 average megawatts of elec- tricity, which is about 36 percent of the region's total electrical consumption. Space heating is the largest single category of consumption in the residential sector; water heating is second. 5-4 Space Heating Conservation in Existing Residential Buildings Figure 5-1 shows the estimated space heat- ing savings available from existing resi- dences at various electricity prices. The technical conservation available with no sin- gle measure exceeding 5.0 cents per kilo- watt-hour is 500 average megawatts. The Council's plan calls for developing up to 85 percent of the technical potential, or about 425 average megawatts. This represents a 37 percent savings in heating loads by 2005. The estimated average cost of insulating and weatherizing existing residences is about 2.9 cents per kilowatt-hour. The Council's assessment of the conserva- tion potential for existing space heating involved four steps. These steps were to: 1. Estimate the cost and potential savings available from improving the thermal efficiency of existing electrically heated dwellings. 2. Develop conservation supply functions that are consistent with the Council's fore- casting model. 3. Compare projected cost and savings esti- mates with historically observed cost and savings data. 4. Estimate realizable conservation potential. Step 1. Estimate the costs and savings from conservation measures. The costs and savings of conservation measures are the primary determinants of the amount of conservation that is available from the supply curves. The Council's estimates of single family weatherization costs are based on information gathered for the 1983 Power Plan and on information provided by Bonneville and utilities on the costs of weatherizing resi- dences. These costs are shown in Table 5-2. Costs from the Hood River Project are prelim- inary, and were only used for triple glazed windows. Costs for caulking and weath- erstripping were taken from Bonneville’s weatherization program and increased to $100 per house. The costs of weatherizing multifamily units are based on the information gathered in the 1983 plan. Cost-effective savings from space heating heat pumps in manufactured homes, estimated to be a total of about 35 average megawatts in the 1983 plan, were not included in the current supply curve in order to reflect some unanswered engineering questions about retrofitting the heat pumps to existing ductwork. The Council used the source in Table 5-2 with the largest sample size to estimate con- servation measure costs for single family houses. The information provided by Bon- neville and Puget Power reflects the cost of installing a conservation measure where that measure is the only insulation installed in that particular building component. This cost con- sequently carries a fixed cost portion with it. For example, Puget's cost of $0.48 per square foot if ceiling insulation is increased from R-19 to R-38 is for precisely that mea- sure added to the ceiling. Had the contractor started with R-11, part of the fixed cost embodied in the $0.48 per square foot would be spread over the costs from R-11 to R-38 instead of just from R-19 to R-38. The fixed cost per R-value added is less, the more insulation is installed. In this discussion, costs that include the fixed portion will be called the “set-up” costs of the weatheriza- tion measures. In Table 5-2, prices that incor- porate the “set-up” costs appear in columns marked with a “2”. Chapter 5 Information from the 1983 Power Plan reflects both the “set-up” costs of insulation measures, and the cost of installing an incre- mental measure of insulation, assuming the contractor is already laying the base insula- tion in that building component. In this dis- cussion these will be called “add-on” costs. In Table 5-2, “add-on” costs appear in col- umns marked with a “1.” It is useful to distinguish between these two types of costs to answer two different ques- tions. “Set-up” costs are included when determining whether any insulation should be added to a building component, given that a certain level already exists. For example, if a ceiling is already insulated to R-38, it turns out that it is not cost effective to the region to Pay for a contractor to come to the house and increase the ceiling insulation level to R-49. “Add-on” costs determine how far a building component should be insulated, assuming the contractor is already set up and has installed some base insulation. It turns out, for example, that it is cost effective to set up a contractor to increase ceiling insulation to R-30 from a base of R-19, and it is also cost effective to continue adding insulation to R-49, if the contractor is already there. Thus the regional cost-effectiveness limit is R-49 if R-19 is the base insulation. Based on the current analysis as described below, the following measures cost less than 5.0 cents per kilowatt-hour: R-49 ceiling insulation if the house has less than R-30; R-11 wall insulation if no insulation currently exists; R-30 underfloor insulation if less than R-19 currently exists; and triple pane win- dows if single panes are present, but not if the windows are already double paned. The cur- tent analysis indicates that if the house is already at R-30 in the ceiling, has some wall insulation, has R-19 or more in the floor and double pane windows, it is not cost effective to weatherize further. These results have important implications for the design of weatherization programs. For example, if a utility runs a weatherization pro- gram that takes the ceiling insulation to R-30 only, the savings from going beyond R-30 are lost to the region, even though it would have been cost effective to go further at the time the house was weatherized. Additionally, these results lead to a weatherization pro- gram design that could be modeled after the Table 5-2 Costs of Weatherizing Single Family Houses (In Dollars Per Square Foot of Component Area—Sample Size = N) SETUP? SET ups? PLAN DD ON® pugeTupeS seTUP® FADD ON Ceiling Insulation RO-R19 _ $0.39(N = 23) — — $0.39 — R19-R38 = $0.39(N=23) $0.22(N=23) $0.48(N=205) $0.48 $0.22 R30-R38 a $0.26(N=23) $0.09(N=23) $0.50(N=6) $0.26 $0.09 R38-R49 a $0.30(N=23) $0.13(N=23) _ $0.30 $0.13 RO-R38 = $0.82(N = 30) _ ~_ $0.79(N = 58) a - Floor Insulation RO-R19 — $0.78(N=36) — $0.58(N=23) _ $0.72(N=1792) $0.72 _ R19-R30 -_ $0.44(N=23) $0.24(N=23) _ $0.44 $0.24 Wall Insulation RO-R11 $0.60(N=2) — $0.63(N=23) _ $0.51(N=270) $0.51 = Doors = $14.05 — = $14.05 = Caulking and Weatherstripping $78/house _ a _ $100/ _ (N=893) house Glass HOOD RIVER PROJECT Add double panes to single $11.59(N = 157) _- $11.59 _- Add single pane to single _ $8.04(N=1522) $8.04 Incremental cost of adding insulation, assuming the contractor is already installing insulation for that building component. >Cost of adding insulation, assuming the contractor is not installing any other insulation in that building component. oil dipstick in a car. If an audit shows that the house already has R-30 in the ceiling, it is only half a quart low and no oil—that is, insulation—should be added. On the other hand, if the audit shows that the ceiling is only at R-19, it is a full quart low, and insulation should be added to the full cost-effectiveness level of R-49, or as close as structural bar- riers permit. Three “typical” building designs were used to estimate the retrofit potential for single family houses in the region. The first is an 850 square foot single-story house built over an unheated basement. The second is a 1,350 square foot house over a vented crawl space, and is similar to the design used in the 1983 plan. The third is a 2,100 square foot two- story house with a heated basement. The multifamily design is a three-story apartment house with four 840 square foot units on each floor. Savings from weatherization measures installed in all four house designs were esti- mated using the SUNDAY computer model, which simulates a building's daily space heat- ing energy needs.® Savings were evaluated using regional average indoor temperature settings, and internal gains consistent with efficient appliances included in the Council's resource portfolio. Savings from insulation measures were evaluated assuming that consumers who now operate their un- weatherized houses at reduced tem- peratures would raise thermostat settings fol- lowing weatherization. (This practice is termed “take-back,” and it reduces savings.) The Council assumed that the worst case for the electrical power system would be if consumers who currently heat their unweatherized houses partially with wood, and partially with electricity, chose not to 5-5 Chapter 5 Costs and Savings of Single Family imate Measures in Zone 3— Missoula Costs and Savings of Single Family este Measures in Zone 2— Spokane LEVELIZED LEVELIZED CAPITAL COST ANNUAL USE COST (85$) CAPITAL COST ANNUAL USE COST (85$) MEASURE uae Total Sisq ft kWhiyr kWh/sq ft mills/kWh MEASURE UA Total ‘$/sq ft kWhiyr kWh/sq ft miliskWh HOUSE SIZE — 850 SQ FT HOUSE SIZE — 850 SQ FT Base Case 674 $ 0 $0.00 29,297 34.5 0 Base Case 674 $ 0 $0.00 25,246 29.7 ° Ceiling 0 to R19 483 $ 332 $0.39 19,068 Ceiling 0 to R19 483 $ 332 $0.39 16,332 19.2 1.60 = 140 Walls 0 to R11 397 $ 813 $0.96 «12,472 14,7 5.38 Walls 0 to R11 397 $ 813 $0.96 14621 172 467 Crantapace 010 R18 i as aide G50 109 as Crawispace 0 to R19 327 $1,425 $1.68 10.939 12.9 747 ACH ees oe Sis uz. lees 7 ‘178 ACH? 60.4 302 $1,525 $1.79 9,724 11.4 10.34 Gonig news 288 «$1712 $201 7,604 89 1282 Ceiling R19 to 30 288 «$1,712 $2.01 9,007. ~S 106 11.25 Coating 201098 ie Sires aS a au Ceiling R30 to 38 285 $1,789 $2.11 8,838 10.4 19.78 Single to Triple Glass 229 $2,879 $3.39 5,119 60 26.40 Single to Triple Glass 29 «$2879 $3.90 6,171 73 23.14 Crawlspace R19 to 30 221 $3,083 «$3.63 «4,806 57 28.17 Crawispace R19 to 30 221 $3,083 $3.63 «5,813 68 24.58 eae coe sas I20) aio 54 74031 Wood to Metal Door 209 $3,645 $4.29 5,253 62 126.04 HOUSE SIZE — 1,350S0 FT HOUSE SZ 1200 S00K Base Case 1043 $ 0 $0.00 41,598 30.8 ° Bass Cons 1043 $ 890 = =$0.00 48,217 S67 7 Ceiling 0 to R19 740 $ 527 $0.39 27,294 20.2 1.59 Comade nis) 8 Sr Shi 288 “= Walls 0 to R11 607 607 1,199 $0.84 21,145 18.7 4.29 Walls 0 to R11 607 $1,199 $0.84 24,769 18.3 3.73 icons see 1200) soma) tose? 148 oar ees eae Size eee; ee GS = Crawispace 0 to R19 456 $2,211 $1.64 14,900 106 8.28 Cramispace 0 0 R19 46 $2,211 : * 64 16,809 125 Ceiling R19 to 30 433° © $2,508 «$1.86 = ‘13,282 98 12.59 Ceiling R19 to 30 433 $2,508 $1.86 15,632 11.6 10.88 Colling R20 t0 38 ae S200) (S186; 19,089 oe, a Ceiling R30 to 38. 428 $2,630 $1.95 15,356 11.4 19.22 Single to Triple Glass mm Me Se Ir id =e Sirois in Wels Ginss 340 $4357 $323 10.891 84 oii Crawispace R19 to 30 328 $4,681 $3.47 «8,651 64 26.86 Crawlspace R19 to 30 328 $4,681 $3.47 10,287 76 23.17 ee en “= 2 Se 8S - — Wood to Metal Door 316 $5,243 $3.88 9,697 72 119.60 —— olen soe --2 W000 FT Base Case 1,208 $ 0 $0.00 43,904 209 o Base Case ioe is 0 ono, sii aa ° Ceiling 00 R19 1,051 $ 273 $013 36620 174 1.62 Celing 0 RI joan 16 os) ois: mors) on 140 ACH 610.4 985 $373 «$0.18 33.619 ~—:16.0 4.19 cH Awa oe srs, Gots, a> 407 ae Walls 0 to R11 787 $1,327 $0.63 24,668 11.7 4.60 Walls 0 to R11 787 $1,327 $0.63 28,964 13.8 3.97 Ceiling R19 to 30 775 $1,481 $0.71 24,148 11.5 12.78 Ceiling RA19 to 30 775 ($1,481 $0.71 —-28,366 13.5 11.12 Coling R30 10 38 ate Stee Sore: (etee uA Zs Ceiling R30 to 38 772 ($1,544 $0.74 28,226 13.4 19.55 ‘Single t0 Triple Glass G86. 95/108) S247 | 15,928 78 =A Single o Tipe Giese a ee. Sar eas i 2008 Wood to Metal Doors 574 $5,757 $2.74 15,425 73 140.13 Wood to Metal Door 574 $5,757 $2.74 18,346 a7 122.48 *Heat loss rate (U-value x area) Air changes per hour use their wood stoves after weatherizing. If the regional cost-effectiveness level were optimized assuming the wood heating use, the house would not be optimally weath- erized if wood heating were slowed or discon- tinued. In order to plan for loads that the electrical system could potentially bear, cost effectiveness for each measure is evaluated assuming no wood heat. 5-6 It is also important to note, however, that the Council used the forecasting model to derive space heating use in unweatherized houses, as described below. This forecasting value is used as the base from which savings are calculated. Since the forecasting figures reflect wood heating use and room closures, as well as other responses and behaviors, the total megawatts of weatherization conser- vation available to the region have been reduced to account for average consumer behavior. Consequently, the average mega- watts available from weatherization consider the effects of wood heat use. Tables 5-3 through 5-6 show the costs, level- ized in mills (tenths of a cent) per kilowatt- hour, and the savings from weatherizing the typical design houses in three representative climate zones in the region. Each measure has its own average, or expected, lifetime. Insulation lasts the lifetime of the residence, which for existing stock is on average about 60 or more years. This was reduced to 50 years to reflect a potential loss of savings from sagging or settling. Storm windows are assumed to last on average about 30 years. Storm doors are assumed to last an average of ten years. Chapter 5 Table 5-5 Table 5-6 Costs and Savings of Single Family Weatherization Measures in Zone 1 — Seattle Costs and Savings of Multifamily Weatherization Measures MEASURE vA Totl Seq kWhyyr kWiveq ft millekWh MEASURE iWhiyr_KWhisq ft = COST COST —_milisykWh = COST/SQ FT HOUSE SIZE — 850SQ FT ZONE 1— PORTLAND Base Case 674 $ 0 $000 18495 218 0 Base Case 9865 1174 8$ 0 STre ° $0 Coiling 0 to R19 483 $ 332 «$039—«11,465 135 203 Walls 0 to R11 7,570 9.01 $342 $ 342 6.43 $0.41 Walls 0 to R11 397 $ 813 $0.96 8,488 10.0 6.98 Ceiling 0 to R38 5,033 5.99 $625 $ 967 10.63 $1.15 Crawispace 0 to R19 327 «$1,425 $1.68 6,074 7A 10.94 Floor 0 to R38 344900 4.11 ‘$625 $1,592 17.03 $1.90 ACH 610.4 302 $1,525 «$1.79 «5,280 62 15.82 ACH 6 to.4 2,797 333 $100 $1,692 19.27 $2.01 Ceiling R19 to 30 288 «$1,712 $2.01 4,812 57 17.22 Single to DoubleGlass 1,602 1.91 $959 $2,651 45.45 $3.16 Ceiling R30 to 38 285 «$1,789 $2.11 4,702 55 30.38 Double toTripleGlass 1,243 148 «= $922 $3,573 145 $4.25 Single to Triple Glass 229 $2,879 $3.39 2,986 35 35.96 Insulated Door 1,121 1.33 $147 $3,720 151 $4.43 Crawispace R19 to 30 221 $3,083 $3.63 2,756 32 38.36 ZONE 2— SPOKANE Wood to Metal Doors 209 «($3,645 «$4.29 2,407 28 202.01 Base Case 14985 178 = $ (0 ST re ° so HOUSE SIZE — 1,350 SQ FT Walls 0 to R11 11,631 1385 $342 $ 342 4.40 $0.41 Gace Cans 1,043 $ © $0.00 31,469 233 ° Ceiling 0 to R38 7,916 942 $625 $ 967 7.26 $1.15 Ceiling 0 to R19 740 $ 527 $0.39 20,124 149 2.00 Floor 0 to R38 5,546 660 $625 $1,592 11.38 $1.90 Walls 0 to R11 607 $1,139 $0.84 «15,262 "3 5.43 ACH 6 to.4 4544 5.41 $100 $1,692 12.54 $2.01 ACH 610.4 568 $1,239 $0.92 13,857 103 8.95 Single toDoubleGlass_ 2,703 3.22 = $959 $2,651 29.50 $3.16 Crawispace 0 to R19 456 $2,211 $1.64 9,880 73 10.55 Double to TripleGiass 2,134 254 = $922 $3,573 91.78 $4.25 Ceiling R19 to 30 433 «$2,508 $1.86 9,099 67 16.42 Insulated Door 1,943 231 $147 $3,720 96.70 $4.43 Ceiling R30 to 38 428 © $2,630 $1.95 8,917 66 29.09 ZONE 3— MISSOULA Single to Triple Glass 340 «$4,357 $3.23 5,974 44 33.24 Base Case 16,784 19.98 so $ 0 ° $0 Crawispace R19 to 30 328 © $4,681 $3.47 5,582 44 35.65 Walls 0 to R11 12,911 15.37 $342 $ 342 3.81 $0.41 ‘Wood to Metal Doors 316 $5,243 $3.88 = 5,204 39 186.80 Ceiling 0 to R38 8,577 10.21 $625 $ 967 6.22 $1.15 HOUSE SIZE — 2,100 SQ FT Floor 0 to R38 5,837 695 $625 $1,592 9.84 $1.90 Bane Casa! 1,208 $ 0 $0.00 32,440 154 o ACH 6 to 4 4680 557 $100 $1,692 10.86 $2.42 Ceiling 0 to R19 1,051 $ 273 $0.13 26,701 127 2.05 Single to DoubleGlass 2,545 3.03 $959 $2,651 25.44 $3.57 ACH 60.4 985 $ 373 $0.18 24,348 11.6 5.34 Double to TripleGiass 1,880 224 = $922 $3,573 78.53 $4.25 Walls 0 to R11 787 $1,327 $0.63 «17,430 83 5.95 Insulated Door 1,658 1.97 $147 $3,720 83.20 $4.43 Ceiling R19 to 30 775 $1,481 $0.71—‘17,034 81 16.81 Ceiling R30 to 38 772 «$1,544 «$0.74 16,941 81 29.55 Single to Triple Glass 586 $5,195 $247 10,839 52 33.88 Wood to Metal Doors 574 $5,757 $2.74 10,633 51 342.59 The levelized costs displayed in tables 5-3 through 5-6 are used to order the conserva- tion measures in terms of the least-cost mea- sure first. For this screening exercise, which is used to rank order the measures, no replacement costs or savings were assumed for the measures with short lifetimes. When average costs of the entire program were evaluated to be used for comparison with other resources in the resource portfolio, all measures costs and savings were taken to the same lifetime as the life for the major measure in the program (in this case, insula- tion). For example, caulking and weath- erstripping were assumed to last ten years, while insulation would last the lifetime of the house or about 50 years. For average pro- gram costs, caulking and weatherstripping would incur an initial cost and replacement costs every ten years until 50 years is reached. The levelized cost calculation allows the application of measures with the least-cost measure first; savings for residual measures are reassessed after each measure is added. Since each measure saves a different amount of energy in each house design and location, an aggregate supply curve must be developed to represent the weighted average savings for all measures in the dwelling types. Accordingly, the savings from each climate zone were combined according to the percentages listed in Table 5-7. For each savings and cost appear in Tables 5-8 and 5-9. 5-7 Chapter 5 Weights Used to Reflect epee for Existing Space Heating CLIMATE ZONE 1 CLIMATE ZONE 2 CLIMATE ZONE 3 Single Family 84% 11% 5% Multifamily 73.1% 22.1% 4.8% Table 5-8 Regionally Weighted Costs and Savings of Single Family Weatherization Measures COST OF SAVINGS USE MEASURE mills/kWh S/sq ft kWh/sq ft 850 SQUARE FOOT HOUSE Base Case 0 $0.00 23.3 Ceiling 0 to R19 2.0 $0.39 14.6 Walls 0 to R11 6.7 $0.96 10.9 Crawlspace 0 to R19 10.5 $1.68 7.8 ACH .6 to .4 15.1 $1.79 6.9 Ceiling R19 to 30 16.4 $2.01 6.3 Ceiling R30 to 38 29.0 $2.11 6.1 Single to Triple Glass 34.3 $3.39 4.0 Crawlspace R19 to 30 36.5 $3.63 3.7 Wood to Metal Door 191.9 $4.29 3.2 1,350 SQUARE FOOT HOUSE Base Case 0 $0.00 24.8 Ceiling 0 to R19 1.9 $0.39 15.9 Wall 0 to R11 5.2 $0.84 12.1 ACH .6 to .4 8.6 $0.92 11.0 Crawlspace 0 to R19 10.1 $1.64 7.9 Ceiling R19 to 30 15.7 $1.86 7.3 Ceiling R30 to 38 27.8 $1.95 7.2 Single to Triple Glass 31.8 $3.23 4.9 Crawlspace R19 to 30 34.1 $3.47 4.6 Wood to Metal Door 178.2 $3.88 43 2,100 SQUARE FOOT HOUSE Base Case 0 $0.00 16.5 Ceiling 0 to R19 2.0 $0.13 13.6 ACH .6 to .4 5.1 $0.18 12.4 Wall 0 to R11 5.7 $0.63 9.0 Ceiling R19 to 30 16.1 $0.71 8.8 Ceiling R30 to 38 28.3 $0.74 8.7 Single to Triple Glass 32.4 $2.47 5.6 Wood to Metal Door 309.3 $2.74 5.5 5-8 The costs of upgrading single pane windows to double and double to triple panes were also evaluated, but do not appear in the tables for single family houses in order not to overcount savings. Single to double upgrad- ing was cost effective for single family houses, but not double to triple, unless the action was part of a one-step upgrade from single. Consequently, only the one step, from single to triple, appears in the tables in order not to count savings from both single to dou- ble and single to triple glazing. In addition, although insulating the ceiling to R-49 was cost effective for single family houses, it was not included in the analysis in order to repre- sent some of the savings lost due to struc- tural barriers. Consequently it also does not appear in these tables, and was not counted as part of the total regional weatherization potential. Step 2. Develop conservation supply functions that are consistent with the Council’s forecast. The Council's supply function for conservation in existing residen- tial buildings was developed for the year 2005. This was done for three reasons. First, the supply of energy available through con- servation in existing buildings is constrained by the rates at which measures can be imple- mented. Second, these rates are con- strained by the need for additional energy supplies. Third, some existing houses will be torn down by the turn of the century. As a result, the conservation savings from existing buildings diminishes with time. By develop- ing its retrofit supply function for the year 2005, the Council was able to account for demolitions and set deployment schedules based on the need for additional supplies. The forecast model was used to determine the number of electrically heated houses built before 1979 that would survive to 2005 and that could be retrofitted. The number of houses known to be retrofitted by utility-spon- sored programs throughout the region, as well as an estimate of the number of house- holds that installed retrofit measures on their own, was subtracted from the number surviv- ing. This calculation resulted in 678,200 sin- gle family electrically heated houses and 281,700 multifamily units that could still be weatherized. Houses built to current practice between 1979 and 1985 are not included as weatherization potential. Current practice Chapter 5 houses represent a lost conservation oppor- tunity because they are insulated well enough that weatherization is not cost effec- tive, yet they are not insulated to the level cost effective for new homes. The cost and savings for each of the three single family houses were merged to esti- mate regional conservation potential by cents per kilowatt-hour. Based on the Pacific Northwest survey, the 2,100 square foot, 1,350 square foot, and 850 square foot houses represented approximately 22, 46, and 32 percent respectively of the regional stock. These weights result in an average house size of 1,355 square feet. Table 5-10 and 5-11 show the curve of regionally weighted costs and savings for single family and multifamily houses. Savings from this curve would be multiplied by the number of eligible units to derive a supply curve that represents regional potential if all houses in the region were uninsulated. However, the vast majority of houses in the region, even those that are not retrofitted, already have some insulation. Therefore, the supply curve for remaining savings cannot be taken from the uninsulated case, but must be estimated based on the average energy consumption or average existing insulation levels in the eligible stock. The ideal solution to this problem would be to know the actual measures already existing in unretrofitted houses so that conservation potential could be directly determined. How- ever, there is currently no reliable data base of such information. The Council relied on its demand forecast to estimate the space heat- ing use of pre-1979 stock that had not been retrofitted by 2005. The number of existing 1979 stock houses that have not yet been retrofitted was estimated. Then the model was run until 2005, allowing removals from this stock to occur and allowing all variables to change except the efficiency level of the shell. Unweatherized single family houses surviving in 2005 are forecast to use about 11,047 kilowatt-hours per year and multi- family units about 5,421 kilowatt-hours per year. These forecasting figures reflect insula- tion levels, wood heating use and room closures, as well as other consumer responses. By using the forecasting figures as the base case from which average mega- watt savings are derived, the Council's weath- erization potential automatically accounts for Tal Regionally Weighted Costs and eee omen Weatherization Measures USE COST OF SAVINGS MEASURE kWh/yr sq ft/yr mills/kWh Cumulative $/sq ft Base Case 11,329 13.49 0 $ 0 Walls 0 to R11 8,724 10.39 5.67 $0.41 Ceiling 0 to R38 5,840 6.95 9.35 $1.15 Floor 0 to R38 4,027 4.79 14.87 $1.90 ACH .6 to .4 3,273 3.90 16.67 $2.01 Single to Double Glass 1,891 2.25 39.28 $3.16 Double to Triple Glass 1,471 1.75 124 $4.25 Insulated Doors 1,329 1.58 130 $4.43 Table 5-10 Regionally Weighted Single Family Weatherization Savings by Cost Category 0 $0.00 22.5 30,436 $ O 5 $0.65 12.3 16,727 $ 886 10 $1.38 8.3 11,312 $1,870 15 $1.57 75 10,212 $2,125 20 $1.68 7.3 9,839 $2,275 25 $1.71 7.2 9,773 $2,318 30 $2.29 6.2 8,374 $3,107 35 $3.25 46 6,199 $4,403 40 $3.31 45 6,103 $4,491 45 $3.33 45 6,090 $4,510 50 $3.34 45 6,077 $4,530 55 $3.36 45 6,063 $4,550 60 $3.37 45 6,050 $4,569 65 $3.39 45 6,037 $4,589 70 $3.40 44 6,024 $4,609 reduced conservation potential from current use of wood heat and room closures. Since forecasting figures are for the year 2005, they incorporate the level of internal gains that results from the appearance of efficient appliances between 1985 and 2005. Conse- quently, no adjustment was necessary for consistency of assumptions about internal gains between the forecast and the individual measure cost-effectiveness evaluations. The weatherization conservation potential available to the region is the difference between the forecast use and the use after all cost-effective measures have been installed. Tables 5-10 and 5-11 show that houses retro- fitted to the regional cost-effectiveness limit of 5.0 cents per kilowatt-hour use about 6,077 kilowatt-hours per year (kWh/yr) for single family (SF) and 1,838 kilowatt-hours per year for multifamily (MF). The total tech- nical potential can be calculated: 5-9 Chapter 5 Regionally Weighted Sen eee Savings by Cost Category mn Soir coneg A cy ouayar® 0 $0.00 13.49 11,329 $ 0 5 $0.36 10.75 9,030 $ 302 10 $1.24 6.70 5,628 $1,040 15 $1.90 4.73 3,974 $1,599 20 $2.18 3.65 3,070 $1,833 25 $2.43 3.29 2,764 $2,045 30 $2.69 2.93 2,458 $2,257 35 $2.94 2.56 2,152 $2,469 40 $3.17 2.25 1,887 $2,659 45 $3.23 2.22 1,862 $2,713 50 $3.29 2.19 1,838 $2,767 55 $3.36 2.16 1,813 $2,821 60 $3.42 2.13 1,788 $2,876 65 $3.49 2.10 1,764 $2,930 70 $3.55 2.07 1,739 $2,984 75 $3.62 2.04 1,714 $3,038 80 $3.68 2.01 1,689 $3,093 Table 5-12 Technical Conservation from Existing Space Heating LEVELIZED CUMULATIVE TECHNICAL POTENTIAL (AVERAGE MEGAWATTS) (contexWh) Single Family Multifamily 1 40 10 2 93 75 3 207 95 4 383 113 5 385 115 6 387 117 7 389 118 678,200 SF households x (11,047 - 6,077 kWh/yr) = 385 average 8,766,000 (kilowatt-hours per average megawatt) megawatts PLUS 281,700 MF households x (5,421 - 1,838 kWh/yr) = 115 average 8,766,000 (kilowatt-hours per average megawatt) megawatts Some eligible houses do not have any insula- tion while others have significant amounts. Thus the supply curve generated from a completely uninsulated house needs to be reduced from the uninsulated base case to a total potential of 500 average megawatts to 5-10 reflect current levels of insulation in eligible houses. The supply curve was reduced by eliminating the cheapest measures, which are assumed to be installed already in exist- ing houses. This was done by excluding the savings and costs between the uninsulated house and the level of consumption pre- dicted by the demand forecast. The adjusted conservation supply function for residential space heating in existing buildings is shown in Table 5-12. Step 3. Compare cost and savings esti- mates with observed cost and savings. The Council compared its estimates of pro- jected energy savings and costs with those observed in current utility weatherization pro- grams. Figure 5-2 shows the relative space heating energy use of electrically heated homes before and after they were retrofitted. It also shows the expenditures to achieve those savings. The curve of this graph repre- sents the Councils estimates of costs and associated savings from weatherizing single family households, based on the models and inputs described above. The plotted points depict utility program experience. The Coun- cil's cost and savings estimates generally agree with existing utility program experi- ence, in terms of relative performance and cost. The principal assumption made in plotting the observed bill changes is that pre-weath- erization space heating electricity use repre- sents the actual thermal efficiency of the house, and is not due to factors such as room closures and wood heat use. The alternative assumption would have been to assume that differences in the observed use pre- and post-weatherization were due to occupant behavior such as lowering thermostat tem- perature, closing off rooms and using wood heat before the house was weatherized. Sur- vey data reveal significant use of wood heat and room closures in unweatherized houses. However, this assumption was not adopted. A lower estimate of the available technical potential for conservation is produced by the assumption that observed use represents the actual thermal efficiency of buildings. It implies that some conservation measures have already been installed that, indeed, may not be in place. Rather than include this tech- nical potential in its assessment of retrofit savings, the Council conservatively assumed that consumers who now operate their unweatherized houses at reduced tem- peratures would raise thermostat settings following weatherization. (This is termed “take-back;” it reduces conservation sav- ings.) Subsequently, the actual bill changes (i.e., savings) expected by the Council assume that consumers will discontinue their use of wood, reopen closed-off rooms and increase thermostat settings. Chapter 5 Step 4. Estimate realizable conservation potential. The final step in the Council's assessment of retrofit potential was to Relative Annual develop an estimate of the share of the 500 Space Heating Use average megawatt potential that could real- Be posts istically be achieved over the next 20 years. 1, @ O FPA Weatherization Given the tools to secure conservation under the Northwest Power Act, the Council esti- mated that 85 percent of the technical poten- 8 tial is achievable. For example, the Hood River Project, which paid fully for all weath- erization measures offered to every house in 6 the community of Hood River, Oregon, and prior weatherization programs operated in the community secured weatherization sat- 4 urations that are similar to this figure. The 15 percent reduction accounts for less than complete market penetration, unanticipated a building barriers beyond those already cred- ALA Pacific Power & Light He VX Puget Sound Power & Light eo Washington Water Power BPA /Oakridge National Lab use use VW V ~~ Seattle City Light 1 f Seattle City Light 2 @< r © Portland General Electric a ited in the estimate, and quality control. The - Idaho Power energy savings available in the Council's plan under its high growth forecast are 425 aver- 0 sea 2,000 3,000 4,000 5,000 6,000 Retrofit Cost age megawatts (500 x 0.85). 1985 Dollars Space Heating Conservation in i New Residential Buildings l Figure 5-2 % i i Comparison of Regional Thermal Integrity Curve Figure 5-3 shows the technical space heating , : : savings available from new residences at Estimated Cost and Savings Compared to Observed Bill Changes various costs. New single family homes rep- in Existing Utility Weatherization Programs resent approximately 770 average mega- watts of technical potential. Multifamily and Cumulative manufactured homes each represent Mecutote approximately 90 average megawatts of technical potential. The Council's plan call for 4,500 il developing 610, 70 and 45 average mega- watts of the technical potential as achievable for single family, multifamily and manufac- tured homes respectively. The total achiev- able conservation potential saves 48 percent of new space heating loads in 2005. The average cost of improving the thermal effi- ciency of new buildings is about 3 cents per kilowatt-hour. Making new houses more efficient is a high priority for securing a least-cost energy future for the region. It is important to insulate houses fully at the time they are built, or cost- effective savings can be lost. In addition, while the number of houses eligible for retro- 3 4 5 6 fitting will diminish over time, the number of Conservation Cost (Cents/kWh) applications that conservation can reach in new houses continues to grow as every new house is built. Figure 5-3 Technical Conservation from Space Heating Measures in New Residences 5-11 Chapter 5 The conservation potential available through improvements in the energy efficiency of new residential buildings was developed in five steps. These steps were to: 1. Establish the characteristics of current new residential construction. 2. Develop construction cost estimates for space heating conservation measures in new dwellings. 3. Assess the cost effectiveness of space heating energy savings produced by effi- ciency improvements in new residential buildings. 4. Estimate the technical potential available from space heating energy conservation in new dwellings. 5. Estimate the achievable conservation potential available from space heating energy conservation in new dwellings. Separate estimates were prepared for single family dwellings (up to four units and less than four stories), multifamily dwellings (five- plex and larger) and manufactured housing (e.g., mobile homes—please see glossary). A description of each of these steps, the data and major assumptions used and their sources follows. Step 1. Establish the characteristics of new residential construction. To deter- mine the potential for improving the energy efficiency of new residential structures it was first necessary to establish their current level of efficiency. In addition to identifying the level of insulation and type of windows commonly installed in new housing, other new home characteristics had to be ascertained, such as average floor area heated, number of sto- ries, window area, “tightness” of the dwelling and foundation type. These characteristics significantly affect the amount of energy needed for space heating. Table 5-13 shows by climate zone and build- ing type the “base case” insulation levels assumed by the Council in its assessment of space heating conservation potential in new dwellings. The information on new single family and multifamily housing charac- teristics shown in this table is derived from three sources. The first was a regional resi- 5-12 dential energy survey conducted for Bon- neville in 1983 (Pacific Northwest Residential Energy Survey 1983, “PNRES '83”). This sur- vey was used to estimate the average size of new dwellings. The second data source was the 1977 through 1983 annual survey of new home characteristics prepared by Housing Industry Dynamics (HID) for Bonneville. The HID survey data was used to determine the typical glass area and foundation types found in new dwellings. For those areas in the region which enforce an energy code, the requirements of such codes served to estab- lish the minimum thermal efficiency levels found in typical new single family and multi- family dwellings. For areas where no energy codes are enforced or where the HID survey data indicated that prevailing practice exceeded current code, the Council as- sumed the level of construction indicated by the survey. The base case characteristics for new man- ufactured housing, shown in Table 5-13, were derived from information submitted to the Council by the Manufactured Housing Institute. The insulation levels assumed reflect the requirements of the U.S. Depart- ment of Housing and Urban Development's rules concerning the eligibility of manufac- tured homes for mortgage insurance under Title Il of the National Housing Act. Once the general characteristics of new dwellings had been identified, “typical” build- ing designs were developed for detailed anal- ysis of space heating conservation potential. Three typical single family detached dwelling designs were developed to represent the mixture of house sizes and foundation types being constructed in the region. A single mul- tifamily building design was chosen to repre- sent new multifamily construction larger than four-plexes. Two manufactured home designs were selected to represent those typically being sold in the region. Table 5-14 summarizes the basic characteristics of the new dwellings used in the Council's assess- ment. These designs were selected as repre- sentative based on features related to their space heating requirements, such as founda- tion type, and not on the basis of their archi- tectural styles. Step 2. Develop construction cost esti- mates for space heating conservation measures in new dwellings. In the devel- opment of the 1983 plan, the Council con- ducted an extensive survey of conservation costs in new residential buildings. Pursuant to the Council's plan, Bonneville, in coopera- tion with the four Northwest states, initiated a regionwide demonstration program on energy efficient new home construction called the Residential Standards Demon- stration Program (RSDP). The Council has analyzed approximately 75 percent of the cost reports submitted by builders in this pro- gram. Except for one measure, infiltration control with mechanical ventilation, the median costs reported by participating builders agreed with those used by the Council in the 1983 plan. Consequently, for all measures except infiltration control with mechanical ventilation, the Council used RSDP median cost in its cost-effectiveness analysis. It appears that the principal reasons for the difference between the Councils estimated cost for infiltration control with mechanical ventilation and the cost reported by builders stem from the limited experience the builders have with the measure and the lack of a competitive market. In Tacoma, where infil- tration control measures coupled with mechanical ventilation are installed more fre- quently, builder costs appear to be $300 to $400 below the median value reported in the demonstration program. Moreover, builders who used an approach to reducing uncon- trolled air leakage that employs common dry wall as a continuous air barrier are reporting significantly lower (75 percent less) costs than builders who used the more conven- tional plastic film. The Council believes the cost reported by Tacoma builders for heat recovery ventilation devices and by those builders who employed the “air-tight” dry wall approach is more repre- sentative of the long-term cost for these mea- sures. Consequently, in updating the esti- mated cost for infiltration control and heat recovery ventilators, the Council used the lower quartile cost reported by the demon- stration program builders. However, median costs for heat recovery ventilators reported by the builders in the demonstration program were used to reflect the cost of these units in climate zone 3, where freeze protection fea- Table 5-13 New Residential Construction Base Case Efficiency Levels and Annual Space Heating Use Assumptions Chapter 5 ; CLIMATE ZONE ; ANNUAL ANNUAL ANNUAL WEIGHTED BUILDING TYPE MTEVEL what) LEVEL awmisgt) LeveL” whist) Nawhveg Single Family 7.9 10.7 9.9 8.4 Ceiling/Roof R-30 R-30 R-38 Walls R-11 R-11 R-19 Underfloor R-11/19 R-19 R-19 Windows Double glazed Double glazed Double glazed (U-.90) (U-.90) (U-.65) Multifamily 5.0 75 9.0 5.5 Ceiling/Roof R-30 R-30 R-30 Walls R-11 R-11 R-11 Underfloor R-11/19 R-19 R-19 Windows Double glazed Double glazed Double glazed (U-.90) (U-.90) (U-.65) Manufactured Homes 9.7 13.8 16.3 10.5 Ceiling/Roof R-11 R-11 R-11 Walls R-11 R-11 R-11 Underfloor R-11 R-11 R-11 Windows Double glazed Double glazed Double glazed (U-.90) (U-.90) (U-.90) Table 5-14 Typical New Dwelling Characteristics CHARACTERISTIC SINGLE FAMILY DETACHED MULTIFAMILY MANUFACTURED HOME Prototype Label A B Cc 12-Units@ A B Size—Gross Floor Area (sq. ft.) 1,344 1,848 2,352 840 sq ft/unit 924 1,344 Foundation Type Crawlspace Crawlspace Basement Skirted Crawlspace Number of Stories 1 2-Split Level 1 w/full basement 3-4/w garage 1 1 Window Area (sq. ft.) 175 240 258 1,140 144 144 Glass Area as a % of Floor Area 13% 13% 11% 11.9% 15.6% 10.8% Gross Wall Area Above Grade 1,376 2,048 1,596 6,422 1,200 1,200 Below Grade _ _ 736 _ -_ _ Total Exterior Envelope Area (sq. ft.) 4,064 4,624 5,244 14,070 3,048 3,888 5-13 Chapter 5 MEE Figure 5-4 Sources of Residential Heating tures must be added. The Council believes that for infiltration control with heat recovery ventilation, this approach reflects both Tacoma's experience and the use of the “air- tight” dry wall infiltration control strategy. As noted earlier, not all space heating conser- vation measures have similar useful lives. Insulation and infiltration control measures (i.e., air/vapor barriers) installed in new single family and multifamily dwellings are antici- pated to last at least 70 years (i.e., the life of the structure). These same measures installed in new manufactured houses are also expected to last the life of the building (i.e., 45 years). However, the Council has assumed that two measures, heat recovery ventilators and energy efficient windows, must be repaired or replaced before the end of the life of the structure. The Council included the cost of repairing and/or replac- ing these two space heating conservation measures when calculating their levelized cost. For the heat recovery ventilator cost, the Council assumed that the ventilation fans would last 15 years, similar to furnace fans, and that the ventilator itself would be replaced every 30 years. Fan replacement was assumed to cost $100. The replacement of heat recovery ventilators was assumed to cost $700 in climate zones 1 and 2, and $900 5-14 in climate zone 3. The difference in replace- ment cost for zone 3 is due to the need to provide for freeze protection features on the heat recovery ventilator. Operational costs (i.e., for energy) were accounted for in the assumed heat recovery efficiency of 60 per- cent. All the windows in new residential struc- tures were assumed to be replaced at 30- year intervals at a cost equivalent to their initial capital cost. The costs of improvements in the space heating efficiency of new manufactured housing were taken from a study prepared for the Manufactured Housing Institute (MHI) and submitted to the Council by MHI. The costs reported in that study and the Bon- neville energy efficient new home demon- stration program were adjusted to 1985 dol- lars from 1984 dollars using the Gross National Product deflator from mid-1984 to January 1985. Tables 5-15 through 5-21 show the retail cost assumed by the Council for potential cost-effective space heating con- servation measures for new single and multi- family dwellings and manufactured housing. Abbreviations: (Tables 5-15 through 5-21) UA— measure of resistance to heat loss Btu/F—British thermal units per degree of Fahrenheit DG to TGTB—double glazed to triple glazed with thermal break Std—standard truss, which compresses insulating bats at ends Adv—advanced truss, which allows more effective use of R-values by not com- pressing bats at the ends ACH— air changes per hour Step 3. Estimate the cost effectiveness of space heating energy savings produced by efficiency improvements in new resi- dential buildings. Once typical new dwell- ing designs were selected, the Council used a computer simulation model to estimate potential space heating energy savings that could be produced by each conservation measure. This model, SUNDAY, is also used to estimate savings from weatherization measures (see discussion above). The absolute value (in kilowatt-hours per year) of the space heating energy savings produced by adding an individual conserva- tion measure is a function of the existing ther- mal efficiency level of the building. The less efficient the existing building, the larger the savings that will be obtained from installing the same measure. To assess the savings that could be pro- duced by installing each space heating con- servation measure, it is necessary to take into account their interaction. This was done by determining each measure's benefit (i.e., change in heat loss rate) and cost (i.e., dol- lars per square foot). The savings produced by each potentially cost-effective measure were then analyzed under the assumption that all measures with higher benefit-to-cost ratios had already been installed in the house. Figure 5-4 illustrates how the heating require- ments of an average current practice house and a model conservation standards house might be met. Heating requirements are met by solar heat, internal gains (the amount of heat released indoors by people and appliances), and the furnace, which can be supplemented by heat from wood burning stoves or other sources. The current practice house reflects average conditions for a house that is primarily heated with electricity. If the house were primarily heated with wood, the contribution from wood would be much larger, but electrical savings would still be significant as long as electricity were the marginal fuel. Chapter 5 Costs and Savings from Canin taueee in New Single Family Houses Zone 1— Seattle ANNUAL LEVELIZED AVERAGE CONSERVATION MEASURE Bur Incremental =e Sisq ft i ale ft Swnye miigkWh RVALUE HOUSE SIZE 1,344 Base Case 487 $ 0 $ 0 $0.00 10,577 7.9 0 0.0 8.35 Floors R11 to R19 470 202 202 0.15 10,006 7.4 571 13.9 8.65 Walls R11 to R19 443 418 620 0.46 9,089 68 917 17.9 9.17 Windows DG to TGTB 374 784 1,404 1.04 6,799 5.1 2,290 20.6 10.86 Insulated Door 363 191 1,595 1.19 6,449 48 350 21.5 11.18 Ceilings R30 to R38 Std 355 188 1,783 1.33 6,187 46 261 28.4 11.44 Floors 19 to R30 341 376 2,160 1.61 5,715 43 472 31.4 11.93 Walls R19 to R25 324 581 2,740 2.04 5,210 3.9 505 45.3 12.53 Ceilings R38 to R49 Adv 311 524 2,264 2.43 4,795 3.6 415 49.7 13.07 Infiltration .6 to .3 ACH 253 1,735 4,999 3.72 3,084 23 1,711 51.9 16.08 HOUSE SIZE 1,848 Base Case 632 $ 0 $ 0 $0.00 14,560 7.9 0 0.0 7.31 Floor R11 to R19 614 193 193 0.10 13,933 75 627 124 7.53 Walls R11 to R19 573 636 829 0.45 12,511 68 1,422 17.6 8.07 Windows DG to TGTB 480 1,075 1,904 1.03 9,372 Su 3,139 20.6 9.64 Insulated Door 469 191 2,095 1.13 9,015 49 356 214 9.86 Ceilings R30 to R38 Std 461 180 2,276 1.23 8,760 47 255 27.8 10.02 Floors R19 to R30 447 361 2,636 1.43 8,301 45 459 31.0 10.34 Infiltration .6 to .3 ACH 367 2,075 4,711 2.55 5,778 3.1 2,523 40.5 12.59 Walls R19 to R25 343 884 5,595 3.03 5,039 rar) 739 47.1 13.50 Ceilings R38 to R49 Adv 330 502 6,098 3.30 4,661 2.5 378 52.3 14.02 HOUSE SIZE 2,352 Base Case 699 $ 0 $ 0 $0.00 15,620 6.6 0 0.0 7.50 Floors R11 to R19 692 84 84 0.04 15,376 65 244 13.6 7.58 Walls R11 to R19 663 460 544 0.23 14,394 6.1 982 18.4 7.91 Windows DG to TGTB 562 1,156 1,700 0.72 11,036 47 3,358 20.7 9.33 Insulated Door 546 287 1,987 0.84 = 10,514 45 522 21.6 9.61 Ceilings R30 to R38 Std 537 204 2,191 0.93 10,235 4.4 279 28.8 9.77 Floors R19 to R30 531 157 2,347 1.00 10,056 43 179 34.5 9.87 Infiltration .6 to .3 ACH 430 2,430 4,777 2.03 6,869 29 3,187 36.5 12.20 Walls R19 to R25 412 639 5,416 2.30 6,331 27 538 46.7 12.73 Ceilings R38 to R49 Adv 397 568 5,984 2.54 5,899 25 432 51.7 13.20 5-15 Chapter 5 CONSERVATION MEASURE Base Case Walls R11 to R19 Windows DG to TGTB Insulated Door Ceilings R30 to R38 Std Floors R19 to R30 Walls R19 to R25 Ceilings R38 to R49 Adv Infiltration .6 to .3 ACH Base Case Walls R11 to R19 Windows DG to TGTB Insulated Door Ceilings R30 to R38 Std Floors R19 to R30 Infiltration .6 to .3 ACH Walls R19 to R25 Ceilings R38 to R49 Adv Base Case Walls R11 to R19 Windows DG to TGTB Insulated Door Ceilings R30 to R38 Std Floors R19 to R30 Infiltration .6 to .3 ACH Walls R19 to R25 Ceilings R38 to R49 Adv 5-16 Table 5-16 Costs and Savings from Conservation Measures in New Single Family Homes Zone 2 — Spokane UA Btu/F 469 442 374 355 341 324 311 253 614 573 480 469 461 447 367 692 562 537 531 430 412 397 COST ANNUAL USE Incremental Cumulative S/sq ft kWh/yr kWh/sq ft HOUSE SIZE 1,344 $ 0 $ 0 $0.00 14,459 10.8 418 418 0.31 13,260 9.9 784 1,202 0.89 10,292 Tel 191 1,393 1.04 9,826 7.3 188 1,581 1.18 9,479 ua 376 1,958 1.46 8,848 6.6 581 2,538 1.89 8,171 6.1 524 3,062 2.28 7,615 5.7 1,735 4,797 3.57 5,258 3.9 HOUSE SIZE 1,848 $ 0 $ 0 $0.00 19,693 10.7 636 636 0.34 17,863 9.7 1,075 1,711 0.93 13,768 75 191 1,902 1.03 13,299 he. 180 2,083 1.13 12,962 7.0 361 2,443 1.32 12,357 6.7 2,075 4,518 2.44 8,988 49 884 5,402 2.92 7,985 4.3 502 5,904 3.20 7,470 4.0 HOUSE SIZE 2,352 $ 0 $ Oo $0.00 21,805 9.3 460 460 0.20 20,523 8.7 1,156 1,616 0.69 16,135 6.9 287 1,903 0.81 15,396 6.5 204 2,107 0.90 15,023 6.4 157 2,263 0.96 14,784 6.3 2,430 4,693 2.00 10,527 45 639 5,332 2.27 9,793 4.2 568 5,900 2.51 9,196 3.9 ANNUAL SAVINGS kWh/yr 0 1,198 2,969 465 347 631 677 555 2,357 1,830 4,095 470 336 605 3,369 1,003 515 1,282 4,388 739 373 239 4,257 734 597 LEVELIZED COST mills/kWh 0.0 13.7 15.9 16.2 21.3 23.5 33.8 37.2 37.7 0.0 13.7 15.8 16.0 Zit 23.5 30.3 34.7 38.4 0.0 14.1 15.2 15.8 21.5 25.9 27.3 34.3 37.4 AVERAGE ACTUAL R-VALUE 8.66 9.19 10.86 11.18 11.44 11.93 12.53 13.07 16.08 7.53 8.07 9.64 9.86 10.02 10.34 12.59 13.50 14.02 7.58 7.91 9.33 9.61 9.77 9.87 12.20 12.73 13.20 Table 5-17 Costs and Savings from Conservation Measures in New Single Family Houses Zone 3 — Missoula Chapter 5 " cost wouaues_ aa LEVELS erage CONSERVATION MEASURE _Btu/F Incremental Cumulative $/sq ft kWh/yr kWh/sq ft kWh/yr mills/kWh R-VALUE HOUSE SIZE 1,344 Base Case 396 $ 0 $ 0 $0.00 13,280 9.9 0 0.0 10.27 Windows DGTB to TGTB 355 $ 599 $ 599 $0.45 11,271 8.4 2,010 17.9 11.44 Floors R19 to R30 341 $ 376 $ 975 $0.73 10,540 78 731 20.2 11.93 Walls R19 to R31 315 $ 813 $1,788 $1.33 9,302 6.9 1,238 25.8 12.90 Ceiling R38 to R49 Adv 302 $ 564 $2,352 $1.75 8,653 6.4 649 34.2 13.48 Infiltration .6 to .3 ACH 243 $1,935 $4,287 $3.19 5,913 4.4 2,740 36.8 16.69 HOUSE SIZE 1,848 Base Case 517 $ 0 $ 0 $0.00 18,149 9.8 0 0.0 8.95 Windows DGTB to TGTB 461 $ 821 $ 821 $0.44 15,371 8.3 2,778 17.8 10.02 Floors R19 to R30 447 $ 361 $1,181 $0.64 14,669 7.9 702 20.2 10.34 Walls R19 to R31 407 $1,238 $2,419 $1.31 12,665 6.9 2,003 24.3 11.37 Infiltration .6 to .3 ACH 327 $2,275 $4,694 $2.54 8,812 48 3,853 29.7 14.15 Ceilings R38 to R49 Adv 314 $ 541 $5,235 $2.83 8,214 44 598 35.6 14.73 HOUSE SIZE 2,352 Base Case 596 $ 0 $ 0 $0.00 20,731 8.8 0 0.0 8.79 Windows DGTB to TGTB 537 $ 882 $ 882 $0.38 17,794 76 2,937 18.0 9.77 Floors R19 to R30 531 $ 157 $1,039 $0.44 17,519 74 275 22.5 9.87 Walls R19 to R31 502 $ 895 $1,934 $0.82 16,114 6.9 1,405 25.1 10.45 Infiltration .6 to .8 ACH 400 $2,603 $4,537 $1.93 11,278 48 4,836 26.5 13.10 Ceilings R38 to R49 Adv 386 $ 612 $5,149 $2.19 10,601 45 677 35.6 13.59 5-17 Chapter 5 CONSERVATION MEASURE Base Case Floors R11 to R19 Ceilings R30 to R38 Std Walls R11 to R19 Windows DG to TGTB Insulated Door Floors R19 to R30 Walls R11 to R25 Infiltration .6 to .8 ACH Ceilings R38 to R49 Adv Base Case Floors R11 to R19 Ceilings R30 to R38 Std Walls R11 to R19 Windows DG to TGTB Insulated Door Floors R19 to R30 Walls R11 to R25 Infiltration .6 to .3 ACH Ceilings R38 to R49 Adv Base Case Floors R11 to R19 Ceilings R30 to R38 Std Walls R11 to R19 Windows DG to TGTB Insulated Door Floors R19 to R30 Walls R19 to R25 Infiltration .6 to .8 ACH Ceilings R38 to R49 Adv 5-18 Table 5-18 Costs and Savings from Conservation Measures in New Multifamily Residences UA Btu/F 2,764 2,691 2,645 2,526 2,081 2,047 1,929 1,434 1,403 2,764 2,691 2,645 2,526 2,081 2,047 2,002 1,929 1,434 1,403 2,764 2,691 2,645 2,526 2,081 2,047 2,002 1,929 1,434 1,403 Dwelling Unit Size 840 square feet ANNUAL _LEVELIZED Incremental a $/sq ft why Kwa ft savy miigkWh ZONE 1— SEATTLE $ 0 $ 0 $ 0 4,168 5.0 0 0.0 $ 48 $ 48 $0.06 3,967 47 201 95 $ 45 $ 93 $0.11 3,844 46 123 14.3 $ 155 $ 248 $0.29 3,525 4.2 319 19.1 $ 426 $ 674 $0.80 2,380 28 1,145 22.4 $ 50 $ 724 $0.86 2,296 27 84 23.4 $ 89 $ 813 $0.97 2,186 2.6 110 31.8 $ 215 $1,028 $1.22 2,017 2.4 170 49.8 $1,135 $2,163 $2.58 899 11 1,118 53.8 $ 124 $2,287 $2.72 838 1.0 61 80.0 ZONE 2— SPOKANE $ 0 $ 0 $ 0 6,337 75 0 0.0 $ 48 $ 48 $0.06 6,069 7.2 268 7.4 $ 45 $ 93 $0.11 5,904 7.0 165 10.7 $ 155 $ 248 $0.29 5,477 65 427 14.3 $ 426 $ 674 $0.80 3,921 47 1,556 16.4 $ 50 $ 724 $0.86 3,806 45 115 17.1 $ 89 $ 813 $0.97 3,654 43 152 22.6 $ 215 $1,028 $1.22 3,418 44 236 35.9 $1,135 $2,163 $2.58 1,849 22 1,569 38.3 $ 124 $2,287 $2.72 1,757 24 92 53.0 ZONE 3 — MISSOULA $ 0 $ 0 $ 0 7,517 8.9 0 0.0 $ 48 $ 48 $0.06 7,208 8.6 308 6.2 $ 45 $ 93 $0.11 7,019 8.4 189 9.3 $ 155 $ 248 $0.29 6,524 78 494 12.4 $ 426 $ 674 $0.80 4,720 5.6 1,804 14.2 $ 50 $ 724 $0.86 4,586 55 134 14.7 $ 89 $ 813 $0.97 4,409 5.2 177 19.8 $ 215 $1,028 $1.22 4,131 49 277 30.5 $1,335 $2,363 $2.81 2,283 27 1,848 39.0 $ 124 $2,487 $2.96 2,174 2.6 109 44.8 AVERAGE ACTUAL R-VALUE 5.09 5.23 5.32 5.57 6.76 6.87 7.03 7.29 9.81 10.03 5.09 5.23 5.32 5.57 6.76 6.87 7.03 7.29 9.81 10.03 5.09 5.23 5.32 5.57 6.76 6.87 7.03 7.29 9.81 10.03 Table 5-19 Costs and Savings from Conservation Measures in New Manufactured Homes Zone 1 — Seattle Dwelling Unit Size 840 square feet Chapter 5 UA cost ANNUAL USE savincs “cost” “ACTUAL CONSERVATION MEASURE Btu/F —_—incremental___Cumulative Sisqft © kWhiyr —«kWhisqft kWh/yr milis/kWh R-VALUE HOUSE SIZE 924 Base Case 440 $ 0 $ 0 $0.00 9,391 10.2 0 0.0 6.92 Ceilings R11 to R19 424 62 62 0.07 8,829 9.6 561 5.2 7.20 Floors R11 to R19 405 162 224 0.24 8,187 8.9 642 11.8 7.52 Ceilings R19 to R27 394 166 390 0.42 7,809 8.5 378 20.5 7.74 Walls R11 to R19 361 568 958 1.04 6,733 7.3 1,076 24.7 8.43 Insulated Door 352 175 1,133 1.23 6,410 6.9 323 25.3 8.66 Floors R19 to R30 334 480 1,613 1.75 5,809 63 601 37.3 9.14 Windows DG to TGTB 302 756 2,369 2.56 4,807 5.2 1,003 44.8 10.11 Ceilings R27 to R38 296 228 2,597 2.81 4,631 5.0 175 60.9 10.30 Infiltration .6 to .3 ACH 259 1,450 4,047 4.38 3,498 3.8 1,133 72.3 11.79 HOUSE SIZE 1,344 Base Case 537 $ 0 $ 0 $0.00 12,493 9.3 0 0.0 7.24 Ceilings R11 to R19 513 90 90 0.07 11,625 8.6 868 4.9 7.58 Floors R11 to R19 486 235 325 0.24 = 10,674 79 951 11.6 8.00 Ceilings R19 to R27 470 242 567 0.42 10,138 75 536 214 8.27 Walls R11 to R19 437 568 1,135 0.84 9,243 6.9 895 29.7 8.90 Insulated Door 428 175 1,310 0.97 8,968 6.7 275 29.7 9.09 Floors R19 to R30 401 699 2,009 1.49 8,045 6.0 923 35.4 9.70 Windows DG to TGTB 369 756 2,765 2.06 6,985 5.2 1,060 42.4 10.53 Infiltration .6 to .8 ACH 315 1,450 4,215 3.14 5,215 3.9 1,770 46.3 12.36 Ceilings R27 to R38 307 332 4,547 3.38 4,961 37 253 61.3 12.68 5-19 Chapter 5 5: Table 5-20 Costs and Savings from Conservation Measures in New Manufactured Homes Dwelling Unit Size 840 square feet Zone 2 — Spokane LEVELIZED AVERAGE commeicneciuns! Oe | iecaane “commen; aiye| ‘wee memes, GRO? | wGceiw | ||nane HOUSE SIZE 924 Base Case 440 $ 0 $ 0 $0.00 13,667 14.8 0 0.0 6.92 Ceilings R11 to R19 424 62 62 0.07 12,937 14.0 730 4.0 7.20 Floors R11 to R19 405 162 224 0.24 12,100 13.1 837 9.1 752 Ceilings R19 to R27 394 166 390 0.42 11,604 12.6 496 15.6 7.74 Walls R11 to R19 361 568 958 1.04 10,182 11.0 1,422 18.7 8.43 Insulated Door 352 175 1,133 1.23 9,753 10.6 429 19.1 8.66 Floors R19 to R30 334 480 1,613 1.75 8,951 9.7 801 28.0 9.14 Windows DG to TGTB 302 756 2,369 2.56 7,606 8.2 1,345 33.4 10.11 Ceilings R27 to R38 296 228 2,597 2.81 7,369 8.0 237 45.0 10.30 Infiltration .6 to .3 ACH 259 1,450 4,047 4.38 5,809 6.3 1,560 52.5 11.79 HOUSE SIZE 1,344 Base Case 537 $ 0 $ 0 $0.00 17,609 13.1 0 0.0 7.24 Ceilings R11 to R19 513 90 90 0.07 16,503 12.3 1,106 3.8 7.58 Floors R11 to R19 486 235 325 0.24 15,350 11.4 1,153, 9.5 8.00 Ceilings R19 to R27 470 242 567 0.42 14,679 10.9 671 16.9 8.27 Walls R11 to R19 437 568 1,135, 0.84 13,167 9.8 1,512 17.6 8.90 Insulated Door 428 175 1,310 0.97 12,776 9.5 390 21.0 9.09 Floors R19 to R30 401 699 2,009 1.49 11,586 8.6 1,190 27.5 9.70 Windows DG to TGTB 369 756 2,765 2.06 10,200 7.6 1,386 32.4 10.53 Infiltration .6 to .3 ACH 315 1,450 4,215 3.14 7,919 5.9 2,281 35.9 12.36 Ceilings R27 to R38 307 332 4,547 3.38 7,648 5.7 271 57.3 12.68 20 Table 5-21 Costs and Savings from Conservation Measures in New Manufactured Homes Zones 3 — Missoula Dwelling Unit Size 840 square feet Chapter 5 UA cost ANNUAL USE sancs “Cost “ACTUAL CONSERVATION MEASURE _Btu/F Incremental Cumulative $/sq ft kWh/yr kWhisq ft kWh/yr mills/kWh R-VALUE HOUSE SIZE 924 Base Case 440 $ O $ 0 $0.00 16,069 17.4 0 0.0 6.92 Ceilings R11 to R19 424 62 62 0.07 15,224 16.5 846 3.4 7.20 Floors R11 to R19 405 162 224 0.24 14,254 15.4 969 78 7.52 Ceilings R19 to R27 394 166 390 0.42 13,681 14.8 573 13.5 7.74 Walls R11 to R19 361 568 958 1.04 12,045 13.0 1,637 16.2 8.43 Insulated Door 352 175 1,133 1.23 11,552 12.5 493 16.5 8.66 Floors R19 to R30 334 480 1,613 1.75 10,629 11.5 923 24.3 9.14 Windows DG to TGTB 302 756 2,369 2.56 9,066 9.8 1,563 28.8 10.11 Ceilings R27 to R38 296 228 2,597 2.81 8,791 9.5 275 38.8 10.30 Infiltration .6 to .3 ACH 259 1,450 4.047 4.38 6,981 7.6 1,810 45.3 11.79 HOUSE SIZE 1,344 Base Case 537 $ 0 $ 0 $0.00 20,875 15.5 0 0.0 7.24 Ceilings R11 to R19 513 90 90 0.07 19,588 14.6 1,287 3.3 7.58 Floors R11 to R19 486 235 325 0.24 18,247 13.6 1,341 8.2 8.00 Ceilings R19 to R27 470 242 567 0.42 17,448 13.0 799 14.2 8.27 Walls R11 to R19 437 568 1,135 0.84 15,627 11.6 1,822 14.6 8.90 Insulated Door 428 175 1,310 0.97 15,144 11.3 482 17.0 9.09 Floors R19 to R30 401 699 2,009 1.49 13,754 10.2 1,390 23.5 9.70 Windows DG to TGTB 369 756 2,765 2.06 12,179 9.1 1,575 28.5 10.53 Infiltration .6 to .3 ACH 315 1,450 4,215 3.14 9,508 71 2,671 30.7 12.36 Ceilings R27 to R38 307 332 4,547 3.38 9,192 6.8 316 49.1 12.68 5-21 Chapter 5 Table 5-22 Weighting Factors Used to Aggregate Table 5-24 Individual Building & Location Savings to Region Regionally Weighted Savings and Costs in New Multifamily Dwellings —- a —_ LEVELIZED COST CAPITAL COST ANNUAL USE RELATIVE USE SAVINGS AVERAGE Single Family (less than five-plex) mills/kWh_ Total Sisq ft kWhiyr = kWh/sq ft % of base kWhiyr ‘R-VALUE 1,344 sq. fl. — Single Story 90% ° $ 0 $000 4649 55 100 0 5.09 TEAS 99: tL — Two Sry w= $ 27 $003 4524 54 97 125 5.17 2,352 sq. ft. — One Story w/Basement 1% 4,400 99. f 10 $ 61 $0.07 4,389 5.2 95 260 5.25 Mutter (ve-plax and targer) 15 $ 184 $022 4014 48 88 635 551 12-Unit 100% 840 sq. ft/unit 20 $ 447 $053 3343 40 73 1,306 6.11 Manutactured Homes 25 $ 766 $091 2569 31 55 2,080 6.94 eet onoie Wie om 30 $ 825 $0.98 2497 30 54 2,152 7.04 1,344 Double Wide 58% 1.170 sq. ft 35 $913 $1.09 2400 29 52 2,248 747 cone a Weraet 40 $1,173 $1.40 2072 25 46 2,577 7.69 Zone 1 — Seattle — ene 45 $1,231 $1.47 2,026 24 45 2,623 7.77 cone 2 ane —_ = 50 $1,329 $158 1,940 23 43 2,709 7.93 goa — ees zane bd 55 $2,200 $2.62 1,085 1.3 23 3,564 9.86 Region ae 60 $2,219 $264 1076 13 23 3,873 9.90 “HDD — Heating Degree Days at 65°F based on Typical Meterological Year (TMY) weather tape used 65 $2,238 $2.66 1,067 1.3 22 3,582 9.93 to estimate savings. TMY weather tapes vary slightly from published long-term averages. 70 $2,257 $269 1057 13 2 3,501 gee Table 5-23 75 $2,276 $2.71 1,048 1.2 22 3,601 10.00 Regionally Weighted Savings and Costs in New Single Family Dwellings 80 $2,295 $2.73 1,039 12 22 3,610 10.03 ANNUAL 85 $2,295 $2.73 1039 1.2 22 3,610 10.03 LEvnilgKWh: otal Seqht kWhyr KWhisgtt “Solves “kWhyr RWALUE 90 92.295 $273 1099 12 22 3610 10.03 . € 0 S00 el Bs ‘00 ‘i a 95 $2,295 $273 1,039 12 22 3,610 10.03 5 $ 91 $007 11479 82 oe 263 eo 100 $2,295 $273 1039 12 22 3,610 10.03 10 $ 181 $0.13 11,216 80 96 527 8.63 15 $ 435 031 10485 75 90 1,257 9.01 Regionally Weighted Savings eee New Manufactured Dwelings 20 $1,309 $094 8112 58 69 3,630 10.56 am 25 $1,800 $1.29 7,117 54 60 4625 11.35 LEVELZEDCOST CAPTTALCOST | ANNUAL USE | RELATIVEUSE SAVINGS AVERAGE 30 $2,114 $1.51 6,708 48 87 5,034 11.73 35 $2,534 $1.81 6,225 45 53 5518 12.22 oO $ 0 $0.00 12,279 106 100 0 7 40 $3,101 $2.21 5,553 4.0 48 6189 13.01 . $ 88 $0.07 11,442 9.9 93 836 1M 45 $3,332 $237 5,332 38 46 6410 13.25 10 $ 248 $0.21 10,768 = 9.3 88 1,511 773. 50 $4048 $289 4708 34 “0 7034 (14.18 15 $ 398 $0.34 10,322 89 84 1,956 7.94 55 $5,047 $3.63 3,731 27 31 8,012 15.89 2 $635 $0.55 9,740 868.4 80 2,530 8.21 60 $5,047 $3.63 3,731 2.7 31 8012 15.89 a $1,014 $0.88 9,086 78 78 3,193 8.63 65 $5,047 $363 3,731 27 31 8.012 15.89 30 $1,577 $137 8165 7.0 67 4113 9.25 70 $5,047 $663 3,731 27 31 8012 15.89 5 $2,015 $1.74 7.539 65 62 4740 oe 75 $5,047 $363 3,731 27 31 8012 15.89 “0 $2,522 $218 6837 59 56 5,442 10.27 80 $5,047 $363 3,731 27 31 8012 15.89 4 $2905 $259 6232 5.4 51 6.047 = 10.82 85 $5,047 $3.63 3,731 27 31 8,012 15.89 50 $3,463 $2.96 5,688 4.9 47 6,591 11.38 90 $5047 $363 3731 27 31 go12 15.89 55 $3,826 $3.28 5317 46 44 6961 11.79 95 $5,047 $363 3,731 27 3 8012 15.89 60 $4,064 $3.52 5,132 4.4 42 7.147 12.08 100 $5,047 $363 3,731 27 31 8012 15.89 65 $4,300 $3.76 4,949 4.3 40 7,330 12.27 70 $4,340 $380 4,918 4.2 40 7,361 12.31 75 $4,340 $380 4918 42 40 7,361 12.31 80 $4,340 $380 4918 42 40 7,361 12.31 85 $4,340 $3.80 4,918 42 40 7,361 12.31 90 $4,340 $380 4918 42 40 7,361 12.31 95 $4,340 $3.80 4918 4.2 40 7,361 12.31 100 $4,340 $380 4918 4.2 40 7,361 12.31 5-22 Chapter 5 When determining the electrical savings of measures applied to a current practice house, at least the following three policy con- siderations must be evaluated: the treatment of wood heating, internal temperature set- tings for the whole house, and internal gains.” The Council assumed no wood heat- ing when evaluating measure savings in new residential buildings. The Council used a constant thermostat setting of 65° for the whole house to represent a combination of higher temperatures when the house was occupied and the occupants active, and a lower nighttime setback. Finally, the Council assumed a cadre of efficient appliances, reflecting appliances that would be in place for the majority of the life of the house, and are present in the region throughout most of the Council's plan. Appliances currently in place in houses are not very efficient, contrib- ute more usable heat to the house, and thus cut space heating loads. This is reflected in Figure 5-4, where internal gains are larger in the current practice house. The Council re-assessed the planning assumptions described above before issuing the current plan and feels that these assump- tions should be maintained based on the following reasons. First, there is no assurance that occupants of houses built to the standards will continue to use wood heat. Changing wood prices, income levels, wood availability and environmental regulations all could reduce the use of wood heating, leav- ing the electrical system vulnerable to mass “fuel switching” to electricity, an action that would be difficult if not impossible to plan resources for. Second, the Act defines con- servation as an efficiency improvement, not a change in lifestyle. Current behavior of con- sumers to close off rooms or lower ther- mostats may represent curtailment rather than conservation as defined in the North- west Power Act. Such behavior is not expected to continue after cost-effective effi- ciency improvements are made. Third, more efficient appliances are clearly cost-effective resources and will be the norm, especially in new houses, in the next decade. Appliance manufacturers have testified that, even with- out appliance standards such as those adopted recently in California and called for in this plan, new appliances will be much more efficient. Therefore, the Council's esti- mates reflect less heat escaping from these appliances to heat the house. Finally, the adoption of planning assumptions different than these would subject the region to greater planning uncertainties than the pres- ent set of assumptions. If the energy effi- ciency requirements of the standards are made less stringent because it is assumed consumers will continue to close off rooms and heat with wood, the degree of uncer- tainty the region must plan for increases. Tables 5-15 through 5-21 show the levelized cost, annual energy use, and energy savings produced by the addition of each measure for each dwelling type, building design and for three representative climate types found in the region (Zone 1-Seattle, Zone 2-Spokane and Zone 3-Missoula). The levelized cost shown for single family and multifamily build- ings is based on a 70-year physical life and a financing cost of approximately 4 percent real.8 Levelization was done using a 3 per- cent real discount rate. The levelized cost shown for manufactured housing is based on a 45-year economic life and levelization ata 3 percent real discount rate. For planning pur- poses, it has been assumed that the effi- ciency improvements in single family and multifamily houses and manufactured hous- ing will be obtained via a marketing and incentive program financed through Bon- neville, public utilities and the region's inves- tor-owned utilities. The Council has established model conser- vation standards for new single family and multifamily houses heated with electricity. The standards are required to achieve all regionally cost-effective conservation sav- ings. As discussed in Volume II, Chapter 4, the Council has found that power savings that can be achieved at a cost in the range of 4.0 to 4.5 cents per kilowatt-hour represent regionally cost-effective resources. Appendix |-B (Volume |) sets forth an illustrative pre- scriptive path for each climate zone that if installed in a typical new house would satisfy the standards. The measures shown in Appendix I-B are all regionally cost effective for the average 1,850 sq. ft. single family house (one and two family dwelling) currently being constructed in the region. In selecting the measures shown in Appendix |-B, the Council chose a typical structure in a typical location in each climate zone, and assumed the building was operated in a typical way. Actual buildings will vary from these typical assumptions. It is not administratively feasi- ble to implement a standard that varies as the actual building conditions vary. Therefore, the specific measures included in the MCS will not perform the same in all houses. For example, the heat recovery ventilator has a higher levelized life cycle cost in the smallest (1,344 sq. ft.) home analyzed by the Council. The Council's model conservation standards retain this measure for three reasons. First, the measure is regionally cost effective for the average new single family house built in the region, since such houses have 1,850 square feet of floor area. Second, as dis- cussed elsewhere, the Council anticipates that the cost of this measure will decline sig- nificantly as more builders develop experi- ence with the heat recovery ventilators and the market for such products matures. Third, the Council believes that some form of mechanical ventilation should be provided in new residences to ensure adequate indoor air quality. Consequently, the Council consid- ers the use of heat recovery ventilation sys- tems necessary to maintain satisfactory ven- tilation while minimizing the energy lost. As shown in Tables 5-16 and 5-17, the installation of R-49 ceiling insulation rather than R-38 ceiling insulation in climate zones 2 and 3 appears to be regionally cost effec- tive. However, the Council has not included this measure in its standards for these cli- mate zones, due to the limited size of the data base on which these costs are based. At the time the Council conducted its analysis, only nine demonstration programs builders had reported costs for this measure. Step 4. Estimate the regional conserva- tion potential available from space heat- ing conservation in new dwellings. The next step in the Council's development of a regional supply curve for space heating con- servation potential requires combining the engineering estimates of individual house savings by climate zone to establish a regional total. Because each measure saves a different amount of energy in each house design and in each location, an aggregate supply curve must be developed that repre- sents the weighted average savings for all measures in comparable dwelling types. 5-23 Chapter 5 Table 5-26 Forecast Model vs. Engineering Estimate for Space Heating in New Dwellings, Regional Average Use FORECASTING MODEL ENGINEERING ESTIMATE BUILDING TYPE kWh/yr kWh/sq ft/yr kWh/yr kWh/sq ft/yr Single family 9,990 71 11,742 8.4 Multifamily 3,590 43 4,650 5.5 Manufactured Home 8,160 8.9 12,280 10.5 Table 5-27 Forecasting Model Dwelling Size vs. Average New Dwellings (square feet) MODEL BUILDING TYPE Single Family 1,400 Multifamily 840 Manufactured Home 920 EXISTING STOCK RATIO OF NEW STOCK TO NEW STOCK MODEL 1,400 1.00 1,030 1.23 1,170 1.27 Each of the three single family dwelling designs was assigned a weight based on its foundation type, size and window area. The specific weight assigned to each design approximately reflects that design's share of the new housing stock additions expected over the forecast period. This was also done for the two manufactured housing designs. Building type weighting was unnecessary for multifamily space heating, because only one multifamily design was used. It should be noted that the Council's forecasting model defines all units up to and including four- plexes as “single family dwellings.” Conse- quently, the weights selected are designed to achieve a much smaller average size for new single family houses (i.e., 1,400 square feet of floor area) than had they been selected on the basis of the more conventional definition of a single family home (one and two family dwellings) used to establish the standards. The average size of typical new one and two family dwellings recently constructed in the region is between 1,600 and 1,800 square feet of floor area. Once each building design's weight had been established, the average savings by climate type was calculated for all designs. These savings were then aggregated to the regional level based on the share of new electrically heated dwellings expected to be constructed in each climate over the forecast period. 5-24 Table 5-22 shows the weight assigned each building design and climate type. Tables 5-23 through 5-25 show the weighted average use, cost and savings available from new single family, multifamily and manufactured houses at levelized costs less than 10 cents per kilowatt-hour (equivalent to 100 mills per kilowatt-hour). Step 5. Estimate the realizable conserva- tion potential from new residential space heating efficiency improvements. In order to establish the proportion of technically available space heating conservation that can realistically be achieved, two adjust- ments must be made to the engineering sav- ings estimates. First, to ensure consistency with the Council's load forecast, the conser- vation resource based on engineering esti- mates of current space heating energy use must be adjusted or scaled to account for the forecasting model's estimate of current space heating use. Table 5-26 compares the aver- age space heating energy use by dwelling type, as estimated by the Council's forecast- ing model for 1985 in the high forecast, and the engineering estimate of space heating use for houses built to current practice. The primary reasons for the differences between each estimate are variations in dwelling unit size, the waste heat released by appliances located in the house, and the use of wood as a substitute for electric heating. The Council's forecasting model does not explicitly assume a specific average dwelling unit size. However, the space heating energy use for each dwelling type in the model implicitly assumes the average dwelling size for existing dwellings in the model's base year (1979). The forecasting model's present implicit assumptions regarding average size for existing dwellings are shown in Table 5-27. Based on survey data, it appears that average new multifamily dwellings (five-plex and larger) and manufactured houses being built today are typically larger than the model assumes for all existing multifamily dwellings and manufactured houses. However, new single family housing (less than five-plexes) appears to be the same size as the existing single family stock. Therefore, engineering estimates of cost and energy savings from conservation actions in new multifamily dwellings and manufactured homes were scaled to match the forecast model's implicit assumptions regarding unit size. This was done by multiplying the engineering esti- mates of use, cost and savings by the ratio of average unit size implicitly assumed in the forecast model to the average floor area of new dwelling units. No size adjustment was made for new single family dwellings, because their size appears to be consistent with the existing stock. Once the adjustment for unit size is made, the forecasting model's estimate of multi- family space heating use is 4,415 kilowatt- hours per year compared to the engineering estimate of 5,720 kilowatt-hours per year for a similar sized unit. Similarly, the forecasting model estimates that space heating use in new manufactured homes is 10,365 kilowatt- hours per year compared to the engineering models estimate for a comparably sized unit of 12,280 kilowatt-hours per year. In addition to differences due to variations in dwelling unit size, the Council's engineering estimates of space heating energy use in new dwellings departs from the forecasting model due to underlying assumptions regarding appliance efficiency and family size. In order to match current (1985) con- sumption, the forecasting model must use current (1985) appliance efficiencies. How- ever, because the Council anticipates sub- stantial efficiency improvements in appliance energy use within the next seven to 15 years, the Council's engineering estimate of space Chapter 5 Table 5-28 Internal Gain Changes from More Efficient Appliances INTERNAL GAINS PROVIDED ENERGY USE PER UNIT (kWh/yr) Ad Cunvertin ann 1Al Fosabaat Saturation At Current At Forecast Percent Efficiencies Efficiencies APPLIANCE/SOURCE (units/household) Efficiencies Efficiencies Indoors (kWh/yr) (kWh/yr) Lighting? 1.00 690 650 90 620 585 Refrigerator 1.083 1,450 675 100 1,570 730 Range/Cooking 1.00 980 880 100 980 880 Freezer 53 1,170 520 50 310 140 Water Heater¢ 1.00 1,200 675 50 600 340 Television 2,000 set-hr/yr 200 200 100 200 200 Clothesdryer a 950 900 10 70 60 Dishwashers, Clotheswashers, & Misc. Appliances _ 1,750 1,500 50 875 750 People? 2.63/2.22 1,920 1,810 100 1,930 1,720 TOTAL 7,135 5,355 @Assumes 1,400 square foot home. For other floor areas, lighting loads should be scaled by floor area. Assumes one refrigerator is located inside the house and 50 percent of .165 refrigerators are located outside the house. cAssumes water tank has R-10 for current efficiencies, and R-20, with R-10 bottom board, and temperature setting of 130°F on 50 percent of tanks. Waste heat from water use is included with contribution from people. “Contribution from people includes 290 kilowatt-hours per year per occupant as sensible heat and 230 kilowatt-hours per year per occupant as latent heat. Also included is 565 kilowatt-hours per year of latent heat provided to the house from the use of warm water for cooking and bathing. heating use assumes the presence of more efficient appliances. Table 5-28 shows the difference in waste heat (i.e., internal gains) released inside typ- ical single family dwellings from people and appliances assumed by the forecasting model in 1985 and in 2005. At current efficiencies and persons per household, approximately 7,100 kilowatt-hours of heat are released each year inside the house by people, lights and appliances. However, with anticipated improvements in appliance effi- ciency and a reduction in the average number of people per household, this will drop to approximately 5,350 kilowatt-hours per year by 2005. Because this waste heat offsets the need for space heating, more efficient appliances mean larger space heating energy require- ments. Had the Council assumed less effi- cient appliances in its engineering estimates, the regional average space heating energy used in new single family houses built in 1985 would fall about 1.2 kilowatt-hour per square foot. This reduction amounts to about 1,600 kilowatt-hours per year in the average new single family house. However, failure to rec- ognize the installation of efficient appliances in this same house by the year 2005 would result in an underestimate of space heating energy needs by 0.9 kilowatt-hours per square foot per year. Therefore, the Council used the lower value, which reflects the appliance efficiency present in new homes over the majority of their useful life. The use of the lower quantity of waste heat in the Council's engineering estimate produces savings for space heating energy that are larger in the near term (when the region is surplus); however, this value results in better estimates of long-term requirements (when the region faces new resource decisions). Nearly 90 percent of the 1,735 kilowatt-hour per year difference between the engineering estimate of space heating use in single family houses and that shown by the forecasting model can be attributed to alternative assumptions regarding appliance efficiency. The Council increased the space heating energy use shown by the forecasting model for single family houses by approximately 1,600 kilowatt-hours per year to account for the interaction of space heating and appliance efficiency. Multifamily space heat- ing use was increased by just over 1,100 kilo- watt-hours per year, and manufactured home space heating use was increased by just under 1,700 kilowatt-hours per year. Although it appears likely that the remaining difference (approximately 500 kilowatt-hours per year) between the forecast model and engineering model’ estimates of space heating use can be accounted for by supplemental heating with wood, the Council has assumed as a conservatism that the difference represents energy efficiency improvements already implemented in new single family dwellings. However, this adjustment and the size adjust- ment result in a base use for space heating of approximately 11,400 kilowatt-hours per year for new single family houses, 5,570 kilowatt- hours per year for new multifamily dwellings and 12,525 kilowatt-hours per year for new manufactured housing. 5-25 Chapter 5 Fe RT hegrcas ee Hence ee ee Pe ee ee Pe Oo NUMBER OF NEW UNITS kimperunt, Cost” | NSitiwas nv Mw NUMBER OF NEW UNITS kanperunt — Scost | PEGONEGAWATTS. Public tou Total Public 10U millskWh = Public OU Total Public tou Total = Public. = 10U millskWh Public [OU Total 482,690 658,980 1,141,670 263 ° 5 15 o 15 60,493 102,655 163,148 0 ° 5 0 ° o 527 0 10 29 o 2 836 88 10 6 1 7 1,257 359 15 69 27 6 1,511 533 15 10 6 7 3,630 2,732 20 200 206 «= 406 1,956 1,107 20 14 13 26 4625 3,727 25 255 280535 2,530 1,770 25 17 21 38 5,034 4,136 30 277,311 589 3.193 2,690 30 22 32 54 5518 4,619 35 304 347652 4.113 3,317 35 28 39 67 6.189 5,291 40 341 398739 4,740 4,019 40 33 47 80 6410 5,512 45 353° 415768 5.442 4,624 45 38 54 92 7,034 6,136 50 388 «462849 6,047 5,168 50 42 61 102 8012 7,113 55 441 535977 6591 5,539 55 46 65 110 8012 7,113 60 441 535° (977 6.961 5,724 60 48 67115 8012 7,113 65 441 535977 7,147 5,907 65 49 69119 8.012 7,113 70 441 535° «977 7,330 5,938 70 51 70 120 8012 7,113 75 441 535977 7,361 5,938 75 51 70 120 8.012 7,113 80 441 535 «977 7,361 5,938 80 51 70 120 8.012 7,113 85 441 535977 7,361 5,938 85 51 70 120 8012 7,113 90 441 535977 7,361 5,938 90 51 70 120 8012 7,113 95 441 535 «(977 7,361 5,938 95 51 70 120 8012 7,113 100 441 535977 7,361 5,938 100 51 70 = 120 Table 5-30 Technical Savings per Unit and Megawatts for New Multifamily Units SAVINGS LEVELIZED —- REGIONAL POTENTIAL mM SH Ty IESG mem nt 129,883 138,016 267,899 154 ° 5 2 ° 2 319 0 10 5 ° 5 780 127 15 12 2 14 1,607 954 20 24 15 39 2,558 1,905 25 38 30 68 2,646 1,993 30 39 31 1 2.765 2,112 35 4 33 74 3.170 2,516 40 47 40 87 3,226 2,573 45 48 4 88 3,332 2,679 50 49 42 92 4,383 3,730 55 65 59124 4,395 3,742 60 65 59124 4,406 3,753 65 65 59124 4417 3,764 70 65 59125 4,429 3,776 75 66 59125 4,440 3,787 80 66 60 126 4,440 3.787 85 66 60 126 4,440 3.787 90 66 60 126 4,440 3,787 95 66 60 126 4,440 3,787 100 66 60 126 5-26 The Council anticipates current research activities in the region will measure actual space heating consumption in new dwell- ings. When this information becomes avail- able it can be used to adjust both the engineering estimate and the forecast model estimate of space heating use in new dwellings. Tables 5-29 through 5-31 show the technical savings per unit and the average megawatts of technical conservation potential from improvements in space heating efficiency in new single and multifamily dwellings and manufactured houses. The achievable con- servation potential for new single family and multifamily dwellings assume a gradually increasing share of new electrically heated residences that install all regionally cost- effective space heating conservation mea- sures between 1986 and 1990. This share is 35 percent in 1986, 45 percent in 1987, 60 percent in 1988 and 75 percent in 1989. After 1989, 85 percent of all new electrically heated single family and multifamily units are assumed to install all regionally cost-effective measures. Similarly, gradual increases in the share of new manufactured houses (10 per- cent in 1986, increasing at an additional 10 percent per year) are assumed to include all regionally cost-effective measures between 1986 and 1989. After 1989, 50 percent of all new electrically heated manufactured houses are assumed to install all regionally cost-effective measures. The combined total of achievable space heating conservation potential in new resi- dences included in the Council's high load forecast is 725 average megawatts. Electric Water Heating Conservation The energy used to heat water represents the second largest end-use of electricity in the residential sector. Figure 5-5 shows the tech- nical potential for improving the efficiency of residential water heating at various costs of electricity. These savings represent better insulated water heaters, pipe wraps, and more efficient appliances that use hot water (e.g., clotheswashers and dishwashers). Chapter 5 Cumulative Average Megawatts 600 500 400 300 200 100 0 1 2 3 4 5 6 7 8 Conservation Cost (Cents/kWh) Figure 5-5 Technical Conservation Potential from Residential Water Heating Measures The cost-effective technical potential identi- fied by the Council for electric water heaters is about 514 average megawatts. The achievable portion of this, about 377 average megawatts, represents about 18 percent of water heating loads in 2005. The average cost of improving the efficiency of electric water heaters is 1.8 cents per kilowatt-hour. The Councils assessment of the conserva- tion potential available from improved resi- dential water heating efficiency involved three steps. These were to: 1. Estimate the cost and savings potential available from improved water heating efficiency. 2. Develop conservation supply functions for technical and achievable potential. 3. Calibrate savings to the Council's forecast. Step 1. Estimate the cost and savings potential available from improved water heating efficiency. The amount of energy consumed for water heating depends on two factors: standby losses and variable use. Standby losses refer to the energy that is used during storage to keep the water hot; they are determined by the temperature of the water and insulation levels of the hot water storage tank and supply piping. Vari- able use is the amount of hot water actually used in the household. Variable use differs substantially among households, depending upon such factors as the habits and number of occupants, and the stock of appliances that use hot water (such as clotheswashers and dishwashers), as well as the temperature of the hot water and the cold water that enters the tank. The base use of water heaters from which conservation potential could be estimated was derived by reviewing current research on the question. Table 5-32 summarizes avail- able data on standby losses from conven- tional (typically R-5) tanks. Water heat was directly submetered in all field studies. Labo- ratory tests on individual units had lower standby losses than those found in field tests. The average value of the full sample is 1,610 kilowatt-hours per year, identical to the Seattle City Light number of 1,610 kilowatt- hours per year, which was used in the 1983 plan. This is the value used for conventional tank standby losses. 5-27 Chapter 5 Table 5-32 Data on Standby Losses from Conventional Water Heater Tanks STANDBY SOURCE (kWh/yr) N NOTES Seattle City Light 1,610 26 ~All unwrapped, submetered Biemer/Auburg '84 1,375 1 Laboratory tests Goldstein/Clear 1,468 Calculated for 1960-1980 vintage tanks Ek ’82 (#36 1,483 1 Laboratory Test Ecotope '82 1,995 91 Some wrapped, many different lecations. Ecotope Heat Pump 1,731 39 Median standby losses in three cities are weighted by Study climate zone's contribution to regional population. Average 1,610 Table 5-33 Variable Demand Use for Hot Water GALLONS/YEAR SOURCE PER PERSON N NOTES Lawrence Berkely 5,582 Laboratories Natural Resources 5,411 Calculated Defense Council Seattle City Light 6,019 26 Calculated Ecotope Heat Pump 7,680 38 Submetered participants selected on basis of Study family size and high water use. Bavir 7,094 Regression results from submetered sample. Long Island Light Co. 6,788 257 Submetered Average 6,429 gallons/person/year At 90° temperature differential this translates to: 1,399 kWh/person/year Table 5-34 Savings from Water Heating Measures (kWhiyr at 80° Temperature Differential) Standby losses for R-5 tank R-20 tank R-11 wrap Bottom board Thermal trap Total percent savings SEATTLE CITY LIGHT BIEMER/AUBURG '84 1,610 1,375 Savings %ofuse Savings %ofuse 700 43.5 550 40.0 100 11.0 192 23.3 40 49 19 3.0 180 23.4 74 12.1 63.3 60.7 5-28 Variable use for the pre-conservation situa- tion was estimated from studies that reported the gallons of hot water used per person or per household. Table 5-33 summarizes the empirical data. Hot water demand was actu- ally measured in some cases, while in others it was calculated. If the figures are converted to kilowatt-hours per person, the average kilowatt-hour use per occupant is approx- imately 1,400 kilowatt-hours per year. Given the tremendous variation inherent in hot water variable use, this number is reasonably close to the value used in the 1983 plan, which is 1,310 kilowatt-hour per occupant for an 80° temperature differential. The Council continued using the 1,310 kilowatt-hours per occupant for base year use, since available data did not dictate a change. The two primary sources for estimating the savings available from various standby con- servation measures were a Seattle City Light (SCL) study, which served as the basis for the 1983 plan figures, and anew laboratory study conducted by Bonneville in 1984 (Biemer and Auburg). Both studies tested R-5 tanks. These studies started with different standby losses (1,610 kilowatt-hour per year for SCL compared to 1,375 kilowatt-hour per year for the Bonneville study) and found different absolute savings estimates. However, the two studies produced comparable results in terms of the relative savings attained for all measures combined, and for two of the four individual measures. The results for each study are shown in Table 5-34. Water heater wraps and thermal traps are the individual measures with the greatest difference. The Council used an average of the percent sav- ings reported in both studies, and applied these to the base year standby use. The cost of efficient tanks is from a survey done by the Pacific Northwest Utilities Con- ference Committee. This cost is the incre- mental cost of purchasing an efficient rather than a conventional tank. Costs for water heater wraps are from Bonneville. Costs for thermal traps, bottom boards, and energy- saving showerheads were adapted from work done at Seattle City Light. Conservation measures for variable use include clotheswashers and dishwashers that use hot water more efficiently, and energy-saving showerheads.'° The costs and savings available from efficient clothes- Chapter 5 washers and dishwashers and costs for showerheads were taken from work done at Lawrence Berkeley Laboratories (LBL). LBL estimated more efficient clotheswashers would save about 355 kilowatt-hours per year and more efficient dishwashers would save 245 kilowatt-hours per year. Estimates of savings made by the Natural Resources Defense Council for dishwashers are some- what lower. Energy-saving showerheads are assumed to save 35 percent of the hot water used for showers. The lifetimes of the measures discussed above are 12 years, except for showerheads at 20 years, and clotheswashers and dish- washers assumed to be ten years. It should be noted that the savings for standby loss conservation measures have been reduced to reflect the interaction between internal gains from water heaters and space heating electricity consumption. This is described in a section that follows the analysis of refrigerator and freezer conserva- tion potential. Base case heat pump water heater costs were taken from work done by the Pacific Northwest Utilities Conference Committee. For a sensitivity analysis described later in this section, costs from an Electric Power Research Institute paper were used. Heat pump water heater savings are from a recent research study conducted for Bonneville. This report indicated that heat pump water heaters saved an average 40 percent of total hot water use. Savings are calculated by assuming that all of the less expensive con- servation measures have been installed first. The lifetime of heat pump water heaters is assumed to be 12 years. The costs of solar water heaters were taken from work done by the Oregon Department of Energy, where the system costs of the solar water heaters were derived from state tax forms. The cost for a system installed by a small contractor was about $4,000. Costs ranged from $3,000 if owner-installed to $5,000 if a large marketing company did the work. Where solar systems were installed as Part of a contract to install numerous systems in one geographic area, the costs could get as low as $2,400. Base case costs used in this analysis are $4,000 per installation plus $10 per year maintenance costs. For a sen- Table 5-35 Measure Costs and Savings for Water Heaters MEASURE SAVINGS WITH LIFE CYCLE MEASURE COST SAVINGS INTERACTION® CENTS/KWH Coste Base Use = 4,454 kWh/Year Base Case $ 0 0 0 0 1,177 Efficient Showerhead $ 34.20 450 450 0.55 1,104 Efficient Clotheswasher $ 22.00 355 355 0.78 1,056 Efficient Dishwasher $ 22.00 245 245 1.13 1,032 Efficient Tank $ 45.00 344 286 1.41 1,005 Thermal Trap $ 22.80 173 144 1.42 992 R-11 Wrap $ 42.40 136 113 3.36 1,006 Bottom Board $ 12.54 26 22 5.19 1,013 Heat Pump $1,488.00 1,090 1,090 14.70 2,274 Solar Water Heater $4,175.00 1,090 1,090 27.40 3,291 aWithout heat pump installed. >This reflects the reduced savings from standby loss measures due to the interaction with electric space heating. ¢Parameters for the life cycle cost calculation are: 10 percent consumer discount rate, zero electricity price escalation and an average residential rate of 3.7 cents per kilowatt-hour. All measures costs and savings are calculated proportionally based on 12 years. sitivity analysis, $2,400 per installation plus maintenance costs were used. Savings are from the interim results of Bonneville’s pro- gram of monitoring solar water heaters and are about 40 percent of total water heating use. Lifetime is estimated to be 20 years. The above assumptions led to the cost-effec- tiveness calculation for each measure shown in Table 5-35. This table assumes an average household with 2.4 occupants. It shows the marginal cost of each water heating conser- vation measure, starting with an estimated average tank"! and water heater use. Except for heat pumps, solar water heaters and bot- tom boards, none of the measures exceed 5.0 cents per kilowatt-hour even after taking into account the interactive effect with space heating. Bottom boards are on the margin of being cost effective and would certainly be so if other measures could not be installed. The analysis suggests that energy efficient tanks, wrapped with insulating blankets and fitted with thermal traps, and all variable reduction measures, are cost effective. Also shown in Table 5-35 is the cost of pur- chasing and operating the water heater over a 12-year period (called the “life cycle cost”). Life cycle costs have been used by the Coun- cil to determine the attractiveness of conser- vation measures to the consumer. All mea- sures that are cost effective to the region— less than 5.0 cents per kilowatt hour—also result in a lower life cycle cost to the con- sumer than the base case. As with most water heating conservation measures, the cost effectiveness of heat pump and solar water heaters depends on the amount of hot water consumed in the household, which can vary significantly even among households of the same size. As shown in Table 5-35, for an average family size with a water heater where all cheaper conservation measures are installed first, the heat pump and solar water heaters are not cost effective. Table 5-36 shows the calcu- lated levelized cost for heat pumps and solar water heaters under various assumptions of family size, pre-installed conservation mea- sures, and capital cost. 5-29 Chapter 5 Table 5-36 Sensitivity Analysis on the Cost Effectiveness of Heat Pump and Solar Water Heaters Base Case: Base costs of heat pump and solar water heaters, all cheaper conservation measures installed first. WATER HEATER WATER HEATERS LEVELIZED LEVELIZED (persons household) COST why) (centakWh) cOsT ow) (centexwn) 2.4 $1,488 1,090 14.7 $4,150 1,090 27.4 3 $1,488 1,299 12.3 $4,150 1,299 23.0 4 $1,488 1,648 9.7 $4,150 1,648 18.1 5 $1,488 1,997 8.0 $4,150 1,997 15.0 6 $1,488 2,346 6.8 $4,150 2,346 Ne? Sensitivity 1: No variable usage conservation measures installed. 2.4 $1,488 1,510 10.6 $4,150 1,510 19.8 3 $1,488 1,824 8.8 $4,150 1,824 16.4 4 $1,488 2,348 6.8 $4,150 2,348 12.7 5 $1,488 2,872 5.6 $4,150 2,872 10.4 6 $1,488 3,396 47 $4,150 3,396 8.8 Sensitivity 2: Less expensive installation costs for heat pump and solar water heaters. 2.4 $ 765 1,090 76 $2,550 1,090 16.9 3 $ 765 1,299 6.3 $2,550 1,299 14.1 4 $ 765 1,648 5.0 $2,550 1,648 tia 5 $ 765 1,997 41 $2,550 1,997 9.2 6 $ 765 2,346 3.5 $2,550 2,346 78 Step 2. Develop conservation supply functions for technical and achievable potential. The savings for each measure were multiplied by the number of units that appear in the forecast between 1992 and 2005 to which that measure applied. The savings from showerheads is limited by the number of new houses likely to be built between 1992 and 2005 with electric water heaters. The number of clotheswashers and dishwashers is assumed to track the number of new electric water heaters with saturations of 78 percent and 50 percent respectively. The number of units was then multiplied by the achievable saturation, also measure-spe- cific, that the Council felt could be secured between 1992 and 2005. The number of units and the achievable saturation for the high demand forecast appear in Table 5-37. The described calculation derives the number of average megawatts that can be secured. Table 5-36 shows that the cost of solar water heaters only approaches the 5.0 cents per kilowatt-hour threshold in high water use households or if low capital costs emerge. Heat pump water heaters are cost effective for a six-person household if no variable con- servation measures costing less than the heat pump, such as energy-saving show- erheads, are installed first. In addition, if the installation and maintenance costs of heat pump water heaters is significantly reduced over the next few years, they might become cost effective for household sizes of four and greater. This estimate however, does not take into account the negative impact on space heating that a heat pump water heater would have if installed in the heated space. Heat pump and solar water heaters are not consid- ered cost effective in the current analysis. However, heat pump water heaters and to some extent solar water heaters appear to be promising resources. The Council will re- evaluate costs and savings for future analy- ses of cost-effective conservation resources. Step 3. Calibrate the supply curve to the Council’s forecast. The engineering and field measurements described above predict a base use of about 4,450 kilowatt-hours per year for an average household. As men- tioned above, this figure represents a mix of unwrapped conventional tanks, wrapped tanks, and some efficient tanks. Standby losses in 1992 are anticipated to differ from the standby losses included in the above described number due to a different mix of conventional and efficient tanks. In order to capture this change, and to be consistent with the Council's forecast of electricity con- sumption, the base case needs to vary with the forecast's prediction of water heating electricity use in 1992. In the demand fore- cast, base case use in 1992 varied between 4,602 kilowatt-hours per year and 4,223 kilo- watt-hours per year, depending on the fore- cast scenario. For purposes of the supply curve, the difference between the forecast base case use and the calculated base case use was assumed to be due to level of water heater insulation. This difference altered the supply curve somewhat to account for the different base case uses. For the high demand forecast, this adjustment reduced the technical supply curve by 50 average megawatts at about the 2 cent per kilowatt- hour level. The amount of conservation available in the high demand forecast at various costs is pre- sented in Table 5-38. Conservation in Other Residential Appliances Approximately one-quarter of the electricity currently consumed in the residential sector is used to operate refrigerators, freezers, stoves and lights. This section describes the conservation assessment for refrigerator/ freezers and freezers. For electric ranges, the most significant conservation potential would stem from the substitution of bi-radiant ovens for electric ovens. This is a technology that should be watched for future assess- ments of conservation potential. The tech- nical conservation potential from replacing traditional incandescent bulbs with fluores- cent bulbs in residential applications repre- sents roughly 50 average megawatts. When the region approaches the end of the current surplus period, the savings from fluorescent bulbs should probably be pursued. No pro- gram should be pursued in the meantime, 5-30 Chapter 5 however, because the bulbs are short-lived. As a conservatism in the conservation assessment for the resource portfolio, the Council did not include an estimate of fluorescent bulb savings in residential applications. The Council has included only 368 average megawatts in its estimates of achievable Potential for refrigerators and freezers, which ismuch less than the amount that could tech- nically be accomplished for a marginal mea- sure cutoff of 5.0 cents per kilowatt-hour as described below. Even so, achievable con- servation through the use of more efficient refrigerators and freezers represents a 12 Percent savings by 2005. At an average cost of 0.8 cents per kilowatt-hour, these savings are the most cost-effective conservation resource available to the region. The savings identified by the Council are based on the level of efficiency improve- ments resulting from revised appliance stan- dards recently adopted in California that become effective in 1992. The new California Standard is phased in starting in 1987, with a more stringent standard becoming effective in 1992.12 The Council's savings reflect the impact of the 1992 standard only, although both the level of the 1987 and 1992 standards are cost effective for this region. The Council found that refrigerators and freezers that go significantly beyond the California 1992 stan- dard are not yet commercially available, although engineering estimates indicate that technologies to beat the 1992 standards are attainable.13 An alternative design re- frigerator that beats the energy requirement of the 1992 standard by about two-thirds is available today, but is quite costly because each unit is handmade. This refrigerator fur- ther corroborates the engineering estimates that refrigerators can be made to beat the 1992 California standards. Savings from going beyond the standard are substantial and represent a promising resource for future evaluations of conservation potential if such units become commercially available. The Council used four steps to evaluate the savings available from refrigerator and freezer efficiency improvements. These were to: Table 5-37 Number of Eligible Units by 2005 and Achievable Conservation Percent for Water Heating Measures High Demand Forecast MEASURE NUMBER ACHIEVABLE PERCENT Efficient Showerheads 1,630,000 90% Efficient Clotheswashers 3,568,500 50% Efficient Dishwashers 2,287,500 50% Efficient Tanks 4,575,000 90% Thermal Trap 4,575,000 85% R-11 Wrap 4,575,000 85% Bottom Board 4,575,000 85% Table 5-38 Conservation Available from Water Heaters CUMULATIVE LEVELIZED COST TECHNICAL POTENTIAL (cents/kWh) (average megawatts) 1 269 2 473 3 504 4 518 5 525 6 526 7 526 1. Estimate the cost and savings potential available from improved refrigerator and freezer efficiency. 2. Develop technical and achievable conser- vation potential. 3. Calibrate the achievable conservation potential to the Council's forecast. Step 1. Estimate the costs and savings potential available from improved refrigerator and freezer efficiency. The potential for energy savings from improved refrigerator and freezer operating efficiencies is well documented. The U.S. Department of Energy (DOE) and the California Energy Commission (CEC) have recently reviewed the option of appliance efficiency standards. The DOE proceeding limited its investigation of efficiency improvement design options to those based on “available” technology. Avail- able technology was defined by DOE as those technologies implemented in units available and sold in 1980. In addition, the DOE analysis only included options that had a payback period of less than five years. The payback period for an energy-saving design option is the length of time it takes an average consumer (in this case, a national consumer) to recover the higher purchase price through the lower cost of energy used to operate the appliance. Both these limits significantly reduced the efficiency options evaluated by DOE.14 The CEC hearings looked at technologies that went beyond the measures analyzed in the DOE hearings. This resulted in a much larger and broader set of designs to reduce refrigerator energy consumption. The CEC proceedings resulted in adoption of revised refrigerator and freezer standards that will lower the current California standards first in 1987 and again in 1992.15 The level of effi- ciency chosen for the most stringent stan- dard—effective in 1992—was set at about the strongest level investigated by DOE. As a consequence, this standard did not include the additional measures that emerged during the CEC hearings. 5-31 Chapter 5 Measure Cost and Savings for Prototype Refrigerators COST OF DISCOUNTED USE MEASURE CUMULATIVE SAVINGS LIFE kWh/yr COST COST (cents/kWh)* CYCLE COST> Base Case in DOE analysis 1,354 $ 0 $ 0 0 $1,000 Foam insulation substituted 1,208 $ 7.38 $ 7.38 0.44 $ 969 in door Compressor EER¢ 3.65 1,072 $ 7.44 $ 14.82 0.47 942 Anti-sweat switch 978 $ 8.17 $ 22.99 0.75 926 Increase door thickness to 940 $ 3.72 $ 26.71 0.84 919 2" 2.4" cabinet insulation, 2/2” 768 $14.82 $ 41.53 0.74 $ 890 freezer insulation High efficiency fan 688 $10.98 $ 52.51 1.18 $ 880 2.4" cabinet insulation, 3” 613 $13.18 $65.69 1.52 $ 874 freezer insulation EER 4.5 518 $27.45 $ 93.14 2.48 $ 877 Evacuated panel 228 $88.40 $181.54 2.63 $ 890 EER 4.8 217 $ 5.49 $187.03 4.10 $ 893 Double freezer gasket 204 $20.60 $207.63 13.8 $ 910 Double gasket-door 186 $34.04 $241.67 16.0 $ 940 Table 5-39 aAdjusted for space heat interaction. >Parameters used for the life cycle cost analysis included: 10 percent consumer discount rate, 22 year lifetime, zero electricity price escalation, and an average residential rate of 3.7 cents per kilowatt- hour. cEER stands for Energy Efficiency Ratio. 5-32 California's 1992 standard is illustrated by taking a frost-free 17 cubic foot refrigerator as an example. In 1992 this unit will be required to use less than 672 kilowatt-hours per year. The current energy use of this appliance will be almost halved, compared to 1,156 kilowatt- hours per year, the average energy use of the same unit sold in 1983 according to the Asso- ciation of Home Appliance Manufacturers (AHAM). The Council used the information that emerged from the DOE and CEC hearings to evaluate the cost effectiveness of efficiency improvements in refrigerators and freezers in the Northwest region. In some cases this meant adjusting savings for the interaction with space heating needs (described in the following section) or re-ordering measures so they were applied with the most cost effective first. In this analysis, the Council used a 17 cubic foot automatic defrost prototype to rep- resent refrigerators, and a 15 cubic foot chest freezer to represent freezers. Automatic defrost units represent approximately 78 per- cent of the refrigerators sold today. Cost effectiveness was analyzed from both the perspective of the region and the indi- vidual consumer. Table 5-39 presents this cost and savings information for the pro- totype 17 cubic foot refrigerator. Savings and levelized costs include the interaction of appliance efficiency improvements with space heating requirements, described more fully in the next section. The costs of measures and their savings were evaluated starting with the base case from the DOE proceedings. However, refrigerators on the market today incorporate some of the measures evaluated in Table 5-39, and consequently a number of models are more efficient than the base case in the DOE analysis. The costs and savings curve, however, can still be used to represent the relative efficiency improvement available for a given cost. The base case is just moved further down the curve to represent currently sold units that have incorporated some of the measures listed in the table. For example, many units sold today have more insulation than the 1.6 inch thick fiberglass insulation assumed in the base case. However, other measures in the table are still viable options for reducing consumption in the average refrigerator. Since a measure'’s levelized cost Chapter 5 is independent of where the base case origi- nates on the curve, the fact that current units exceed the DOE 1980 base case does not mean that the measures in the table are any less cost-effective. Improving the efficiency of the prototype refrigerator to the level where the last mea- sure installed has a marginal cost of 5.0 cents Per kilowatt-hour, a new prototypical 17 cubic foot refrigerator would save 939 kilowatt- hours per year beyond 1983 current average use of 1,156 kilowatt-hours per year. This results in a total consumption of about 217 kilowatt-hours per year. The purchase and Operation costs of the refrigerator over its lifetime (life cycle cost) at 10 percent discount rate is less at the cost-effectiveness limit (a consumption of 217 kilowatt-hours per year) than at the base case. However, the 5.0 cents per kilowatt-hour cost-effectiveness limit results in a much lower energy use than the 1992 California standard. The 1992 Califor- nia standard results in refrigeration use of about 672 kilowatt-hours per year for the pro- totype, which represents a marginal cost of about 1.2 cents per kilowatt-hour, and a net reduction in life cycle cost. For the average stock of refrigerators instead of the prototype, the level of the 1992 standard is about 675 kilowatt-hours per year. The level of the 1992 standard for average refrigerators was used for the Councils conservation assessment. The costs and savings for measures that can be applied to the prototype chest freezer appear in Table 5-40. Less extensive analy- sis was done on freezer conservation poten- tial than on refrigerator potential in both the Department of Energy and the California hearings. The last measure analyzed has a marginal cost of only 1.7 cents per kilowatt- hour—even after the cost of the last mea- sure was increased significantly to account for the fact that it represents advanced tech- nology. Consumption is reduced from 720 kilowatt-hours per year, which is average 1983 consumption for this type of freezer according to AHAM, to 342 kilowatt-hours per year, a savings of 378 kilowatt-hours per year if all measures are used. All measures investigated resulted in lower purchase and operation costs than the base case over the life of the freezer. The 1992 California stan- dard for average freezers is about 519 kilo- watt-hours per year. The level of the 1992 Table 5-40 Measure Cost and Savings for Prototype Freezers USE MEASURE CUMULATIVE COST OF SAVINGS DISCOUNTED LIFE kWh/yr COST cost (cents/kWh)* CYCLE COST Base Case in DOE analysis 851 $ 0 $ 0 0 $739 Compressor EER 3.5 788 $ 4.58 $ 458 0.58 $726 Foam Insulation sub- 704 «=$ 8.25 $12.83 0.78 $710 stituted in door Increase door thickness to 678 $ 3.12 $15.95 0.95 $706 2" Increase cabinet thickness 571 $19.34 $35.29 1.43 $696 to 2.5” Advanced technology 342 $50.00 $85.29 1.73 $681 alncludes interaction with space heater. >Parameters used for the life cycle cost analysis are: 10 percent consumer discount rate, 22 year lifetime, zero electricity price escalation, and an average residential rate of 3.7 cents per kilowatt- hour. standard for average freezers was used to establish the Councils limit of available and reliable conservation for this appliance. Step 2. Develop conservation supply functions for technical and achievable potential. The savings resulting from the level of the 1992 California refrigerator and freezer standards were multiplied by the number of refrigerators and freezers pur- chased between 1992 and 2005. Since the energy load that has to be met by thermal plants after conservation actions are taken is determined by the forecast, the savings from conservation measures in refrigerators and freezers has to be evaluated consistently with the values carried in the forecasting model. There is reason to suspect that the forecast- ing model's projections of electricity use by these appliances are too high. The model projects refrigerator use of about 1,400 kilo- watt-hours per year in 1984, which is signifi- cantly above the average use of 1,140 kilo- watt-hours per year estimated by AHAM based on actual 1984 sales. This discrep- ancy in 1984 suggests that the model's pro- jections in 1992 could be too high. This would have two effects: First, estimates of savings due to the California standards would be too high, and second, projected demands would be too high by an equivalent amount. This evidence raises several questions. Is the DOE test procedure, on which the AHAM estimates are based, an accurate simulation of actual use, which the forecast tries to cap- ture? Are the technology curves used in the forecasting model accurate representations of available technology? Would it be reason- able to adjust the consumers implicit dis- count rates in the forecasting model to make the model match AHAM-estimated appli- ance efficiencies in 1984? A thorough analysis of this issue will take a significant amount of work. This analysis has not been given first priority in the preparation of this plan, because the policy impact of the possible error is negligible. The amount of generating resources required to serve these appliances after the effect of the standard would be unaffected by such an error. Like- wise, the marginal cost effectiveness of mea- sures included in the standard is unchanged. Consequently, the Council's forecasting model was used to estimate the base case use of refrigerators and freezers in 1992. In the high demand forecast in 1992, new refrigerators used 1,402 kilowatt-hours per year and freezers used 1,147 kilowatt-hours per year. For refrigerators, a base use of 1,402 kilowatt- hours per year and a standard in 1992 of 675 kilowatt-hours (kWh) per year resulted in a total technical potential: Chapter 5 [4,483,000 refrigerators purchased 1992-2005 x (1,402 - 675 kWh/year) x (1 - .2 space heat interaction)] 8,766,000 kWh per average megawatt(MWa) For freezers, a base case use in 1992 of 1,147 kilowatt-hours per year and a standard of 519 kilowatt-hours per year resulted in a total technical potential: = 297MWa [1,796,000 freezers purchased 1992-2005 x (1,147 - 519 kWh/year) x (1 - .13 space heat interaction)] 8,766,000 kWh per average megawatt (MWa) These technically achievable savings were reduced by 10 percent to account for noncompliance. Should current shipment-weighted usage for refrigerators and for freezers be used as base year usage instead of the forecasting values, total savings would be about 190 average megawatts for refrigerators and 50 average megawatts for freezers. The cost effectiveness, or desirability, of savings from refrigerator and freezer efficiency improve- ments is not reduced by using either the base case from the forecasting model or the case from shipment-weighted average efficiencies. The Interaction between Internal Gains and Electric Space Heat A house is warmed by a combination of inter- nal and external heat sources. Internal heat comes from incidental or waste heat given off by appliances and people (usually called “internal gains’) and from the space heater. The external source of heat is primarily radi- ant energy from the sun (usually called “solar gain”). These heating sources are in balance, and if the heat produced by any one of them decreases, more heat must be added from the other components to keep the house at the same temperature. This section deals with the interaction between the waste heat given off by appliances and the heat supplied by the space heater. 16 5-34 112 MWa If the efficiency of an appliance located inside the heated space, such as a refrigerator, is improved, the unit both uses less energy and gives off less waste heat. This in turn causes the space heater to use more energy in order to keep the house at the same temperature it was before the refrigerator's efficiency was improved. . The balance between how much energy is saved by the refrigerator and how much extra heating is done by the space heater depends on many factors. A prominent factor is the insulation level of the house. The better insu- lated a dwelling is, the less useful the waste heat from the appliance. For example, the space heater must produce about an addi- tional 5 kilowatt-hours per year for every 10 kilowatt-hours per year saved by the appliance efficiency improvement, assuming all of the following: the appliance is located in the heated space, electricity is the space heating fuel, no air conditioning is installed, and the house is fairly uninsulated. In other words, only 50 percent of the savings from improving appliance efficiency would be realized. This estimate accounts for periods of the year, such as summer, when additional space heat is not necessary. This estimate must be tempered by other intervening vari- ables to calculate the average expected impact on the Northwest electrical system from improved appliance efficiencies. First, the appliance must be one that produces internal gains. Many do not; for example, about half the electric freezers in the region are located outside heated areas. Waste heat generated from freezers (and other appliances) that are outside the heated shell of the house does not contribute to internal gains. Consequently, any efficiency improve- ments in appliances located outside the house would be fully realized as 100 percent energy savings and would not require that additional heat be provided by the furnace. Second, a number of electrical appliances that do produce internal gains, such as refrigerators, are located in houses that do not use electricity for their space heating. In this case, the full amount of electricity saved by improving the appliance's efficiency is realized by the region's electrical system. Finally, the reduction of internal gains is a benefit to the house if air conditioning equip- ment is installed. In this case, less cooling needs to be provided in the summer to offset the internal gains from inefficient appliances. For water heaters, only the standby use of holding hot water in the tank (for units located in the house) is an internal gain. Variable hot water demand does not contribute signifi- cantly to internal gains, even though it uses electricity.17 Consequently, only efficiency improvements in standby use for tanks located in the house increase the heat needed from the space heater. When all of these factors are considered, electricity used for space heating must make up, on average in the region, about 17 per- cent, 20 percent and 13 percent of the sav- ings from standby losses on water heaters, refrigerators, and freezers, respectively. These figures were used to devalue the sav- ings obtainable from these appliances in the preceding cost-effectiveness evaluations. Primary Sources for the Residential Sector Space Heating: Bonneville Power Administration, Pacific Northwest Residential Energy Survey, 1983. Letter from Reidun Crowley, Puget Power, May 1, 1985, on costs of residential conserva- tion retrofits. Gates, Howard, Manufactured Housing Institute, Optimum Thermal Insulation for Manufactured Homes, September 1984, revised October 1984. Goldman, C.A., Technical Performance and Cost-Effectiveness of Conservation Retrofits in Existing U.S. Residential Buildings: Analy- sis of the BECA-B Data Base, LBL-17088, October 1983. Hirst, Eric, Richard Goeltz, Dennis White, Benson Bronfman, David Lerman, Kenneth Keating, Evaluation of the BPA Residential Weatherization Program, ORNL/CON-180, June 1985. Housing Industry Dynamics, Special Report, December 1984, prepared for the Bonneville Power Administration. Palmiter, Larry and David Baylon, Assess- ment of Electric Power Conservation in the Pacific Northwest, Volume 1, Residential Building Conservation, June 1982 (Draft), Submitted to Battelle Pacific Northwest Laboratories by Ecotope Group. Palmiter, Larry and Mike Kennedy, Assess- ment of Electric Power Conservation and Supply Resources in the Pacific Northwest, Volume 1, Supplement A — Heat Pumps in Residential Buildings, January 1983 (Draft), Submitted to Battelle Pacific Northwest Laboratories by Ecotope Group. Palmiter, Larry and Mike Kennedy, Assess- ment of Electric Power Conservation and Supply Resources in the Pacific Northwest, Volume 1, Supplement B — Passive Solar, Internal Gains and Monthly Loads of Resi- dential Buildings, February 1983 (Draft) Submitted to Battelle Pacific Northwest Laboratories by Ecotope Group. Water Heating and Appliances: Letter from David Robison, Oregon Depart- ment of Energy, March 13, 1985, on the costs of solar water heaters. Letter from Fred Avril, Long Island Lighting Company, April 2, 1984, on annual hot water usage from solar demonstration program. Memorandum to File from Alan Cooke, Pacific Northwest Utilities Conference Com- mittee, April 18, 1984, “Conservative Assumptions: The Plan Assumes No Use of Most Efficient Water Heater Heat Pumps.” Memorandum to Residential Supply Curve Work Group from Alan Cooke, Pacific Northwest Utilities Conference Committee, January 8, 1985, “Water Heater Cost Comparison.” Hanford, Jim, Mike Kennedy, Mary Jane DeLaHunt, Larry Palmiter, Ecotope Group, Heat Pump Water Heater Field Test, DRAFT Final Report, April 3, 1985, Contract with Bonneville. Biemer, Jon, C. Douglas Auburg, Calvin Ek, Bonneville Power Administration, “Domestic Water Heating — Summary Research Find- ings for Conventional Systems’ in Conserva- tion in Buildings: Northwest Perspective, May 19-22, 1985, in Butte, Montana. Harris, Jeff and Chris Dent, “Measured Results of 75 Solar Water Heating Systems in the Northwest; Interim Results’ in Conser- vation in Buildings: Northwest Perspective, May 19-22, 1985, in Butte, Montana. U.S. Department of Energy, Consumer Prod- ucts Efficiency Standards Engineering and Economic Analysis Document and Supple- ment, March 1982. Natural Resources Defense Council, Appliance Survey, 1981. Natural Resources Defense Council, A Model Electric Power and Conservation Plan for the Pacific Northwest, November 1982, Prepared for the Northwest Conserva- tion Act Coalition. Chapter 5 Messenger, Michael, and R. Michael Martin, California Energy Commission, Technical Analysis of the Energy Conservation Poten- tial for Refrigerators, Refrigerator-Freezers and Freezers: Part |-Recommended Effi- ciency Levels, and Part II-Final Staff Recom- mendations for Revised Standards/Fleet Average Goals, Docket 84-AES-1, Item Code P400-84-013, Revised August 1984. Association of Home Appliance Manufactur- ers, 1983 Energy Consumption and Effi- ciency Data for Refrigerators, Refrigerator- Freezers and Freezers, June 1, 1984, Revised July 1, 1984. Geller, Howard S., Analysis of Minimum Efficiency Standards for Domestic Refrigerators and Freezers in the Pacific Northwest, Draft, February 1985, Prepared for Bonneville Power Administration. Sherman, Max, Mark Modera, Dariush Hekmat, Energy Impacts of Efficient Refrigerators in the Pacific Northwest, Feb- ruary 20, 1985, Prepared for Bonneville Power Administration. Reese, S.P., and H.A. Wall, Residential Elec- tric Water Heater Conservation Potential, 1981, Seattle City Light. Dobyns, J.E., and M.H. Blatt, Science Applications, Inc., Heat Pump Water Heat- ers, EM-3582 Research Project 2033-5, May 1984, Prepared for Electric Power Research Institute. Bavir, et. al., “Hour Use Profiles for Solar Domestic Hot Water Heaters in the National Solar Network,” Solar Engineering, 1981. Ek, Calvin, The Effects of External Insulation on Electric Water Heater: A Laboratory Report (Revised Edition), 1982, Bonneville Power Administration. Seattle City Light, Conservation Planning Process: A Status Report, 1982. Ek, C.W., and R.J. Miller, The Effectiveness of Anti-Convection Devices in Reducing Standby Losses from Domestic Water Heat- ers, 1982, Bonneville Power Administration, Division of Laboratories. 5-35 Chapter 5 Cumulative Average Megawatts 1,500 1,000 500 0 1 2 3 Conservation Cost (Cents/kWh) 4 5 6 7 Figure 5-6 Technical Conservation Potential from the Commercial Sector Commercial Sector The commercial sector consumed approx- imately 20 percent of the region's total energy sales in 1983, or about 2,936 average mega- watts. This sector's energy consumption is dominated by space heating, cooling, and lighting. The commercial sector consists of many diverse buildings that use electricity in myriad ways. Because of the complexity of electricity use, much less precision is possible for estimating the conservation potential in this sector compared to the residential sector. However, projects are currently underway in the region that will enable analysts to better understand the end-uses of electricity and better evaluate the conservation potential in this sector. This section evaluates the conservation potential from the array of traditional com- mercial buildings, such as offices and schools, as well as from less well known sources, such as pumping in municipal waste water treatment plants. Because of 5-36 their unique nature, waste water treatment plants are discussed in a separate section at the end of the text on commercial buildings. This section includes savings from both pri- vately and publicly-owned buildings. The Council's current conservation assess- ments for existing commercial buildings are based on experience from current regional programs and on engineering estimates. The conservation estimate for new commer- cial buildings is based on engineering esti- mates of how much electricity will be saved by the commercial model conservation stan- dards within each building type. Figure 5-6 shows the amount of commercial sector con- servation available at various costs in existing and new buildings and waste water treatment facilities. In the high demand forecast, the Council estimates 780 average megawatts of technical conservation potential in existing commercial buildings, 514 average mega- watts from new commercial buildings, and 15 average megawatts from waste water treat- ment plants. The potential from new com- mercial buildings includes only those savings resulting from the commercial model conser- vation standards. For existing commercial buildings, 732 aver- age megawatts of the technical potential is achievable in the high forecast. For new com- mercial buildings, about 430 average mega- watts of the technical potential is achievable in the high forecast. Achievable savings from existing commercial buildings represent about 25 percent of electrical load from these buildings in 2005, and are available at an average cost of 2.3 cents per kilowatt-hour. Achievable savings from new commercial buildings represent a savings of about 12 percent of load in the year 2005 and are available at an average cost of about 2.0 cents per kilowatt-hour. Similar to new resi- dences, new commercial buildings will last longer than the current electrical surplus. It is important to build these structures efficiently in order to avoid losing a cost-effective con- servation resource. The Council's estimate of conservation sav- ings from the commercial sector involved the following three steps: 1. Identify the current regional average con- sumption for typical existing commercial building categories and typical new com- mercial buildings. 2. Evaluate conservation potential in existing and new commercial buildings. 3. Develop estimates of realizable potential for conservation at various costs in new and existing commercial buildings. Step 1. Identify the current regional aver- age consumption for typical existing commercial building categories and typ- ical new commercial buildings. The Coun- cis commercial sector forecasting model contains representations of ten building cate- gories. Table 5-41 shows the annual energy use for all-electric commercial buildings existing in 1985, as estimated by the Coun- cils forecast. This table also presents esti- mates from several recent analyses of exist- ing commercial building consumption, which can be compared to the forecast values. To convey the relative importance of each build- ing type in the analysis, the percent of total electricity in 1985 consumed by each build- ing type is also shown. Chapter 5 In comparing the data shown in Table 5-41 and the forecast model assumptions, three factors should be kept in mind. First, the buildings shown in Table 5-41 were not selected to be statistically representative of the average. Second, the annual use figures in Table 5-41 from research on buildings in this region represent each building's total energy use regardless of the fuel source; total energy use is then converted to kilowatt- hours per square foot. Since many of these buildings use natural gas or fuel oil for some end-uses, the conversion efficiencies of these fuels are included in the figures. In contrast, the forecasting figures assume that all the energy requirements of the building are supplied by electricity. Third, the year of operation for the buildings in the sample is not 1985, while the forecast figures use 1985 as the operating year. Even given these caveats, the sample data available to the Council on actual energy use per square foot in existing commercial build- ings are reasonably close to the Council's forecasting estimates. The building category with the most variance is restaurants. Res- taurants are particularly sensitive to the prob- lem of fuel conversion efficiencies, because some energy intensive end-uses in this build- ing type are non-electric. Additionally, energy use in restaurants, especially fast food estab- lishments, is related more to the number of customers served than to the square footage of the building. Restaurants, however, are a relatively small portion of commercial elec- tricity use. The table also shows that office buildings, retail establishments and the mis- cellaneous category are by far the dominant users of electricity in the commercial sector. These three categories make up about 60 percent of the total commercial sector elec- trical use. Much less data are available on the actual energy use of newly built commercial build- ings in the region. Table 5-42 shows some of the data for new commercial buildings. The Council's forecast assumptions on new com- mercial buildings built to current practice appear first in Table 5-42. These buildings are assumed to meet the level of ASHRAE 90-80.18 The next column shows the level of efficiency assumed by the forecast resulting from the commercial model conservation standards, which are essentially ASHRAE Table 5-41 Summary of Annual Energy Use for Existing Commercial Buildings Located in the Region ANNUAL ENERGY USE COUNCIL'S BUILDING TYPE’S Ty toh MESS” wean ene) “CONST ON Office (N = 157) 108 6 27 27 27% Retail (N= 581) 281 5 22 30 21% Grocery (N = 336) 86 50 61 51 7% Restaurant (N = 220) 375 49 116 45 7% Hotel (N = 6) 32 16 23 19 3% Hospital (N = 30) _ _- 29 30 4% School (N = 146) 49 2 20 23 8% College Included in “Schools” 28 3% Warehouse (N = 77) 107 2 20 11 7% Other (N= 41) 45 7 22 26 13% Table 5-42 Summary of Annual Energy Use for New Commercial Buildings Located in the Region (Kilowatt-Hours per Square Foot) SAMPLE OF CURRENT CHRORTGREASEE® —eMORESRCESSy Sopot Office 21 az, 19 (N=14) Restaurant 39 39 — Retail 22 20 22 (N=8) Grocery 46 46 44 (N=1) Warehouse 8 7 18 (N=1) School 19 17 16 (N=3) College 23 22 22 (N=1) Health 26 26 - Hotel/Motel 15 14 _ Miscellaneous 22 22 28 (N=2) 5-37 Chapter 5 Table 5-43 Retrofit Savings from Existing Commercial Buildings: Puget Power's Program* AVERAGE USE OF PROGRAM BUILDINGS COUNCIL BUILDING TYPE PERCENT SAVINGS (PRE-RETROFIT) FORECAST (Sample Size = N) FROM AVERAGE USE (kWh/sq ft/yr) (kWh/sq ft/yr) Office (N= 62) 30% 26 27 Retail (N= 11) 16% 25 30 Grocery (N =36) 23% 62 51 Restaurant (N= 10) 22% 89 45 Hotel (N = 2) 16% 24 19 Hospital (N = 30) 28% 29 30 School (N = 28) 17% 24 23 Warehouse (N= 4) 26% 16 11 Other (N=8) 21% 22 26 Average savings = 22% Average savings weighted by building type = 22% “Program offers measures such as heating, ventilating and air conditioning modifications, glazing and insulation, lighting measures and some process modifications. Table 5-44 Technical Conservation from Commercial Buildings LEveUZED cost CUMULATIVE eee 1.0 293 421 2.0 457 684 3.0 509 769 4.0 514 773 5.0 514 780 6.0 519 788 90-80 with lighting improvements in some building types (modeled from ASHRAE 90-80E). The third column shows available data from work done by a Bonneville contrac- tor and from work at the Oregon Department of Energy that documents actual energy use in a few recently built (post-1980) commer- cial buildings. These figures need to be qualified. First, the forecast figures for both current practice and the model conservation standards assume an all-electric building; consequently, fuel conversion efficiencies are not required. In contrast, the average use figures for current practice buildings are for total energy and include fuel conversion efficiencies. Sec- ondly, the sample size of current practice buildings is very small and buildings were not selected to represent the region. Even given these caveats, there is general agreement among the sources. Step 2. Evaluate the conservation poten- tial in existing and new commercial build- ings. For existing buildings, the Council used engineering estimates of conservation potential conducted for the 1983 plan and estimates of conservation from Puget Sound Power & Light's commercial retrofit program. The modeling done for the 1983 plan was on a limited number of building types, but resulted in about 30 percent savings beyond current demand forecast usages for a cost less than about 5 cents per kilowatt-hour. Puget's program corroborates this engineer- ing estimate. Table 5-43 displays the current savings information available from Puget Power and shows the base year use figures from both Puget and the Council's forecasting model. Percent savings are from the engineering estimates done at the time of the audit. Preliminary results of post-retrofit bill- ing data suggest that, on average, the engineering estimates are reasonably accu- rate, although estimates for particular build- ings may vary significantly. Puget's program is aimed at measures that cost less than 3 cents per kilowatt-hour,19 and more conser- vation would be expected if marginal mea- sures cost up to 5.0 cents per kilowatt-hour. On average, all the savings from Puget's pro- gram would produce electricity at a cost sig- nificantly less than 3 cents per kilowatt-hour. The savings that appear in Table 5-43 reflect heating, ventilating, air conditioning, lighting, and insulation measures, and, in some cases, process improvements for the com- bination of measures that were installed. Any given building may have installed only one, or all, of the measures. Total percent savings are about 22 percent whether a simple aver- age of the savings is calculated or the sav- ings are weighted based on 1985 electricity consumption by building type. If lighting improvements, which include some outdoor lighting, and process improvements, which aren't related to square footage, are removed from the estimates of savings, the total sav- ings is reduced to about 20 percent. Puget's base consumption figures are close to those used by the Council's forecast. In addition, Puget's overall savings figure of 22 percent for a marginal measure cost of 3 cents per kilowatt-hour compares favorably with the Council's estimate of 30 percent sav- ings in existing buildings for a marginal mea- sure cost of about 5 cents per kilowatt-hour. Thirty percent savings was used as the tech- nical potential for conservation in existing commercial buildings. For new commercial buildings, the Council modeled the improvement over current prac- tice (ASHRAE 90-80) that would result from the model conservation standards. These efficiency changes, primarily lighting effi- ciency improvements, were modeled based on ASHRAE 90-80E. The main impact of the standards is to change the connected light- ing load, which particularly affects offices and Chapter 5 retail stores. The measures evaluated are available for less than 5.0 cents per kilowatt- hour. Table 5-42 shows the Council's forecast estimate of the energy consumption of cur- rent practice new buildings and the estimate of use from commercial buildings meeting the standards. Savings for all commercial buildings from the standards range between 0 and 19 percent of current use. Savings beyond current practice to the stan- dards level are small compared to savings evaluated on prototype buildings for the Council in the 1983 plan. Data from the pro- totype buildings indicate that, with the last measure installed costing about 5.0 cents per kilowatt-hour, savings of 38 percent in Offices, 19 percent in retail stores, 68 percent in schools, and 31 percent in motel/hotels are possible beyond current practice for new construction. These prototype savings indi- cate that the resource available from making commercial buildings more efficient than the standards is quite promising and should be investigated further. This is further supported by additional infor- mation provided by Bonneville. The 1983 Power Plan asked Bonneville to develop energy use and cost data on energy efficient commercial buildings in climates similar to those found in the region. Figure 5-7 shows the results of this work. The contractor selected by Bonneville collected energy use data on buildings in the region that were reputed to be energy efficient. Slightly less than half the buildings found were more energy efficient than the commercial model conservation standards. A few bettered the standards by 30 percent. The contractor noted, however, that it was difficult to quantify why one building met the standard and another did not. While the Council is not currently counting as areliable and available resource any savings from constructing buildings more efficiently than the commercial standards, such sav- ings do reflect a very promising resource. In the Action Plan, the Council identifies ways Bonneville can help make these additional savings more reliable and achievable. The actual measures that beat the standards and that can be generically recommended for average buildings need to be identified, as well as their cost. Mechanisms to secure this Annual Energy Intensity (% Of Standard ) @ All Electric © Fuel and Electricity A Average New All-Electric Stock Average Existing Stock Office Small Office College Schools Warehouse Figure 5-7 Annual Energy Use of New Commercial Buildings in the Northwest as a Percentage of the Model Conservation Standards resource need to be developed aggressively to bring the resource into the portfolio. Step 3. Develop estimates of realizable potential for conservation in new and existing commercial buildings. As described above, the Council's estimates of the technical potential for conservation were based on a 30 percent savings in existing commercial buildings and on a range between 4 and 19 percent savings, depend- ing on the building type, for new commercial buildings. The total regional savings available from this average level of improvement were estimated using the Council's commercial sector forecasting models as described below. First, this sector's demand was forecast assuming no further improvements in effi- ciency. Demand was forecast separately for new and existing buildings. Then the percent of improvement over current efficiencies rep- resented by the commercial standards was imposed on the model, and the demand was re-estimated. The difference between pro- jected demand at current efficiencies and demand with the improvements from the commercial model conservation standards represented the total technical conservation. Achievable potential was then estimated by re-running the model with percent efficiency improvements that reflected achievable sat- urations. The achievable level of saturation in both existing and new commercial buildings was 85 percent. In the Council's high forecast, 732 average megawatts are achievable in existing build- ings, and 430 average megawatts in new commercial buildings. Table 5-44 shows the achievable conservation that is available at a given cost in the high demand forecast. This curve was estimated using the breakdowns by cost given in the 1983 plan. It should be noted that the current estimate does not separate the conservation potential in governmental buildings from the rest of the commercial sector. Bonneville sponsored a project that attempted a census of institu- tional buildings and extrapolated the results 5-39 Chapter 5 from respondents to non-respondents. Some results from this census produced anomalies when compared to the forecast assumptions. For example, the floorspace reported in schools exceeded the floorspace allocated to this building type in the Council's commercial sector forecast by 24 percent. When additional information becomes avail- able to enable a reasonable calibration, the Council will separate the conservation poten- tial in government buildings from the general commercial sector. Waste Water Treatment Recent information is available for estimating the conservation potential from waste water treatment facilities. A report on waste water facilities produced by a Bonneville contractor provides some of the information used to estimate conservation potential in this sector. In addition to this work, the Council con- ducted a telephone survey of municipal water systems in the region's major population cen- ters to determine the approximate size of pre- conservation loads. The Council's assessment of conservation relies on data collected in the telephone sur- vey and on a review of Environmental Protec- tion Agency data on the 550 waste water treatment plants in the Pacific Northwest. In addition, energy use and energy conserva- tion audit information from plants outside the region were used to assess the costs and potential energy savings from 15 cost-effec- tive conservation measures. Table 5-45 Technical Conservation from Waste Water Treatment Facilities LEVELIZED COST CUMULATIVE (cents/kWh) MEGAWATTS 1.0 8 2.0 8 3.0 10 4.0 14 5.0 15 6.0 15 5-40 In waste water treatment plants the treatment processes themselves account for the largest use of energy. Energy required for lighting and heating, ventilating and air condi- tioning equipment is less significant than the energy required for pumping, aeration and sludge treatment. Of these in-plant pro- cesses, the electrical energy used to operate pumps and motors accounts for the largest energy demand. The conservation potential estimated here does not include potential generation of electricity from methane cogeneration potential. Of the 15 energy conservation measures analyzed, only one, the installation of high efficiency motors, was found to exceed 5.0 cents per kilowatt-hour and was therefore not considered in the analysis. Table 5-45 shows the total estimated technical savings at about 15 average megawatts based on an esti- mated load of 68 average megawatts. Achievable savings were estimated to be about 85 percent of the technical potential, or about 13 average megawatts. Primary Sources for the Commercial Sector Battelle Pacific Northwest Laboratory, Rec- ommendations for Energy Conservation Standards and Guidelines for New Com- mercial Buildings Volume III: Description of the Testing Process, October 1983, Prepared for the U.S. Department of Energy. Letters from Reidun Crowley, Puget Sound Power & Light, February 12 and 15, 1985, on commercial retrofit costs and savings. Energuard Corporation, Summary of End Use Data in Commercial Buildings, December 1984. Charlie Grist, Oregon Department of Energy Commercial Building Survey and personal communication. JRB Associates, Assessment of the Poten- tial for Conservation in Waste Water Treat- ment Plants, JRB No. 2-815-02-796, Febru- ary 1985, Prepared for the Bonneville Power Administration. Lerman, David, John Weigant, Benson Bronfman of Evaluation Research Corpora- tion, DRAFT Institutional Buildings Program Census Extrapolation and Analysis, ERC/ PO-7, February 1985, Prepared for the Bonneville Power Administration. Mazzuchi, Richard P., Assessment of Elec- tric Power Conservation and Supply Resources in the Pacific Northwest, Volume ll — Commercial Building Conservation, June 1982 (Draft), Battelle Pacific Northwest Laboratories. Piette, Mary Ann, Denise Flora and Scott Crowder, Energy Efficient New Commercial Buildings in the Northwest Region: A Com- pilation of Measured Data, March 1985, Prepared for the Bonneville Power Administration. Portland Energy Conservation Incorporated, personal communication, January 1983, on energy use of monitored commercial buildings. City of Tacoma, Jake Fey, personal commu- nication, January 1983, on energy use of commercial buildings. Al Wilson, Seattle City Light, personal com- munication, July 1, 1985. Industrial Sector In 1983, firm sales to the industrial sector were 5,659 average megawatts, which is about 39 percent of firm loads. About 34 percent of firm industrial electricity use was consumed by firm load requirements of the direct service industries, which are mainly the aluminum industry and some chemical and other primary metal producers. The largest consumers among the non-direct ser- vice industries, representing about 85 per- cent of non-direct service industry demand, are lumber and wood products, pulp and paper, chemicals, food processing and pri- mary metals. Chapter 5 The Council used 500 average megawatts as the technical and achievable conservation potential from the direct service and non- direct service industries. This conservation saves about 5 percent of projected industrial use in the year 2005. Industrial sector sav- ings cost an average of about 3.1 cents per kilowatt-hour. Figure 5-8 depicts this conser- vation at various costs. Assessing the technical and economic potential for industrial conservation presents a more difficult problem than in any other sector. Not only are industrial uses of elec- tricity more diverse than the commercial sec- tor, but the conservation potential is also more site-specific. Moreover, because energy use frequently plays a major role in industrial processes, many industries con- sider energy-use data proprietary. For new industrial plants, estimating conservation potential is not yet possible, because incom- ing plants are quite specific in their energy use, and a “base-case” plant from which to estimate savings has not been established. All these factors make it difficult to estimate conservation savings. In the past, industrial representatives have been skeptical of studies that estimate the potential of industrial conservation based on a “typical plant” within an industry. Such stud- ies extrapolate results from a typical plant to estimate the potential for the whole industry. Industry spokespeople argued that typical plants do not exist for most industries. Among other reasons, differences in product lines and the age of plants do not allow com- parison of individual plants within the same industry. Industrial representatives were con- cerned that even though their plant was not like the typical plant used in the analyses, policies and programs affecting them would be developed based on those analyses. For these reasons, in the 1983 Power Plan the Council did not attempt to draw upon or redo studies based on the typical plant approach. Instead, the Council relied on esti- mates supplied by industry in response to a Council survey. The Council also conducted an analysis of its own which attempted to estimate industrial conservation potential by specific end-uses, such as motors, lights, etc. This approach had some of the same problems of the typical plant analysis—lack of information about how electricity was used in the various plants. Cumulative Average Megawatts 600 500 400 300 200 100 0 1 2 3 4 5 6 aw 8 Conservation Cost (Cents/kWh) Figure 5-8 Technical Conservation Potential from the Industrial Sector In preparation for the 1986 Power Plan, the Council considered ways to estimate conser- vation potential in the region's industries that would have the support of industrial repre- sentatives. The approach that received such support was a survey asking individual plant managers to estimate conservation poten- tials in their specific plant. The surveys were coordinated by industry trade associations such as Northwest Pulp and Paper Associa- tion and the Industrial Customers of North- west Utilities. Data from specific firms were masked to protect proprietary data. Each firm was asked how much conservation would be available at specified prices in each of four areas: 1) motors, 2) motor controls, 3) light- ing, and 4) other, a category that depended on the nature of the firm. The firm was also asked to estimate the lifetime of equipment in each of the four categories. Finally, since the Council and industrial representatives did not want to follow this survey with yet another, firms were asked to estimate how much cogeneration would be available to the region at specified prices per kilowatt-hour. The survey was sent to over 200 industrial firms in the Northwest. Forty-seven of the surveys were returned, representing 70 per- cent of industrial electricity use in the region. Non-direct service industries which returned surveys represent 52 percent of the non- direct service industry regional load. All of the direct service industries responded through Direct Service Industries, Inc. The results of survey respondents were extrapolated to nonrespondents in order to capture regional conservation potential in the industrial sector. The results of this survey are presented in Table 5-46. The Council's plan includes developing 500 average megawatts of the currently identified conservation potential in the industrial sector at an average cost of 3.1 cents per kilowatt-hour. This conservation is both technical and achievable, since the sur- vey identified what could and would be done for given prices. In forecasts lower than the high case, conservation from the direct ser- vice industries was reduced to reflect the demand forecast assumptions concerning reduced load from these industries. 5-41 Chapter 5 Table 5-46 Industrial Sector Technical Conservation Potential LEVELIZED CUMULATIVE COST POTENTIAL (cents/kWh) (Megawatts) 1.0 130 3.0 466 5.0 500 7.0 529 5-42 Cumulative Average Megawatts 150 2 3 4 Conservation Cost (Cents/kWh) Figure 5-9 Technical Conservation Potential from Irrigated Agriculture Primary Sources for the Industrial Sector Andrews, Laurel, Neil Leary and Craig McDonald, Synergic Resources Corpora- tion, Survey of Industrial Conservation and Cogeneration Potential in the Pacific North- west, SCR Report No. 7193-R3, 1984, Pre- pared for the Northwest Power Planning Council. Letter from Laurel Andrews, Synergic Resources Corporation, April 23, 1985. Irrigation Sector In 1983, the region's irrigated agriculture con- sumed 615 average megawatts of electricity, less than 5 percent of the region's total con- sumption. Figure 5-9 shows the estimated irrigation savings available from existing and new irrigation systems at various electricity prices. The technical potential, evaluated with a marginal measure not exceeding a cost of 5.0 cents per kilowatt-hour, is 146 average megawatts. The Council's plan calls for developing up to 85 percent of this poten- tial, or 124 average megawatts. This repre- sents about 14 percent of electricity use for irrigation in 2005 and is available at an aver- age cost of about 1.8 cents per kilowatt-hour. The Councils assessment of conservation potential for this sector involved the following two steps: 1. Evaluate the end-use conservation mea- sures to be included in the supply curve analysis. 2. Estimate realizable conservation potential by using the cost and potential savings data available from the Irrigation Sector Energy Planning Model, and compare the relationship of the cost and savings data derived from the base load forecast used by Bonneville with the Council's load forecasts. Step 1. Evaluate the end-use conservation measures to be included in the analysis. In the 1983 Power Plan, the Council relied on estimates of conservation potential in irri- gated agriculture provided by a Council con- tractor. At the time, the research represented the most complete picture of energy conser- vation opportunities in the region's irrigation sector. Since that time, Bonneville has hired various contractors to continue analytical studies in order to better characterize the irrigation sector. These efforts have produced improved baseline data and analytical tools, which the Council used to prepare its assess- ment of the conservation potentials in this sector. The conservation opportunities considered in the irrigation supply curve estimates include: * low pressure irrigation on center-pivot, side- roll, and handmove systems; * fittings redesign; * mainline modifications; * improved scheduling. Low pressure irrigation involves using sprin- kler or spray application devices designed to operate at lower pressures than conventional sprinkler devices. These low pressure devices can be divided into three major types: low pressure spray heads, low pres- sure impact sprinklers and drop tubes. The fittings of an irrigation system include valves, elbow joints and other components used to connect the irrigation pump to the pipes of the system and to connect the pipes within the system to each other. Fittings redesign involves using larger tapered fittings to replace valves and elbows that are too small or that change abruptly in size and direction. Mainline modification involves increasing the size of the system's mainline, resulting in decreased energy losses due to friction. This redesign generally can be accomplished most economically by installing a second mainline pipe parallel to the existing one. Improved scheduling involves the improve- ments in both timing and amount of water applications. This reduces water use without reducing crop yields, and energy use is reduced due to a decrease in pumping requirements. Scheduling is the cornerstone of a basic comprehensive management approach to efficient water and energy man- agement, with all other conservation mea- sures being necessary components. The supply curve analysis does not address two major options which were included in the 1983 plan: very low pressure systems2° and pump improvements. Very low pressure water application systems are not unlike existing low pressure center- pivot systems equipped with drop tubes. It appears, however, that the application of this very low pressure system on slope condi- tions typical in the Northwest works best as part of an agriculture practice known as res- ervoir or basin tillage. This practice creates small circular furrows which hold the applied water, reducing the problem of runoff. Field tests, planned over the next two years, will assess the hydraulic characteristics of very low pressure (5-15 pounds per square inch) system components, evaluate reservoir tillage and associated soil/irrigation manage- ment practices, and determine energy sav- ings. Until results are analyzed, there appears to be no reliable estimate of poten- tial conservation savings from this technology. There is broad recognition of the need to operate pumps at efficiencies that match the original operating specifications. Worn pump components, improper sizing, dirty water conditions, scale build-up in the mainline, improper fittings, pumps taking in air, and multiple versus single pump usage, can all affect pump efficiency. In addition, bringing pump efficiencies to original specifications may increase horsepower and result in increased energy use. Pump testing pro- grams have been conducted in the region for several years. However, Bonneville aban- doned the testing program after acknowledg- ing the need for a more comprehensive approach to irrigation efficiency. There is also information to suggest that some of the pump testing conducted not only on irrigation systems, but also on municipal water systems, has been incomplete. Pumps operate optimally when the clearance between the impeller and pump housing is properly adjusted. According to some profes- sionals in the field, this adjustment is not done enough, especially prior to testing. It is analogous to attempting to determine the fuel efficiency of a car without first tuning the engine. Consequently, there are currently no reliable estimates of conservation potential for pump efficiency improvements. However, both very low pressure systems and pump efficiency improvements are promising resources and deserve further research and analysis. Chapter 5 Step 2. Estimate realizable conservation potential. Conservation supply estimates for the irrigation sector were developed using the Irrigation Sector Energy Planning Model (ISEP). The model combines both engineer- ing and economic principles to derive energy savings and levelized costs per kilowatt-hour for conservation investments. The model uses a number of baseline data inputs, including estimates of crop-specific acreages in 11 subbasins in the region; type of irrigation systems used; pumping lift; pumping plant efficiencies; estimates of water application volumes to specific crops by irrigation system type; and system operating pressures. The model also uses rough estimates of conservation measures believed to have been applied on existing acreages and subtracts these estimated sav- ings prior to calculating the remaining con- servation potential. In atest of the model to estimate the baseline energy use for regional irrigation loads, the ISEP model estimates were within 7 percent of the load estimated from billing records. This indicates a high degree of confidence for this part of the model. The irrigation savings information provided to the Council by Bonneville is based on a 1984 final Bonneville base case load forecast and Bonneville acreage forecasts. The average megawatt estimates used in this plan are adjusted from the estimates provided by Bonneville with a ratio proportionate to the difference between the Council's 1985 high case irrigation load forecast and the 1984 Bonneville base irrigation load forecast. The Council's 1985 agricultural forecasts were adjusted downward to reflect the subtraction of loads from U.S. Bureau of Reclamation projects in the region. This is necessary in order to adjust irrigation savings from the ISEP, which are based on Bonneville loads, because the Bonneville forecast does not include Bureau loads. The Council assumes a total of 116 average megawatts of Bureau load in the region, of which 55 average mega- watts are directly related to pumping facilities at Grand Coulee Dam. The rest of the load is scattered in the region. 5-43 Chapter 5 Table 5-47 Technical Conservation Potential from the Irrigation Sector 1 18 25 2 81 29 3.2 106 29 4.2 116 30 47 116 30 Table 5-47 summarizes the estimates of con- servation potentials on existing and new acreages that result from the above- described models and adjustments. The conservation estimate for existing acreages is 116 average megawatts of technical poten- tial. The conservation estimate for new acreage is 30 average megawatts, which is only included in the high demand forecast. Primary Sources for the Irrigation Sector Gordon, Fred, Irrigation Technical Supply Curve Project Research Summary, Bon- neville Power Administration, December 1984. Harrer, B.J., Lezberg, A.J., Wilfert, G. L., An Integrated Assessment of Conservation Opportunities in the Irrigated Agriculture Sector of the Pacific Northwest, Battelle Pacific Northwest Laboratory, February 1985. Harrer, B.J., Draft: A Sensitivity Analysis of Conservation Opportunities in the Irrigated Agriculture Sector of the Pacific Northwest, Battelle Pacific Northwest Laboratory, Febru- ary 1985. Pharayra, Barbara, Summary of 1984 Pacific Northwest Irrigation Conservation Potential: 1984 BPA Final Load Forecast—Base Case, Bonneville Power Administration, May 1985. 5-44 1./ These savings must be increased by line losses of 7.5 percent to be consistent with evaluations in the resource portfolio, as described later in this chapter. 2./ A“measure” means, as appropriate, either an individual measure or action or a combination of actions. 3./ Levelized life cycle cost is the present value of a resource's cost (including capital, financing and operating costs) converted into a stream of equal annual payments; unit levelized life crc costs (cents per kilowatt-hour) are obtained by dividing this payment by the annual kilowatt-hours saved or produced. Unlike installed cost, levelized costs that have been corrected for inflation permit com- parisons of resources with different lifetimes and generating capabilities. The term “level- ized cost” as generally used in this chapter refers to unit levelized life cycle cost 4./ Least costly is defined in terms of a measure's levelized life cycle cost, stated in terms of mills or cents per kilowatt-hour. 5./ The system models are the Decision Model and the System Analysis Model. These are briefly described later in this chapter and fully described in Volume II, Chapter 8. 6./ The SUNDAY model simulates space heating needs based on heat loss rate, daily access to solar energy, daily inside and outside tem- peratures, thermal mass, and the amount of “waste heat” given off by lights, people and appliances. 7./ These items are discussed here in terms of the calculated savings per measure. Under Step 5, these items are discussed in terms of differences between the demand forecast estimates of space heating loads and esti- mates from the engineering model. 8./ As noted in the introduction, finance costs are taken from the system models and reflect a sponsorship mixed among Bonneville and investor-owned utilities. 9./ This assumes a 90°F temperature differential between the incoming water and the tank setting. 10./ In the 1983 plan, these efficiency improve- ments were credited to the miscellaneous appliance sector. Since they pertain to the amount of hot water used by these appliances, they are more accurately included as water heating savings. 11./ This assumes about 50 percent of current tanks are wrapped, about 5 percent are effi- cient tanks and there is some saturation of other standby loss measures. 12./ The California standard represents a mini- mum efficiency level for various product classes (for example, a refrigerator with top-mounted freezer and automatic de- frost capability) within the grouping of refrigerators, rei jerator/freezers and freez- ers. This standard also varies by the size of the refrigerated space and the freezer space. 13./ For example, those technologies included in the supply curves presented in this section. 14./ As aconsequence of litigation over the DOE standards that resulted from the described evaluation, DOE has been required to return to the evaluation and investigate all tech- nologically feasible measures. 15./ Appliance manufacturers have sued the CEC over the revised standard. 16./ Solar gains are considered constant in this discussion. 17./ A recent American Society of Heating, Refrigerating and Air-Conditioning Engineers’ publication suggests that the minor internal gain from variable use should be ignored. The gain from the hot water in the pipes is offset by heat used to heat cold water brought inside the heated shell through other pipes. 18./ ASHRAE stands for the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. This organization sets vari- ous standards for building practices based on consensus. 19./ The program allows measures costing more than 3 cents per kilowatt-hour to be installed if financed independently from the program. 20./ This technology was called “low energy pre- cision application” or “LEPA” in the 1983 Power Plan. Chapter 6 Generating Resources This chapter describes the selection of gen- erating resources for development of the resource portfolio. Resources selected were further assessed to develop the detailed information required by the system planning models. The chapter concludes with a dis- cussion of how existing resources are allo- cated between public agencies and investor- owned utilities for the purpose of determining the Administrator's obligations. This chapter considers alternatives using coal, geothermal, hydropower, municipal Solid waste, natural gas, nuclear power, solar insolation, wind, wood residue and waste heat as sources of energy for electrical power generation. Both stand-alone and cogenera- tion opportunities are considered. The chap- ter also examines opportunities for increas- ing the capability of existing regional generation projects and for reducing losses in the regional power transmission and dis- tribution system. The chapter includes a description of existing contracts for the import and export of power, and a discussion of potential future imports. The focus is on central station generation of electricity; however, consumers might use certain resources considered in this chapter to generate electricity at the point of end use, or to offset the need for electrical power through direct resource applications. An example of the former might be the use of solar photovoltaics for local electricity pro- duction. The latter would include the use of low temperature geothermal energy for space heating. End-use applications are considered in the Council's planning as con- servation resources. Selection of Available Resources The Northwest Power Act requires that the Council's plan give priority to resources that the Council determines to be cost effective. For resources of equal cost effectiveness, priority is given first to conservation, second to renewable resources, third to generating resources using waste heat or generating resources of high fuel conversion efficiency, and finally to all other resources. The Northwest Power Act defines “cost effec- tive” to mean that a measure or resource must be forecast to be reliable and available within the time it is needed and to meet or reduce electrical power demand of consum- ers at an estimated incremental system cost no greater than that of the least-cost similarly reliable and available alternative or combina- tion of alternatives. System cost, in turn, is defined as an estimate of all direct costs of a measure or resource over its effective life, including, if applicable, distribution and trans- mission costs, waste disposal costs, end-of- cycle costs, fuel costs and quantifiable environmental costs. The Council is also required to take into account projected real- ization factors and plant factors, including appropriate historical experience with similar measures or resources. Finally, the North- west Power Act provides a 10 percent advan- tage in calculation of system costs for conser- vation resources. Resources were classified as “cost effective,” or “promising,” using screening criteria based on the resource requirements of the North- west Power Act described above. Cost-effec- tive resources were incorporated into the sys- tem planning models as part of developing the resource portfolio. Promising resources are candidates for future resource portfolios. Research, development or demonstration activities and other actions will better estab- lish the role of promising resources in future power plans. Criteria used to judge the availability, reliabil- ity and cost effectiveness of resources are as follows: 1. Commercially Available Technology: The technology for producing electrical power or for increasing the efficiency of the exist- ing power system must be commercially available. 2. Predictable Cost and Performance: It must be sufficiently demonstrated that the tech- nology’s cost and performance charac- teristics are predictable. 3. Competitive Cost: The resource must be cost-competitive using currently available technology. Because of the complexity of the regional power system, it is not possi- ble to forecast cost-competitiveness accu- rately without using the system planning models. However, a preliminary estimate of cost-competitiveness can be made using levelized life cycle costs.’ A level- ized life cycle energy cost of 4.5 cents per kilowatt-hour (January 1985 dollars) is used as a criterion for generating resources. A levelized life cycle energy cost of 5.0 cents per kilowatt-hour is used for conservation resources. These levels were chosen on the basis of the estimated cost of energy from new conventional coal projects, as described in Chapter 3 of Vol- ume |. 4. Demonstrated Resource Base: The esti- mates of the amount of capacity and energy available from a given resource require a confirmed primary energy source (e.g., coal, falling water, wind). 5. Institutionally Feasible: Development of the resource must not be currently con- strained by legal, financial, regulatory or other institutional barriers. 6. Environmentally Acceptable: The resource must be environmentally acceptable and capable of complying with current environmental policies, laws and regulations of the federal, state and local governments, and the Council's Columbia River Basin Fish and Wildlife Program. Further discussion of the environmental effects of resources is provided in Chapter 9 of this volume. 6-1 Chapter 6 The conclusions described below represent the best judgment of the Council given the information presently available. Development of Detailed Planning Information Resources judged to be available using the criteria described above were further assessed to fully develop the cost, technical performance and other information required for planning. This information was obtained by assessing actual projects, either pro- posed or under construction, or by assessing generic projects representative of actual pro- jects that might be developed in the region. Detailed planning information is provided in Appendices 6-B through 6-J to this chapter. Transmission and Distribution System Efficiency Improvements For the 1984-85 operating year, itis estimated that Bonneville losses will be 135 megawatts (exclusive of losses attributable to serving nonfirm load). Losses for the balance of the regional system are estimated to be about 1,200 megawatts. Measures available to reduce transmission and distribution losses include conductor and transformer replacement, voltage upgrade, addition of parallel transmission lines or distribution feeders, power factor cor- rection, capacitor replacement and system reconfiguration. These measures are com- mercially available and well demonstrated. Promising measures include advanced design transformers, improved voltage reg- ulation and optimization of generating pat- terns. Because these measures reduce the consumption of electrical energy, they are considered to be conservation resources under the Regional Act. Measures that improve the efficiency of transmission and distribution have several attractive characteristics. Because losses vary with the square of the load, loss reduc- tion measures are effective in reducing peak loads. Older transformers and capacitors 6-2 often contain PCB fluids which can be dis- posed of if the equipment is replaced to improve system efficiency. Because system efficiency improvements do not affect sales, they do not reduce revenue to the implement- ing utility. Cost and Availability Because much of the regional transmission and distribution system was designed when the cost of losses was much lower than at present, there are many cost- effective oppor- tunities to upgrade the system. Often, where efficiency improvement measures cannot be justified solely on the basis of loss reduction, cost effectiveness can be achieved when upgrades are required to increase capacity or to improve reliability. Although the cost and performance of individual loss-saving com- ponents are typically well understood, the complexity of the transmission and distribu- tion system is such that analyses of specific applications are required to assess accu- rately the cost-effective loss reduction poten- tial of these improvements. Bonneville System Efficiency Improvements Bonneville has established a Loss Savings Task Force to assess potential loss savings projects on the Bonneville system. This task force has identified 38 possible loss reduc- tion projects, providing in excess of 41 mega- watts total energy savings at estimated costs of 10 cents per kilowatt-hour or less. Thirty- six of these projects, totaling 34 megawatts, are estimated to be available at costs of less than 5.0 cents per kilowatt-hour. The Loss Savings Task Force identified addi- tional measures with the potential for cost- effective reduction of Bonneville system losses. These include reconductoring with compacted conductors, addition of subcon- ductors, insulating groundwires, transposi- tions, and modification of system operating practice. The feasibility and cost effective- ness of these measures require further study. Utility System Efficiency Improvements Bonneville has established the Customer System Efficiency Improvement (CSEl) pro- ject in response to Action Item 11.2 of the 1983 Power Plan. The purpose of this project is to perform an assessment of the potential for loss reduction on non-Bonneville regional transmission and distribution systems, including those of Bonneville's federal and direct service industrial customers. The study is assessing the technical potential, economic potential and likely achievable level of system efficiency improvements. A summary progress report of the CSEI pro- ject, issued in January 1985, estimates tech- nically available loss reduction to be 350 to 585 megawatts (approximately 30 to 50 per- cent of current system losses). Preliminary estimates of the availability and cost effec- tiveness were prepared for two specific loss reduction measures—reconductoring sub- transmission and primary distribution lines and transformer replacement. These esti- mates indicate that approximately 115 mega- watts of energy could be obtained from high efficiency distribution transformers at costs of 5.0 cents per kilowatt-hour or less. An additional 30 to 35 megawatts of energy could be obtained at similar cost by recon- ductoring subtransmission lines and distribu- tion feeders. Conclusion The 36 Bonneville loss reduction projects, totaling 34 megawatts at 5.0 cents per kilo- watt-hour or less, are considered available for the resource portfolio. The Council encour- ages continued assessment of the cost and availability of loss reduction potential on the Bonneville system. The cost and availability of these projects are shown in Table 6-1. Table 6-1 Cost and Availability of Transmission and Distribution System Efficiency Improvements TRANSMISSION & CEES?” (average meguvrats) PI ccs dc 0-1.0 7 1.0-2.0 1 2.0-3.0 22 3.0-4.0 4 4.0-5.0 _0 TOTAL 34 —— Chapter 6 Because of the preliminary nature of the esti- mates of loss reduction potential on non- Bonneville systems, the Council does not consider these savings to be available for the resource portfolio. Because of the potential magnitude and cost effectiveness of this resource, the Council seeks additional infor- mation regarding availability, cost effective- ness and methods of acquisition of this resource. The Council also encourages confirmation of the cost and performance of advanced trans- mission and distribution loss reduction mea- sures, Hydropower Efficiency Improvements Hydropower efficiency improvement mea- sures offer the potential for cost-effective increases in capacity and energy from exist- ing regional hydropower projects. This poten- tial is due to advanced designs, materials and equipment that have become available since many of the region's hydroelectric pro- jects were built. Additionally, electrical energy costs, and therefore the cost of electrical losses, are much higher now than when much of the regional hydropower system was designed. Because the cost of losses used for the original designs was lower than if these projects were being designed today, designs and equipment were often chosen that are of lower efficiency than those that would be selected today. Preliminary estimates of at least 270 average megawatts of potential savings through improvements to hydropower generation Prompted the Council to include Action Item 11.2 in the 1983 Power Plan. It called for the Bonneville Power Administration to conduct studies of potential improvements that could be made in the efficiency of power genera- tion, transmission and distribution. Substan- tial progress in assessing these resources has occurred since. A December 1984 report, prepared by the U.S. Army Corps of Engineers, assessed ongoing and potential improvements in the efficiency of the Corps hydropower projects. A January 1985 report, prepared by Raymond Kaiser Engineers for Bonneville, is the first attempt at a regionwide assessment of hydropower efficiency improvement potential. This study estimates the costs and energy savings attributable to a variety of efficiency improvement measures applied to a generic 100 megawatt hydro- power unit. The generic estimates are aug- mented by a case study of the 774 megawatt Wells hydropower project. Regionwide esti- mates are developed by extrapolating generic plant estimates. During preparation of the plan, Bonneville, the Pacific Northwest Utilities Conference Committee and regional hydropower operators worked to refine the estimates of hydropower efficiency improve- ments appearing in the Raymond Kaiser study. The results of this refinement are used in this plan. Efficiency Improvement Measures The principal measures available to improve hydropower project efficiency are the following: Turbine Improvements. Turbine runners (blade and hub assembly) of improved design and materials, air injection, contour reshaping and seal improvement may improve turbine reliability and efficiency beyond original design specifications, especially for older units. In addition, improvements in the efficiency of turbine operation and design will often reduce the mortality of fish passing through the units. Turbine Governor Improvements. Many of the region's hydropower projects use tur- bines of the Kaplan type. The blade angle of a Kaplan turbine is adjustable to improve efficiency as load and water head vary. On early units, the blade angle was controlled by a two-dimensional mechanical cam. As res- ervoir level fluctuated, cams were to be changed to maintain optimum efficiency. Because of the effort required, these cams have typically been changed only when it is anticipated that the reservoir will be main- tained at a constant level for some time. As a result, these turbines are often operated at less than optimum efficiency. In the early 1970s, a three-dimensional mechanical cam was developed. The three- dimensional cams incorporate the contours of the set of two-dimensional cams ina single cam, eliminating the need to change cams manually to follow operating head. Three- dimensional mechanical cams have been retrofitted to The Dalles units 15 - 22, Bon- neville units 1 - 10, and three units each at Little Goose, Lower Monumental and Lower Granite. More recently, a microprocessor-based blade control system has been developed in which the relationships between blade angle, gate opening and operating head are elec- tronically programmed. The second Bon- neville powerhouse employs this system. Unit 7 at Wells and all John Day and McNary units have been retrofitted with this system. To maintain optimum performance, a Kaplan turbine should have an “index” test per- formed that determines the optimal rela- tionship among blade angle, gate opening and operating head. This relationship is unit- specific and varies over the unit life. An advanced microprocessor-based blade con- trol system has been proposed, that would provide automatic index testing and update of the electronic cam program. The expected increase in efficiency has been estimated to be from 0.5 percent to 3 percent. Develop- ment and demonstration of governors incor- porating automatic index testing is required before the potential of these devices can be assessed. Generator Windage Loss Reduction. Improvements in the design of generator cooling systems have reduced “windage” losses due to air friction. Retrofit of older generators with improved cooling systems has been demonstrated; however, not all older machines lend themselves to retrofit- ting. The general feasibility of cooling system retrofits has also been questioned because of interference with access to generator inter- nals. Additional assessment of this measure is required before the cost and availability of potential energy savings can be determined. Generator Rewinding. Modern conductor insulation is thinner than that available in the past, allowing a greater amount of conduct- ing material to be placed in each stator slot in a generator rewind. This reduces resistance losses and may increase the rated capacity of the machine to fully use the increased generator capacity; however, turbine improvements may also be required. Addi- 6-3 Chapter 6 Table 6-2 Generic Hydropower Efficiency Improvement Measures? NEW KAPLAN NEW FRAN! MEASURE RUNNERS RUNNERS. eo cAm Capacity (Net MW) 100 100 100 Capacity Factor (%) 64 64 64 Efficiency Gain (%) (Typical) 1.50 1.50 1.0 Annual Energy (MWa) (Typical) 0.96 0.96 0.6 Option Lead Time (mos)> 12 12 12 Construction Lead Time (mos) u2 12 12 Option Cost (million $) 0.14 0.08 <0.01 Construction Cost (million $) 1.27 0.68 0.04 Fixed O&M¢ Cost (million $/yr) 0.00 0.00 0.00 Variable O&M Cost (cents/kWh) 0.0 0.0 0.0 Capital Replacement (million $/yr) 0.00 0.00 0.00 Net Decommissioning Cost (million $) 0.00 0.00 0.00 Amortization Life (yrs) 30 30 30 Operating Life (yrs) 30 30 30 Energy Cost (cents/kWh)?¢ a) 0.6 0.1 WINDAGE LOSS GENERATOR MEASURE REDUCTION REWIND SOXCNERS Capacity (Net MW) 100 100 100 Capacity Factor (%) 64 64 64 Efficiency Gain (%) 0.20 0.05 0.06 Annual Energy (MWa) 0.13 0.03 0.04 Option Lead Time (mos)> 12 12 12 Construction Lead Time (mos) 12 12 12 Option Cost (million $) <0.01 0.15 0.02 Construction Cost (million $) 0.05 1.35 0.21 Fixed O&ME Cost (million $/yr) 0.00 0.00 <-0.01 Variable O&M Cost (cents/kWh) 0.0 0.0 0.0 Capital Replacement (million $/yr) 0.00 0.00 0.00 Net Decommissioning Cost (million $) 0.00 0.00 0.00 Amortization Life (yrs) 30 30 30 Operating Life (yrs) 30 30 30 Energy Cost (cents/kWh)?¢ 0.3 39.8 43 (table continued on next page) 6-4 tional assessment of this measure is required before the cost and availability of potential energy savings can be determined. Solid State Exciters. Solid state exciters are now available that feature lower losses and reduced maintenance costs compared to earlier designs. Additional assessment of this measure is required before the cost and availability of potential energy savings can be determined. High Efficiency Transformers. Transform- ers are inherently high efficiency devices. Losses are affected by the number of core laminations, the number of windings and other design (and cost) factors. Older units were selected upon the basis of energy costs much lower than those experienced at pre- sent, and therefore may be less efficient than designs that would be selected based upon current and forecast energy costs. The cost and availability of energy savings through replacement of main power transformers should be assessed as part of the Customer System Efficiency Improvement study (see preceding section). This is a conservation measure under Regional Act criteria. Improved Water Use. Water bypassing tur- bines at existing hydropower projects may include water used for fishway attraction, navigation lock operation, fish ladders and juvenile fish bypass systems. Small quan- tities of bypass water are necessary to oper- ate fish ladders, navigational locks and juve- nile fish bypass systems, and cannot be reduced beyond certain practical limits. How- ever, bypass losses can be reduced at cer- tain projects through improved spillway gate seals, spillway gate position indicators, bypass water energy recovery facilities and other measures. Increased Operating Head. Increasing the operating head of hydraulic turbines can increase the turbine power output. Turbine modifications and generator rewind may be required to fully use the additional power. Methods available for increasing operating head include raising reservoir levels and reducing head losses due to hydraulic fric- tion. The feasibility of raising reservoir levels is highly site-specific and requires considera- tion of the social and environmental effects of the increased pool level, possible impacts on the output of upstream projects due to Chapter 6 increase in tailwater levels and the cost of Modifying turbine generator units to exploit the increased operating head. The Chief Joseph pool level was raised successfully; conversely, the proposed High Ross project was terminated, largely on environmental grounds. Head losses result from friction in water intakes, canals, penstocks and other water conveyance structures. These losses can be reduced by several means, including enlarging the existing water conveyance structures and constructing parallel struc- tures. The Council does not consider any energy from these measures to be available at this time, because of the site-specific feasi- bility of potential increases in reservoir levels and reductions in operating head, and the lack of regionwide assessments of the avail- ability, cost and potential environmental and social impacts of these measures. Reduction in Station Service Loads. Hydropower station loads may be reduced through typical industrial conservation mea- sures. These include efficient motors, high efficiency lighting and controls, load balanc- ing, power factor correction, high efficiency station service transformers, removal of unnecessary voltage regulators, heating, ventilating and air conditioning (HVAC) improvements, and weatherization. No energy from this source is considered to be available at this time. This is a conservation resource under Regional Act criteria. Resource Cost The Council has assessed the cost of hydro- power system efficiency improvements, using as its principal source the study pre- pared for Bonneville by Kaiser Engineers. The Kaiser study develops estimates of the cost and performance characteristics of each of the hydropower efficiency improvements described above, with the exception of bypass water energy recovery facilities. These are too site-specific to be estimated generically. Typical cost and performance estimates, based on a generic 100 megawatt Capacity hydropower unit operating at 64 per- cent capacity, are shown in Table 6-2. Table 6-2 (Continued) IMPROVED GATE POSITION HIGH-EFFICIENCY IMPROVED MEASURE INDICATORS MOTORS LIGHTING Capacity (Net MW) 100 100 100 Capacity Factor (%) _— ee aes Efficiency Gain (%) _ 0.002 0.015 Annual Energy (MWa) 0.08 0.002 0.015 Option Lead Time (mos)> 12 12 12 Construction Lead Time (mos) 12 12 12 Option Cost (million $) <0.01 <0.01 <0.01 Construction Cost (million $) 0.07 <0.01 0.01 Fixed O&M¢ Cost (million $/yr) 0.00 0.00 <0.01 Variable O&M Cost (cents/kWh) 0.0 0.0 0.0 Capital Replacement (million $/yr) 0.00 0.00 0.00 Net Decommissioning Cost (million $) 0.00 0.00 0.00 Amortization Life (yrs) 30 30 30 Operating Life (yrs) 30 30 30 Energy Cost (cents/kWh)¢ 0.1 ey 1.0 IMPROVED HIGH-EFFICIENCY MAIN MEASURE _ POWERHOUSE HVAC _ TRANSFORMERS __ Capacity (Net MW) 100 100 Capacity Factor (%) _ 64 Efficiency Gain (%) _ 0.12 Annual Energy (MWa) 0.038 0.08 Option Lead Time (mos)? 12 12 Construction Lead Time (mos) 12 12 Option Cost (million $) 0.01 0.14 Construction Cost (million $) 0.07 1.22 Fixed O&M¢ Cost (million $/yr) <0.01 0.00 Variable O&M Cost (mills/kWh) 0.0 0.0 Capital Replacement (million $/yr) 0.00 0.00 Net Decommissioning Cost (million $) 0.00 0.00 Amortization Life (yrs) 30 30 Operating Life (yrs) 30 30 Energy Cost (cents/kWh)¢ 3.1 13.0 AAll costs are incremental; January 1985 dollars. >Feasibility study and engineering design to equipment order. ¢Operation and maintenance. 4Levelized lifetime energy costs for 1995 inservice date. Real interest rates, public ownership. 6-5 Chapter 6 Table 6-3 Availability of Energy from Hydropower Efficiency Improvements MEASURE Cost ENERGY (Average Megawatts) Effective Conditional Promising COMMENTS Turbine Runner Replacement Electronic 3-D Cams Windage Loss Reduction Generator Rewinding Solid-state Excitors High Efficiency Transformers Improved Water Usage Reduced Station Service Totals 85 0 27 _ 0 0 5.0 0 0 16 112 21 28 46 23 123 Includes energy from the refined Bonneville Power Administration estimate for all pre-1960 units with the exception of Wells. The planned Wells upgrade is con- sidered to be an assured resource. Available energy is from units presently not equipped with mechanical 3-D cams. Promis- ing energy is from units equipped with mechanical 3-D cams. 0.2 percent efficiency gain from all pre-1980 units. Energy from all pre-1975 units thought not to have been rewound since 1975 is consid- ered as conditionally promising. Assumes 0.05% efficiency gain. Energy from all units thought not to have solid-state excitors is promising. Assumes 0.06% effi- ciency gain. Energy from all pre-1975 pro- jects thought not to have post-1975 transformer replace- ment is conditionally promising. Energy from new gate position indicators at all projects is prom- ising. Bypass water turbines are considered as a new hydro- power resource. Energy from improved lighting and HVAC and high-efficiency motors at all pre-1980 units is promising. 6-6 The levelized life cycle costs of the hydro- power efficiency improvement measures appearing in Table 6-2 were estimated using the financial assumptions described in Chap- ter 4 of this volume. These costs indicate that all measures, with the exception of generator rewinding and main transformer replace- ment, meet the 4.5 cent per kilowatt-hour cost criterion. Certain measures, particularly electronic governors and improved gate posi- tion indicators, are so low cost that it may be desirable to install these measures during the current surplus. Generator rewinding and replacement of main transformers do not appear to be cost effective if undertaken for the sole purpose of improving unit efficiency. Rewinds are occa- sionally required to restore degraded insula- tion. If rewinds are undertaken for this pur- pose, the energy savings shown in Table 6-1 can be obtained for a much lower cost than indicated. Similarly, if main transformer replacement is required for reasons other than improved efficiency, a high efficiency transformer can be obtained for a much smaller incremental investment than indi- cated in Table 6-2. The incremental capital costs and resulting levelized energy costs of other measures will likewise be lower than the full costs shown in the table. Resource Availability The Bonneville-Kaiser report derived poten- tial regionwide energy savings resulting from hydropower efficiency improvements. This calculation was attempted through extrapola- tion of the estimated savings from the 100 megawatt generic unit, using an inventory of regional hydropower units. The resulting regionwide estimates of savings were ques- tioned, however, because of inaccuracies in the inventory of regional units and, in certain cases, incorrect extrapolation. The Council has received comments from Bonneville, the Pacific Northwest Utilities Conference Committee (PNUCC) and util- ities identifying errors in the regional hydro- power inventory. In 1985 a joint effort was undertaken involving the Council, Bon- neville, PNUCC and regional hydropower operators to review and revise the inventory of hydropower units upon which the estimate of availability of regional savings is based. Chapter 6 The resulting revised estimates of regionwide potential savings appear in Table 6-3. Energy classified as “cost effective” in Table 6-3 is a discretionary resource that can be obtained when needed. Energy classified as “promis- ing” appears to be cost effective, but requires further confirmation of availability or cost. “Conditional” energy is from measures (transformer replacement and generator rewind) that are likely to be cost effective only if the measure is undertaken for reasons other than efficiency improvement. This resource, therefore, may be a lost oppor- tunity resource. Because a comprehensive inventory of candidates for transformer rewind and generator replacement is not available, energy from these measures is considered promising, not available. Energy from potential hydropower efficiency improvements is an attractive resource because of its low cost and generally negligi- ble environmental effects. Improvements in turbine design and operation allowing better Operating efficiency may reduce the mortality of fish passing through the turbines. System efficiency improvements have promising optioning characteristics because of short lead times and the direct control of many hydropower projects by the region's utilities. However, much of the region's hydropower capacity is controlled by federal agencies, and improvements to these projects are sub- ject to the federal budgeting process. Ways should be explored to better control the tim- ing of improvements to federal projects. The Council encourages further assessment of the cost and availability of the conditional and promising resources identified in Table 6-3. Methods of controlling the timing of the development of this resource should be investigated, and ways should be explored of transferring the resource to utilities likely to need additional capacity. The Council also encourages development and demonstra- tion of advanced technologies leading to fur- ther improvements in the efficiency of hydro- power units. Conclusion The Council has concluded that energy sav- ings from turbine runner replacement and electronic governors should be included in the resource portfolio. All measures, with the exception of generator rewinds, transformer replacement, reservoir raising and bypass energy recovery units, are likely to be cost effective. Generator rewinds and transformer replacements are incrementally cost effec- tive if required for reasons other than effi- ciency improvement. Measures involving increases in operating head (reservoir raising and head loss reduction) must be assessed individually. Bypass energy recovery units are considered in the hydropower assess- ment. The cost and availability of cost-effec- tive hydropower efficiency improvements are summarized in Table 6-4. Table 6-4 Cost and Availability of Hydropower Efficiency Improvements LEVELIZED COST HYDROPOWER (cents/kWh) (average megawatts) 0-1.0 112 1.0-2.0 0 2.0-3.0 0 3.0-4.0 0 4.0-5.0 _o TOTAL 112 Thermal Plant Efficiency Improvements The efficiency of existing thermal plants may be upgraded to an extent depending upon age and design. This upgrading may reduce operating costs and increase plant capacity and energy output. The extent of upgrades may range from minor component replace- ment to complete repowering using advanced design heat sources such as fluid- ized bed combustors. Major process modifi- cations such as repowering are unlikely to be cost effective at present because of the con- temporary design of most of the region's ther- mal plants. Component upgrades typical of industrial conservation efforts, such as effi- cient motors, variable-speed motor control- lers, efficient pumps and efficient lighting, may be cost effective. Bonneville has proposed a pilot study of a typical regional coal-fired thermal plant, to be carried out subject to budget approval. The Council is not aware of any existing assess- ment of the regional potential for thermal plant upgrades. Because of the lack of information regarding the availability and cost of thermal plant upgrades, the Council does not consider energy from this resource to be available at present. The Council encourages further study to establish preliminary estimates of the availability and cost effectiveness. The study should draw upon assessments that may have been performed elsewhere and would apply to regional coal plants. Geothermal Electric Power No geothermal-electric power plants are presently operating in the Pacific Northwest. Because the quality and extent of Northwest geothermal resources had not been demon- strated, the regional geothermal resource potential was not incorporated in the resource portfolio of the 1983 plan. However, the resource assessment prepared for Bon- neville by the four Northwest states (The “Four State Study”) indicates that approx- imately 4,400 megawatts of potentially cost- effective electrical energy could potentially be obtained through development of regional geothermal resource areas. 67 Chapter 6 Table 6-5 Pacific Northwest Geothermal-Electric Resources POTENTIAL POTENTIAL ENERGY PROJECT COUNTY STATE caw) etaWe). : (centeikWh) Cove and Crane Creek Washington ID 224 179 3.4 Big Creek Hot Springs Lemhi ID 29 23 3.5 Newberry Volcano Deschutes OR 1,946 1,557 3.8 Wart Peak Caldera Lake OR 145 116 41 Glass Buttes Lake OR 348 278 4.2 Raft River Area Cassia ID 15 12 4.3 Cappy-Burn Butte Klamath OR 473 378 4.3 Mickey Hot Springs Harney OR 138 110 4.4 Bearwallow Butte Deschutes OR 763 610 45 Melvin-Three Creek Buttes Deschutes OR 1,380 1,104 4.5 Boulder Hot Springs Jefferson MT 3 2 4.7 Vale Hot Springs Malheur OR 163 130 4.9 Klamath Falls Area Klamath OR 453 362 5.0 Olene Gap Hot Springs Klamath OR 26 21 5.0 Neal Hot Springs Malheur OR 43 34 5.0 Klamath Hills Area Klamath OR 366 293 5.1 Lakeview Area Lake OR 10 8 5.1 Crump Hot Springs Lake OR 79 63 5.1 Deer Creek Hot Springs Boise ID 3 2 5.2 Summer Lake Hot Springs Lake OR 5 4 5.2 Borax Lake Hot Springs Harney OR 83 66 5.2 Alvord Hot Springs Harney OR 35 28 5.3 Hallinan Hot Springs Lake OR 2 5.3 Trout Creek Area Harney OR 4 3 5.3 Fischer Hot Springs Lake OR 1 1 5.4 Ennis Madison MT 3 le 5.4 Barry Ranch Hot Springs Lake OR 1 1 5.4 Jackson Hot Springs Beaverhead MT 3 2 5.5 Crater Lake Area Klamath OR 45,300 36,240 5.5 Generic High Cascades4 Linn OR 34 27 5.6 Mt. McLoughlin Jackson OR 17,598 14,078 5.6 Roystone Hot Springs Gem ID 3 2i/ 5.6 Baker Hot Springs Whatcom WA 3 2 5.9 Mt. Hood Clackamas OR 6 5 6.2 (Table Continued) Generation Technology Acentral station geothermal plant consists of a wellfield, transmission piping and a power plant. Unit capacities typically range from 25 to 100 megawatts, with multiple units employed for larger fields. Integrated wellhead units are also available, ranging in capacity from tens of kilowatts to several megawatts. Three basic generation technologies are in use: dry steam plants, flashed steam plants and binary cycle plants. The choice of tech- nology for a given development depends upon the quality of the resource. Pacific Northwest geothermal resources are antici- pated to be of the low and intermediate tem- perature hydrothermal type. The intermedi- ate temperature resources (150 - 210°C), offer the greatest potential for electricity gen- eration. It is likely that flashed steam plants would be used for resources greater than 175°C and binary plants would be used for resources less than 175°C. The choice of central station versus wellhead units would depend upon the capacity of the field and the development strategy. Wellhead units might be used for small site develop- ment, for serving small remote loads, for the early phases of development of large fields, and for reservoir confirmation work. Flashed steam plants are a well-demon- strated, commercially available technology. Binary cycle plants are commercially avail- able and demonstrated at wellhead scale. Commercial demonstration of central station binary cycle technology is underway at Heber, California. Project Cost and Performance The resource assessment prepared by the four states for Bonneville contains prelimi- nary estimates of the cost of development and operation of 92 promising geothermal resource areas, based upon the limited infor- mation currently available. These cost esti- mates yield levelized energy costs as low as 3.4 cents per kilowatt-hour using the Coun- cil’s revenue requirements methodology. Ten sites, producing a potential 4,370 megawatts of energy, are estimated to have levelized energy costs of 4.5 cents per kilowatt-hour or less. Chapter 6 The Four State Study evaluated and ranked the known and suspected geothermal resource sites of the region. Of the 1,265 known or suspected geothermal sites evalu- ated, 245 warranted detailed analysis, indi- cating either development potential, cost effectiveness, or both. Sites were ranked on the basis of development potential and on Cost. Seventy-eight electric power sites were identified as having “good” or “average” Potential for development. The Four State Study report contains level- iZed life cycle cost estimates for each of the electrical generating sites passing the study's site screening process. These costs, how- ever, are not comparable to the levelized energy costs used elsewhere in this plan, largely because of somewhat different finan- cial assumptions adopted by the Council. To obtain comparable cost estimates, the Coun- cil first compiled basic cost and performance data from the Four State Study. The cost estimates were adjusted to the January 1985 base used in this plan. Minor adjustments were made to the construction schedule and Payout assumptions provided in the Four State Study to improve consistency with other resources assessed in this plan. The resulting data for a typical site are shown in Appendix 6E to this chapter. The resulting cost and performance estimates were used to calculate levelized life cycle costs using the Council's revenue requirements meth- odology. Investor-owned utility financing was assumed, as described in Chapter 4 of this volume, for consistency with the levelized life cycle cost estimates prepared for other resources. The resulting levelized life cycle Costs are shown in Table 6-5 for all resource areas having an estimated electrical genera- tion potential of one average megawatt or greater, and “good” or “average” potential for development. The locations of these areas are shown in Figure 6-1. Ten of these electric power sites, estimated to be technically capable of generating approx- imately 4,370 megawatts of electrical energy, could be competitive with busbar costs of new conventional coal plants. (Table Continued) White Arrow Gooding ID 1 1 6.8 Indian Creek Hot Springs Owyhee ID 1 1 6.9 Squaw Creek Hot Springs Franklin ID 3 2 6.9 Norris Hot Springs Madison MT 1 1 7.0 Sharkey Hot Springs Lemhi ID 1 1 7.4 Crane Hot Springs Harney OR iJ 2 74 Battle Creek Hot Springs Franklin ID 1 1 7.7 Umpqua Hot Springs Douglas OR 1 1 8.1 Magic Hot Springs Camas ID 3 2 8.1 Little Valley Area Malheur OR 1 1 8.2 O. J. Thomas Well Harney OR 1 1 8.3 Maple Grove Hot Springs Franklin ID 1 1 9.2 Murphy Hot Springs Owyhee ID 1 1 9.3 Marysville Well Lewis & Clark = MT 5 4 9.9 Latty Hot Springs Elmore ID 1 1 10.8 aSites having an estimated developable capacity of 1 MW or greater and assessed as having “good” or “average” development potential in the Four State Study. >From Bonneville Power Administration, 1985, Evaluation and Ranking of Geothermal Resources for Electrical Generation or Electrical Offset in Idaho, Montana, Oregon and Washington. ©Levelized life cycle costs calculated using the estimates of development and operating costs appear- ing in Bonneville Power Administration, 1985, Evaluation and Ranking of Geothermal Resources for Electrical generation or Electrical Offset in Idaho, Montana, Oregon and Washington. These cal- culations assumed investor-owned utility financing, using the financial assumptions described in Chapter 4 of Volume II. 4Selected as a typical High Cascades site. Many additional sites of this type may exist. Resource Availability Current interpretations of available data sug- gest that a substantial and cost-effective geothermal resource is potentially available to the region. Appropriate conversion tech- nology is available. However, the geothermal systems of the region must be further charac- terized, and reservoirs tested, before this resource can be considered to be available. Generic conceptual models of the physical characteristics of the region's geothermal systems must be confirmed. Characterization of the cost, availability, and environmental effects of a geothermal resource is a multi-step process. First, basic geologic, geochemical, geophysical, and hydrological data must be gathered and ana- lyzed to develop conceptual models of geo- thermal systems. Much of this preliminary effort is accomplished by what are referred to as surface reconnaissance techniques. Sub- sequently, models are created to explain the data. These conceptual models lay the foun- dation for directing exploratory drilling toward potential geothermal production sites. Explo- ration drilling defines the thermal, lithologic, and geochemical environment with the aim of pinpointing specific sites for geothermal fluid production wells. Finally, production wells are drilled and tested. Only at this stage is the resource confirmed. The conceptual generic models of the geothermal systems are also confirmed in this process. The characteristics of geothermal energy— in terms of costs, geology, brine charac- teristics, and environmental concerns— cannot necessarily be extrapolated from one reservoir to another. Although geothermal resources of one geologic province (such as the Cascades) may have commonalities, each reservoir will be unique. Determination 6-9 Chapter 6 Cappy Burn Butte of regionwide geothermal energy availability and potential will require resource assess- ment on a reservoir-by-reservoir basis. Strategies for assessing and developing geothermal resources have been developed for the Basin and Range geologic province. These strategies would apply to south- eastern Oregon, southern Idaho, northern Nevada and northwestern Utah—a large geographic portion of the region within Bon- neville’s service area. However, no basic models exist yet for the Cascades province, where a significant resource is projected to exist. Conclusion The Council considers the geothermal resource potential of the region to be promis- ing. The estimated cost and magnitude of this resource suggest that geothermal could play a major role in providing future energy needs of the region. However, because the information regarding the character and 6-10 Newberry Volcano Glass Buttes Wart Peak Figure 6-1 Pacific Northwest Geothermal Resource Areas extent of regional geothermal resource areas used to prepare the estimates of cost and availability is very preliminary, this resource cannot yet be considered as available for the resource portfolio of this power plan. Because of the apparent magnitude and cost effectiveness of this resource, the Council encourages continuation of activities leading to confirmation of regional geothermal resource areas. The Four State Study should be maintained and refined as a repository and clearinghouse of information regarding the availability and costs of geothermal resource development. Effort should be made to obtain broad regional review and concurrence on this study. Due to the large investment required to com- plete confirmation activities, it is clear geo- thermal resource developers will not under- take a full confirmation program without assurance of a substantial market for their resource to recover investment costs. Under conditions of extended power surplus, large markets for energy are not likely to emerge. Because it may be desirable to confirm this resource as an alternative to making deci- sions to construct new large conventional thermal resources, alternative methods of promoting the confirmation of geothermal resources should be explored. These meth- ods might include offers similar to those called for by Action Item 17.1 of the 1983 plan (perhaps using well-head scale plants), addi- tional resource assessment and research, development and demonstration activities, or granting developers conditional access to out-of-region markets. A geothermal resource research and devel- opment agenda that includes ongoing scien- tific and technical investigations should be prepared to identify and sequence specific actions required to confirm this resource. Hydroelectric Power Existing and assured Pacific Northwest hydropower projects provide 29,800 mega- watts of firm capacity and about 12,300 megawatts of firm energy (Appendix 6-A). Most environmentally acceptable large- scale hydropower sites have been devel- Oped. Remaining potential includes irrigation and flood control and other non-power water Projects that could be retrofitted with genera- tion equipment; addition of generating equip- ment to existing hydropower projects; plus many undeveloped sites potentially suitable for small-scale development. The Council included 920 megawatts of firm energy from new hydropower in the 1983 Power Plan. These resources included approximately 255 megawatts representing the addition of generation to non-power pro- jects and capacity additions to existing power Projects. The balance consisted of projects at currently undeveloped sites. Generation Technology Hydropower projects extract energy from fall- ing water and require operating head (vertical drop) and water flow. Project configuration may be of the instream, diversion or canal or Conduit types. Instream projects use operat- ing head created by a dam, which backs water up the stream channel. Sometimes the dam may impound sufficient water to permit regulation of streamflow so power can be generated when needed. Such projects are called storage projects. If sufficient reservoir storage is not present to allow streamflow regulation, power is generated as streamflow Permits. Such projects are called run-of-river projects. In a diversion project, water is diverted from the stream by a diversion structure (dam or weir) and transmitted to a downstream Powerhouse by canals, conduits or other conveyance structures. The operating head is developed by the difference in elevation between the diversion structure and the powerhouse. Sometimes the diversion struc- ture is a dam that may provide additional operating head and storage to permit regu- lated power production. Chapter 6 A canal or conduit project involves the con- struction of a powerhouse using potential operating head present on existing non- power water conveyance structures such as irrigation canals and water supply conduits. Hydropower is a renewable energy source and is free from toxic emissions. However, its effect on stream characteristics may present environmental problems. Dams and reser- voirs transform a portion of the stream from a free-flowing stream to a lake-like impound- ment. This results in inundation of land and biologically significant changes to the stream. Water is diverted from the natural stream channel in a diversion project. Con- sideration must be given to maintaining ade- quate streamflows for biological and aes- thetic purposes. Dams, diversions and powerhouses may form barriers to the natu- ral movement of anadromous and resident fish. Provisions for fish passage and protec- tion from turbines may be required. Canal and conduit projects are typically environ- mentally benign; however, because conduits and canals are themselves conveyance structures in a diversion project, considera- tion must be given to effects of project opera- tion on instream flow. Although hydropower technology has been in use for a century, improvements in turbine and generator design, materials and control systems have increased the efficiency of newer plants. These improvements create a potential for cost-effective upgrades of older plants. This potential is considered in the section concerning hydropower efficiency improvements. Project Cost and Performance The cost and performance and environmen- tal effects of hydropower projects are highly site-specific and not amenable to generic assessment. Information currently available to the Council is from an assessment done for the 1983 Power Plan. This information was limited to sites for which the results of individual engineering studies were available. To improve estimates of the cost and energy production potential of regional hydropower resources, the Council, in cooperation with the Corps of Engineers and the Bonneville Power Administration, is assembling the Pacific Northwest Hydropower Data Base and Analysis System. This data base will contain locational, cost, performance and other information on all Northwest hydro- power projects that have been submitted to the Federal Energy Regulatory Commission permitting and licensing process, plus addi- tional sites appearing in the Corps of Engineers National Hydropower Survey. In addition, cost and energy production estimating methods under development will allow costs and energy production to be esti- mated for projects without such estimates. This data base will be operational in 1986. A review of recent hydropower development and proposed projects suggests that most new development will be sponsored by inde- pendent developers and purchased by con- tract. For this reason, until improved informa- tion is available, the Council will assume that new hydropower will arrive at a cost of 4.0 cents per kilowatt-hour, slightly less than the avoided cost of the marginal thermal resource (new coal). Resource Availability Because of concerns regarding the environ- mental impact of hydropower projects, especially the potential impact on fish and wildlife resources, hydropower resource potential may not approach the estimates appearing in the 1983 Power Plan. The Council and Bonneville are sponsoring the Pacific Northwest Hydropower Assessment Study to improve the ability to identify environmentally acceptable hydropower pro- jects. This study is scheduled to be complete in 1986. Until the Hydropower Data Base and the Hydropower Assessment Study are avail- able, the Council will use a conservative esti- mate of 200 megawatts of firm energy poten- tially available from future hydropower development. This estimate is based on the estimate of 255 megawatts of cost-effective hydropower appearing in the 1983 Power Plan, representing potential development at 6-11 Chapter 6 Table 6-6 Planning Assumptions: New Hydropower Design: Sponsor: Capacity: Firm Energy: Seasonality: Cost: Operating Life: Maximum Ramp-in: Dispatch: Run-of-River Additions to Existing Developments Independent Developers 800 megawatts 200 megawatts January 6 percent February 7 percent March 8 percent April 12 percent May 12 percent June 12 percent July 13 percent August 7 percent September 6 percent October 5 percent November 6 percent December 6 percent 4.0 cents/kilowatt-hour* 30 years 20 megawatts/year None *Levelized lifetime, real existing non-power projects and additions to existing hydropower projects. The 255 megawatt figure was reduced to 200 mega- watts to account for approximately 50 mega- watts of firm energy from projects included in the 255 megawatt estimate that are now operating or under construction. As described above, this energy is assumed to be available at a real levelized cost of 4.0 cents per kilowatt-hour, slightly less than the Cost of new coal. The Council recognizes that the addition of power generating equipment to certain non- Power projects may have adverse effects on fish and wildlife. It is likely, however, that the energy otherwise available from these sites can be obtained from environmentally acceptable projects at undeveloped sites. 6-12 Conclusion The Council concludes that at least 200 megawatts of firm energy are available to the region from future cost-effective and environ- mentally acceptable hydropower develop- ment. This energy has been incorporated into the resource portfolio of this plan. The planning assumptions used for this resource are shown in Table 6-6. Because of the uncertainty of this estimate, revised estimates based on improved inventories and cost estimating tools could show a greater potential. The Council encourages the continued development of information and planning tools leading to improved estimates of regional hydropower availability. The Council also encourages actions leading to better understanding of the possible environmental effects of hydro- power development and available mitigation measures. Municipal Solid Waste Electric Generation Studies prepared for the Council for the 1983 Power Plan indicated that municipal solid waste (MSW) projects potentially producing an aggregate of 22 megawatts of energy were planned; projects potentially producing an additional 48 megawatts were under con- sideration. The Council did not include energy from MSW in the 1983 plan, given the uncertain public acceptance of these pro- jects. No electric generation projects using MSW for fuel are yet operating in the region. Three, however, are scheduled to come into service within the next two years. The 13 megawatt Ogden-Martin (Trans-Energy) plant located near Salem, Oregon, is planned to come into service in January 1987. A small (less than 1 megawatt) unit in Coos County, Oregon, is scheduled to come into service in 1986. Tacoma, Washington, is converting a retired steam plant into a cogeneration facility that will co-fire coal, wood residue and refuse-derived fuel obtained from MSW. Generation Technology Direct-fired steam-electric plants continue to be the principal technology for generating electricity with municipal solid waste. Cogeneration improves their cost effective- ness. Direct-fired plants use shredded waste or refuse-derived fuel fired alone or co-fired with fuels such as coal. Mass-burning is the most common combustion technology in cur- rent use. Gasified and pelletized MSW fuels are being investigated. Production of elec- tricity from municipal solid waste using direct- fired steam-electric power plants is an estab- lished process, especially in Europe, and to an increasing extent in the United States. MSW-fired power plants can be a cost-effec- tive and environmentally acceptable method of disposing of municipal solid wastes. These plants, however, present the potential for air pollution, noise, odor and traffic impacts, impairing their public acceptance. Air pollu- tion remains a concern and a significant impediment to public acceptance of MSW generation facilities. Potential pollutants cre- ated by combustion of MSW include particu- Chapter 6 lates, carbon monoxide, hydrocarbons, heavy metals, dioxin, chlorides and fluorides. Dust and microorganisms can also be released from fuel storage and handling sys- tems. Control mechanisms are available, however. Conventional flue gas particulate removal technologies may be used to control Particulates and heavy metals. Carbon mon- oxide, hydrocarbons and dioxins can be con- trolled by maintaining specific combustion Conditions and by provision of afterburners. It is important to consider that many of these Potential pollutants are present when munici- Pal solid wastes are disposed by alternative means, such as in landfills, and that properly designed and operated MSW power plants may offer improved control. Project Cost and Performance The cost and performance characteristics of generic 1, 10 and 20 megawatt stand-alone, direct-fired steam-electric plants fired with MSW are described in the resource assess- ment prepared for the 1983 plan. For the 10 and 20 megawatt projects—a size range Suitable for the waste load of larger metro- Politan areas— energy costs were estimated to range from approximately 0.5 to 7.4 cents per kilowatt-hour (1980 dollars), depending upon project size and the tipping fee.$ Equiv- alent 1985 costs would range from approx- imately 0.7 to 10 cents per kilowatt-hour. The cost of electrical energy from these plants is highly sensitive to the tipping fee and to the size of the facility. The tipping fee is highly dependent on the local solid waste disposal situation, and can range from less than $1 to more than $5 per million British thermal units (Btu) of waste. This fee is deter- mined by the avoided cost of alternative methods of waste disposal. Landfill is often the principal alternative to use of waste in an MSW generation plant. Landfill disposal costs for metropolitan areas will likely in- crease over time at a rate exceeding general inflation as the availability of landfill sites decreases. Costs for more rural areas, with a greater availability of suitable disposal sites, will likely pace general inflation. The more cost-effective projects will therefore likely be located near metropolitan areas. An addi- tional factor contributing to the potentially greater cost effectiveness of MSW plants near major metropolitan areas is the greater volume of available waste. This will allow con- Struction of larger and therefore more cost- effective projects. Resource Availability Estimates prepared for the 1983 Power Plan indicated that MSW sufficient to support 147 megawatts of electrical energy production would be available in the region by 1985. Regional MSW availability was expected to increase, due to population growth and other factors, to an amount capable of supporting 169 megawatts by the year 2000. These esti- mates remain the most current estimates of regional generation potential using MSW. Conclusion Although certain applications of MSW gener- ation appear cost effective, the Council has not specifically included the electrical energy potential of the region's solid waste in the resource portfolio because of uncertainties regarding air quality effects and the difficulty of siting MSW generation projects. The three projects currently under development in the region (Ogden-Martin, Tacoma and Coos County) were included in the inventory lead- ing to the estimated availability of cogenera- tion.4 The remaining resource is considered promising, subject to resolution of questions regarding air quality effects and site selec- tion. Development and implementation of sit- ing and emission control standards for MSW plants throughout the region will assist confir- mation of this resource. Solar-Electric Power Although gaps remain in regional coverage of solar data, enough is known to say that the potential is large. As an example, the south- eastern Oregon and southwestern Idaho areas receive about 83 percent of the direct normal solar insolation received by Phoenix, Arizona. Areas west of the Cascades receive far less solar insolation. Western Oregon, for example receives only about 52 percent of the insolation received in Phoenix. The high cost of solar-electric generation technology precluded it from consideration as an available resource in the 1983 plan. Recognizing its potential, however, the Coun- cil called for continuing the insolation data collection in the region. The Council also recommended that Bonneville monitor technology changes that may lead to cost breakthroughs. Generation Technology The two basic types of solar-electric tech- nology are solar-thermal generation and solar-photovoltaic generation. Solar-thermal plants use heat engines® to produce power from solar insolation. Major types of solar thermal systems include cen- tral receivers, parabolic troughs, parabolic dishes and solar ponds. Central receivers and solar ponds have not yet been demon- strated in the size and form required for commercial installations. One 10 megawatt prototype central receiver is operating in Bar- stow, California; a second, larger prototype was planned but has not been constructed due to funding difficulties. Because the size and cost of research prototypes and demon- stration units is substantial, the rate of deployment of central receiver units and solar ponds for testing is likely to be slow. Parabolic trough and dish designs, on the other hand, are inherently small scale and modular. Being smaller, and inherently less costly, these designs are likely to evolve more rapidly than central receiver or solar pond technologies. The modularity of parabolic trough and dish designs also presents the potential advantages of factory fabrication, short lead time and development to match rates of load growth. At present, the leading design appears to be the parabolic trough design. A power plant employing parabolic troughs and developed under the auspices of Public Utility Reg- ulatory Policies Act contracts and third party financing is in operation at Daggett, Califor- nia. Phase lis on line with an installed capac- ity of approximately 14 megawatts; Phases II and Ill are under construction with an ulti- mate aggregate capacity of all three phases in excess of 74 megawatts. Several different prototype parabolic dish type units are being evaluated by Southern California Edison (SCE) for their respective manufacturers. Some of these designs have exhibited effi- ciencies in excess of 27 percent, greatly exceeding photovoltaic efficiencies but with the modular advantages of photovoltaics. The prototype parabolic dish units use Stir- ling engines to convert heat to mechanical energy. Since the Stirling engines can use any external heat source for power, hydrogen or natural or synthetic gas could be used to operate the units at night or during periods of limited solar insolation. 613 Chapter 6 Table 6-7 Generic Solar Generating Projects: Cost and Performance Summary TYPE THERMAL PHOTOVOLTAIC PHOTOVOLTAIC PHOTOVOLTAIC THERMAL Design Central Receiver Flat Plate (fixed) Flat Plate (track) | Concentrating Stirling Dish Location SE Oregon SE Oregon SE Oregon SE Oregon SE Oregon Sponsor 1OU lou 1OU lou 10U Capacity (Net MW) 100 10 10 10 10 Heat Rate (Btu/kWh) n/ap n/ap n/ap n/ap Wap Availability (%)@ 83 95 95 95 95 Capacity Factor (%) 50 28 40 28 28 Option Lead Time (mos)® 24 24 24 24 24 Construction Lead Time (mos) 48 24 24 24 24 Option Cost (million $) 17 1 1 1 1 Construction Cost (million $) 843 136.0 144.5 48.5 38.8 Fixed Fuel Cost (million $/yr) n/ap n/ap n/ap n/ap nap Variable Fuel Cost ($/MMBtu) n/ap n/ap n/ap n/ap n/ap Fixed O&M¢ Cost (million $/yr) 0.0 0.05 0.05 0.05 0.0 Variable O&M Cost (cents/kWh) 1.0 0.2 0.2 0.2 11 Capital Replacement (million $/yr) nav n/av n/av n/av niav Decommissioning Cost (million $) 3.0 0.0 0.0 0.0 0.0 Amortization Life (yrs) 30 30 30 30 30 Operating Life (yrs) 30 30 30 30 30 Energy Cost (cents/kWh)4 15 38 29 14 12 @Equivalent Annual Availability. Site and license acquisition. ¢Operation and maintenance. dLevelized lifetime energy costs for 1995 inservice date. Solar photovoltaic electricity is produced directly from solar insolation by using pho- tosensitive materials. Photovoltaic electricity generation is attractive for several reasons: 1) It makes use of total solar radiation, not only direct sunlight; 2) Unlike solar-thermal designs, water is not required for cooling; and 3) Tracking devices are not required (though performance may be improved through the use of tracking). The modularity and solid state nature of photovoltaic technology sug- gests that rapid improvements in the tech- nology are probable. Modularity also pro- vides potential advantages of short lead time, factory fabrication and synchronization with load growth. Substantial basic physics and material science problems must be resolved if the efficiencies and per unit costs, which 6-14 photovoltaics must achieve in order to become cost effective, are to be realized. Because the Pacific Northwest's marginal energy costs and solar insolation are both lower than the Southwest's, the technology will be cost effective in the Southwest before it becomes cost effective in this region. Project Cost and Performance The Oregon State Department of Energy has supplied current cost and performance esti- mates for five generic solar-electric central station units. These include a solar-thermal central receiver, a solar-thermal Stirling dish, and fixed, tracking and concentrating pho- tovoltaic stations. Key cost and performance characteristics of these units are shown in Table 6-7. The costs shown in Table 6-7 rep- resent present-day costs and do not reflect possible future cost reductions. As is evident from Table 6-7, solar-electric technology is not yet cost-competitive with other resource alternatives for central-station electricity generation. Resource Availability The solar insolation received by this region, primarily in southeastern Oregon and south- western Idaho, could support a large base of solar-electric generation. The resource potential has not been estimated. Conclusion Solar energy is renewable and has relatively benign environmental effects. Furthermore, many of the leading solar-electric technolo- gies have desirable planning characteristics such as small module size and short lead time. On the other hand, solar is an intermit- tent resource andis at its prime in areas of the region remote from major load centers. Despite these problems, solar electricity generation may be highly desirable if costs can be reduced. Only the currently high Cost of solar electricity generation keeps the Council from further considering this resource in this plan. Significant reductions in price for solar tech- nologies have occurred in the past, and the region should continue to monitor further development of solar-electric technologies. The Action Plan calls for development of a solar resource research, development and demonstration agenda. Planning tools are needed to better assess the value of intermit- tent resources to the regional power system. Using these tools, the possible role of solar- electric generation and other intermittent resources can be assessed in future plans. Mountain “i Columbia Figure 6-2 Pacific Northwest Wind Resource Areas Wind-Electric Power Two experimental wind projects and one windfarm are currently operating in the region. The experimental projects include the 7.5 megawatt capacity U.S. Department of Energy project at Goldendale, Washington, and the 0.2 megawatt capacity Whiskey Run Project on the Oregon coast. A 1.25-mega- watt windfarm operated by private devel- opers and contracted to Pacific Power and Light is also located at Whiskey Run. An 80 megawatt (capacity) commercial windfarm has been proposed at Cape Blanco. Many other locations in the Pacific Northwest have awind resource sufficient for electrical power generation by wind turbines. The wind resource assessment conducted for the 1983 plan identified seven prime areas for the large scale development of windpower. Pos- sible development scenarios were postu- lated, and potential energy production and levelized energy costs estimated for each. These costs, which were especially sensitive to wind turbine hardware costs, wind regime and deployment scenarios, ranged from 4.5 to 7 cents per kilowatt-hour in 1980 dollars. Current (1985) dollar equivalents would be Chapter 6 approximately 6 to 9.3 cents per kilowatt- hour. Although the estimated cost of several of these sites was close to the cost of new coal, wind was not included in the resource portfolio of the 1983 plan, primarily because of the uncertain cost and performance char- acteristics of the turbines available at that time. Generation Technology There has been rapid development of wind- power in California since 1981. This is due largely to the fortuitous combination of sev- eral key factors, including attractive state and U.S. income tax regulations, abundant in- state investment capital, high avoided cost for power purchased under the Public Utility Regulatory Policies Act provisions and a favorable wind resource near load centers. It is estimated that as of the end of 1985, approximately 13,000 turbines having an aggregate nameplate rating of approximately 1,100 megawatts will have been installed, representing about 98 percent of installed U.S. wind capacity. Because this develop- ment has been tax-shelter driven, it is not clear that it will continue with the pending expiration of federal energy tax credits. 6-15 Chapter 6 Table 6-8 Representative Wind Turbine Cluster: Cost and Performance Summary Design Fuel Location Sponsor Capacity (Installed MW) Capacity (Net MW) Heat Rate (Btu/kWh) Availability (%)> Capacity Factor (%) Option Lead Time (mos)¢ Construction Lead Time (mos) Option Cost (million $) Construction Cost (million $) Fixed Fuel Cost (million $/yr) Variable Fuel Cost ($/MMBtu) Fixed O&M? Cost (million $/yr) Variable O&M Cost (cents/kWh) Capital Replacement (million $/yr) Net Decommissioning Cost (million $) Land Royalty (% net cost of energy) Amortization Life (yrs) Operating Life (yrs) Energy Cost (cents/kWh)e 100 — Nordtank Model 65/13 WTGs@ nap Columbia Hills East, Washington 10U 65 6.2 "Wap 95 35 12 24 0.1 9.7 nap nap Inc. in variable O&M ue Inc. in variable O&M Inc. in variable O&M 5 20 20 5.5 @Wind turbine generators. >Equivalent Annual Availability. Site and license acquisition. Operation and maintenance. Levelized lifetime energy costs for 1995 inservice date. 6-16 The California experience, however, has stimulated the evolution of the wind turbine from a novel machine of questionable reliabil- ity to a fairly well-proven generation tech- nology. A somewhat unexpected develop- ment has been the evolution of the intermediate-scale machine (50 to 500 kilo- watts) as the machine of preference. This contrasts with the utility-oriented research of the late 1970s that focused on megawatt- scale machines. Although multi-megawatt, utility-operated machines may become com- mon in the future, the present trend is to intermediate-scale machines developed in windpark settings by independent devel- opers. Windpower can now be considered as having a commercially-available and demon- strated technology. Project Cost and Performance The Oregon Department of Energy has pre- pared cost and performance estimates for a Nordtank 65/13 wind turbine generator. This European machine is representative of the better intermediate-scale horizontal axis machines available on the current market. Its cost and performance characteristics, adjusted to be comparable with other resources, are summarized for a represen- tative wind resource area in Table 6-8 and described in detail in Appendix 6F. Resource Availability Regional wind resource areas have been sur- veyed and monitored by Oregon State Uni- versity under the Bonneville Regional Wind Energy Assessment Program. Under this program, Oregon State University (OSU) has identified 46 areas in and adjacent to the region having good wind resource potential (Figure 6-2). The Oregon State Department of Energy has estimated the number of Nor- dtank turbines that could be installed at eight of the better areas and the resulting energy production (Table 6-9). The levelized cost of energy at the more favorable areas was esti- mated by the Council using the financial assumptions of Chapter 4 of this volume. Financing was assumed to be at investor- owned utility rates. An in-service date of 1995 was assumed, using present day machine cost and performance data. Estimated level- ized energy costs ranged as low as 5.5 cents per kilowatt-hour. Conclusion Estimates of energy cost and availability from the better regional wind resource areas indi- cate that wind, though not presently cost effective, offers a large resource potential. Continuation of the cost reductions that have Occurred in the wind generating equipment industry over the past several years may make this resource cost effective in the future. These projects would be highly modu- lar and would likely be environmentally acceptable if properly developed. On the other hand, the resource is intermittent and the larger sites are extremely remote from load centers and located in harsh climates. In addition, many of these areas are not suffi- ciently well understood to allow their resource be considered confirmed. The Council has not included wind in the resource portfolio, primarily because of cost, and to a lesser extent because of the limited information regarding site characteristics. However, the Council believes there is a good chance that wind will become cost effective in the future. For this reason, the region should take actions to ensure that the resource can be developed if cost effective. These actions include developing and implementing a wind resource research and development agenda, developing tools to assess the value to the regional power system of intermittent resources, and developing and implement- ing a resource acquisition policy by Bon- neville to include intermittent resources. State and local governments are encour- aged to implement siting and performance standards that ensure environmentally acceptable wind resource development. Wood One utility-operated generation plant using wood residue (the 45 megawatt Kettle Falls Generating Station) is currently operating in the region. In addition, the output of several small stand-alone wood-fired plants oper- ated by small power producers is contracted to regional utilities (Appendix 6-A). The total energy output of these plants is about 60 megawatts. Council studies for the 1983 plan estimated the total wood residue resource of the region to be sufficient to support genera- tion of about 215 megawatts of energy, exclusive of projects then under construction. Chapter 6 Table 6-9 Cost and Availability of Energy From Better Pacific Northwest Wind Resource Areas man "ene Realy PROJECT COUNTY STATE (MW) (MWa) (cents/kWh) Columbia Hills East 1 Klickitat WA 7 2 5.5 Albion Butte Cassia ID 38 12 5.6 Rattlesnake Mountain 1 Benton WA 20 7 5.8 Sieban 1 Lewis and Clark MT 120 36 6.4 Bennett Peak Elmore ID 4 1 6.4 Goodnoe Hills Addition Klickitat WA 10 3 6.4 Sevenmile Hill Wasco OR 56 15 6.7 Blackfeet Area 1 Glacier, Pondera MT 15,800 4,370 6.7 *Because insufficient information is available concerning the amount of developable land in each wind resource area, actual potential may vary + 40 percent or more from values shown. No stand-alone wood-based generation was included in the 1983 plan other than the Ket- tle Falls Generating Station. However, wood accounted for a portion of the 400 megawatts of energy from renewable-based cogenera- tion included in the 1983 plan. Previous studies by the Council have indi- cated that both stand-alone and cogenera- tion plants fired by wood are cost effective. There is however, considerable uncertainty regarding the cost and availability of this resource. This uncertainty is created by changing and competing uses of the resource (such as the apparent increase in the use of wood for residential heating in recent years) and changing economics within the forest products and pulp and paper industries. Better definition of the cost and availability of this resource, and the factors that impact cost and availability over time, are required. Until a better understanding of this resource is achieved, the Council will con- tinue to include it in the promising category. Cogeneration Cogeneration is the simultaneous production of electricity and useful heat energy. The heat energy is typically used for industrial process or space heating applications. Cogeneration providing about 230 megawatts of capacity and 130 average megawatts of energy is cur- rently contracted to regional utilities.6.7 Addi- tional projects, providing 80 megawatts of capacity and 60 megawatts of energy, are scheduled to come into service by 1989 (Appendix 6-A).7 The Council included 500 megawatts of future cogeneration in the 1983 Power Plan under the high and medium-high load growth forecasts. Cogeneration Technology Cogeneration installations may employ either topping cycles or bottoming cycles. Topping cycles use heat energy first to pro- duce electricity. Exhaust heat from the gener- ation process is then used for industrial pro- cesses or other heating applications. Topping cycles may use steam-electric tur- bine equipment, gas turbines or internal combustion engines. Bottoming cycles recover waste heat from industrial or other processes to use in gener- ating electricity. This type of cogeneration installation consists of a heat recovery boiler powering a turbine, employing either steam or an organic working fluid. Cogeneration technology is commercially available and mature. Development con- tinues on advanced cogeneration concepts, such as fuel cells with waste heat recovery, and packaged units for both general and spe- cialized applications. 6-17 Chapter 6 Table 6-10 Planning Assumptions — New Cogeneration (Sponsors: Independent Developers) LOAD GROWTH Low Medium-Low Medium-High High Capacity: 200 290 290 510 Energy: 130 190 190 320 Seasonality: None Cost: 4.0 cents/kilowatt-hour* Operating Life: 20 years Maximum Ramp-In: 32 megawatts/year Dispatch: None *Levelized lifetime, real Project Cost and Performance Resource Availability Because of the variety of potential applica- tions, technologies and unit sizes, the cost and performance characteristics of cogeneration installations is highly site-spe- cific. Common Pacific Northwest industrial applications include wood-fired steam tur- bine topping cycles and natural gas-fired combustion turbine topping cycles. The cur- rent pattern of cogeneration development suggests that this resource will continue to be developed by independent small power pro- ducers— either industries having cogenera- tion opportunities, or third party developers contracting to both the utility purchasing the plant output and the industry served by the cogenerated energy. For this reason the Council assumes that most cogeneration will be marketed under the provisions of the Pub- lic Utility Regulatory Policies Act of 1978 (PURPA), at the avoided cost of new utility resources as required by PURPA. Because of this assumption and because few potential cogeneration projects inventoried in the PNUCC study were less costly than new coal plants, the Council has assumed that cogeneration will be acquired at a levelized cost of 40 mills per kilowatt-hour—slightly less than the cost of new coal. 6-18 The Councils assessment of the availability of cogeneration is based upon The Pacific Northwest Utilities Conference Committee (PNUCC) report, Cogeneration Potential in the Pacific Northwest, dated December 1984. The PNUCC assessment, which iden- tifies approximately 1,000 megawatts of potential cogeneration, is based upon an inventory of regional industrial cogeneration potential. In a survey conducted by PNUCC member utilities, industries were asked how much cogeneration capacity they would con- sider providing at a given first year purchase price for power. Responses were grouped into categories of “assured,” “planned,” “pro- spective,” “under consideration” and “poten- tial,” depending upon the current status of the proposed cogeneration project. The Council estimates an available cogeneration resource of approximately 320 megawatts of energy. This includes, first, all projects reported as “assured” in the PNUCC. inventory. To avoid double-counting, projects included as “assured” in the Northwest Regional Forecast of March 1985 (the source of the inventory of existing regional resources used for the system planning models) were deleted from the estimate. To the remaining projects were added all “planned,” “prospec- tive,” “under consideration” and “potential” projects having estimated levelized life cycle costs of 5.5 cents per kilowatt-hour or less.8 Excluded from this estimate were any pro- jects reporting use of municipal solid waste, since this resource potential is considered separately. Because of the dependence of cogeneration on industrial operation, adjustments should be made for different load scenarios: the higher the load growth, the more cogenera- tion is likely to be available. Based on the approach of the PNUCC Cogeneration Work Group to quantifying the constraints and uncertainties of resource development, the most likely amount of cogeneration that will be available to the region is estimated to be 190 megawatts of energy for medium rates of growth. Under low rates of growth, industries such as lumber products and pulp and paper are projected to decline. A 50 percent decrease in the potential for these industries is taken, reducing the estimate to approx- imately 130 megawatts of energy. For high rates of growth, industrial expansion will pro- vide additional cogeneration opportunities. Adding an increment for industrial growth, as projected by the work group, increases the cogeneration estimate to approximately 320 megawatts. These cogeneration assumptions result in a total available and cost-effective resource of approximately 320 megawatts of energy for high load growth conditions. This resource consists of approximately 210 megawatts of energy supplied by renewables (of which approximately 140 megawatts are supplied by wood), and 110 megawatts of energy sup- plied by non-renewables, primarily coal. One major project is included in the inventory of projects upon which this estimate was based. This project is the proposed coal-fired Crown Zellerbach project at Wauna, Oregon, of 100 megawatts installed capacity. Addition of power generating equipment to the Fast Flux Test Facility (FFTF) located on the Hanford Reservation in Washington would provide an additional 101 megawatts of capacity and 65 to 70 average megawatts of energy. Because this project would use cur- rently wasted heat, it would likely qualify as a cogeneration facility under provisions of PURPA and asa third-priority resource under provisions of the Northwest Power Act. Assuming that operation of the FFTF itself would continue to be funded by the federal government for research purposes, the cost of energy from the power addition would be well under the 4.5 cents screening criterion. Restrictions on the timing of the power addi- tion give this resource some characteristics of a lost opportunity resource. Conclusion The Council concludes that approximately 320 megawatts of energy may be available to the region from future cogeneration develop- ment. Because of the sensitivity of this resource to economic activity, lesser amounts of cogeneration will likely be avail- able under lower load growth conditions. The Council currently considers approximately 190 megawatts of energy to be available under medium levels of load growth and 130 megawatts to be available under low levels of load growth. The planning data assumptions regarding Cogeneration are shown in Table 6-10. Future assessments of cogeneration poten- tial should consider the effect of economic activity on cogeneration potential. These assessments should also consider the effect Of utility financing of cogeneration equipment upon the availability of this resource. Coal-Fired Electric Generation The Pacific Northwest power system cur- rently includes 12 coal-fired units totaling 5,924 megawatts of nameplate capacity, capable of supplying about 2,480 megawatts of energy to the region. One additional unit is scheduled to come into service in 1986. This is Colstrip 4, at Colstrip, Montana. The regional share of this unit will be 490 mega- watts of capacity, producing about 370 mega- watts of energy (Appendix 6-A). Additional coal-fired projects have been proposed or licensed. These include additional units at Boardman, Oregon; the Creston Generating Station, at Creston, Washington; the Salem project, near Great Falls, Montana; an addi- tional unit at the Wyodak site, near Gillette, Wyoming; and an eight-unit project at Thou- sand Springs, Nevada. Proven reserves of coal, far in excess of those required to meet electricity needs for the foreseeable future, are available to the region. Coal resources sufficient to support electrical power generation are found in Mon- tana, Utah, Wyoming, Washington, Alberta and British Columbia. Out-of-region coal resources could be used by generation Plants located at the minemouth, with the Chapter 6 electrical power transmitted into the region. Alternatively, out-of-region coal could be transported to power plants located nearer major load centers. In-region reserves could provide a future source of coal for electric power generation. However, because of the uncertainties associated with the availability and cost of in-region coal, the cost of poten- tial future coal plants is based on Northern Great Plains coal delivered by unit-train to plant sites located in eastern Washington or Oregon. Generation Technologies The direct-fired steam-electric power plant is the established technology for producing electricity from coal. Although considered a mature technology, enhancements in plant control, efficiency and reliability have improved the cost and performance of new plants compared with earlier designs. A range of unit sizes is available, allowing capacity additions to be matched to load growth. Smaller plant sizes have somewhat shorter construction lead times and greater reliability, but are generally more costly to build and operate than larger units. Alternative coal-based generating technolo- gies under development include fluidized bed combustion designs, integrated coal gasification combined-cycle plants and magnetohydrodynamics. Fluidized bed designs burn coarsely ground coal (or other fuel) in a bed of limestone particles suspended by air injection. The limestone removes sulfur from the coal, reducing or eliminating the need for flue gas desulfurization. In atmospheric fluidized bed combustion (AFBC), the fluidized bed oper- ates near atmospheric pressure. The hot combustion gases power a steam boiler, as in a conventional power plant. In pressurized fluidized bed (PFBC) designs, fuel is burned at elevated pressure. This allows the hot combustion gases to power a gas turbine prior to final heat recovery in a steam boiler. This combined cycle design results in higher energy conversion efficiencies. Cogeneration applications of AFBC tech- nology are commercially available, and cen- tral-station electric generation atmospheric fluidized bed combustion power plants are being demonstrated. Two utility-scale AFBC units are currently under construction in the U.S., and many in the industry believe that the next generation of central-station coal plants will be largely of AFBC design. Devel- opment of PFBC designs is not as advanced; however, a 170 megawatt PFBC demonstra- tion plant is expected to enter service in 1986. Integrated coal gasification combined-cycle (IGCC) plants consist of a coal gasification plant powering gas turbines. Turbine exhaust heat recovery, using steam boilers, creates a highly efficient combined cycle. These plants feature a high degree of modularity, improved control of atmospheric emissions and high energy conversion efficiencies. Integrated coal gasification combined-cycle plants entered the demonstration stage in May 1984 when the 100 megawatt Cool Water plant was brought on-line in southern California. Magnetohydrodynamics (MHD) is a process for converting heat energy directly into elec- tricity. High combustion temperatures, com- bined-cycle operation and direct conversion of thermal energy to electrical energy offer the advantages of high energy conversion efficiency. The MHD concept also promises improved control of atmospheric emissions. The components of a MHD power plant include a combustor, a MHD “channel,” a heat recovery boiler and a steam turbine- generator. Pulverized coal would be burned at high temperature and pressure in the com- bustor. Potassium “seed” is injected to ionize the hot gas, creating an electrically conduc- tive plasma. The plasma passes through the MHD channel, where a strong magnetic field would be established by use of supercon- ducting magnets. The ionized gasses, mov- ing rapidly through the magnetic field, create an electrical potential across electrodes installed in the channel. After exiting the channel, the hot plasma is passed through a heat recovery boiler. Steam from this boiler drives a conventional steam turbine-gener- ator, augmenting the power production of the MHD channel. The potassium seed not only serves to ionize the combustion gasses but also scavenges sulfur by chemical reaction of sulfur and potassium. 6-19 Chapter 6 Table 6-11 Generic Coal Projects: Cost and Performance Summary Design Fuel Location Sponsor Capacity (Net MW) Heat Rate (Btu/kWh) Availability (%)@ Capacity Factor (%) Option Lead Time (mos)® Construction Lead Time (mos) Option Cost (million $) Construction Cost (million $) Fixed Fuel Cost (million $/yr) Variable Fuel Cost ($/MMBtu) Fixed O&M Cost (million $/yr) Variable O&M Cost (cents/kWh) Capital Replacement (million $/yr) Net Decommissioning Cost (million $) Amortization Life (yrs) Operating Life (yrs) Energy Cost (cents/kWh)° 2-603 MW Direct-Fired Units 2-250 MW Direct Fired Units 1-110 MW Atmospheric FBC Unit Subbituminous Coal Subbituminous Coal Subbituminous Coal Eastern Oregon Eastern Oregon Eastern Oregon 1oU lou lou 603 250 110 10,080 10,190 11,200 75 77 75 70 72 70 48 48 48 729 60 72 47.3 27.6 4.2 1,466.2 858.7 197.2 0.0 0.0 0.0 2.00 2.00 2.00 11.41 9.1 3.6 0.1 0.2 0.1 14.5 6.0 12.0 0.0 0.0 0.0 30 30 30 40 40 40 4.6 5.4 6.0 @Equivalent Annual Availability. ‘Site and license acquisition. Levelized lifetime energy costs for 1995 inservice date. dLead time shown is to completion of first unit; additional 12 months (minimum) required to complete second unit. Limited proof-of-concept design work has been carried out over the past several years at two government-funded test facilities. The U.S. Department of Energy, after failing to promote MHD research for three years, pro- posed, in 1984, a five-year MHD research and development program that would culmi- nate in the repowering of an existing conven- tional fossil plant with a MHD power train. Conceptual design of the project would start in 1986 with test operations complete by 1995. The Montana Power Company Frank Bird plant has been proposed as a candidate for this retrofit. The DOE proposal, though stricken from FY-86 budget recommenda- tions, has been restored by Congress. 6-20 Project Cost and Performance The Council, assisted by its Coal Options Task Force, assembled cost and perform- ance characteristics for representative plants of each commercially-available design. These included a large conventional plant, consisting of two units of 603 megawatts of capacity each; an intermediate size conven- tional plant, consisting of two units of 250 megawatts each; and a small AFBC plant consisting of a single 110 megawatt unit. These plants are described in detail in Appendices 6B through 6D, and are summa- rized in Table 6-11. As evident in Table 6-11, the AFBC plant is clearly not yet cost-com- petitive with the two conventional designs. Although the 250 megawatt unit appeared to be somewhat more expensive than the 603 megawatt unit, these two units were com- pared using the decision model, since the somewhat shorter lead time of the 603 megawatt unit may have compensated for its greater construction and operating cost. This, however, was not found to be the case, and the 603 megawatt unit was subsequently used in the resource portfolio. Chapter 6 Resource Availability Three sites, within or adjacent to the region, are either currently licensed for construction of new coal plants or appear to be capable of being readily licensed. These sites are as follows: Boardman: Boardman, Oregon, is the site of the existing Portland General Electric 530 megawatt coal-fired Boardman Generating Station. The Site Certification Agreement for the Boardman site, issued in 1975, allows Construction and operation of two additional thermal power units. The additional units are limited in capacity from 450 to 1,350 mega- watts each, and are to be completed by August 31, 1991, and August 31, 1993, respectively. The site certification requires eight months notice prior to commencement of construction, followed by a state and local government comment period and a public hearing. The state siting council would then issue a new or revised certificate, or revoke the certificate, as appropriate. The warranted completion date may be extended in this process. Certain facilities that would be common to the existing and additional units have been constructed in conjunction with the existing Boardman unit. No specific design relative to the additional units has been undertaken. Because it is unlikely that construction of an additional coal unit could now be completed at this site by the time requirements cited in the Site Certification Agreement, the Council chose not to consider this as a licensed site for purposes of developing the resource Portfolio. Creston Generating Station: The Wash- ington Water Power Company (WWP) has Pursued siting and licensing of a four-unit coal-fired project at a site near Creston, Washington. A Site Certification Agreement (SCA) for construction and operation of the Creston Generating Station was issued by the Washington State Energy Facility Site Evaluation Council (EFSEC) in February 1983. A Prevention of Significant Deteriora- tion (PSD) permit for the project was issued by the U.S. Environmental Protection Agency in January 1983. Completion of the first unit was originally scheduled for 1988, but has since been indef- initely deferred. The SCA requires construc- tion of the first unit to commence within five years of issue (February 1988) and the fourth unit within 15 years of issue (February 1998). Similar time limitations apply to the design of the emissions control system. The PSD per- mit was issued for the standard period of 18 months and was extended by the Environ- mental Protection Agency for an additional 18-month period in December 1984. Thirty- day written notice to EFSEC is required prior to commencement of construction; however, no additional public hearings are required unless it is necessary that the SCA be amended. The PSD permit requires a review of the emission control technology six months prior to beginning construction. Ambient air quality considerations limited the original proposal to four units of approx- imately 508 (net) megawatts of capacity each. Proposed reclassification of the Spokane Indian Reservation airshed to PSD Class | will limit the capacity of the site to approximately half the originally planned capacity. The Washington Water Power Company (WWP) has secured the site and has per- formed project feasibility studies. Detailed project engineering has not commenced. WWP and Puget Sound Power and Light Company are currently maintaining the site and associated permits, although these com- panies, commenting on the draft of this plan, indicated that they would likely not maintain the site and permits indefinitely. No con- straints to continued maintenance of the site and permits for the first two units have been identified by the Council. Wyodak: In 1981, Pacific Power and Light Company (PP&L) was granted all necessary permits for construction of a second unit at the Wyodak site. The proposed second unit would be of 332 megawatts rated capacity and would include a flue gas desulfurization system. Construction was originally sched- uled to commence in late 1981 or early 1982, with completion scheduled for 1986. In November 1981, completion of the unit was deferred to 1989. An application to the Wyo- ming Department of Environmental Quality to extend the PSD permit was denied, and the permit has been allowed to expire. Cur- rent PP&L projections show no need for the plant until 1997. Two additional projects have been proposed adjacent to the region: Salem: A 330-megawatt coal-fired power plant has been proposed for a site near Great Falls, Montana, by the Montana Power Com- pany (MPC). A portion of the site was acquired by the Company and an application for site certification was submitted to the Montana Department of Natural Resources (MDNR). Completion was originally sched- uled for 1989, with full output intended for MPC loads. Processing of the application has since been suspended by mutual agree- ment of MPC and MDNR. Thousand Springs: Sierra Pacific Resources has proposed construction of eight 250 megawatt coal units at Thousand Springs in northeastern Nevada. The com- pany intends to market the output of the plant to customers throughout the West. Applica- tions for permits have been initiated. Comple- tion of the permitting process could be as early as 1988 with construction of the first unit complete by 1992. The plant is proposed to be developed by several non-utility partners plus Sierra Pacific Resources, the parent company of Sierra Pacific Power Company. Conclusion Coal resources adequate to meet any likely future need of the region for electricity are found in and adjacent to the region. Conven- tional and AFBC technologies are available to use this resource, if needed, and other promising technologies are under develop- ment. Conventional direct-fired steam-elec- tric technologies continue to enjoy a cost advantage compared with AFBC technology. Comparison of large (603 megawatts) and small (250 megawatts) conventional units, using the Council's planning models, indi- cates that the larger units remain more cost effective even though they require a some- what longer lead time for construction. 6-21 Chapter 6 Table 6-12 Generic Combustion Turbine and Combined-Cycle Projects: Cost and Performance Summary COMBUSTION TURBINE COMBINED CYCLE Design Fuel Location Sponsor Capacity (Net MW) Heat Rate (Btu/kWh) Availability (%)# Capacity Factor (%) Option Lead Time (mos)® Construction Lead Time (mos) Option Cost (million $) Construction Cost (million $) Fixed Fuel Cost (million $/yr) Variable Fuel Cost ($/MMBtu) Fixed O&ME Cost (million $/yr) Variable O&M Cost (cents/kWh) Capital Replacement (million $/yr) Net Decommissioning Cost (million $) Amortization Life (yrs) Operating Life (yrs) Open Cycle Industrial Natural Gas/No.2 Fuel Oil In-region 1OU 2@105 ea. (nominal) 10,700 85 19¢ 24 304 0.8 52.5 1.0 5.10 (gas)/5.70 (oil) 0.3 0.2 0.3 0.0 20 30 2 Units, ea w/2 CTs & HRSG? Natural Gas/No.2 Fuel Oil In-region 1OU 2@286 ea. (nominal) 9,800 83 19¢ 24 8 6.4 362.2 2.4 5.10 (gas)/5.70 (oil) 48 <0.1 2.0 0.0 20 30 aEquivalent Annual Availability. ‘Site and license acquisition. cAverage capacity factor when operated in hydrofirming mode, assuming sufficient turbines to firm about 700 MW of nonfirm energy. dHeat recovery steam generator. ©Operation and maintenance. Three sites in the region are either currently licensed for construction of new coal plants or appear capable of being readily licensed. Planning for two additional sites has been initiated. After reviewing the status of poten- tial coal sites, the Council believes that sites in an essentially licensed condition could support construction of approximately 1,250 megawatts of new coal-fired generating capacity. This capacity could be expected to produce about 950 megawatts of energy that could be available to the region. This resource could be developed at a cost of approximately $1,216 per kilowatt—the cost of new 603 megawatt coal-fired units, exclud- ing the cost of siting and licensing. 6-22 Additional partially licensed sites could sup- port the construction of approximately 2,700 megawatts of new coal-fired generating capacity. This capacity would be capable of supplying about 2,025 megawatts of energy to the region. These sites could be developed for about $1,255 per kilowatt—the full cost of new 603 megawatt, coal-fired units. Additional coal development, if required, would have to be located at new sites. In this plan, it is assumed that these sites could be developed at the full cost of new 603 mega- watt, coal-fired units (about $1,255 per kilo- watt). Actual development costs of these sites is, however, less certain than develop- ment costs at fully or partially licensed sites. Chapter 6 Gas-Fired Electric Generation One combined-cycle plant and several com- bustion turbine plants in the region have access to natural gas (Appendix 6-A). (The remaining combustion turbines use fuel oil.) As these plants typically have been used only to meet peaking loads, natural gas has not played a substantial role in meeting the region's electrical loads. However, natural gas generating plants may be attractive for Cogeneration, firming of secondary hydro- Power during low water years, and meeting unexpected high rates of load growth until more cost-effective alternatives can be devel- oped. The high load growth case for the 1983 Power Plan included 1,050 megawatts of combustion turbines to meet unexpected load growth, and 100 megawatts of non- renewable cogeneration, a portion of which would be gas-fired. With the exception of a small producing field in Oregon, no natural gas is produced in the region. All major load centers in the region are, however, served by natural gas distribu- tion systems receiving gas from both domes- tic and Canadian sources. Prospects for a continued supply of natural gas appear to be good, with the principal technical question relating to the adequacy of the regional gas transmission and distribution system. At 1983 consumption levels, recoverable domestic reserves should be adequate for more than 60 years. The Canadian supply should remain secure due to long-term con- tracts, an established transmission system, abundant Canadian resources and a close Political relationship, although curtailments of Canadian exports have been experienced in the past due to internal political problems. Future gas supplies can be obtained by developing unconventional gas resources, including tight sands, Devonian shale gas, coalbed methane, synthetic gas (from coal) and geopressured gas. Longer-term gas resources may include gas hydrates and abiogenic methane. Generation Technology Generation technologies using natural gas have several attractive characteristics. The clean combustion characteristics of natural gas result in low maintenance costs, good reliability, siting flexibility and a modest environmental impact. Many have short lead times and low capital costs, and are available in unit sizes that can be closely matched to load growth. Further development of natural gas generating technologies is leading to improved fuel efficiency and reliability. The conventional natural gas generating technologies include direct-fired steam-elec- tric plants, combustion turbines and com- bined-cycle plants. All are commercially available and mature technologies. Develop- ment continues, however, especially for com- bustion turbines and combined cycle plants. A major objective of current work is to increase the efficiencies of these machines. The fuel cell is the principal emerging gener- ation technology using natural gas. Fuel cell plants emit no by-products other than water and carbon dioxide, are relatively quiet, highly modular and are forecast to be more efficient than combustion turbines. Capital costs, however, will have to decrease consid- erably for fuel cells to compete with combus- tion turbines or combined-cycle plants, especially for intermittent duty applications. Natural gas may find application in the longer term as a secondary fuel for solar thermal units using Stirling engine-driven generators. Project Cost and Performance The Council assembled cost and perform- ance characteristics for a twin unit combus- tion turbine plant and a twin unit combined- cycle plant. These are described in detail in Appendices 6G and 6H and are summarized in Table 6-12. Because of the high cost of fuel, the levelized life cycle costs of these plants is much greater than the 4.5 cent screen (Table 6-12). As described in Chapter 7 of Volume |, when used in hydrofirming applications, how- ever, the net cost of firmed power would be much less than the levelized costs of Table 6-12. Resource Availability The near-term availability of new natural gas electric generating projects is potentially lim- ited by provisions of the Powerplant and Industrial Fuel Use Act of 1978 and the regional natural gas transmission, distribu- tion and storage system. Siting and environ- mental concerns do not generally appear significantly constraining. Limitations on urban siting may arise due to noise consid- erations; however, plants could likely be built at existing or licensed thermal sites in the region. Provisions of the Fuel Use Act appear to be the most significant constraint on use of gas- fired generation. The Fuel Use Act has the objective of curtailing use of natural gas and petroleum-derived fuels for generation of electricity where acceptable substitutes are available; it prohibits the use of natural gas or petroleum-derived fuels as primary fuels for new electric generating plants, except under special exemptions subject to approval of the U.S. Department of Energy. Exemptions allowing plants to be built and operated for cogeneration or peak loads appear to be the best opportunities for qualifying new gas- fired generating plants for firming secondary hydropower (See Chapter 7 of Volume 1). In the long term, natural gas supplies could be augmented, if necessary, by gasification of coal. The cost of electricity generated by coal-gasification combined-cycle plants appears to impose a long-term cap on the cost of gas-fired generation. Conclusion Natural gas-fired generating plants use com- mercially available and well-demonstrated technology. These plants are not cost effec- tive for baseload generation applications, nor are new plants permitted to be constructed for this application. Analysis using the Sys- tem Analysis Model indicates that these plants would be a cost-effective alternative for firming secondary hydropower. Potential constraints to the development of these plants for secondary hydropower firm- ing include the future cost and availability of fuel, and provisions of the Fuel Use Act. Site availability appears to present a less signifi- cant constraint. 6-23 Chapter 6 Table 6-13 Cost and Performance Characteristics of WNP-1 and WNP-3 WNP-1 WNP-3 Type Pressurized Water Reactor Pressurized Water Reactor Design Babcock and Wilcox Model 205 Fuel Assembly + Combustion Engineering System 80 Location Richland, Washington Satsop, Washington Sponsor 100% - 106 Public Utilities (net-billed) 70% - 103 Public Utilities (net-billed) 30% - 4 Investor-owned utilities Capacity (Net MW) 1,250 1,240 Availability (%) 65% 65% Expected Shelf Life 15 years, minimum 15 years, minimum Construction Lead Time (mos) Minimum Preservation Cost (million $/yr)# Remobilization Cost Construction Cost to Complete (million $) Financing (Construction) Fixed Fuel Cost (million $/yr) Variable Fuel Cost ($/MMBtu) Fixed O&MP Cost (million $/yr) Variable O&M Cost (cents/kWh) Capital Replacement (million $/yr) Net Decommissioning Cost (million $) Amortization Life (yrs) Operating Life (yrs) 54 + 9 mos remobilization 12 32 1,383 Bonds at 10.2% nominal 35.4 0.0 71.0 0.11 21.0 3.5 30 40 54 + 9 mos remobilization 12 34 1,310 Bonds at 10.2% nominal (Public Share) 80% debt at 13.4% nominal (IOU Share) 20% equity at 14.95% nominal (IOU Share) 38.9 0.0 71.0 0.11 21.0 (mature plant) 3.5 30 40 Recent estimates of the supply system indicate minimum preservation costs to be $10 million per year for WNP-1 and $14 million per year for WNP-3. >Operation and Maintenance. Nuclear Three nuclear power plants of 2,980 mega- watts aggregate installed capacity presently operate in the region (Appendix 6-A). The energy production potential of these plants is approximately 1,930 megawatts. Five years ago, eight additional commercial nuclear plants were in various stages of planning or construction. At present, all have been termi- nated with the exception of Washington Pub- lic Power Supply System Nuclear Projects 1 and 3 (WNP-1 and WNP-3). 6-24 The 1983 Power Plan included WNP-1 and WNP-3 in the resource portfolio since the public share of the plants had been acquired by Bonneville prior to the Northwest Power Act. In the 1983 Power Plan, the Council assumed that the projects would be com- pleted as then scheduled—in 1991 and 1987, respectively. Although no special cost- effectiveness assessment was performed, comparisons of the levelized costs of these projects with the cost of alternative resources indicated the projects would be cost effective. Events since adoption of the 1983 Power Plan have altered the status and potential cost effectiveness of WNP-1 and WNP-3. Construction has been suspended indefi- nitely, based upon the findings of a Bon- neville study completed in November 1984. Because of the indefinite suspension of con- struction, the plants have become potential resource options to the region. Because the costs of preservation and completion of con- struction, and associated uncertainties, have affected the availability, reliability and cost effectiveness of these projects, it was neces- sary to reassess these projects for the 1986 Power Plan. This section focuses on the cost, performance and schedule characteristics of WNP-1 and WNP-3, and associated uncer- tainties. These assumptions are used for the analyses of Chapter 8 of this volume. Chapter 6 WNP-1 WNP-1 is a Babcock and Wilcox Model 205 Fuel Assembly pressurized water reactor nuclear power plant of 1,250 megawatts (net) Capacity located on the Hanford Reservation in Washington. The plant, owned by 106 pub- lic utilities, is being constructed and will be operated by the Washington Public Power Supply System (Supply System). The plant is 100 percent net-billed? by Bonneville. The plant was scheduled for commercial operation in June 1986 prior to the decision of May 1, 1982, to defer plant completion for up to five years. This decision was based on revised load forecasts showing lower elec- trical load growth than previously anticipated, and upon perceived difficulties in marketing bonds for continued construction financing. Construction was estimated to be 63 percent complete when suspended. WNP-3 WNP-3 is a Combustion Engineering Sys- tem 80 pressurized water reactor nuclear power plant of 1,240 megawatts (net) capac- ity located near Satsop, Washington. Sev- enty percent of the plant is owned by 103 Public utilities and is net-billed by Bonneville. The remaining 30 percent is owned by four investor-owned utilities.1° This plant is also being constructed and is to be operated by the Supply System. The plant was scheduled for commercial operation in December 1986 prior to a slow- down order in February 1983. The slowdown was prompted by revised electrical load growth forecasts showing lower growth than previously estimated. On July 8, 1983, due to the inability of the Supply System to continue to market construction bonds, construction was suspended for three years or until financ- ing became available. Construction is esti- mated to be 76.5 percent complete. Follow- ing suspension of construction on WNP-3, the Council studied the cost effectiveness of WNP-3. That study, adopted by the Council on November 21, 1983, found that eventual completion of the project would be cost effec- tive to the region and the plant should be preserved. Bonneville and the Supply System assumed construction restart dates for the two plants of July 1985 for WNP-3, and July 1986 for WNP-1. In June 1984, Bonneville announced it would study the assumptions regarding the construction schedules and methods of financing WNP-1 and WNP-3 to determine what assumptions should be used in the rate proposal for the period extending from July 1, 1985, through September 30, 1987. The results of the study were also used in prepar- ing Bonneville budgets for fiscal years 1986 and 1987. The Bonneville study considered three courses of action for each project: restart of construction in accordance with current assumptions, additional two-year delay, and termination. The study found that further deferral of these plants promised the greatest benefit to the region. WNP-1 and WNP-3 Cost and Performance The Council, with the cooperation of the Sup- ply System, has assembled cost and per- formance characteristics for WNP-1 and WNP-3. This information, and the Council's analysis, has received extensive public review prior to incorporation into this plan. Detailed cost and performance charac- teristics of WNP-1 and WNP-3 are provided in Appendices 61 and 6J. These characteristics are summarized in Table 6-13. The cost effectiveness of these projects, using base case planning assumptions, is estimated to be 3.5 cents per kilowatt-hour for WNP-1 and 3.6 cents per kilowatt-hour for WNP-3. The estimate is based on an assumed inservice date of 2000 and includes preservation costs until construction resumes in 1995. Previous estimates assumed an earlier inservice date and therefore less preservation cost. WNP-1 and WNP-3 Availability Anumber of key uncertainties and their effect on project costs, schedules and perform- ance were considered in assessing the role of WNP-1 and WNP-3 in the 1986 Power Plan. Several uncertainties were found to create barriers to the ability to preserve and complete the plants. The principal uncertainties considered by the Council included the following: Continued Ability to Finance Preservation Preservation of WNP-1 is currently being financed from reinvested funds remaining from construction bond sales. These funds are expected to suffice for the planned level of preservation ($36 million per year) until 1988. Adoption of a reduced preservation budget will extend the period of time for which funds will be available for preservation financ- ing. When funds are exhausted, preservation funding must come from another source, likely from Bonneville rates as is presently done for WNP-3 preservation. Because WNP-3 has no residual funds from construction bond sales, WNP-3 preserva- tion costs are funded from Bonneville rates. Under terms of the WNP-3 Settlement Agreement, the preservation costs of the investor-owned utility share are paid by Bon- neville from rates. Future load growth is more likely to occur in the investor-owned utility service territories than in the service territories of publicly- owned sponsors of WNP-1 and WNP-3. It is therefore likely that the utilities that are cur- rently paying the preservation costs for these projects are not the utilities that will need the capability of these projects. For this reason, it is not clear that financing of preservation through Bonneville rates will continue to be politically feasible. The Council regards this possibility as a very significant potential barrier to successful preservation of the projects. While it appears that physical preservation of the plants for an extended period is feasible (see below), the continued ability to fund preservation is questionable. Resolution of this uncertainty will require the development and implementation of a policy for allocation of option acquisition and main- tenance costs to those benefiting from the options. 6-25 Chapter 6 Availability and Cost of Construction Financing A significant potential constraint to comple- tion of WNP-1 and WNP-3 is the lack of avail- able and cost-effective financing. Because of current litigation and other institutional ques- tions affecting these projects, it is unlikely that bonds to finance these projects could be marketed at present. Alternatively, construc- tion of the projects could be financed directly through Bonneville rates. However, it is unlikely that Bonneville customers would accept the rate impacts of this alternative. Some progress on the settlement of litigation has been made. Bonneville and the investor- owned utility sponsors of WNP-3 have nego- tiated a settlement regarding the utilities breach-of-contract suit, although this settle- ment is being challenged in court. A settle- ment master has been appointed for the WNP-4/5 litigation. Litigation may be resolved by the time resumption of construc- tion is required. It is likely, however, that a residual perception of investment risk will remain even following full settlement of out- standing litigation. The Council has assumed that a risk perception premium of 1.0 percent be assigned to capital borrowed to complete construction. The Council has concluded that it is unlikely that conventional financing for completion of WNP-1 and WNP-3 could be found at this time. Physical Preservation Prolonged suspension of construction could result in unacceptable deterioration of struc- tures and equipment. Particular concern has been expressed regarding WNP-3, which is located in an area of high humidity. Specific concerns include the adequacy of the tempo- rary roof and wall enclosures of the reactor building, temporary wall enclosures of the turbine building, and corrosion of exposed reinforcing steel. Preservation programs are in place at both WNP-1 and WNP-3. Structures have been closed to the weather with temporary enclosures where necessary, and weather- sensitive material has been stockpiled within structures. Humidity in sensitive equipment is controlled by shrouding, heaters and dehumidification. Preservation maintenance procedures have been established for each item of equipment and are being imple- 6-26 mented through a computerized monitoring system. Corrosion coupons have been placed throughout the plants to monitor cor- tosion of materials. Results to date indicate corrosion rates well within acceptable limits for long-term preservation. An assessment of long-term preservation of equipment and structures, drawing upon the U.S. Navy's experience with its mothballed fleet, indicates that, with proper controls, excellent preservation of equipment and structures is possible. Certain products, such as rubber goods, have a limited shelf life and will have to be replaced, even under condi- tions of controlled humidity. Replacement costs are not expected to be significant, as these products are typically designed for periodic replacement during operation. The current preservation program was estab- lished to preserve the plants for a relatively brief period. Continuation of this program appears generally adequate to ensure long- term preservation, although several specific problems remain to be resolved. These include the adequacy of the temporary roof of the WNP-3 reactor building and the need to protect exposed reinforcing steel from exces- sive corrosion. The current situation is acceptable for the near term, and there is time to resolve these problems. The Council concludes that the projects can likely be physically maintained so that com- pletion could be deferred through the 20-year planning period. Maintenance of Site Certification Agreement The Washington State Energy Facility Site Evaluation Council (EFSEC) has informed the Council of concerns regarding mainte- nance of the Site Certification Agreements. The principal concern is renewal of the National Pollutant Discharge Elimination System (NPDES) permit for WNP-3. EFSEC believes the NPDES permit has charac- teristics of a water right. Thus, a permit notin current use could be challenged by a com- peting beneficial use. Because there is no evidence of a potentially competing use, the Council views renewal of the WNP-3 NPDES permit and, consequently, maintenance of the site certification agreement, to be highly probable. Claims Against WNP-1 or WNP-3 Assets by WNP-4/5 Bondholders Successful prosecution of claims by bond- holders of WNP-4 and WNP-5 for losses Sus- tained upon default of the WNP-4/5 bonds could result in their reaching other assets of the Supply System, including WNP-1 and WNP-3. The ultimate effects are unclear, although successful claims would be sunk costs to the region. However, this unresolved litigation continues to preclude conventional financing of the projects. NRC Construction Permit and Operating License There appear to be no fundamental technical or legal considerations that would preclude maintaining the construction permits or obtaining the operating licenses of either WNP-1 or WNP-3. The principal licensing uncertainties concern the extent of unplanned design changes ultimately needed to obtain the operating licenses, and the effect of design changes on current esti- mates of costs and schedules to complete. The Supply System has reviewed pending Nuclear Regulatory Commission (NRC) reg- ulatory actions that might result in design changes, and has incorporated the likely effects into the costs and schedules shown in Appendices 61 and 6J. The potential impact of design changes mandated during the preservation period is tempered by the opportunity to complete engineering plans and specifications prior to resuming construction. Questions remain concerning the com- pliance of current design with NRC stan- dards, and the possibility of new design changes being mandated prior to granting of the operating licenses. Given the experience of the past decade, it could be expected that major new backfits would be required prior to receipt of operating licenses. The effects of such actions could range from minor to severe. Estimates of the historical impact of backfits on U.S. plants average $55 million per plant, but range upward to $1 billion for the worst cases. There is, however, evidence that past experience may not be typical of the future. Commercial nuclear power is a matu- ring technology, and the rates of change experienced during the developing phases of the technology will likely diminish as the tech- Chapter 6 nology matures. The appearance of stan- dard plant designs—WNP-3 is one—is evi- dence of this trend. Also, efforts are underway to improve the nuclear regulatory process. For example: + Application of the new “readiness review” concept to WNP-1 and WNP-3 is underway. Under “readiness review,” completed con- struction will be reviewed for compliance with NRC standards. * Standardized nuclear steam supply system (NSSS) designs are being adopted. For example, WNP-3 employs a standardized Combustion Engineering (CE) System 80 NSSS. This design received Final Design Approval (FDA) from NRC in December 1983. The FDA allows a standard safety analysis to be referenced for operating license approval. Other factors that will contribute to continued licensability of these projects are the following: + Advanced reactor design. The WNP-3 CE System 80 NSSS and the Babcock and Wilcox (B&W) 205 Fuel Assembly NSSS of WNP-1 are state-of-the-art pressurized light water reactor designs. + Existing lead plant experience. The lead CE System 80 plant is Palo Verde 1, which has received its low-power operating license and is scheduled for service in 1986. Palo Verde 2 and 3 are of similar design and scheduled for service in 1986 and 1987, respectively. These plants should serve to “shake out” the System 80 design. The Tennessee Valley Authority's Belle- fonte 1 and 2 units were the lead U.S. Model 205 units. Construction of the Bellefonte units has recently been suspended, and WNP-1 could become the lead U.S. Model 205 unit. The WNP-1 NSSS design, how- ever, is substantially the same as the Muelheim-Kaerlich nuclear power plant, located in West Germany. This plant has completed hot functional tests and is scheduled for commercial operation in August 1986. + Formalized NRC treatment of plant pres- ervation. Informal discussions are under- way between the U.S. Department of Energy, the Supply System and NRC tegarding means by which the continued licensability of mothballed plants could be ensured. Though no formal NRC policy cur- tently exists to ensure continued licen- sability of mothballed plants, a generic NRC policy statement is anticipated. This may lead to a more formal system involving NRC certification of preservation programs. Based upon the above findings, the Council concludes there is an acceptable probability that WNP-1 and WNP-3 construction permits can be maintained and that operating licenses can be obtained if construction were completed. More Stringent Seismic Design Criteria for WNP-3 The design-basis seismic event for WNP-3 was established on probable seismic activity from faulting in the Puget Sound Basin. Sub- sequent analysis of the relative motions of the Pacific, Juan de Fuca and North Ameri- can crustal plates has raised the possibility of seismic events of greater magnitude result- ing from interplate motion. Historically, the plate boundary has been quiet. One school of thought attributes the historical lack of seismicity to strong coupling between the plates. If present, this could lead to a future earthquake of great magnitude. A second school holds that an aseismic subduction" of the Juan de Fuca plate is occurring and, although the plates are in relative motion, no large-magnitude earthquakes will result from this motion. A program to assess the aseismic hypothesis has been developed by the Supply System in response to NRC queries. Implementation of this program is provided for in the Supply System budget for 1985, and in the proposed WNP-3 preserva- tion program. The costs have not been assessed for retro- fitting and redesigning WNP-3, if required, to more stringent seismic design criteria. The present design of much of the structure and equipment may be adequate for a larger magnitude design seismic event, because of overdesign of existing equipment and because seismic resistance is often not the controlling design factor for equipment and structures. Redesign, replacement or rework of some structures and equipment would likely be required if the design-basis seismic event were changed significantly. The poten- tial effect of this change on completion costs is not known. WNP-1 would not be affected by a finding of a seismically-active plate interface. Continued Availability of Nuclear Components With the hiatus in U.S. orders for new nuclear plants, and the completion, suspension or abandonment of plants under construction, nuclear plant component and equipment production could dwindle to the point that completion of the projects would be affected by lack of design-specific equipment and materials. Several arguments weigh against this event. First, the bulk of equipment for WNP-1 and WNP-3 has been procured. Second, a sub- stantial inventory of plants is presently oper- ating or nearing operation, including designs similar to WNP-3. The market for spares and replacements provided by these plants will encourage the continued availability of com- ponents and materials. Third, the Naval nuclear program will ensure the continuation of a nuclear component manufacturing industry. In addition, the foreign commercial nuclear power industry will provide a continu- ing market for U.S. manufacturers, as well as a potential source of equipment for the domestic industry. Finally, it will remain possi- ble to retool and requalify for production, although components produced in limited production runs would be more expensive. Additional insurance could be provided by identifying and procuring critical equipment and material during the preservation period. The Council concludes that there is an acceptable probability that nuclear plant components and materials will remain available. Technical Continuity Loss of technical continuity would require additional effort prior to resuming construc- tion to reestablish the engineering status of the projects. Loss of technical continuity can be prevented by proper documentation, con- tinuation of engineering and licensing efforts during the preservation period, and provision of a technical “ramp-up” prior to resumption 6-27 Chapter 6 of construction. The Supply System preser- vation proposals provide for technical con- tinuity through a continuing engineering and licensing program and by a “hands-on” plant maintenance program. The Council concludes that preservation pro- grams incorporating continued licensing, engineering and maintenance will ensure technical continuity. Litigation Regarding Shared Assets The Participants Agreement for WNP-4 and WNP-5 allowed cost sharing for certain joint services and facilities on the basis of respec- tive benefit to the projects. Representatives for the WNP 4/5 bondholders argue the full costs of the shared services and facilities should be assumed by WNP-1 and WNP-3, because the WNP-4/5 interests are receiving no benefit. If successful, this suit could result in additional costs of $131 million for WNP-1 and of $269 million for WNP-3. These costs are not included in the capital costs-to-com- plete appearing in Appendices 61 and 6J. However, these costs may be “sunk” to the region, because their assignment to the region does not depend on completion of WNP-1 and 3. Operating Life It is possible that the plants, though com- pleted, might not operate as designed for their intended 40-year physical life. Events leading to this result include: 1) Disqualifica- tion, or extended shutdown of a plant design for safety-related reasons; 2) Derating, for safety or environmental reasons; or 3) Per- manent or extended outage due to major accident or equipment failure. The probability of such events is thought to be relatively low. Events 1 and 3 are accounted for by the data on plant performance used to develop the availability assumptions of Appendices 61 and 6J to the extent that they have occurred during the early part of the operating lives of large nuclear plants. The Council concludes that the assumptions regarding unscheduled outage rates (22 percent) adequately account for potential factors affecting the operating life of these plants. The cost effec- tiveness of the two plants is, however, highly sensitive to their operating availability. For this reason, the Council will continue to monitor the performance of similar plants. 6-28 Conclusion The Council has assessed the cost, sched- ule and performance characteristics of WNP-1 and WNP-3 and has concluded the planning data provided in Appendices 6-1 and 6-J are reasonable base case assump- tions regarding the characteristics of those projects. As described in Chapters 7 of Vol- ume | and Chapter 8 of Volume II, the Council has performed sensitivity analyses on cost and performance characteristics for which there is substantial uncertainty. The Council has also identified and assessed uncertain- ties possibly affecting the region's ability to preserve, complete and operate these plants. Based on this assessment, the Coun- cil has concluded that it is likely that the plants can be preserved, completed and operated even if completion were to be defer- red until late in the planning period. The principal constraints to the completion of these projects, given a need for power, are institutional. The most significant of these are uncertainty regarding continued preservation funding and litigation involving the plants that currently precludes conventional financing. The Council has concluded that these uncer- tainties are so significant that they currently preclude the projects from consideration in the resource portfolio. Assessment of the value of those projects to the region (Volume I, Chapter 7, and Volume ll, Chapter 8) indicates they could have sig- nificant present value. Because of this value, the Council recommends continued preser- vation of the plants and resolution of the bar- riers to preservation and construction. Imports The Northwest region is not an isolated sys- tem. Interconnecting transmission lines with neighboring systems allow power to be transferred between regions. Total resources available to this region include these trans- fers. Transfers can involve the sale or pur- chase of firm energy, or the sale or purchase of peaking capacity. Energy Transfers Transfers can take the form of transfers between utilities in different regions, intra- company transfers by utilities that serve both regional and extra-regional loads, and trans- fers of portions of thermal resources that are outside of the region's boundaries, but are intended to serve regional loads. Transfer agreements can include combinations of firm energy and peaking capacity transfers. Gen- erally, three types of arrangements are made: + A peaking capacity exchange in which the agreement is to return not only the bor- rowed energy, but also energy to “pay” for the cost of the exchange. This type of arrangement represents an energy import into the region. + Apeaking capacity sale in which the pay- ment for the capacity is made in dollars instead of energy. This type of arrangement represents no long-term exchange of energy. - A firm energy sale or purchase in which payment is made in dollars for long-term delivery or receipt of firm energy. This type of arrangement will affect the load/resource balance in the region. Tables 6-14 through 6-17 summarize the extra-regional transfers of firm power and peaking capacity used for both the System Analysis Model and the Decision Model. In general the region imports more firm power than it exports. This is primarily due to imported energy from Pacific Power & Light's thermal resources outside the region, which are used to meet regional loads. The sum of all the power exchanges represents a net energy import to the region of about 1,200 megawatts in 1986. This amount decreases to a net of about 200 megawatts by the year 2005. These types of exchanges should not be confused with sales of surplus power to Southwest utilities. Table 6-14 Summary of Firm Energy Exports (Average Megawatts) Chapter 6 CONTRACTS 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 MPC EXPORT 0} || || @) | a, || 10.) || Ko 1 | to ||| (or | || (6 || | Le 0 0)! | |i"z. 71 | 7 711 | 2) Wm) ie | 21 || 2 EWEB TO SCM 27 | | e7'| | :27| | 27'|| er) Kee, || | 2 | ier, | |2a,| || 27" || 27" | ier) || er! | 'on'|| | ‘ri | iz) | 27) | er lier Lor WwP To SCE 6 || | | 8) | 8 | 18) || te} | fe | | el] a | | 10 0 0: | jio o 0 ©} | 0) | |-O+) | || | | | Jo PP&L TO PG&E 29 29 290 9 sa—sstSsCi‘i 0 0 i | | 0) | (0) | 70)! | |e} | BPA TO MPC RESTORATION 6 | || || 6) | 6) || |) || 1 | | ce: ||| 6) | || 6 | 6 6 6 6 6|||| 6 @) | |e) | 6} | 6! | |e | | BPA TO MPC 68 68 68 68 65 65 6 6 6 6 6 0 0 o Oi] | ||| | 0 | ||| (|| | 0 BPA TO BC HYDRO 0} ||| 0} | |r@) | o') || te || | fo ||| (0; | | to || | io! | | 10 0 0 5 47 124 119 115 137 204 201 TCL TO WAPA #1 28 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 19 TCL TO WAPA #2 2) ||| 2) | ie) || 2s) | Wt | ee || fs | | re | ie 2 2) || 12 21 || 2 il | ||| | 2 | 2 | 2 | 2 TCL TO WAPA #3 2) ||| 2 | ie || ye || 2) | oe || fe, | || 2 | | 2 2 2) || |2 2 ||| 2 2) W2i | F211 | V2) wai | 2 BPA TO WAPA 100, || $3i| | 01 | |o:||| ;0)| | | || | || || |; || | to 0 0 oO 0 0 | | |fo|| | to] || |@!) | fie} | LONGVIEW TO WAPA 3636 86 86 GK —K——_K——_ K—6———_ K_—6—K—K_— K_—6— 8—“ 18 TOTAL 306 241 208 208 205 203 198 197 197 197 183 110 115 157 234 229 225 242 309 277 Table 6-15 Summary of Firm Energy Imports (Average Megawatts) CONTRACTS 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 SW TO BPA 193'| | /184,| | 18) | 0 ||| 01! | 0 ||| 0;|| o | | 0 0 0 0 o ||| |e) | ko} | Ho) | 10:| | 6 S. DIEGO TO Wwe 32 a2 | | (32) | 5 ||| 10 | || 0 | | 10 | || 0) | |10 0 0 0 0 o ‘0| | |.0| | || | | 0°) || (o|| | o MPC IMPORT 14 14: | 14) | 14 | |-48) | 43) | 43) | /43) | /43/ | |43 | | 43 | | 0 0 o 0 | | | 'o) | |e) | |e) | Fe) | 46 MPC TO BPA 29 29 29 (299 sss COCti PP&L(WYO)TOPP&L 879 806 743 705 660 664 609 617 567 552 489 484 426 438 392 382 323 323 276 419 SCE TO WWP 8 8. || 8) je ||| | | a | || | | 0) | |e 0 0} || |: ||| |O'|||| (0 © | | || O} | 10) | 6:) || fox |40 SCM TO EWEB 14 1414 «140«*144 1414141414141 14 1 1 14 1414188 ARIZONA TO WWP 5 ||| |i) | fe || 10" || || 1) | | 10) | || 10) | | 6 0 0} || |-O1) ||| (o|| || |e io | | | 0) | fo} | 1:0') || Fo:| | 0 BC HYDRO TO SCL 23 3636 386 86 866K KK — 86 SKK ——_6—_ K—_ K— K_—EK— 86 TOTAL 1,197 1,123 894 811 760 764 701 709 659 644 581 563 505 517 471 461 388 379 332 475 Abbreviations: (Tables 6-14 through 6-17) BC HYDRO—British Columbia Hydro Power Authority BPA—Bonneville Power Administration EWEB—Eugene Water and Electric Board LONGVIEW—Longview Fiber MPC—Montana Power Company MPC RESTORATION—Due to coordination agreement PGE—Portland General Electric PG&E—Pacific Gas and Electric PP&L—Pacific Power and Light S. DIEGO—San Diego SCE— Southern California Edison SCL—Seattle City Light SCM— Southern California Municipalities SW— Southwestern Utilities TCL— Tacoma City Light WAPA— Western Area Power Agency WWP—Washington Water Power WwYO— Wyoming Chapter 6 6-16 Summary of Peaking Capacity Exports (Megawatts) CONTRACTS 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 BPA TO SW 677 624 32 O08 0 0 0 0 0 60 0 0 0 0 60 0 0 0 0 0 WWP TOS. DIEGO 12 «112 «112 «0 0 0 0 0 0 0 0 0 0 0 0 0 60 0 0 0 PP&L to PG&E 100 100 100 100 100 100 100 100 100 10 0 0 0 0 0 0 0 0 0 0 BPA TO MPC 80 80 8 8 8 7 7 7 «7 «7 «7 +O 0 0 0 0 0 0 0 0 BPA TO MPC 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 0 0 0 0 BPA TO MPC 0 0 0 0 60 60 0 0 0 82 8 0 0 0 0 0 0 0 0 0 BPA TO BC HYDRO 0 0 0 0 60 60 0 0 60 60 60 60 0 41 242 235 226 204 330 287 TCL TO WAPA #1 15 15 15 15 15 15 1 15 15 15 1 1 15% 15 #15 15 15 15 15 15 BPA TO WAPA 100 0 0 0 60 60 0 60 0 60 0 60 0 6 0 0 60 60 0 0 LONGVIEWTOWAPA 45 45-4545 45 45 45 45 5 5 5545 5 5 5 5 5 5 TOTALS 1,229 1,076 484 340 340 337 337 337 337 362 287 160 160 201 402 395 286 264 390 302 Table 6-17 Summary of Peaking Capacity Imports (Megawatts) CONTRACTS 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 PP&L (WYO) TO PP&L 1,363 1,278 1,197 1,168 1,136 1,108 1,081 1,043 1,006 974 941 908 875 847 809 781 754 731 704 683 SCE TO PGE 100 100 100 100 100 100 0 oO 0 0 0 0 0 60 60 0 60 60 0 0 SCE TO WwP so 80 80 80 8 8 0 0 Oo 0 0 0 0 0 0 0 60 60 0 0 BCHYDROTOSCL 195 195 195 195 195 195 195 195 195 195 195 195 195 195 195 195 195 195 195 195 TOTAL 1,738 1,653 1,572 1,543 1,511 1,483 1,276 1,238 1,201 1,169 1,136 1,103 1,070 1,042 1,004 976 949 926 899 878 Out-of-Region Imports from closer interaction of regional power sys- _ further analysis should be pursued. Cooper- The previous section describes existing agreements for extra-regional power trans- fers. Potential exists for additional agree- ments between utilities within the region and utilities outside the region. An assessment of resources available to the region, therefore, should include an evaluation of the cost and availability of these potential out-of-region imports. Based on previous analysis, ‘2 it appears that substantial benefits could result tems. These potential benefits, however, may be constrained by inadequate interregional transmission capacity. In addition to this, the realization of out-of-region imports depends on complex agreements being reached with out-of-region suppliers. Because of these uncertainties, the Council has assumed, for the development of the resource portfolio, that existing contracts will not be renewed and that no new contracts will be available. Because of the potential benefits, however, ation with other regions can only occur if both regions perceive that effort to be in their best interests. The Council encourages detailed discussions with out-of-region suppliers to evaluate potential benefits, especially when it appears that out-of-region resources are more cost effective than resources devel- oped in the region. The Council plans to con- duct a West Coast energy study in order to gain better understanding of the potential benefits of broader regional resource exchange and development. 1./ Levelized life cycle costs: The present value of a resource's cost (including capital, financing and operating costs) converted into a stream of equal annual payments. Unit levelized life cycle costs (cents per kilowatt-hour) are obtained by dividing this payment by annual kilowatt-hours saved or produced. Levelized life cycle costs permit comparisons of resources having different patterns of cash flow over their lifetimes. The term “levelized life cycle cost” as gener used in this chap- ter refers to unit levelized life cycle costs. 2./ Published as: Bonneville Power Administra- tion, Evaluation and Ranking of Geothermal Resources for Electrical Generation or Elec- trical Offset in Idaho, Montana, Oregon and Washington, June 1985. 3./ The fee assessed for disposal of waste. 4./ The Transenergy (Ogden-Martin) plant has subsequently been redesigned as a stand- alone (non-cogeneration) plant. 5./ Heat engines are devices that convert ther- mal energy to mechanical energy. Examples include steam turbines, gas turbines, internal combustion engines, and Stirling engines. 6./ Additional cogenerators normally producing power for on-site use will occasionally con- tract for short-term power sales to local util- ities when avoided cost electricity prices and fuel prices are favorable. 7./ Excluding Montana Power Company. 8./ The estimates of available cogeneration were prepared using a 5.5 cent per kilowatt-hour screen. Use of the current 4.5 cent per kilo- watt-hour screen would lower the estimated availability of this resource. The PNUCC inventory upon which the estimates were based is implicitly based on industrial financ- ing of cogeneration projects. Utility financing, using the financial assumptions of this plan, would likely reduce the cost of new cogenera- tion and increase its availability. For this rea- son, the Council considers the foregoing esti- mates to be reasonable. 9./ Net-billing is a Chapter 6 ss that allows Bonneville to underwrite the costs of electric generating projects. Under net-billing, the project partici- pants purchase a percentage of a project's generating capability, for which the partici- pant pays to the Supply System a pro rata share of the costs of constructing and operat- ing the project. The participants assign their share of generating capability to Bonneville, which, in turn, reimburses the respective par- ticipants by crediting the participant's whole- sale power bill. If the participant's obligation to the Supply System exceeds the participant's wholesale power bill, the balance is paid to the participant as cash. 10./ By the terms of the settlement negotiated between Bonneville and the investor-owned utility owners of WNP-3, in response to the breach of contract suit filed by these inves- tor-owned utilities, Bonneville may, in the future, acquire the capability of the investor- owned utility share of WNP-3 in accordance with the provisions of section 6(c) of the Northwest Power Act. 11./ Subduction is the movement of one crustal Plate beneath another. 12./ “Out-of-Region Imports/Exports,” North- west Power Planning Council issue rr, March 1985. aT 6-31 Ap) Existing and Assured Regional Generating Key to Tables of Appendix 6-A Utilities/Operators Baker City of Baker Bonners Ferry City of Bonners Ferry BPA Bonneville Power Administration Centralia City of Centralia Chelan Chelan County PUD #1 Clark Clark County PUD #1 Cowlitz Cowlitz County PUD #1 CPN CP National Douglas Douglas County PUD #1 EWEB Eugene Water and Electric Board GECC General Electric Credit Corporation Grant Grant County PUD #1 Idaho Falls City of Idaho Falls IPC Idaho Power Company Lower Valley Lower Valley Power and Light Company MPC Montana Power Company Pend Oreille Pend Oreille County PUD #1 PGE Portland General Electric Company PNGC Pacific Northwest Generating Cooperative PP&L Pacific Power and Light Company PSPL Puget Sound Power and Light Company Seattle Seattle City Light Snohomish Snohomish County PUD #1 SPPC Sierra Pacific Power Company Tacoma City of Tacoma Light Division USBI U.S. Bureau of Indian Affairs USBR U.S. Bureau of Reclamation USCE U.S. Corps of Engineers USTC United States Trust Company WPPSS Washington Public Power Supply System Wwe The Washington Water Power Company Status Pe Preliminary Permit Lc Licensed EX Exempted (from Federal Energy Regulatory Commission license) POL Power on Line (Inservice) UNC Under Construction PND Pending GTD Granted ndix 6-A sources 6-A-1 Appendix 6-A Table 6-A-1 Federal Hydropower Projects NAMEPLATE PEAK AVERAGE CRITICAL orenaron tur” Same” mp’ Sse” srarus MENACE Federal Columbia River Power System: Albeni Falls USCE 43 27 27 26 POL 1955 Anderson Ranch USBR 4 c c c POL 1950 Big Cliff USCE 18 4 12 10 POL 1954 Black Canyon USBR 8 c c c POL 1986 Boise Diversion USBR 2 c c c POL 1912 Bonneville USCE 1,077 1,148 771 622 POL 1938 Chandler USBR 12 9 8 7 POL 1956 Chief Josepht USCE 2,069 2,688 1,405 1,117 POL 1955 Cougar USCE 25 6 17 13 POL 1964 Detroit USCE 100 99 47 36 POL 1953 Dexter USCE 15 6 99 84 POL 1955 Dworshak USCE 400 460 240 181 POL 1974 Foster USCE 20 6 14 12 POL 1968 Grand Coulee USBR 6,580 6,632 2,286 1,870 POL 1941 Green Peter USCE 80 78 29 22 POL 1967 Hills Creek USCE 30 30 19 15 POL 1962 Hungry Horse9 USBR 285 328 108 99 POL 1952 Ice Harbor USCE 603 693 309 207 POL 1961 John Day USCE 2,160 249 1,232 911 POL 1968 Libby USCE 525 461 219 180 POL 1975 Little Goose USCE 810 930 320 209 POL 1970 Lookout Point USCE 120 71 38 22 POL 1954 Lost Creek USCE 49 18 35 22 POL 1977 Lower Granite USCE 810 930 326 214 POL 1975 Lower Monumental USCE 810 930 321 206 POL 1969 McNary USCE 980 1,128 700 635 POL 1953 Minidoka USBR 13 c c c POL 1909 Palisades USBR 119 49 73 68 POL 1957 Roza USBR 11 3 96 4 POL 1958 The Dalles USCE 1,807 2,076 1,005 747 POL 1957 6-A-2 (table continued on next page) Appendix 6-A Table 6-A-1 (cont.) Federal Hydropower Projects NAMEPLATE PEAK AVERAGE CRITICAL CAPACITY CAPACITY ENERGY ENERGY INSERVICE OPERATOR (Mw) (mw) (MWa)> (MWa)> STATUS YEAR Other Federal Hydropower: Big Creek USBI 0 d d d POL 1916 Green Springs® USBR 16 18 7 A POL 1960 Savage Rapids Diversion USBR N/A N/A <1 <1 POL 1955 Wapato Drop 2 USBI 2 N/A 1 1 POL 1942 Wapato Drop 3 USBI 1 N/A <1 <1 POL 1932 @From PNUCC “Northwest Regional Forecast,” March 1985. »Average of estimated values for operating years 1986 through 2005 from PNUCC “Northwest Regional Forecast,” March 1985. Peak capacity is for January. cJoint peak capacity, average energy and critical period energy for Anderson Ranch Black Canyon, Big Cliff and Minidoka are 26 MW, 38 MWa, and 30 MWa, respectively. STotals for Flathead Irrigation Projects: 4 MW peak capacity; 2 MW average energy; and 2 MW critical period energy. Contracted to Pacific Power and Light Company. t Includes uprating, scheduled for completion by September 1986. includes uprating, scheduled for completion by August 1992. Table 6-A-2 Investor-owned Utility Hydropower Projects NAMEPLATE PEAK AVERAGE CRITICAL i a a i Albany PP&L 1 c c c POL 1923 American Falls IPC 92 0 42 34 POL 1978 Bend Power PP&L 1 c c c POL 1913 Big Fork PP&L 4 c c c POL 1910 Black Eagle MPC 16.8 k k k POL na Bliss IPC 75 75 47 44 POL 1949 Brownlee IPC 585 675 267 198 POL 1958 Bull Run PGE 21 22 12 10 POL 1912 C.J. Strike IPC 88 88 57 53 POL 1952 Cabinet Gorge wwe 200 227 132 112 POL 1952 Cascadei IPC 13 5 4 2 POL 1926 Cochrane MPC 48.0 k k k POL na Clear Lake IPC 3 d d d POL 1937 Clearwater 1 PP&L 15 e e e POL 1953 Clearwater 2 PP&L 26 e e e POL 1953 Cline Falls PP&L 1 c c c POL 1913 Condit PP&L 10 c c c POL 1913 Copco 1 PP&L 20 f f f POL 1918 Copco 2 PP&L 27 f f f POL 1925 (table continued on next page) 6-A-3 Appendix 6-A Eagle Point East Side Electron Fall Creek Faraday Fish Creek Flint Creek Hauser Hell's Canyon Holter Iron Gate John C. Boyle Kerr Lemolo 1 Lemolo 2 Little Falls Long Lake Lower Baker Lower Malad Lower Salmon Falls Madison Merwin Meyers Falls Milltown Monroe Street Moroney Mystic Lake Naches Naches Drop Nine Mile Nooksack North Fork Noxon Rapids Oak Grove Oxbow Pelton Post Falls Powerdale 6-A-4 UTILITY PP&L PP&L PSPL PP&L PGE PP&L MPC MPC IPC MPC PP&L PP&L MPC PP&L PP&L wwe wwe PSPL IPC IPC MPC PP&L wwpP MPC wwpP MPC MPC PP&L PP&L wwe PSPL PGE WwwpP PGE IPC PGE wwe PP&L Table 6-A-2 (cont.) Investor-owned Utility Hydropower Projects NAMEPLATE PEAK AVERAGE CRITICAL came” Swap’ Sip” suru MCE 3 h h h POL 1957 3 f f f POL 1924 26 c c c POL 1904 2 c c c POL 1903 35 43 23 17 POL 1907 11 e e e POL 1952 14 k k k POL 1901 17.0 k k k POL na 392 389 214 157 POL 1967 38.4 k k k POL n/a 18 f f f POL 1962 80 f f f POL 1958 168.0 k k k POL 1938 29 e e e POL 1955 33 e e e POL 1956 32 g g g POL 1910 70 g g g POL 1914 64 66 45 39 POL 1925 14 d d d POL 1911 68 68 35 32 POL 1910 9.0 k k k POL n/a 136 136 792 626 POL 1931 1 1 63 50 POL 1915 3.0 k k k POL 1906 D g g g POL 1890 45.0 k k k POL na 10.0 k k k POL n/a 6 c c c POL 1909 1 c c c POL 1914 12 g 9 9 POL 1908 2 c c c POL 1906 38 54 26 19 POL 1958 397 530 215 156 POL 1960 51 45 26 147 POL 1924 190 220 109 81 POL 1961 97 97 36 31 POL 1957 15 9 9 9 POL 1906 6 c c POL 1923 (table continued on next page) Appendix 6-A Table 6-A-2 (cont.) Investor-owned Utility Hydropower Projects NCAPACITY CAPACITY ANEnGY ENERGY INSERVICE UTILITY (MW)* (mw)? (MWa)> (MWa)> STATUS YEAR Prospect 1 PP&L 4 h h h POL 1912 Prospect 2 PP&L 32 h h h POL 1920 Prospect 3 PP&L 7 h h h POL 1932 Prospect 4 PP&L q h h h POL 1944 Rainbow MPC 36.5 k k k POL va River Mill PGE 19 23 13 10 POL 1911 Round Butte PGE 247 290 96 83 POL 1964 Ryan MPC 48.0 k k k POL Wa Shoshone Falls IPC 12 12 1 9 POL 1907 Slide Creek PP&L 18 e e e POL 1951 Snoqualmie Falls 1 PSPL 12 i i i POL 1898 Snoqualmie Falls 2 PSPL 29 i i i POL 1910 Soda Springs PP&L 1 e e e POL 1952 Stayton PP&L " c c c POL 1937 Swan Falls IPC 10 10 1 10 POL 1910 Swift 1 PP&L 204 189 74 55 POL 1958 T.W. Sullivan PGE 15 14 14 14 POL 1985 Thompson Falls MPC 30.0 k k POL 1915 Thousand Springs IPC 9 d d d POL 1912 Toketee PP&L 43 e e POL 1950 Twin Falls IPC 10 10 6 a POL 1935 Upper Baker PSPL 94 84 4 34 POL 1959 Upper Falls WWwP 10 9g g 9g POL 1922 Upper Malad IPC 7 d d d POL 1948 Upper Salmon A IPC 18 19 18 18 POL 1937 Upper Salmon B IPC 17 15 15 15 POL 1947 Wallowa Falls PP&L 1 c c c POL 1921 West Side PP&L 1 f f f POL 1908 White River PSPL 70 48 36 28 POL 1912 Yale PP&L 108 117 64 53 POL 1953 aFrom PNUCC “Northwest Regional Forecast,” March 1985. Average of estimated values for operating years 1986 through 2005 from PNUCC “Northwest Regional Forecast.” Peak capacity is for January. Totals for Pacific Power and Light small projects: Peak, 33; Average, 27; Critical 26. Totals for Idaho Power Company Spring projects: Peak, 30; Average, 28; Critical, 29. ®Totals for Pacific Power and Light Umpqua River projects: Peak, 175; Average, 129; Critical, 97. ‘Totals for Pacific Power and Light Klamath projects: Peak, 92; Average, 41; Critical, 42. STotals for Washington Water Power Spokane River projects: Peak, 154; Average, 112; Critical, 91. Totals for Pacific Power and Light Rogue River projects: Peak, 25; Average, 43; Critical, 35. ‘Totals for Puget Sound Power and Light small projects: Peak, 72; Average, 54; Critical, 47. ilncludes 1984 expansion. kApproximately 40% of the capability of Montana Power Company projects is available to serve regional load. In accordance with Northwest power planning convention, the output of these resources used to serve regional load is treated as import to the region. Ne Appendix 6-A Table 6-A-3 Publicly-owned Utility Hydropower Projects NAMEPLATE PEAK AVERAGE CRITICAL Prosser vrury Yap AS” seus MERE Alder Tacoma 50 34 24 18 POL 1945 Baker Baker <1 N/AV N/AV N/AV POL 1934 Boundary‘ Seattle 635 655 508 367 POL 1967 Box Canyon Pend Oreille 60 71 48 46 POL 1955 Calispel Creek Pend Oreille 1 c c c POL 1920 Carmen-Smith EWEB 80 40 27 20 POL 1963 Cedar Falls Seattle 20 d d d POL 1905 Chelan Chelan 48 52 42 38 POL 1928 City Idaho Falls 8 e e e POL 1982 Cushman 1 Tacoma 43 27 1 10 POL 1926 Cushman 2 Tacoma 81 88 25 23 POL 1930 Diablo Seattle 122 159 95 75 POL 1936 Gorge Seattle 138 175 115 94 POL 1924 Henry M. Jackson Snohomish 112 89 53 4 POL 1984 Idaho Falls Lower Idaho Falls 11 e e e POL 1904 Idaho Falls Upper Idaho Falls 8 e e e POL 1938 LaGrande Tacoma 64 65 41 33 POL 1912 Leaburg Dam EWEB 14 7 13 12 POL 1930 Mayfield Dam Tacoma 162 172 97 65 POL 1963 Mossyrock Tacoma 300 303 114 87 POL 1968 Moyie Falls 1-Upper Bonner’s Ferry <1 c c c POL 1921 Moyie Falls 2-Lower Bonner's Ferry 2 c c c POL 1941 Newhalem Creek Seattle 2 d d d POL 1921 Packwood Lake WPPSS 26 30 11 7 POL 1964 Priest Rapids Grant 789 912 580 506 POL 1959 Rock Island Chelan 620 544 330 271 POL 1933 Rocky Reach Chelan 1,212 1,266 693 560 POL 1961 Ross Seattle 360 364 88 70 POL 1952 So. Fork Tolt River Seattle 15 7 9 8 Assured 1989 Strawberry Creek Lower Valley 2 e e e POL 1951 Swift 2 Cowlitz 70 76 25 20 POL 1958 Trail Bridge EWEB 10 3 6 4 POL 1963 Walterville EWEB 8 5 8 7 POL 1911 (table continued on next page) 6-A-6 Appendix 6-A Table 6-A-3 (cont. Publicly-owned Utility Hydropower Projects NAMEPLATE PEAK AVERAGE CRITICAL CAPACITY CAPACITY ENERGY ENERGY INSERVICE PROJECT UTILITY (mw) (mw) (MWa)> (MWa)> STATUS YEAR Wanapum Grant 831 986 611 514 POL 1963 Wellso Douglas 774 820 457 386 POL 1967 Yelm Centralia 10 10 9 9 POL 1930 @From PNUCC, “Northwest Regional Forecast,” March 1985. Average of estimated values for operating years 1986 through 2005 from PNUCC “Northwest Regional Forecast,” March 1985. Peak capacity is for January. Totals for Big Creek, Calispel Creek, Moyie Falls 1 and 2 (Flathead Irrigation Projects are: Peak, 4 MW; Average, 2 MWa; Critical, 2 MWa. Totals for Cedar Falls and Newhalem Creek are: Peak, 32 MW; Average, 13 MWa; Critical 8 MWa. eTotals for City, Idaho Falls Upper, Idaho Falls Lower, and Strawberry Creek are: Peak 21 MW; Average, 21 MWa; Critical, 16 MWa. ‘Includes Units 55 and 56. 9 Includes upgrades scheduled for completion by 1989. Table 6-A-4 Contracted Resources# NAMEPLATE AVERAGE PROJECT FUEL Grn NS cAMW) eM) STATUS INvEAR Wind: Whiskey Run PP&L (R&D Contract) 1.25 0.01 POL 1981 Subtotal, Wind 1.25 0.01 Thermal: (* = cogeneration; ? = not known whether project is cogeneration) AEM Corporation (?) Coal MPC 12.0 n/av POL 1985 Afton Generating Company (*) Wood IPC 6.0 5.8 POL va Big Horn Energy (?) Coal MPC 15.0 nav Planned 1986 Biomass One (*) Wood PP&L 25.0 18.3 POL 1986 Biosolar (*) Biomas PP&L 25.0 17.5 Planned 1987 Blue Mountain Forest Products (*) Wood CPN 3.5 3.2 Planned 1986 Boeing (Auburn) (*) Gas PSPL 9.0 8.0 POL na Boise Cascade (Emmett, ID.) (*) Wood IPC 9.1 5.0 POL na Boise Cascade (Medford) Wood PP&L 8.5 0.3 POL Wa Bozeman Woodwaste (7?) Wood MPC 12.0 nav POL 1985 Cristad Enterprises (*) Wood CPN 3.0 27 POL Wa Daw Forest Products Wood PP&L 10.0 0.9 POL na Evergreen Forest Products (*) Wood IPC 5.0 5.0 POL na Gorge Energy (*) Wood PP&L 8.5 29 POL na Great Western Malting (*) Gas Clark 20.1 15.9 POL na Husky Industries (*) Biomass PP&L 5.0 3.8 Planned 1989 D. R. Johnson (CPN) (*) Biomass CPN 75 5.6 Planned 1986 D. R. Johnson (PP&L) (*) Biomass PP&L 75 5.7 Planned 1987 Kinzua (*) Wood PGE 10.0 74 POL na (table continued on next page) 6-A-7 Appendix 6-A Table 6-A-4 (cont.) Contracted Resources? NAMEPLATE AVERAGE PROJECT FUEL Gruny NS caw) ean) STATUS NSEAR Lakeview Power Company (*) Biomass PP&L 15.0 11.3 Planned 1987 Lane Plywood (*) n/av EWEB 0.8 n/av POL na Longview Fibre (*) n/av BPA 45 35.9 POL na Metro West Point (*) Sewage Methane Seattle 3.9 2.0 POL na Pacific Crown (Woodpower, Inc.) (*) Wood WwP 6.0 45 POL n/a Perkins Power (?) Coal MPC 12.0 n/av POL 1985 Potlatch (Lewiston #1) (*) n/av WwP 36.5 9.1 POL n/a Red Lodge (?) Coal MPC 10.0 n/av Planned 1986 Roseburg Lumber Wood PP&L 52.0 26.0 POL n/a St. Regis (Libby) Wood PP&L 13.3 1.8 POL n/a Vaagen Brothers Lumber (*) Wood WWP. 4.0 2.0 POL n/a Warm Springs Forest Products Wood PP&L 9.0 0.5 POL n/a Weyco (*) Pulping Liquor EWEB 51.2 14.0 POL n/a Weyerhauser (Everett) (*) n/av Snohomish (Negotiating) 12.5 10.0 POL n/a Ogden-Martin MSW PGE 13.1 _74 UNC 1986 Subtotal, Thermal 497.0 227.3 NAMEPLATE AVERAGE PROJECT PERMITNO. = OMIT caw) fw) = staTuS. YEAR Hydropower: Bend Diversion 3473 PP&L 2.8 1.4 LC-GTD 1986 Big Sheep Creek 5118 WwP 1.0 0.6 EX-UNC 1985 Cedar Draw Creek 8278 IPC 1.4 0.6 LC-POL 1984 Elk Creek 3503 IPC 2.0 1.9 EX-GTD 1986 Elk Creek Falls 6524 WwP 46 1.5 LC-PND 1986 Eltopia Branch Canal Mi. 4.6 3842 Seattle & Tacoma 2.2 1.0 LC-POL 1983 Falls Creek 6661 PP&L 44 17 EX-POL 1984 Farmers Irr. Dist. Project 2 7532 PP&L 2.5 1.0 EX-UNC 1986 Farmers Irr. Dist. Project 3 6801 PP&L 1.7 0.7 EX-UNC 1987 Galesville 7161 PP&L 1.8 0.7 LC-UNC 1986 Jim Boyd 7269 PP&L 11 0.5 LC-GTD 1986 Jim Ford Creek 7986 wwpP 1.5 n/av Lc 1987 Kasel-Witherspoon 6410 IPC 1.0 1.2 EX-POL 1983 Koyle Ranch 4052 IPC 1.3 0.8 EX-POL 1983 L.Q. and L.S. Snake Drains 5767 IPC 2.0 1.3 EX-POL 1984 Lacomb Irrigation 6648 PP&L 1.0 0.6 EX-UNC 1986 Lateral #10 6250 IPC 1.6 1.2 EX-POL 1985 (table continued on next page) 6-A-8 Appendix 6-A Table 6-A-4 (cont.) Contracted Resources NAMEPLATE AVERAGE PROJECT PERMIT NO. Gruny NS caw) eva). STATUS INvEAR Little Wood River 7427 IPC 2.4 0.5 EX-POL 1985 Lowline Canal Drop 3216 IPC 8.0 n/av EX-POL 1985 Lucky Peak 2832 Seattle 87.0 32.2 LC-UNC 1988 Main Canal Headworks 2849 Seattle & Tacoma 26.0 9.8 LC-UNC 1986 Middle Fork Irrigation District 4458 PP&L 3.3 2.5 EX-GTD 1986 Mitchell Butte 5357 IPC 1.5 0.6 LC-PND 1987 N-32 Hydro (Marco Ranch) 7170 IPC 1.9 nav POL 1985 Opal Springs 5891 PP&L 5.0 3.7 LC-POL 1984 Owyhee Dam 4354 IPC 3.7 1.4 LC-UNC 1985 Owyhee Tunnel No. 1 4359 IPC 5.0 2.7 PP-GTD 1991 Pelton Reregulating 2030B PP&L 19.6 9.3 LC-POL 1982 Portland Hydro 2821 PGE 35.6 12.6 LC-POL 1982 Portland Wellfield 7052 PGE 45 23 EX-UNC 1985 Potholes East Canal Headworks 2840 Seattle 7.5 2.8 LC-GTD 1984 Potholes East Canal Mile 66 3843 Seattle & Tacoma 2.4 1.3 LC-POL 1983 Rock Creek 6450 IPC 21 1.3 EX-POL 1983 Rocky Brook 3783 Seattle 1.5 n/av EX-UNC 1986 Russell D. Smith 2926 Seattle & Tacoma 6.1 2.8 LC-POL 1981 Sandy Creek (Koma Kulshan) 3239 PSPL 17.6 9.2 LC-PND 1989 Shellrock Creek (L.M. Baker) niav PGE 21.8 13.4 PP-GTD 1986 South Dry Creek 8831 MPC 1.8 c EX-POL 1985 Summer Falls 3295 Seattle & Tacoma 90.0 37.0 LC-POL 1984 Twin Falls 4885 PSPL 20.0 8.8 LC-GTD 1989 Valsetz 7217 PP&L 3.9 1.9 EX-GTD 1987 Week Falls 7563 PSPL 3.4 1.6 LC-GTD 1988 Winchester 6775 PP&L _12 _ 06 LC-POL 1984 Subtotal, Hydropower 416.3 174.7 Total, Contracted Resources 915 402 Exclusive of projects of less than 1 MW capacity. >From various sources compiled by the Council, including PNUCC Thermal Resources Data Base, October 1984; PNUCC Northwest Regional Forecast, March 1985; Pacific Northwest Hydropower Data Base; Idaho Public Utility Commission, Oregon Public Utility Commissioner, Montana Power Company, Washington State Energy Office. cApproximately 40% of the capability of Montana Power Company resources is available to serve regional load. In accordance with Northwest power planning convention, the output of these resources used to serve regional load is treated as import to the region. 6-A-9 Appendix 6-A Table 6-A-5 Large Thermal Units NAMEPLATE PEAK AVERAGE PROJECT & CAPACITY CAPACITY ENERGY INSERVICE UNIT FUEL UTILITY (MW)? (Mw)? (MWa)> STATUS YEAR Boardman Coal PGE-65%; IPC-10%; PNGC-10%; GECC-15% 560 530 3579 POL 1980 Centralia 1 Coal PP&L-47.5%; WWP-15%; PSPL-11%; Snohomish-8%; 665 640 448 POL 1971 Tacoma-8%; Seattle-8%; PGE-2.5% Centralia 2 Coal PP&L-47.5%; WWP-15%; PSPL-11%; Snohomish-8%; 665 640 448 POL 1972 Tacoma-8%; Seattle-8%; PGE-2.5% Colstrip 1 Coal MPC-50%; PSPL-50% 358 165¢ 110¢ POL 1975 Colstrip 2 Coal MPC-50%; PSPL-50% 358 165¢ 110¢ POL 1976 Colstrip 3 Coal MPC-30%; PSPL-25%; PGE-20%; WWP-15%; 778 490¢ 368° POL 1984 PP&L-10% Colstrip 4 Coal USTC-30%; PSPL-25%; PGE-20%; WWP-15%; 778 4goch 368M = LC-UNC. 1986 PP&L-10% J.E. Corette Coal MPC 172 ¢ e POL 1968 Jim Bridger 1 Coal PP&L-66%4%; IPC-331%% 509 1679 1134 POL 1974 Jim Bridger 2 Coal PP&L-66%%; IPC-331/4% 509 1674 113¢ POL 1975 Jim Bridger 3 Coal PP&L-66%%; IPC-3314% 509 1679 1134 POL 1976 Jim Bridger 4 Coal PP&L-66%%; IPC-3314% 509 1674 113¢ POL 1979 Valmy 1 Coal IPC-50%; SPPC-50% 254 127 89 POL 1981 Valmy 2 Coal IPC-50%; SPPC-50% 250 138 96 POL 1985 Hanford! Nuclear WPPSS 800 oe 380 POL 1966 Trojan Nuclear PGE-67.5%; EWEB-30%; PP&L 2.5% 1,216 1,080 648 POL 1976 WNP-2 Nuclear WPPSS 1,154 1,100 656 POL 1984 Kettle Falls Wood WWP. 51 42.2 31.6 POL 1983 aFrom PNUCC Thermal Resources Data Base, October 1984. >Declared (by sponsors) to be available to the region (from 1985 PNUCC Northwest Regional Forecast, March 1985). cApproximately 40% of the capability of Montana Power Company resources is available to meet regional load. In accordance with Northwest power planning convention, the output of these resources used to serve regional load is treated as import to the region. ‘The portion of the Pacific Power and Light Company share of Jim Bridger is treated as an import to the region in accordance with Northwest power planning convention. Operation of the N-reactor for plutonium production has priority over production of steam for electricty. Therefore, the firm capacity of Hanford Generating Project is zero. ‘The Hanford Generating Project operating contract extends through June 1993 and can be terminated on a one-year notice. The resource is considered to be available until June 1993. 9General Electric Credit Corporation share to be sold to San Diego Gas and Electric on a 25-year contract beginning in 1989. hUnited States Trust Company share of Colstrip 4 is leased back to Montana Power Company. 6-A-10 Appendix 6-A PRIMARY FUEL AND UNIT Combustion Turbine Bethel 1 Gas Bethel 2 Gas Frederickson 1 Gas Frederickson 2 Gas Fredonia 1 Gas Fredonia 2 Gas Libby Oil Northeast Gas Othello Oil Point Whitehorn 1 Oil Point Whitehorn 2 Gas Point Whitehorn 3 Gas Whidbey Island Oil Wood River Gas Diesel Bonners Ferry 1 Oll Bonners Ferry 2 Oil Bonners Ferry 3 Oil Crystal Mountain Oil Summit 1 Oil Summit 2 Oil Steam-Electric Lake Union 1 Oil Lake Union 2 Oil Lake Union 3 Oil Shuffleton 1 Oil Shuffleton 2 Oil Combined Cycle Beaver Gas TOTALS: UTILITY PGE PGE PSPL PSPL PSPL PSPL PP&L (leased) wwe wwe PSPL PSPL PSPL PSPL IPC Bonners Ferry Bonners Ferry Bonners Ferry PSPL PGE PGE Seattle Seattle Seattle PSPL PSPL PGE Table 6-A-6 Reserve Units “CAPACITY «CAPACITY. «—sENERGY ENERGY INSERVICE (Mw)? (Mw)? (MWa)= (MWa)? STATUS YEAR 56.7 75.0 9.5 12.9¢ POL 1973 56.7 75.0 9.5 12.9¢ POL 1973 85.0 89.0 2.1 14.2! POL 1981 85.0 89.0 2.1 14.2! POL 1981 129.0 124.0 29 17.8! POL 1984 129.0 124.0 29 17.8! POL 1984 24.0 24.0 0.0 16.2 POL 1972 61.2 68.0 2.0 57.8 POL 1978 28.2 33.0 1.0 28.1 POL 1973 61.0 68.0 1.0 §2.7 POL 1974 85.0 89.0 1.0 14.2° POL 1981 85.0 89.0 1.0 14.21 POL 1981 27.0 29.0 1.0 23.0 POL 1972 50.0 50.0 1.0 42.5 POL 1974 2.4 2.4 0.0 = POL 1930 2.4 2.4 1.0 _ POL 1930 2.4 2.4 1.0 _ POL 1973 28 27 0.1 _ POL 1969 2.8 3.0 0.5 _ POL 1970 2.8 3.0 0.5 a POL 1973 36.0 n/av 0.0 n/av POL 1921 36.0 n/av 0.0 n/av POL 1921 36.0 n/av 0.0 n/av POL 1921 35.0 44.0 1.0 26.99 POL 1930 35.0 44.0 1.0 26.99 POL 1930 545 601 53 443 POL 1977 1,129.9 419 845.1. MW MW MW aFrom PNUCC Thermal Resource Data Base, October 1984. >From PNUCC Northwest Regional Forecast, March 1985. Declared by sponsor to be available as firm energy. From PNUCC “Northwest Regional Forecast,” March 1985. ‘Based on base load capacity (may be less than peak capacity) and the Council's assumptions regarding availability except as noted. The Council's availablity assumptions are as follows: combustion turbines-85%; diesel generators-87%; combined cycle plants-83%. Restricted to 2,000 hours of operation during any year operated. ‘Constrained to a maximum of 1,500 hours per year operation by the Powerplant and Industrial Fuel Use Act of 1978. 9From PNUCC Northwest Regional Forecast. 6-A-11 APPENDIX 6-B PLANNING ASSUMPTIONS — GENERIC CONVENTIONAL COAL PROJECT GENERAL CHARACTERISTICS TWO 603-MEGAWATT UNITS Two units, 650 megawatts (gross), 603 megawatts (net) nominal capacity each; pulverized coal fired; 2,400 psig, 1000°F/1000°F reheat, Site Eastern Oregon Plant design 3.5 HgA backpressure. Fuel type Fuel transport Unit train Heat rejection Emission control SO2—wet scrubbers NOX —combustion control Wyoming subbituminous, 8,445 Btu/Ib. Mechanical draft cooling towers; makeup from the Columbia River Fly ash— electrostatic precipitator Transmission Ten-mile, 500 kV double circuit interconnection. TECHNICAL PERFORMANCE (MW, net) Heat Rate Operating State per Gn Project (Btu/kWh, net) Peak 633 1,266 10,210 Maximum sustainable 633 1,266 10,210 Rated (Least cost) 603 1,206 10,080 Minimum sustainable 151 302 11,940 Transition Times Cold start— Minimum sustainable 12 hours Hot start—Minimum sustainable 1 hour Minimum sustainable— Rated 1 hour Operating Availability Equivalent Annual Availability 75% ‘Annual Maintenance Outage Period Normal 30 days Major Overhaul (every fifth year) 60 days Average 36 days Other Planned and Unscheduled Outages (Equivalent) 17% PROJECT DEVELOPMENT ee aa Phase |: ail Phase it: Period 48 mos 9 mos Cash Flow Year 1 $ 11.8 million $ 30.3 million Year 2 11.8 million Year 3 11.8 million Year 4 11.8 million Total $ 47.3 million ($39/kW) $ 30.3 million ($25/kW) Hold cost (excl. of return on investment) Expected shelf lite Option close-out cost Construction period (Unit 1) Lag to Unit 2 completion Cash Flow Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Construction Total $ 0.3 million/yr ($0.3/KW/yr) 5 years negligible Construction trom Option Phase | 72 mos 12 mos $ 58.6 million $169.6 million $250.9 million $415.7 million $400.5 million $155.1 million $ 15.6 million $1,466.2 million ($1,216/kW) $ 0.3 million/yr ($0.3/kWiyr) 2 years negligible Option Phase I 60 mos 12 mos $174.5 million $258.6 million $428.0 million $412.0 million $161.1 million $ 16.8 million $ 0.0million $1,451.0 million ($1,203/kW) OPERATION Fuel Inventory Fuel price (delivered) Fixed O&M Variable O&M Capital Replacement Cost ‘Amortization Life Operating Lite General Inflation (nominal) Capital Escalation (real) O&M Escalation (real) Delivered Fuel Escalation (real) 90 days @ 1,206 MW $ 2.00/MMBtu $11.1 million/yr ($9.20/kW/yr) $ 0.11 cents/kWh $14.5 million/yr ($12.00/KW/yr) 30 years 40 years 5.0 %/yr 0.4 %/yr 0.0 %/yr 1.0 %/yr 6-B-1 Appendix 6-B Site: As assumed in Kaiser Engineers Power Corporation, Bonneville Power Administration Comparative Electric Gener- ation Study, January 1983 (KEPC, 1983). Plant design: Per KEPC (1983). Fuel type: Per KEPC (1983). Fuel transportation: Per KEPC (1983). Heat rejection: Per KEPC (1983). Emission control: Per KEPC (1983). Transmission: Assumes a site near the existing Boardman project, consistent with KEPC (1983). Peak capacity: Capacity and heat rate per performance curves supplied by Bonneville, April 10, 1985. Maximum sustainable capacity: Capacity and heat rate per performance curves sup- plied by Bonneville, April 10, 1985. Rated capacity: Capacity and heat rate per performance curves supplied by Bonneville, April 10, 1985. Minimum sustainable capacity: Capacity based on recommendation of the Coal Options Task Force. Heat rate per perform- ance curves supplied by Bonneville, April 10, 1985. Transition times: Cold start and hot start transition times are based on times esti- mated for Creston in Pacific Northwest Utilities Conference Committee. Thermal Resources Data Base. October 1984. (PNUCC, 1984a). Transition time from mini- mum sustainable to maximum sustainable is estimated. 6-B-2 Equivalent annual availability: Based on the North American Electric Reliability Coun- cil (NERC) Generating Availability Data Sys- tem records for subcritical coal-fired plants, scrubbed, of 400 to 600 + megawatts capac- ity. In analyzing the NERC data, it was assumed that a typical well-maintained plant operating in a baseload mode would receive routine annual maintenance plus a major overhaul every five years of operation. The first ten years of plant life, encompassing two such five-year cycles, should therefore be representative of the operating availability of the plant during the portion of its life during which it is expected to operate in baseload mode. Equivalent annual availability was cal- culated by weighting the NERC equivalent annual availability data for operating years 1 through 10 in proportion to the number of Operating years represented by the compiled statistics (i.e., statistics for operating years 1, 2 and 3 (compiled individually) each received a weight of “1;” statistics for years 4 through 10 (compiled in aggregate) received a weight of “7”). The resulting equivalent annual avail- ability was 78 percent. The 78 percent was rounded down to 75 percent—first because of the limited operating history (6 unit-years) available for large (600 + megawatts) scrub- bed units, and second, the expectation that a major plant rebuild or retrofit occurring at least once during the 40-year baseload life of the plant would reduce the lifetime availability below that observed during the first ten years of operation. Annual maintenance outage period: The outage periods shown are assumed values, slightly more conservative than Pacific North- west experience, as reflected in PNUCC (1984a). Other planned and unscheduled out- ages: The rate of other planned and unscheduled outages is calculated from the equivalent unscheduled outage rate (based on NERC data, as described above) and the assumed average annual maintenance out- age period. Option development periods: Phase | includes conceptual development, comple- tion of permitting and licensing, and site acquisition. The Phase | development period is based on the recommendation of Robert E. Henriques, The Washington Water Power Company, letter to Jeff King, Northwest Power Planning Council, dated August 15, 1984. Phase II includes engineering to major component order. The Phase II remobiliza- tion period is based upon estimates prepared for the Creston project by The Washington Water Power Company, provided in a letter from R. E. Henriques, The Washington Water Power Company, to J. C. King, Northwest Power Planning Council, dated February 27, 1985 (Henriques, 1985). The Phase II remobilization would occur only with a hold between Phases | and Il. The Phase II devel- opment period is based on estimates pre- pared for the Creston project by the Wash- ington Water Power Company, and provided in a letter from R. E. Henriques, The Wash- ington Water Power Company, to J. C. King, Northwest Power Planning Council, dated August 15, 1984 (Henriques, 1984). Option development cost: Phase | devel- opment costs are the sum of licensing costs plus land acquisition costs. Licensing costs are taken as 3 percent of total overnight cap- ital costs, per Pacific Northwest Utilities Con- ference Committee “Working Paper, Devel- opment of Generic Resource Data,” October 1984 (PNUCC, 1984b). Land costs are from KEPC (1983) (1984 update). Phase II remobilization costs are those estimated for Creston by R. E. Henriques, The Washington Water Power Company; telephone conver- sation on April 30, 1985. Phase || develop- ment costs are taken as 40 percent of engineering costs as estimated in KEPC (1983) (1984 update), escalated to 1/85 using the Handy Whitman Index. Appendix 6-B Option hold cost: Based upon estimates prepared for Creston (Henriques, 1985). Expected shelf life: Phase |: Expected shelf life is estimated time until one or more of the advanced coal technologies becomes fully established, requiring new feasibility and BACT studies. Phase II: Estimated time until the bulk of Phase II engineering would have to be reworked. Option close-out cost: Close-out costs are assumed to be covered by the sale of the site. Construction schedule: Based on discus- sions of the Coal Options Task Force. Con- sistent with Creston estimates. Lag time to next unit: As estimated in KEPC (1983). Construction cash flow: From Phase |: Based on KEPC (1983), (1984 cost update) as adjusted in PNUCC (1984b); further adjusted as follows: Land costs deleted (included in option development costs); coal inventory based on 90 days continuous oper- ation at rated capacity (603 megawatts per unit at 10,080 Btu/kilowatt hour); cost of a ten- mile section of 500kV double circuit trans- mission line added (based on costs cited in letter from W. D. Beebe, Bonneville Power Administration, to J. C. King, Northwest Power Planning Council, dated April 26, 1985); costs escalated to January 1985 using appropriate Handy-Whitman cost indices. Payout schedule based on KEPC (1983) (licensing activities deleted). From Phase Il: Costs derived as from Phase I, less 20 per- cent of engineering costs, per discussion of the Coal Options Task Force. Fuel inventory: Inventory used in KEPC (1983); continuous operation at rated capac- ity assumed. Fuel price: As recommended by the Coal Options Task Force. Midpoint of a range of coal costs obtained from technical reports and observation of current coal prices. Fixed O&M cost: Seventy percent of labor and maintenance materials from KEPC (1983), (1984 cost update), escalated to 1985 using GNP deflator (1.05). Variable O&M cost: Raw materials and chemicals, utilities, 30% of labor and mainte- nance materials and sludge and ash dis- posal from KEPC (1983), (1984 cost update), escalated to 1985 using GNP deflator (1.05). Capital replacement cost: Based upon estimated routine annual replacements plus the levelized cost of a major refurbishment at year 20. The cost of routine replacements is estimated to be $5.00/kW/yr based upon his- torical replacement costs for Centralia and Bridger. Major refurbishment costs are esti- mated to be $300/KW, based upon recent estimates prepared for the U.S. Congress, Office of Technology Assessment (U.S. Con- gress, Office of Technology Assessment, 1985, New Electric Power Technologies). This cost is levelized over the operating life using a 3% discount rate. Amortization life: Based on Electric Power Research Institute Technical Assessment Guide, May 1982 (EPRI, 1982) recom- rendatior 1S. Operating life: Design life of major compo- nents (boiler and turbogenerator). Assumes a major refurbishment about year 20. Refur- bishment costs are included in interim capital replacement costs. Cost escalation: Values adopted by the Council for the 1986 Energy Plan. 6-B-3 APPENDIX 6-C PLANNING ASSUMPTIONS — GENERIC CONVENTIONAL COAL PROJECT GENERAL CHARACTERISTICS PROJECT DEVELOPMENT Option Phase |: Option Phase I Site Eastern Oregon License, Site Acquisition __nit. Detailed Engineering Plant design ‘Two units, 270 MW (gross), 500 MW (net) nominal capacity each; Period 48 mos 9 mos pulverized coal fired; 2,400 psig, 10000F/10000F reheat, 3.5 HgA Cash Flow Year 1 $ 6.9 million $ 17.9 million backpressure. Year 2 6.9 million Fuel type Wyoming subbituminous, 8,445 Btu/Ib. Year 3 6.9 million Fuel transport Unit train Year 4 6.9 million Heat rejection Mechanical draft cooling towers; makeup from the Columbia River Total $ 27.6 million ($55/kW) $ 17.9 million Emission control Fly ash—electrostatic precipitator ($36/kW) SO2—Wet scrubbers NOX—combustion contro! Hold cost (excl. of return on investment) $ 0.3 million/yr ($0.6/KW/yr) $ 0.3 million/yr) ($0.6 kW/\ Transmission Ten-mile, 500 kV single circuit interconnection. : ” Expected shelf life 5 years 2 years Option close-out cost negligible negligible Construction from Construction from: TECHNICAL PERFORMANCE ' . 7 Operating State Por Gay HM ct (eunwn net) Construction period (Unit 1) 60 mos 48 mos Peak 262 524 10,320 Lag to Unit 2 completion 12 mos 12 mos Maximum sustainable 262 524 10,320 Cash Flow Year 1 $ 17.2 million $ 42.9 milion Rated (Least cost) 250 500 10,190 Year 2 $ 42.9 million $291.7 million Minimum sustainable 62.5 625 11,670 Year 3 $292.0 million $360.7 million Transition Times Year 4 $360.7 million $120.2 million Cold start—Minimum sustainable 12 hours Year 5 $120.2 million $ 26.1 million Hot start—Minimum sustainable 1 hour Minimum sustainable—Rated <1 hour Year 6 $ 25.8 million Operating Availability Construction Total $858.7 million $841.6 million Equivalent Annual Availability 77% ($1,717/kW) ($1,683/kW) Annual Maintenance Outage Period Normal 30 days Major Overhaul (every fifth year) 60 days OPERATION Average 36 days Fuel Inventory 90 days Other Planned and Unscheduled Outages (Equivalent) 15% Fuel Cost (delivered) $2.00/MMBtu Fixed O&M $9.1 million/yr ($18.25/kWiyr) Variable O&M 0.2 cents/kWh Capital Replacement Cost $6.0 million/yr ($12.00/kW/yr) Amortization Life 30 years Operating Life 40 years General Inflation (nominal) 5.0%lyr Capital Escalation (real) 0.4%Jyr O&M Escalation (real) 0.0%/yr Delivered Fuel Escalation (real) 1.0%lyr 6-C-1 Appendix 6-C Site: As assumed in Kaiser Engineers Power Corporation, Bonneville Power Administration Comparative Electric Gener- ation Study, January 1983 (KEPC, 1983). Plant design: Per KEPC (1983). Fuel type: Per KEPC (1983). Fuel transportation: Per KEPC (1983). Heat rejection: Per KEPC (1983). Emission control: Per KEPC (1983). Transmission: Assumes a site near the existing Boardman project, consistent with KEPC (1983). Peak capacity: Capacity and heat rate per performance curves supplied by Bonneville, May 1, 1985. Maximum sustainable capacity: Capacity and heat rate per performance curves sup- plied by Bonneville, May 1, 1985. Rated capacity: Capacity and heat rate per performance curves supplied by Bonneville, May 1, 1985. Minimum sustainable capacity: Capacity based on recommendation of the Coal Options Task Force. Heat rate per perform- ance curves supplied by Bonneville, May 1, 1985. Transition times: Cold start and hot start transition times are based on times reported for Valmy | in Pacific Northwest Utilities Con- ference Committee. Thermal Resources Data Base. October 1984. (PNUCC, 1984a). Transition time from minimum sustainable to maximum sustainable is estimated. 6-C-2 Equivalent annual availability: Based on the North American Electric Reliability Coun- cil (NERC) Generating Availability Data Sys- tem records for subcritical coal-fired plants, scrubbed, of 200 to 399 MW capacity. In analysing the NERC data, it was assumed that a typical well-maintained plant operating in a baseload mode would receive routine annual maintenance plus a major overhaul every five years of operation. The first ten years of plant life, encompassing two such five-year cycles, should therefore be repre- sentative of the operating availability of the plant during the portion of its life during which it is expected to operate in baseload mode. Equivalent annual availability was calculated by weighting the NERC equivalent annual availability data for operating years 1 through 10 in proportion to the number of operating years represented by the compiled statistics (i.e., statistics for operating years 1, 2 and 3 (compiled individually) each received a weight of “1;” statistics for years 4 through 10 (compiled in aggregate) received a weight of “7"). The resulting equivalent annual avail- ability was 79 percent for the type of plant described above. An equivalent availability of 77 percent was selected in consideration of the following: 1) Smaller units consistently show better availability than larger units; this relationship is preserved between the 77 per- cent recommended for smaller units and the 75 percent recommended for larger units. 2) Substantial operating experience (48 unit- years) is available for smaller units, lending credibility to the high reliability indicated for these size units by the FERC data. 3) The lifetime equivalent availability is likely to be somewhat lower than that documented for the first ten years of experience because of the possibility of major rebuilds later in plant life. Annual maintenance outage period: The outage periods shown are assumed values, slightly more conservative than Pacific North- west experience, as reflected in PNUCC (1984a). Other planned and unscheduled out- ages: The rate of other planned and unscheduled outages is calculated from the equivalent unscheduled outage rate (based on NERC data, as described above) and the assumed average annual maintenance out- age period. Option development periods: Phase | includes conceptual development, comple- tion of permitting and licensing, and site acquisition. The Phase | development period is based on the recommendation of Robert E. Henriques, The Washington Water Power Company, letter to Jeff King, Northwest Power Planning Council, dated August 15, 1984 (Henriques, 1984). Phase II includes engineering to major component order. The Phase II remobilization period is based upon estimates prepared for the Creston project by The Washington Water Power Company, provided in a letter from R. E. Henriques, The Washington Water Power Company, to J. C. King, Northwest Power Planning Council, dated February 27, 1985 (Henriques, 1985). The Phase II remobilization would occur only with a hold between Phases | and II. The Phase II development period is based on estimates prepared for the Creston project by the Washington Water Power Company, and provided in Henriques (1984). Option development cost: Phase | devel- opment costs are the sum of licensing costs plus land acquisition costs. Licensing costs are taken as 3 percent of total overnight cap- ital costs, per Pacific Northwest Utilities Con- ference Committee “Working Paper, Devel- opment of Generic Resource Data,” October 1984 (PNUCC, 1984b). Land costs are from KEPC (1983) (1984 update). Phase II remobilization costs are those estimated for Creston by R. E. Henriques, The Washington Water Power Company; telephone conver- sation on April 30, 1985. Phase II develop- ment costs are taken as 40 percent of engineering costs as estimated in KEPC (1983) (1984 update), escalated to 1/85 using the Handy Whitman Index. Option hold cost: Based on estimates pre- pared for Creston (Henriques, 1985). Expected shelf life: Phase |: Expected shelf life is estimated time until one or more of the advanced coal technologies becomes fully established, requiring new feasibility and BACT studies. Phase II: Estimated time until the bulk of Phase II engineering would have to be reworked. Option close-out cost: Close-out costs are assumed to be covered by the sale of the site. Construction schedule: Per KEPC (1983). Lag time to next unit: Per KEPC (1983). Construction cash flow: From Phase |: Based on KEPC (1983), (1984 cost update) as adjusted in PNUCC (1984b); further adjusted as follows: Land costs deleted (included in option development costs); coal inventory based on 90 days continuous oper- ation at rated capacity); cost of a ten-mile section of 500kV single circuit transmission line added (based on costs cited in letter from W. D. Beebe, Bonneville Power Administra- tion to J. C. King, Northwest Power Planning Council, dated April 26, 1985); costs esca- lated to January 1985 using appropriate Handy-Whitman cost indices. Payout sched- ule based on KEPC (1983) (licensing activities deleted). From Phase Il: Costs derived as from Phase |, less 20 percent of engineering costs, per recommendation of the Coal Options Task Force. Fuel inventory: Inventory used in KEPC (1983); continuous operation at rated capac- ity assumed. Fuel price: As recommended by the Coal Options Task Force. Midpoint of a range of coal costs obtained from technical reports and observation of current coal prices. Fixed O&M cost: Seventy percent of labor and maintenance materials from KEPC (1983), (1984 cost update), escalated to 1985 using GNP deflator (1.05). Variable O&M cost: Raw materials and chemicals, utilities, 30% of labor and mainte- nance mateials and services, and sludge and ash disposal from KEPC (1983), (1984 cost update), escalated to 1985 using GNP deflator (1.05). Appendix 6-C Capital replacement cost: Same as unit cost estimated for the 603 MW generic plant. Includes cost of a major refurbishment at year 20. Amortization life: Based on Electric Power Research Institute Technical Assessment Guide, May 1982 (EPRI, 1982) recom- mendations. Operating life: Design life of major compo- nents (boiler and turbogenerator). Assumes a major refurbishment at year 20. Refurbish- ment costs are included in interim capital replacement costs. Cost escalation: Values adopted by the Council for the 1986 Power Plan. 6-C-3 pendix 6-D Ap Planning Assumptions — Generic AFBC Coal Project Single 110-Megawatt Unit (January 1985 dollars) GENERAL CHARACTERISTICS PROJECT DEVELOPMENT Phase! Phase Il Site Eastern Oregon Option noption Plant design _ Single unit, 110 megawatt (net) capacity; coal-fired, atmospheric fluidized bed, Period ‘a6 ines ~~ _ = steam-electric power plant; 1,500 psig, 1000°F steam, 2.0 HgA turbine back- pressure; zero makeup. Cash Flow Year 1 $ 1.1 million nav Fuel type Wyoming subbituminous, 8,445 Btu/Ib, Year 2 1.1 million Fuel transport Unit train Year 3 1.1 million Heat rejection Mechanical draft cooling towers; makeup from the Columbia River Year 4 1.1 million Emission control Particulates — cyclone separators and baghouse. Total $ 4.4 million ($40/KW) nav SOX — Crushed limestone injection. Hold cost (excl. of return on investment) $0.1 million’yr ($0.9/KWiyr) av NOX — Combustion temperature control aides 7 lite 5 years nav Transmission _Located on existing regional grid. = Option close-out cost negligible nav Construction From Construction From TECHNICAL PERFORMANCE Option Phase! Option Phase I Gapacty Veet Rate Construction period 72 mos nav Sate f ? aa Cash Flow Year 1 $ 0.2milion nav Peak nav nav Year 2 15.6 million Maximum Sustainable nav nav Year 3 35.5 million Rated (Least Cost) 110 11,200 Year 4 92.7 million Minimum Sustainable 39 nav Year 5 43.4 million Transition Times Year 6 9.8 million Cold start — Minimum Sustainable nav ee Total $197.2 million ($1,793/kW) nav Hot start — Minimum Sustainable nav Minimum sustainable — Rated niav Operating Availability OPERATION Equivalent Annual Availability 75% Fuel Inventory 90 days @ 110 MW Annual Maintenance Outage Period Fuel Cost (delivered) $2.00 MMBtu Normal 35 days Fixed O&M $3.6 milli ($33/kW/ ix millior’yr r) Major Overhaul nav monet mn Variable O&M 0.1 cents/kWh — _— Capital Replacement Cost $1.3 million/yr ($12.00/kW, al ment 1.3 mili 12.00/KW/ Other Planned and Unscheduled Outages (Equivalent) 17% he Amortization Life 30 years Operating Life 40 years General Inflation (nominal) 5.0%lyr Capital Escalation (real) 0.4%/yr O&M Escalation (real) 0.0%lyr Delivered Fuel Escalation 1.0%lyr 6-D-1 Appendix 6-D Site: As assumed in Kaiser Engineers Power Corporation. 1985. Bonneville Power Admin- istration Comparative Electric Generation Study (Supplemental Studies). KEPC (1985). Plant design: Per KEPC (1985). Fuel type: Per KEPC (1985). Fuel transport: Per KEPC (1985). Heat rejection: Per KEPC (1985). Emission control: Per KEPC (1985). Transmission: Assumes a site near the existing Boardman site, consistent with KEPC (1985). Peak Capacity: Not available. Maximum sustainable capacity: Not available. Rated capacity: Capacity and heat rate per KEPC (1985). Minimum sustainable capacity: Based on the value reported for an AFBC plant in Elec- tric Power Research Institute Technical Assessment Guide, May 1982 (EPRI, 1982). Transition times: Not available. Equivalent Annual Availability: Computed from annual maintenance outage period and other planned and unscheduled outages. Annual maintenance outage period: From estimates reported in EPRI, 1982. 6-D-2 Other planned and unscheduled out- ages: From estimates reported in EPRI, 1982, rounded to nearest percent. Option development period: Assumed to be similar to other generic coal units. Option development cost: Assumes devel- opment of option to Node 2 (completion of permitting and licensing, acquisition of options on site). Cost to achieve Node 2 is based upon 1 percent to achieve Node 1 (conceptualization, feasibility study and environmental baseline data) and 1 percent for completion of permitting and licensing, as estimated in Battelle (1982) for coal steam- electrical power plants. To the above is added estimated land cost from KEPC (1985). (Land is assumed not have escalated in value between 1984 and 1985). Option hold cost: Estimate includes project management ($100,000 per year); environ- mental monitoring ($15,000 per year), rounded to nearest 0.1 million. Expected shelf life: The estimated time until significant advances might occur in a new technology such as AFBC, requiring new feasibility and BACT studies. Option close-out cost: Sale of site assumed to offset close-out costs. Construction schedule: Per KEPC (1985). Construction cash flow: Based on KEPC (1985), adjusted in a manner consistent with the PNUCC Thermal Resources Data Base, further adjusted as follows: Land costs deleted (included in option development costs); coal inventory based on 90 days con- tinuous operation at rated capacity (110 megawatts at 11,200 Btu/kilowatt-hours); costs escalated to January 1985 using appropriate Handy-Whitman cost indices. Payouts based on KEPC (1985). Fuel inventory: Inventory used in KEPC (1985); continuous operation at rated capac- ity assumed. Fuel Price: Council generic coal price. Fixed O&M cost: KEPC (1985), escalated to 1985 using estimated general inflation deflator 1984-1985 (1.05). Variable O&M cost: KEPC (1985), esca- lated to 1985 using estimated general infla- tion deflator 1984-1985 (1.05). Capital replacement cost: Assumed to be similar to conventional station. Amortization life: Based on EPRI, 1982. Operating life: Design life of major plant components (boiler and turbogenerator). Assumes a major plant refurbishment at year 20. Refurbishment costs are included in interim capital replacement costs. Cost escalation: Values adopted by the Council for the 1986 Power Plan. cP “amen 6-E Planning Assumptions — Representative Geothermal-Electric Area (Newberry Volcano, Oregon) GENERAL CHARACTERISTICS PROJECT DEVELOPMENT Site Newberry Volcano, Deschutes County, Oregon License, Site ‘hequisition Init. Detailed _~ + Plant design 5 MW Central station, single lash, steam-electric geothermal plant and Period (Completion of licensing) 36 months Wav Fuel type Intermediate temperature hydrothermal geothermal resource. Cash Flow: Year 1 $2.9 million wav Fuel transport —_Not applicable Year 2 $2.9 million Heat rejection Mechanical draft cooling ru $2.9 milion Emission control _n/av Transmission _115 KV transmission interconnect to nearest line of 115 KV or greater icon $8.8 million (517040) — Hold cost (excl. of return on invest- ment) nav TECHNICAL PERFORMANCE Expected shel life we Capacity (MW, net) Heat Rate Operating State Pee Oy eM i bect (BtukWh, net) Coen cose-cut cost = aT iit Construction from Construction from Peak nav nav nav Option Phase | Option Phase I! Maximum sustainable niav nav nav Construction period 36 months bid Pia cose san a 1640 waa Cash Flow: Year 1 $ 29.3 million wav Minimum sustainable Wav av nav ae bee aude Tr Times Year 3 $ 64.5 million Cold start — Minimum sustainable nav Construction Total $137.8 million ($2,756/KW) Wav Hot start — Minimum sustainable nav Minimum sustainable — Rated nav OPERATION Operating Availability Equivalent Annual Availability 80% Fuel Inventory wap ‘Annual Maintenance Outage Period: nav Fuel price (delivered) ‘ap (included in capital cost) Other Planned and Unscheduled Outages (Equivalent) n/av Fixed O&M $2.5 million’/yr ($50/KW/yr) Variable O&M Included in fixed O&M Capital Replacement Cost nav Amortization Life 30 years Operating Lite 30 years General Inflation (nominal) S%lyr Capital Escalation (real) 0.4%/yr O&M Escalation (real) 0.0%/yr Delivered Fuel Escalation (real) rvap 6-E-1 Appendix 6-E Plant design: From Bonneville Power Administration, 1984, Evaluation and Rank- ing of Geothermal Resources for Electrical Generation or Electrical Offset in Idaho, Montana, Oregon and Washington. (“Four State Study”) (Binary and double flash units assumed for some locations). Fuel type: From Four State Study. Fuel transportation: Not applicable. Heat rejection: From Four State Study. Emission control: Not available. Transmission: From Four State Study. Peak capacity: Not available. Maximum sustainable capacity: Not available. Rated capacity: From Four State Study — varies by area. Minimum sustainable capacity: Not available. 6-E-2 Transition times: Not available. Equivalent annual availability: From the Four State Study. Annual maintenance outages: Not available. Equivalent unscheduled outage rates: Not available. Option development period: Corresponds to 72-month total development period sug- gested in the Four State Study. 24 months used for locations developed with wellhead units. Option development costs: Not available. Expected shelf life: Not available. Option close-out cost: Not available. Construction period: From the Four State Study. Construction cash flow: Estimated con- struction cost, which varies by site, was taken from the Four State Study. Figures were adjusted to January 1985 dollars. The con- struction payout was that recommended in the Four State Study, adjusted to account for the 36-month option development phase. A 12-month construction period, (one year payout) was used for sites assumed to be developed with wellhead units. Fuel inventory: Not applicable. Fuel price: Not applicable. Fixed O&M cost: All operating and mainte- nance costs are fixed, as calculated in the Four State Study. Costs, which vary by site, were taken from the Four State Study and escalated to January 1985 dollars. Variable O&M cost: Included in fixed O&M. Capital replacement cost: Cost of replace- ment production and injection wells (includ- ing a “dry hole” allowance) is incorporated into plant capital costs. Amortization life: Consistent with operating life assumptions. Operating life: From the Four State Study. Cost escalators: Values adopted by the Council for the 1986 Power Plan. Appendix 6-F Planning Assumptions — resentative Windpark (Columbia Hills East, Washington) GENERAL CHARACTERISTICS PROJECT DEVELOPMENT Site Columbia Hills East-1, Washington ee 1: lekiotone strain uw: Plant design Windpark, consisting of approximately 65 to 150 Nordtank 65/13 horizontal Period (Completion of licensing) 42 months Wav wind turbine generators Cost: Year 1 $0.06-0.14 million ($15/kW) nav Fuel type None Total $0.06-0.14 million ($15/kW) nav Fuel transport None Hold cost (excl. of return on investment) n/av nav Heat rejection None Expected shelf life nav nav Emission control None Option close-out cost nav nav Transmission _ One mile intertie to existing grid a ‘ ae Option Phase | ‘Option Phase Ii Construction period 24 months TECHNICAL PERFORMANCE a Cash Flow: Year 1 $2.5-5.6 million ($625/kW) nav Capacity (MW, Heat Rate Operating State Per Unit Project (Btu/kWh, net) Year 2 $3.7-8.4 million ($930/kW) Peak 0.068 ANO4 Wap Construction Total $6.2-14.0 million ($1,555/ Maximum sustainable 0.068 41-94 nap kW) nav Rated 0.065 4.0-9.0 nap Minimum sustainable 0.0 0.0 map OPERATION Transition Times Cold start — Minimum sustainable map Fuel Inventory map Hot start — Minimum sustainable nap Fuel price (delivered) nap Minimum sustainable — Rated nap Fixed O&M Included in variable O&M Operating Availability Variable O&M 1.2 cents/kWh Equivalent Annual Availability (Maturity) 95% Capital Replacement Cost iided Sloahers Capacky Factor Se Royalty (Wind Right) 5% of total production cost Annual Maintenance Outage Period. nav Aenoricanon te hues Other Planned and Unscheduled Outages (Equivalent) nav Operating Lite nei General Inflation (nominal) S%/yr Capital Escalation (real) 0.4%/yr O&M Escalation (real) 0.0%/yr Delivered Fuel Escalation (real) nvap 6-F-1 Appendix 6-F Plant design: A Danish production machine, representative of the better machines currently on the market. Range of number of installed turbines reflects uncer- tainty regarding the percent of these sites suitable for turbine installation. A range of 40%-90% surface area is used for each case, as recommended by Oregon Department of Energy. Fuel type: None. Fuel transportation: None. Heat rejection: None. Emission control: None. Transmission: Approximate distance from Columbia Hills East site to nearest substation as reported by Oregon State University. Peak capacity: Per Unit: As reported for the Nordtank 65/13 by the manufacturer (Nor- dtank, Inc., Pacific Palisades, CA). Plant: Peak capacity of the fully developed site as estimated by Oregon State Department of Energy (ODOE). Site capacity derated by 5 percent to account for internal site electrical losses, as recommended by ODOE. Maximum sustainable capacity: Same as peak. Rated capacity: Per Unit: As reported for the Nordtank 65/13 by the manufacturer. Plant: Rated capacity of the fully developed site as estimated by ODOE. Site capacity derated by 5 percent to account for internal site elec- trical losses. Minimum sustainable capacity: Output at cut-in speed (8.3 MPH). Transition times: Not applicable (Plant is not dispatchable). Equivalent annual availability: Ninety- eight percent availability was recommended by ODOE based on field experience to date with the Nordtank 65/13 and warranty insur- ance offered by manufacturer. Reduced to 95% by the Council to increase conservatism of the estimates. 6-F-2 Capacity factor: Calculated for each wind resource area by ODOE. Area wind data was acquired from Oregon State University. Cor- rected for altitude, integrated with the Nor- dtank power curve and adjusted for the 95% equivalent annual WTG availability. Annual maintenance outages: Specific information not available. Equivalent unscheduled outage rates: Specific information not available. Option development period: As recom- mended by ODOE, based on California wind development experience. Option development costs: One percent of total development costs as recommended by ODOE, based on California wind develop- ment experience. Expected shelf life: Not available. Option close-out cost: Not available. Construction period: Twelve month con- struction period was recommended by ODOE based on California wind develop- ment experience. Extended to 24 months by the Council to account for remobilization time incurred if the development occurs as a two- phase option process, with the first phase being site selection and licensing, and the second phase being design, procurement, construction and testing. Construction cash flow: Construction costs include wind turbine generator (WTG), balance of plant (BOP), transmission, access and contingency. WTG costs, including pur- chase, foundation, warranty installation and shipping were estimated by ODOE to be $1,023/kW (nameplate). Balance-of-plant (BOP) costs, exclusive of contingency and permit costs were estimated by ODOE to be 13% of WTG costs. Transmission intercon- nect and access development costs were estimated by the Council to be 2% of WTG + BOP costs. Contingency was assumed (by the Council) to be 25% of total development costs (including option development). The total costs were increased by 5 percent to adjust to net rated capacity of windpark. Con- struction costs were allocated at 40% for first year, 60% for second year. Fuel inventory: Not applicable. Fuel price: Not applicable. Fixed O&M cost: All operation and mainte- nance costs are included as variable costs. Variable O&M cost: As recommended by ODOE based on experience at better Califor- nia wind developments. Capital replacement cost: Capital replace- ment for first five years covered by manufac- turer's warranty. Capital replacement for bal- ance of plant life included in variable O&M estimate. Royalties: As recommended by ODOE based on California experience. Amortization life: Set equal to operating life. Operating life: Design life of the major plant components (wind turbine generators), as reported by ODOE. Cost escalators: Values adopted by the Council for the 1986 Power Plan. Planning Assumpti Appe' ions — Generic Combustion Turbine Pro Two 105-Megawatt (Nominal) Units (January ndix 6-G 1985 dollars) GENERAL CHARACTERISTICS PROJECT DEVELOPMENT tion Phase |: ion Phase Il: She (Oregon or Washington Liconte the ‘Acquistion _ Init. Shtahed Enpinesring Plant design Two units, each a single shaft, industrial-grade, open cycle combustion turbine- Period 24 mos Twav generator of 105 megawatt nominal capacity. Cash Flow Year 1 $ 0.4 million nav Fuel type Primary — Natural gas; 950 Btu/sct (LHV). Secondary — No. 2 Fuel oil; 18,100 Btu/Ib (LHV). Year2 0.4 million Fuel transport —_ Natural gas — High pressure pipeline. Total $ 0.8 million ($4/kW) nav Fuel oil — Pipeline, rail or truck Hold cost (excl. of return on investment) Heat rejection To atmosphere. $ 0.1 million/yr ($O.S/kWiyr) wav Emission control Particulates — None required Expected sheff life 5 years nav SOX — Low sulfur fuel oil. Option close-out cost Negligible nav NOX — Water injection sah es a sa eas Transmission Ten-mile, 230 kV single circuit grid connection Option Phase! _ Option Phese t_ Construction period 30 mos nav Lag to Unit 2 completion none nav TECHNICAL PERFORMANCE : : Cash Flow Year 1 $18.4 million nav Capacity (NW, ,— ae Heat Rate @ LHV Operating State Per Unit (Btu/kWh, net) Year 2 $31.5 million Peak(January) 124 Ber 10,530 Year3 $2.6 million amas 104 208 10,710 Construction Total $52.5 million ($250/KW) nav Minimum sustainable 5 10 62,000 Transition Times OPERATION Cold start — Minimum sustainable 0.5 hour Hot start — Minimum sustainable 0.5 hour Fuel Inventory Gas: none Oil: 14 days @ 208 MW Minimum sustainable — Rated <0.5 hour Fixed Fuel Cost Gas: $0.53 million’/yr ($2.50/kW/yr) Oil: $0.53 million/yr ($2.50/KW/yr) Operating Availability Variable Fuel Cost Gas: $5.10/MMBtu Oil: $5.70/MMBtu Equivalent Annual Availability 85% Fixed O&M $0.27 milion/yr ($1.30/KW/yr) Annual Maintenance Outage Period 42 days Variable O&M 0.21 cents/kWh Normal 30 days Capital Replacement Cost $0.27 million’yr ($1.30/kW/yr) Major Overhaul (every fifth year) 90 days ‘Amortization Lite 20 years Average 42 days Operating Life 30 years Other Planned and Unscheduled Outages (Equivalent) 4% General Inflation (nominal) 5.0 %/yr Capital Escalation (real) 0.4 %iyr O&M Escalation (real) 0.0 %iyr Natural Gas Escalation (real) 1.8 %lyr Fuel Oil Escalation (real) 1.6 %lyr 6-G-1 Appendix 6-G Site: Assumes that combustion turbines could be built at existing thermal plant sites. Plant design: Based on twin Westinghouse W501D units as used for the Puget Power Fredonia project. Fuel type: Typical combustion turbine fuel characteristics. Fuel transportation: Typical of potential sites. Heat rejection: Typical of an open-cycle combustion turbine. Emission control: As practiced at the Fre- donia project. Transmission: Assumes a site near the regional transmission grid. Peak Capacity: Capacity: Maximum sus- tainable capacity during cold weather condi- tions. Based on values reported for the Puget Power Fredonia project (Westinghouse W501D units) in the Pacific Northwest Util- ities Conference Committee Thermal Resources Data Base. October 1984 (PNUCC 1984a). Heat rate: Based on values reported for Fredonia in PNUCC (1984a). Val- ues given are based on lower heating value of fuel. Base load capacity: Capacity: Rating of Fredonia units as reported in PNUCC (1984a). Heat rate: Based on values reported for Fredonia in PNUCC (1984a). Values given are based on lower heating value of fuel. Minimum sustainable capacity: Capacity: Rating of Fredonia units as reported in PNUCC (1984a). Heat rate: Based on values reported for Fredonia in PNUCC (1984a). Val- ues given are based on lower heating value of fuel. Transition times: As reported for Fredonia in PNUCC (1984a). Values from minimum sustainable to maximum sustainable are estimated. 6-G-2 Equivalent annual availability: Based upon National Electric Reliability Council (NERC) Generating Availability Data System (GADS) records for combustion turbines. Equivalent annual availability was calculated by weighting the NERC equivalent annual availability data for operating years 1 through 10 in proportion to the the number of operat- ing years represented by the compiled statis- tics. (i. e., statistics for operating years 1, 2 and 3, which are compiled individually, each received a weight of “1”; statistics for years 4 through 10, which are aggregated, received a weight of “7”). The resulting equivalent annual availability for all combustion turbine units is 85.6 percent. A value of 85 percent was chosen, considering the following: 1) The NERC data base represents a large number of unit-years of operating experience (8,261 unit-years for operating years one through ten); confidence in the statistics is therefore good. 2) The NERC data base includes data for both aircraft-derivative and industrial-type units, The generic turbine is an industrial-type unit, generally considered to be more reliable than the aircraft-derivative units; therefore, its performance can be expected to be as least as good as averages of aircraft-derivative and industrial units. 3) The NERC data base includes units subject to all modes of operation from peaking to continuous duty. Continuous duty operation — thought to be more typical of Northwest units employed for firming secondary hydro — generally results in more reliable operation. Annual maintenance outage period: The schedule shown assumes an annual com- bustor inspection, a “hot path” inspection every fifth year and a major overhaul every tenth year. Periods are derived from esti- mates appearing in J.H. Borden “Outage Management Improves Turbine Availability,” Diesel and Gas Turbine Worldwide, April 1982, and could be shortened by improved outage management. Other planned and unscheduled out- ages: Calculated from the 85 percent equiv- alent annual availability and the average annual maintenance outage rate. The result- ing value (4 percent) is somewhat more con- servative than the 3 percent recommended by Westinghouse Electric Corporation Com- bustion Turbine Systems Division. Option development schedule: Assumes development of option to Node 2 (Concep- tualization, completion of permitting and licensing, acquisition of site). Based on Fre- donia experience, as reported in the PNUCC “Working Paper, Development of Generic Resource Data,” October 1984 (PNUCC 1984b). Option development cost: Assumes devel- opment of option to Node 2 (Conceptualiza- tion, completion of permitting and licensing, acquisition of site). Cost to achieve Node 2 is based upon 1 percent of total capital costs as estimated in Battelle, Pacific Northwest Lab- oratories Development and Characterization of Electric Power Conservation and Supply Resource Planning Options, plus estimated purchase cost of 100 acres of land at $2,500 per acre. Option hold cost: Estimate includes project management, EFSEC, environmental base- line and indirect costs, as follows. Project management taken as one engineering staff at $50,000 per year. EFSEC and environ- mental baseline costs scaled from 1984 Creston hold costs, as presented by The Washington Water Power Company to the Northwest Power Planning Council on July 17, 1984. Indirect costs taken at 11 percent as appearing in the Creston presentation. Expected shelf life: Expected shelf life is estimated time until fuel cell technology becomes fully established, requiring new feasibility and environmental studies. Option close-out cost: Revenues from the sale of land are assumed to cover close-out costs. Construction schedule: Based on Puget Power experience for Fredonia as reported in PNUCC “Working Paper, Development of Generic Resource Data,” October 1984 (1984b). Construction cash flow: As estimated by Westinghouse Electric Corporation, Com- bustion Turbine Systems Division for twin Westinghouse W501D units, installed and ready to operate. Payout schedule based on actual timing of construction expenditures for Fredonia as reported in the PNUCC (1984b). Appendix 6-G Fuel contract: Provisions stated are similar to Fredonia. Fuel inventory: Similar to Fredonia. Fuel cost (service charge): Natural gas—Fixed fuel oil service charge experienced by the Fredonia project, from PNUCC (1984a). Thought to be typical of a contract with provision for short-term inter- ruptibility. The service charge covers the cost of pipeline service to the project. Fuel oil— Fixed fuel oil service charge expe- rienced by the Fredonia project, from PNUCC (1984a). Thought to be typical of a contract with provision for delivery with advance notice. The service charge covers the cost of pipeline service to the project. Fuel cost (variable): Natural gas—NWPPC planning assump- tions for industrial gas sold in Washington, medium-low and medium-high load growth cases. Fuel oil—NWPPC planning assumptions for industrial oil sold in Washington, medium-low and medium-high load growth cases. Fixed O&M cost: As reported for Fredonia in PNUCC (1984a), escalated to 1985 using the 1984-1985 GNP deflater (1.05). Variable O&M cost: As reported for Fre- donia in PNUCC (1984b), escalated to 1985 using the 1984-1985 GNP deflater (1.05). Capital replacement cost: Based on one major overhaul every ten calendar years. Cost of overhaul assumed to be ten percent of original equipment cost ($37.8 million at 180 dollars per kilowatt), rounded to nearest million dollars. Levelized at 3 percent dis- count rate over the project life. Amortization life: Based on Electric Power Research Institute Technical Assessment Guide, May 1982 (EPRI, 1982) recom- mendations. Operating life: Likely physical life with major overhauls at ten-year increments and oper- ated primarily for secondary firming. Cost escalation: Assumptions are taken from the Council decision regarding financial variables. 6-G-3 pendix 6-H Planning Assumptions — Generic Combined Cycle Project Two 286-Megawatt Units (January 1985 dollars) GENERAL CHARACTERISTICS PROJECT DEVELOPMENT ste Oregon or Washington eae Tenen wn Salad age Plant design Two combined-cycle units, 286 megawatt nominal capacity each. Each unit Period 24 mos va eee cee chp a — = generators and one steam turbine generator of 84 megawatts gross capacity. Year 2 3.2 million Steam conditions are 1,210 psig and 950°F. asl rT wav ee tonto en — investment) $0.2 million/yr ($0.4/kWiyr) nav Fuel transport Netra gas Hoh preeaur pipeine. Sceteiaulin Sous fae Heat rejection _To atmosphere via heat recovery steam generators and mechanical draft Option close-out cost negligible nav i or Emneeron corer tor ae are Const. period (Unit 1 insr) 45 mos nav NOX — Water injection. Lag to second unit 3mos nav Transmission Ten-mile, 500 kV single-circuit grid connection Cash Flow: Year 1 $ 14.5 million nav Year 2 $115.9 million TECHNICAL PERFORMANCE Year3 $155.8 million Operating States — CT Mode: Gopechy Gi noth aia Year 4 $ 76.0 — Operating State Per Unit’ “Project (BtukWh, net) Construction Total $362.2 milion ($633/kW) nav Peak (January) 248 496 10,530 Base Load 208 416 10,710 OPERATION Minimum sustainable 10 20 62,000 Operating States — Combined Cycle Mode: Fuel Inventory Gas: none Oil: 14 days 9 572 MW One CT, undired boler <= 4 Fixed Fuel Cost Gas: $1.4 million/yr ($2.50/KW/yr) Oil: $1.4 million/yr ($2.50/KW/yr) One CT, fired boller —— 9270 Variable Fuel Cost(LHV) Gas: $5.10/MMBtu Oil: $5.70/MMBtu Two CTs, unfired boiler 240 480 9,350 om SES neo Me COREY) Two CTs, fired boiler 283566 9,810 Mates CaM DO come a oa Capital Replacement Cost $2.4 million/yr ($4.20/KW/yr) Transition Times — CT Mode: aa so yenrs) Cold start — Minimum sustainable 0.5 hour Operating Lite 30 years Hot start — Minimum sustainable 0.5 hour General Inflation (nominal) 5.0 %lyr Minimum sustainable — Rated <0.5 hour Capital Escalation (real) 0.4 %lyr Transition Times — Combined Cycle Mode: O&M Escalation (real) 0.0 %iyr Cold start — Minimum sustainable 9 hours Natural Gan Escalation (ree) aie seve Hot start — Minimum sustainable 3 hours Fuel Ou Eecatenon eal eae Minimum sustainable — Rated 3 hours Operating Availability Equivalent Annual Availability 83% Annual Maintenance Outage Period Normal 30 days Major Overhaul (every fifth year) 90 days Average 42 days Other Planned and Unscheduled Outages (Equivalent) 6% 6-H-1 Appendix 6-H Site: Assumes that combined cycle plants could be constructed at existing licensed thermal sites. Plant design: As described in Kaiser Engineers Power Corporation Bonneville Power Administration Comparative Electric Generation Study (Supplemental Studies), February 1985 (KEPC, 1985). Fuel type and source: Typical of potential sites. Heat rejection: per KEPC, 1985 Emission control: As practiced at the Fre- donia project. Transmission: Assumes a site near the regional transmission grid. Capacities and Heat Rates: CT Mode—Capacities and heat rates cited for the combustion turbines operating inde- pendently are as derived for the generic stand-alone combustion turbine plant. CC Mode—Capacities and heat rates for combined cycle operation are based upon the four modes of operation reported for the EI Paso Electric Newman station (Gas Tur- bine World, July 1981). The capacity states for the Newman station were adjusted by the ratio of nominal gross capacities of the New- man station (220 megawatts) and the station used in KEPC (1985) (296 megawatts) (Both are Westinghouse PACE package plants using W501D gas turbines). The resulting ratioed capacity was further adjusted by the ratio of net to gross plant output reported in KEPC (1985) to arrive at net capacity states. Transition times: CT Mode—As reported for Fredonia in PNUCC (1984a). Values from minimum sus- tainable to maximum sustainable are estimated. 6-H-2 CC Mode—As reported for El Paso Electric Newman Station in Gas Turbine World, July 1981. Equivalent annual availability: A West- inghouse study (“Gas turbine combined cycle reliability has made impressive pro- gress.” Modern Power Systems. October 1982) reports average annual availabilities for combined cycle plants to be approximately 1 percent lower than for individual combustion turbines (86.5 percent vs. 87.6 percent). The recommended equivalent annual availability of 83 percent is derived from a more conser- vative 2 percent reduction of the 85 percent availability selected for the generic stand- alone combustion turbine. Annual maintenance outage periods: The generic combined-cycle plant is assumed to operate in a manner similar to the generic stand-alone combustion turbine (i.e., as a unit primarily for firming secondary hydro- power), operated approximately one year in four or five with fairly continuous operation when needed. The resulting maintenance schedule will therefore be similar to that developed for the stand-alone combustion turbine, consisting of an annual 30-day rou- tine maintenance period and a major inspec- tion following each year of secondary firming operation (estimated to be one in five years). The durations for these inspections are taken from “Outage management improves gas turbine availability,” in Diese! and Gas Tur- bine Worldwide, April 1982. The outage durations are conservative and could be sub- stantially shortened by increasing the spare parts stock. Rate of other planned and unscheduled outages: Calculated as an equivalent out- age rate from the equivalent annual availabil- ity and the annual maintenance outage schedule. Option development schedule: Based on Fredonia experience as reported in PNUCC. Working Paper — Development of Generic Resource Data, October, 1984 (PNUCC, 1984b). Option development cost: Assumes devel- opment of option to include conceptualiza- tion, completion of permitting and licensing and acquisition of the site. Cost is based upon 1.7 percent of total capital costs as esti- mated in Battelle, Pacific Northwest Labora- tories Development and Characterization of Electric Power Conservation and Supply Resource Planning Options, August 1982 (Battelle, 1982), plus the estimated cost of 100 acres of land at $2,500 per acre. Option hold cost: Estimate includes project management ($100,000 per year), EFSEC ($25,000 per year), environmental baseline ($75,000 per year) and indirect costs at 11 percent of the foregoing. Rounded to the nearest 0.1 million per year. Return on invest- ment is not included. Expected shelf life: Expected shelf life is the estimated time until fuel cell technology becomes fully established. This is assumed to require new feasibility and environmental studies. Option close-out cost: The value of the land is assumed to offset option close-out costs. Construction schedule: Based on KEPC, 1985. Construction cash flow: Based on KEPC (1985), adjusted in a manner consistent with PNUCC (1984p), further adjusted as follows: Land costs deleted (included in option devel- opment costs); fuel inventory based on two weeks operation at maximum sustainable capacity (572 megawatts at 8,030 Btu/kWh); costs escalated to January, 1985 using appropriate Handy-Whitman cost indices. Payout rate based on KEPC (1985). Fuel inventory: As assumed in KEPC (1985), commensurate with natural gas contracts with provision for short-term interruption. Fuel cost (service charge): Natural gas—Fixed fuel oil service charge experienced by the Fredonia project, from PNUCC (1984a). Thought to be typical of a contract with provision for short-term inter- ruptability. The service charge covers the cost of pipeline service to the project. Fuel oil—Fixed fuel oil service charge expe- rienced by the Fredonia project, from PNUCC (1984a). Thought to be typical of a contract with provision for delivery with advance notice. The service charge covers the cost of pipeline service to the project. Fuel cost (variable): Natural gas—NWPPC planning assump- tions for industrial gas sold in Washington, medium-low and medium-high load growth cases. Fuel oll—NWPPC planning assumptions for industrial oil sold in Washington, medium-low and medium-high load growth cases. Fixed O&M cost: As estimated in KEPC (1985), escalated to 1985 using the 1984-1985 GNP deflator (1.05). Variable O&M cost: As estimated in KEPC (1985), escalated to 1985 using the 1984-1985 GNP deflater (1.05). Appendix 6-H Capital replacement cost: Based on one major overhaul every ten calendar years of operation. Cost of overhaul assumed to be 10 percent of original plant cost (362 million), rounded to nearest million dollars. Levelized at a 3 percent discount rate over the project life. Amortization life: Based on value used for stand-alone generic combustion turbines (the limiting major component). Operating life: Based on value used for stand-alone generic combustion turbines (the limiting major component). Cost escalation: Assumptions are taken from the Council decision regarding financial variables. 6-H-3 pendix 6-1 Planning Assumptions Washington Public Power Supply System Nuclear Project No. 1 (January 1985 dollars) GENERAL CHARACTERISTICS CONSTRUCTION Site Richlend, Weshinglon Construction Period 54 months Ownership Public utilities (net-billed) 100% hua Plant design 1,338 megawatt (nameplate)/1,250 megawatt (net) capacity; pres- surized water nuclear power plant. Babcock and Wilcox Model 205 iarductaaailed oon Fuel Assembly. 2nd 12 months $ 422 milion Heat rejection Mechanical draft cooling towers; makeup from the Columbia River 3rd 12 months $ 402 million Transmission Located on existing regional grid (Ashe substation). 4th 12 months $ 244 million Last 6 months $ 78million TECHNICAL PERFORMANCE Total $1,383 million ($1,106/kW) Heat Rate Financing Bonds at 9.2% (nominal) plus 1% risk premium. Operating Siete ane Gens Maximum sustainable 1,250 9,829 Rated (least cost) 1,250 9,829 Een Minimum sustainable 500 wav Termination Period 24 months Transition Times Termination Costs Cold start — Minimum sustainable 24 hours Termination program $ 33 million Hot start — Minimum sustainable 5 hours Nominal site restoration $ 15 million Minimum sustainable — Maximum sustainable nav Full site restoration $100 million Operating Availabilty Sale of assets (receipts) $125 million Equivalent Annual Availability 65% Funding Existing funds ($125 million) reinvested at 9.2% (nominal). Bon- Annual Maintenance and Refueling Outage Period 60 days neville rates if existing funds are depleted. (No credit for existing ; ne funds taken for comparison of alternative resources.) Annual Maintenance and Refueling Outage Timing June-July Other Planned and Unscheduled Outages (Equivalent) 22 percent OPERATION PRESERVATION Fixed fuel cost $35.4 million/yr ($28.30/KW/yr) Operating Costs Expected Shelf Lite 15 years, minimum 1 ced " c ed 37 1.0 mill -B0/k\ Bal eas. ix si lion/yr ($56.80/kW/yr) Variable O&M $ 0.0011/net kWh CY 1985 $60 million Decommissioning fund 3.5 milli .80/KW/ Jan 86 to Restart $36 million/yr (Earned value at $24 million per year beginning oun hicabiuandtehs _ in July 1986 to be credited against costs to Capital Replacement Cost complete) Operating year 1 $ Smillion Cash Flow (minimum) Operating year2 $11 million CY 1985 $60 million Operating year $16 million cy 1986 $20 million} _ (Includes $8 milion for additional ramp-down to 4 1 rit 16. 80/KW/ pokey ca ean Operating year andon $21 miillion/yr ($16.80/kW/yr) Jan 87 to Remobilization $12 million/yr ee al = Remobilization Year $44 million ($32 million for ramp-up plus $12 million Sete _ preservation) Funding Existing funds ($ 125 million) reinvested at 10.3% (nominal) rate. COST ESCALATION Bonneville rates if existing funds are depleted (No credit for existing funds taken for comparison of alternative resources) General inflation (nominal) 5.0%/yr Capital (real) 0.4%Iyr 08M (real) 0.0%Iyr Fuel (real) None to 1993; 1%/yr thereafter Appendix 6-1 Maximum sustainable capacity: Capaci- ties and heat rates are from Pacific Northwest Utilities Conference Committee Thermal Resources Data Base. October 1984 (PNUCC, 1984). Rated capacity: Capacities and heat rates are from PNUCC, 1984. Minimum sustainable capacity: Capaci- ties are from PNUCC, 1984. Transition times: From PNUCC, 1984. Operating availability: The principal oper- ating availability parameters are equivalent annual availability, planned outage rate and equivalent unscheduled outage rate. Equiv- alent annual availability represents the frac- tion of the year that a unit is available to operate at full power. Because a unit may occasionally be available for derated (partial power) operation, annual availability is expressed in equivalent full power hours. Equivalent annual availability is a function of planned outages (for maintenance, repair or refueling) and unplanned outages. Of the three availability parameters, only planned outages (listed in this appendix as “Annual Maintenance and Refueling Out- ages’) is readily available. One 60-day planned annual outage is scheduled for WNP-1. The Council, for the 1983 cost-effectiveness assessment of WNP-4 and 5, undertook an extensive analysis of the equivalent annual availability and equivalent unplanned outage rate of large (1,000+ MW) nuclear power plants. That analysis included examination of performance data maintained by the Nuclear Regulatory Commission (NRC) and the North American Electric Reliability Council (NERC). The Council concluded, based on that analysis, that a 22 percent equivalent unplanned outage rate, a 60-day annual planned outage, and a 65 percent equivalent annual availability were representative of large nuclear plants. In developing assumptions for this plan regarding the performance of WNP-1 and WNP-3, the Council chose to rely upon the assumptions developed earlier for the WNP-4 and 5 study, unless persuasive evi- dence was available suggesting that the ear- lier values should be modified. To determine if the earlier values should be modified, the Council examined operating histories of all large (1,000 + MW) pressurized water reac- tors with one year or more operating history, compiled in the NRC Grey Book. A com- parison of the availabilities of these units through October 1984 with the Grey Book data through September 1982 studied earlier indicated no significant changes in the avail- ability of these units. The Council therefore concluded that equivalent unplanned out- age, and equivalent annual availability assumptions for WNP-1 and WNP-3 should remain unchanged from the assumptions used earlier for WNP-4 and WNP-5. The planned annual maintenance and refueling outages are scheduled to coincide with the period of seasonal hydropower surplus. Preservation shelf life: The Council con- cludes that the projects can be preserved for a minimum of 15 years (see discussion in this chapter). Preservation cash flow: The Supply Sys- tem has provided cash flows for “currently planned” and “minimum level” preservation programs. The planned preservation program, at $36 million per year, includes licensing and reg- ulatory activities leading to “earned value” credit against costs-to-complete. Following completion of the currently planned ramp- down to about 400 staff by July of 1986, the planned preservation program would con- tinue to restart of construction at a rate of $36 million per year. Earned value of approx- imately $24 million per year would begin to accrue about mid-1986. The minimum level preservation program at $12 million per year contains no provision for ongoing engineering and licensing activities and would evidently forego certain record update activities and maintenance staffing. Incremental ramp-down costs of $8 million would be experienced due to additional staff layoffs, and additional ramp-up costs of $32 million would be required prior to restart of construction to restore engineering and licensing staff to planned preservation levels. The Supply System has recently reestimated minimum preservation to be $10 million per year. Preservation financing: Preservation at Project 1 would be financed initially from the current account balance of $125 million. These funds are reinvested and are expected to cover planned preservation costs at Pro- ject 1 through the first quarter 1988. Financ- ing from Bonneville rates would follow exhaustion of this fund. In comparing the cost effectiveness of WNP-1 with other alternatives, the Council chose not to credit the project with the current account balance, reasoning that these funds are not yet sunk, and could be recovered by the region if the project were terminated. An interest rate premium (discussed under con- struction financing) is added to the “stan- dard” equity rate chosen by the Council for all other resources. Construction period: The construction period for Project 1 has been reduced from 60 months to 54 months in accordance with updated estimates received from the Supply System in January 1985. The Council dis- cussed with the Supply System the likelihood of maintaining the revised schedule. Learn- ing that the critical path is hiring and training of operators, and considering the excellent construction rates achieved on Project 3 in the year prior to the decision to slow con- struction, the Council concluded that the schedule appears reasonable. Appendix 6-1 Construction cash flow: The Supply Sys- tem has provided revised construction cash flows for the project. The project shows a modest decrease compared to earlier esti- mates. This is attributable to incorporation of earned value through January 1985; deduc- tion of planned earned value activities through July 1985; minor scope changes; and use of consolidated construction meth- ods using capped cost, risk-sharing con- tracts. The estimate to complete includes contingencies of approximately 9 percent and incorporates known and probable changes resulting from pending regulatory actions. Construction financing: Financing is assumed to be by bonds. Direct financing by Bonneville rates was considered; however, in view of the considerable impact of rate finan- cing on Bonneville rates, the likelihood of a prolonged preservation period (allowing time for the WNP-4/5 settlement to proceed), and equity questions regarding rate financing of capital investment, the Council concluded that financing should be assumed to be bonds. A risk premium of 1 percent is added to all public and private debt and equity financing. This premium is based on statements by Seattle Northwest Securities Commission (financial advisors to the Supply System) and Salomon Brothers Inc.; Goldman, Sachs and Company; Merrill Lynch Capital Markets and Smith, Barney, Harris Upham and Com- pany, Inc. (Senior Managing Underwriters). These firms concluded that the risk premium would not likely exceed 1/2 to 1 percent. Termination period: As estimated by the Supply System. Termination costs: As estimated by the Supply System. Termination financing: Termination financ- ing is assumed to be similar to preservation financing. Fuel costs: Fuel cost estimates were pro- vided to the Council by the Supply System and are based on operation at 65 percent capacity factor for a 40-year plant life. Operation costs: Fixed operating costs are as provided by the Supply System. Variable operating costs consist of the federal fuel disposal charge. This latter charge is based on gross energy production and has been adjusted to represent costs based on net energy production. Capital replacement costs: Capital replacement costs are as currently estimated by the Supply System. Amortization life: The 30-year amortization life is based upon the recommendation of the Electric Power Research Institute and is con- sistent with values used by the Council for other resources. Actual bond maturity peri- ods might vary. Operating life: The Council has chosen a 40-year operating life to be consistent with design life of 40 years. Arguments regarding nuclear plant operating life were examined in some detail. First, the principal plant compo- nents—for example, the reactor vessel— are conservatively designed to withstand the conditions imposed by normal plant opera- tion for a period of 40 years or more. Second, NRC operating licenses are for 40 years duration, based on the 40-year design life. Third, plant preservation should not impact the expected operating life. Finally, efforts are underway in the industry to extend plant operating life beyond 40 years. On the other hand, there are arguments that 40 years is an optimistic assumption regard- ing operating life. Commercial nuclear plants have been operating only since 1957. More- over, many of the commercial plants coming into service prior to 1969 were demonstration plants, such that operating experience on mature commercial plants is available only since the late 1960s. Given the relatively brief history of the technology, there is currently no Statistically sound basis for judging the expected operating lives of the current gener- ation of commercial nuclear plants. It is also argued that the number of retire- ments of plants with operating lives of 25 years or less indicate that a 40-year operat- ing life cannot be expected. Many of the retirements have been plants that were origi- nally designed as demonstration plants that were not intended to remain economically competitive with later commercial units. Sev- eral commercial plants have been retired early (largely due to the economic conse- quence of required backfits) or shut down for prolonged periods for repair or regulatory reasons. The Council, weighing these arguments, chose to consider a base case operating life of 40 years, but to explore the efforts of shorter operating lives on cost effectiveness through sensitivity analysis. Cost escalation: Values for general infla- tion, capital and operation and maintenance are as adopted by the Council for the 1986 Power Plan. The Supply System recom- mended that the nuclear fuel escalation rate be equivalent to general inflation through 1993, citing the soft nuclear fuel market, and 1 percent real thereafter. The Council con- curs with this argument. 6-1-3 Appendix 6-J Planning Assumptions Washington Public Power Supply System Nuclear Project No. 3 GENERAL CHARACTERISTICS (January 1985 dollars) PRESERVATION Site Satsop, Washington Ownership Public Utilities (net-billed) 70% Investor-owned Utilities (capability proposed for 30% acquisition by Bonneville) Plant design 1,324 megawatt (nameplate)/1,240 megawatt (net) capacity; pres- ‘surized water nuclear power plant. Combustion Engineering System 80. Heat rejection Chehalis River. Transmission Located on existing regional grid Natural draft cooling tower; makeup from the wells adjacent to the TECHNICAL PERFORMANCE Operating State Maximum sustainable Rated (Least cost) Minimum sustainable Transition Times Cold start — Minimum sustainable Hot start — Minimum sustaintable Minimum sustainable — Maximum sustainable Operating Availability Equivalent Annual Availability Annual Maintenance and Refueling Outage Period Annual Maintenance and Refueling Outage Timing Other Planned and Unscheduled Outages (Equivalent) Seeoty Heat Rate net) (BtwkWh, net) 1,240 10,459 1,240 10,459 500 nav 24 hours 5 hours av 65% 60 days May-June 22 percent Expected Shelf Life Cash Flow (as planned) CY 1985 CY 1986 Jan 87 - Restart Cash Flow (minimum) CY 1985 CY 1986 Jan 87 to Remobilization Remobilization Year Funding Publicly-owned utility share Investor-owned utility share 15 years, minimum $61 million $48 million (Includes balance of planned ramp-down ($30 million) plus 6 months planned pres: ervation ($18 million).) $36 milion/yr (Earned value of $24 million per year beginning in July 1986 to be credited against costs to complete.) $61 million $57 million (Includes balance of planned ramp-down ($30 million), 6 months minimum preserva- tion ($6 million), $13 million capital expen- ditures, and $8 million for ramp-down to minimum preservation levels.) $12 millionyr $46 million (Includes $34 million for ramp-up plus $12 million for preservation.) Bonneville rates. Bonneville rates. CONSTRUCTION Construction Period 54 months Cash Flow 1st 12 months $ 133 million 2nd 12 months $ 373 million 3rd 12 months $ 396 million 4th 12 months $ 322 million Last 6 months $ 86 million Total $1,310 million ($1,056/«W) Financing POU Share Bonds at 9.2% plus 1% risk premium. JOU Share Similar to preservation financing. TERMINATION Termination Period 24 months Termination Costs Termination program $ 31 million Nominal site restoration $ 20 million Full site restoration $100 million Sale of assets (receipts) $ 70 million Funding Similar to preservation financing. OPERATION Fixed Fuel Cost $38.9 million/yr ($31.40/kW/yr) Operating Costs Fixed O&M $71.0 million’yr ($57.30/kW/yr) Variable O&M $ 0.0011/net kWh Decommissioning fund $ 3.5 million/yr ($2.80/KW/yr) Capital Replacement Cost Operating year 1 $ Smillion Operating year 2 $11 million Operating year 3 $16 million Operating year 4 and on $21 million/yr ($16.90/KW/yr) Amortization Life 30 years Operating Life 40 years COST ESCALATION General Inflation (nominal) 5.0%lyr Capital (real) 0.4%lyr O&M (real) 0.0%/yr Fuel (real) None to 1993; 1%/yr thereafter 6-J-1 Appendix 6-J Maximum sustainable capacity: Capaci- ties and heat rates are from Pacific Northwest Utilities Conference Committee Thermal Resources Data Base. October 1984 (PNUCC, 1984). Rated capacity: Capacities and heat rates are from PNUCC, 1984. Minimum sustainable capacity: Capaci- ties are from PNUCC, 1984. Transition times: From PNUCC, 1984. Operating availability: (See discussion provided in Appendix 6-I for WNP-1.) Preservation shelf life: With the exception of exposed rebar, no significant deterioration appears to have been experienced at WNP-3. At current rates of corrosion, the most severely corroded rebar could be expected to last 16 years without exceeding allowable metal loss. Application of protec- tive coatings would prevent further deteriora- tion of rebar, and in any event the rebar could be replaced at relatively minor cost. The Council concludes that the project can be preserved for a minimum of 15 years (see discussion in this chapter). Long-term mini- mum level preservation of WNP-3 would likely require improved closure of the reactor building and the turbine building. 6-J-2 Preservation cash flow: The Supply Sys- tem has provided cash flows for “currently planned” and “minimum level” preservation programs. The planned preservation program, at $36 million per year, includes licensing and reg- ulatory activities leading to “earned value” credit against costs-to-complete. Following completion of the currently planned ramp- down to about 400 staff by July of 1986, the planned preservation program would con- tinue to restart of construction at a rate of $36 million per year. Earned value of approx- imately $24 million per year would begin to accrue about mid-1986. The minimum level preservation program at $12 million per year contains no provision for ongoing engineering and licensing activities and would evidently forego certain record update activities and maintenance staffing. Additional demobilization and remobilization costs would be incurred. Approximately $13 million would be required to place Project 3 into a condition suitable for long-term low- manpower preservation. Twelve million dol- lars of this would be earned value. Addi- tionally, incremental ramp-down costs of $8 million would be experienced due to addi- tional staff layoffs, and additional ramp-up costs of $32 million would be required prior to restart of construction to restore engineering and licensing staff to planned preservation levels. The Supply System has recently reestimated the cost of minimum preserva- tion to be $14 million per year. Preservation financing: Funding of the Supply System share of Project 3 would con- tinue from Bonneville rates. Bonneville is assuming the preservation funding of the investor-owned utilities share of Project 3 in accordance with the WNP-3 settlement. Construction period: The Council dis- cussed with the Supply System the likelihood of maintaining the 54-month schedule, esti- mated by the Supply System. Learning that the critical path is hiring and training of oper- ators, and considering the excellent con- struction rates achieved on Project 3 in the year prior to the decision to slow construc- tion, the Council concluded that the schedule appears reasonable. Appendix 6-J Construction cash flow: The Supply Sys- tem has provided revised construction cash flows for the project. The project shows a modest decrease compared to earlier esti- mates. This is attributable to incorporation of earned value through January 1985; deduc- tion of earned value activities through July 1986; minor scope changes; and use of con- solidated construction methods using cap- ped cost, risk-sharing contracts. The esti- mates to complete include contingencies of approximately 9 percent and incorporate known and probable changes resulting from pending regulatory actions. Construction financing: Financing of the Supply System share of the project is assumed to be by bonds. Alternative financ- ing by Bonneville rates was considered; how- ever, in view of the considerable impact of rate financing on Bonneville rates, the like- linood of a prolonged preservation period (allowing time for the WNP-4/5 settlement to proceed), and equity questions regarding rate financing of capital investment, the Council concluded that financing should be assumed to be bonds. The investor-owned utilities share of Project 3 is assumed to be capitalized at 50 percent debt and 50 percent equity. A risk premium of 1 percent is added to all public and private debt and equity financing. This premium is based on statements by Seattle Northwest Securities Commission (financial advisors to the Supply System) and Salomon Brothers Inc.; Goldman, Sachs and Company; Merrill Lynch Capital Markets and Smith, Barney, Harris Upham and Com- pany, Inc. (Senior Managing Underwriters). These firms concluded that the risk premium would not likely exceed 1/2 to 1 percent. Termination period: As estimated by the Supply System. Termination costs: As estimated by the Supply System. Termination financing: Termination financ- ing is assumed to be similar to preservation financing. Fuel costs: Fuel cost estimates were pro- vided to the Council by the Supply System and are based on operation at 65 percent capacity factor for a 40-year plant life. Project 1 bonding resolutions prohibit transfer of Proj- ect 1 fuel to Project 3 were Project 1 to be terminated. Operation costs: Fixed operating costs are as provided by the Supply System. Variable operating costs consist of the federal fuel disposal charge. This latter charge is based on gross energy production and has been adjusted to represent costs based on net energy production. Capital replacement costs: Capital replacement costs are as currently estimated by the Supply System. Amortization life: The 30-year amortization life is based upon the recommendation of the Electric Power Research Institute and is con- sistent with values used by the Council for other resources. Actual bond maturity peri- ods might. vary. Operating life: (See discussion provided in Appendix 6-1 for WNP-1). Cost escalation: Values for general infla- tion, capital and operation and maintenance are as adopted by the Council for the 1986 Power Plan. The Supply System recom- mended that the nuclear fuel escalation rate be equivalent to general inflation through 1993, citing the soft nuclear fuel market, and 1 percent real, thereafter. The Council con- curs with this argument. 6-J-3 Introduction: The Regional Power System The electrical power system in the Pacific Northwest is dominated by hydropower. The Northwest system is unique in the United States because of this characteristic. Cur- rently the hydropower system produces approximately 70 percent of the total elec- tricity used by the region. Even with demand growth at the Council's high level, hydro- power would still produce almost half the region's electricity at the turn of the century. There are two key characteristics to the Northwest hydropower system. First, it varies widely in annual energy capability, depend- ing upon rainfall and the snowpack accumu- lated in the region each year. The average annual output of the hydropower system since recordkeeping began in 1879 (and including the effect of the Council's water budget) is approximately 16,400 megawatts. This is about 4,100 megawatts, or 33 percent, greater than the critical period energy capability. During a good year the annual capability can be as much as 50 percent greater than critical period capability. “Critical period” refers to that sequence of low water conditions during which the lowest amount of firm load can be carried. The energy that can be generated during the critical period is called “firm” energy. Energy that can only be generated when water conditions are both better than critical conditions and sufficient to refill system reservoirs is called “nonfirm” energy. Asecond characteristic, equally important, is that the variation within the year can be even greater than the variation across the water conditions from year to year. Over half the annual firm energy from the Northwest hydropower system comes from natural streamflows; less than half comes from reservoir storage. Figure 7-1 shows the variation in natural streamflow at The Dalles on the lower Columbia. The relatively low amounts and low variability of natural stream- flows between August or September and the onset of the spring runoff in March or April are important in considering the risks that can be Chapter 7 Better Use of the Hydropower System Flow in 1,000 Cubic Feet/Second % = Percent of Time Flow is Exceeded Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Figure 7-1 Average Daily Columbia River Natural Flow at The Dalles, Oregon taken in using the reservoir storage. (The 10, 25, 50, 75 and 90 percent lines represent percentage of time the flow is equalled or exceeded on that particular day. These lines are based on ten-day mean values.) Historically, the Columbia River discharges about 73 percent of its natural runoff between April and October, and only 27 percent in the November to March winter period when elec- trical loads are highest. This ratio of 73:27 has been altered by upstream storage proj- ects so that the regulated flow matches the pattern of the region's loads. However, the river and its storage system are managed for multiple purposes besides electricity genera- tion. Flood control, irrigation, fish and wildlife requirements, recreation and navigation may limit the availability of upstream storage for power generation. The reservoir storage itself is significantly limited. A large part of the hydropower sys- tem water supply comes from the snowpack inthe upper Columbia and upper Snake river basins, in the mountains of British Columbia, Montana and Idaho. However, only 40 per- cent of even the average January to July runoff is storable in the system's reservoirs. This means large portions of the total annual water supply come during the spring runoff from April through July. Moreover, most of the water from the melting snow must pass through the generators or over the spillways if it cannot be used in the springtime, because it cannot be stored for use in the following fall and winter when demand is higher. Figure 7-2 shows the amounts of electrical energy available at various probability levels above the critical period quantities over the 102-year historical record. The variability of the hydropower system has maior effects on the economics of other existing and new resources, because it influences the way they operate.’ 7-1 Chapter 7 Monthly Average Megawatts Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr % = Percent of Time Nonfirm Availability is Exceeded May Jun Figure 7-2 Probability of Nonfirm Energy Availability Critical Period Planning Power system planning is currently con- ducted on a critical period basis. To deter- mine the total amount of energy resources required, it is assumed the hydropower sys- tem will not produce more energy than it did during the worst conditions of the past. Criti- cal periods run from the beginning of the reservoir drawdown season in August or September to the beginning of the refill sea- son—the onset of the spring runoff in March or April. The number of annual cycles of sequential dry years the system can support is a function of the amount of reservoir stor- age and generating resources (both existing facilities and new ones that come on-line dur- ing the period being studied). Currently, the worst conditions are either the four-year sequence from August 1928 to March 1932, or the more severe but shorter two-year sequence from September 1943 through April 1945. 7-2 However, the common reference to a four- year critical period in regional planning docu- ments does not mean that the region will necessarily have four years to work out prob- lems before the system's reservoirs are empty. An exact repetition of the water sequence from 1928 through 1932 is not required for the system to be in trouble. The wet fall of 1976 left reservoirs high enough to sell nonfirm energy, but the following spring runoff was the worst since recordkeeping began in 1879. By November 1977, reser- voirs were within five months of going empty under continued low-water conditions. Planning to critical water as described above does not guarantee that demands will always be met. Even worse water conditions could occur. The Northwest Power Pool uses the 1928-78 water sequence for planning, and the critical period currently being used is the worst four-year, two-year or occasionally three-year sequence of flows during that period. To determine if that period is representative of the longer term, University of Washington researchers made an independent statistical examination of the complete historical record from 1879 to the present. They concluded that the currently used critical period is not an unlikely event; in fact, it is by no means the worst possible sequence that could occur. For instance, a two-year sequence worse than the 1943-45 water sequence (two-year critical period) could occur with approx- imately a 2.3 percent probability and would have a recurrence interval of approximately every 45 years. Moreover, approximately 16 percent of the years in the 102-year record start with reservoirs less than 95 percent full, indicating potential critical-water problems for system operators. There will generally be no nonfirm energy in the years that do not refill, except for that generated by water bud- get? flows from mid-April to mid-June. Within the confines of system planning, the hydropower system can take some advan- tage of the increased energy expected above that available during the critical period. This flexibility in the hydropower system's opera- tion means that, although the total number of megawatts of energy resources planned will be determined by the critical period energy capability, the kinds of resources used should take into account the various water conditions and the seasonal pattern of water availability. Because the hydropower system's storage capability is only about 40 percent of the average annual runoff, the system's flexibility is limited. Moreover, the Canadian reservoirs, which constitute a major portion of the sys- tem storage capability, are only available to U.S. operators for limited use. The ability of the system to take advantage in the fall of the large quantities of nonfirm energy expected in the spring is limited by the risk that system operators are willing to take before the spring runoff begins. The maximum drawdown of the reservoirs for energy generation in the fall and winter is limited by the natural stream- flow during the one-year critical period, 1936-37, the lowest single natural streamflow in the historical record. The hydropower sys- tem ran against this one-year critical period until the late 1960s when the Canadian treaty projects increased the storage capability of the system and increased the length of the critical period beyond one drawdown season. Increasing drawdown in the fall and winter to the limits allowed by the 1936-37 critical period adds approximately 1,300 megawatts to the hydropower system's average firm energy capability over that eight-month period before the first year’s runoff. This addi- tional drawdown may be generically called “provisional draft,” because it is borrowed from the following spring against the expec- tation of greater than critical runoff from the snowpack. Going beyond 1,300 megawatts moves the risk of emptying the reservoirs one year forward in time, from before the second year's runoff to before the first year's runoff. In recent years, up to about 1,000 megawatts of additional first-year drawdown were taken. Over the last several years, however, in response to the water budget and to utility efforts to maintain the high drawdown levels, the Corps of Engineers and the Bureau of Reclamation have sought to maintain reser- voir refill levels. In order to protect these lev- els at the end of the first year, the Corps and Bureau have instituted further drawdown lim- its that restrict this “provisional draft” to approximately 300 megawatts in the first year. The hydropower system has one additional characteristic that is very important for the analysis of resources. The total amount of water available to the system establishes a limit to the amount of energy that can be produced. In a thermal-based power system, energy is not limited. If energy demand exceeds projections, it can still be met, if adequate capacity is available, simply by providing more fuel to the power plants. In such a system, capacity is the most critical component, and providing sufficient capacity is the major consideration in generation plan- ning. Conversely, in the hydropower system, if energy loads exceed the firm energy capability of the system during a period of adverse flows, there is no way in which demands can be met, regardless of how much installed hydropower capacity is avail- able. Hence, firm energy capability is the critical quantity in planning the hydropower system. In the Pacific Northwest power system, hydropower plants have been expanded to ensure that system peak loads can be met, that system capacity reserves will be ade- quate, and that a substantial portion of the nonfirm energy potential can be used. How- ever, this capacity would be of limited usefulness unless system firm energy resources were sufficient to meet energy demand. Although the regional power sys- tem is evolving from a hydropower-based system to a hydrothermal system, hydro- power will continue to be the dominant source. The experience has been, and fur- ther investigation is indicating that it will con- tinue to be, that the binding constraint on the Northwest power system is the total firm energy load rather than the maximum peak load. Nonfirm Strategies Figure 7-2 shows the general distribution of nonfirm energy in time. The rights to nonfirm are roughly a function of ownership or con- tractual rights to output of the dams and are generally distributed as follows: 67 percent Bonneville, 10 percent generating public util- ities and 23 percent investor-owned utilities. There are three major uses of the region's nonfirm energy at this time. The first is direct service load. The primary user is the first or top quartile of the direct service industry load which is served by nonfirm energy. There are also several smaller nonfirm loads such as electric boilers in industrial plants. The sec- ond major use is sales to extraregional mar- kets, primarily California. The third major use is to shut down thermal plants in the North- west to save the fuel cost. The large amount of hydropower available in most years, in excess of critical period amounts, offers the Northwest a resource which could be put to better use than it has been previously. Council studies for the 1986 Power Plan assessed the risks and benefits of different strategies for doing this. This chapter discusses these strategies for using more nonfirm energy to meet some of this region's firm loads: Chapter 7 1. Intermittent use of energy imported from other regions, or of combustion turbines to back up the nonfirm energy; 2. Reducing demand for electricity when necessary, through rate surcharges or contract arrangements including arrange- ments with the direct service industries. The benefit expected from these strategies would be regional savings due to reduced need for new thermal plants. Using about 700 megawatts of combustion turbines rather than the same amount of coal plants, for instance, could save the region approx- imately $175 million. The 700 megawatt value was derived using the Decision Model, which showed additional scheduling benefits not captured in the System Analysis Model. These savings take account of the existing uses of the nonfirm power. The Council believes there is potential value in using more of this nonfirm energy in the region to serve firm loads. Several strategies could be used in conjunction with nonfirm energy to meet firm loads. This section of the plan will describe several of them. The Council recognizes that these strategies will affect existing uses of the nonfirm energy; this has been considered in the analysis done by the Council. For instance, increased use of nonfirm energy to meet firm loads would decrease the amount of nonfirm energy that could be sold to the Southwest. While sales of nonfirm to the Southwest under the new Bonneville Intertie Access Policy have achieved higher prices than pre- viously, itis not clear that market conditions in California, including competition from the other Southwest states, will allow much of an increase in prices from now on. Other exist- ing uses, such as service to the direct service industries and displacement of thermal plants, have also been taken into account. 7-3 Chapter 7 System Reliability Its Implications An essential beginning to making better eco- nomic use of nonfirm energy is to examine the power system's major reliability criterion, critical water. This examination gives insights into the system's current reliability level and the effects of changes in that level. This infor- mation will suggest which strategies might be more attractive for more economic use of nonfirm energy. Following this discussion, several strategies will be discussed in depth. and The Council reviewed the effects of the criti- cal water criterion in the issue paper entitled “Critical Water Planning.” This discussion looks at some non-dollar measures of the costs of relaxing the critical. water criterion, such as the frequency and magnitude of failures to meet load, effects on reservoir refill and effects on the water budget. Summary of Results The results of the review suggest several conclusions. The first conclusion is that critical water is a somewhat arbitrary point on a continuum. It does not yield 100 percent reliability. It is important and is used as the base in the following discussion because it is embedded in a number of contracts and institutional arrangements, not because it is completely reliable. The second is that critical water is probably too stringent a criterion against which to operate the system. The hydro system could reasonably provide another 500 megawatts of energy for firm loads without backup gen- eration before it is significantly stressed. A further 500 megawatts would not impose insurmountable problems but would be the limit of reasonable operation. The major caveat to this conclusion is that changing the critical water standard will significantly affect service to nonfirm loads such as the direct service industry top quartile, especially if the interruptible loads are increased. They are direct competitors for the water. Moreover, combustion turbines can operate eco- nomically up to the level of about 700 mega- watts in the absence of firm deficits. The decision as to whether it is desirable to move 7-4 off of critical water is very sensitive to the imputed cost of meeting or curtailing load, particularly top quartile loads. The Council has maintained the critical water standard in its analysis for the 1986 plan. The third conclusion is that if the system were planned to operate using nonfirm energy to meet firm loads, some institutional mecha- nisms would need to be put in place to restrict or meet load during periods of extended low streamflow. This could be quite complicated institutionally, especially to the extent it involves adjusting loads down to available resources. Moreover, as was noted during public com- ment on the original issue paper, the model- ing may not convey the extent of this difficulty. System managers would naturally be risk conscious. They would call for load restriction measures in advance of actual need, with the expectation that longer, smaller restrictions would be preferable to shorter, more severe restrictions. Because of the uncertainty in the fall about the spring runoff, this action would often lead to restrictions that would not be justified by later events, just as most people's expenditures for home fire insurance are never “justified” by a fire. The model was risk neutral, in the sense that it simply failed to meet load, but only when resources were actually not available. The fourth conclusion is that if the region were to plan to operate using nonfirm to meet firm loads, it would probably be wiser to limit the adaptive changes in system operation to avoid too much risk. The fifth point of the analysis suggests that planning to take more risks with the hydro- power system for higher and less likely loads would probably pay off. This conclusion stems from the Councils Decision Model, which shows the appropriate build level for resources is about the mean value, 50 per- cent of the load range, rather than some significantly higher level. In this model the costs of operating to higher than expected loads, given the build level, are weighted by the low probability of the loads. Finally, a cautionary note is needed about the institutional matrix. Critical water is bound up in anumber of institutions and will need to be examined more extensively by other parties before there is a possibility of widespread acceptance of any departures from it. Background In the 1983 plan the Council assumed critical water as the basis for all its studies, although it indicated it intended to review that decision for the 1986 plan. Since then, studies by other agencies have skirted the issue of criti- cal water planning, while not addressing it directly. Bonneville did such a study as part of its 1984 resource strategy decision not to budget for options against high load growth. That study did not plan resources to meet firm loads in high load growth cases, but assumed that all curtailments indicated by the System Analysis Model were met by high cost purchases from some undetermined source. For clarity, this analysis assumes a situation where no emergency resources are available to meet load. It focuses on potential failures to meet load. If emergency resources are presumed to be available, they can be used to meet the load or as a base from which further steps away from critical water can be taken. The value of combustion turbines and the availability of other backup resources are explored later in this chapter. Finally, this dis- cussion is not a study of potential capacity problems but only energy problems. Analysis The analysis looked at the results from just one operating year (rather than the usual 20 years) and at five levels of loads and resources—a balanced case, and depar- tures from critical water (“firm deficits’) of 500, 1,000, 1,500 and 2,000 megawatts— to show in detail the effects of going off critical water by varying amounts. As described above, the system is operated in relation to the “critical period” —the historical period in which the lowest water sequences occurred. Usually the critical period is either four or two years (42 months or 20 months) long. The shorter period had more severe droughts than the longer. The two periods are almost identical Chapter 7 in the amount of hydropower generation (Firm Energy Load Carrying Capability, or FELCC) they can produce from natural flows plus complete draft of the reservoirs from full to empty. Atwo-year critical period is more likely, easier to conceptualize and, for convenience, was the period modeled in the System Analysis Model. Any operating year which starts with reservoirs full in September is a “first” year for operating purposes. Other years are “sec- ond” years. The single operating year for which results are shown below was modeled as the third operating year in a sequence of four. Depending on the water conditions drawn by the model, it either started full or not full and thus sometimes was a “first” year and sometimes was a “second” year. Thus, if the four water conditions drawn for the four-year sequence were all good, the system refilled at the end of each year, and each year was a “first” year. If the four water conditions were bad, the system started full with a “first” year and failed to refill three times, giving three succeeding “second” years. Most of the studies used three different oper- ating strategies that successively increased the adaptability of the system to operate with a firm deficit. The first strategy embodies the provisions of the Pacific Northwest Coordina- tion Agreement,3 which do not allow borrow- ing of water from later years of the critical period to cover firm deficits in the first year. The second strategy allows such borrowing but maintains the current restrictions imposed by the Corps of Engineers and the Bureau of Reclamation on maximum hydro- power energy generation in the first year of the critical period. These restrictions limit the amount of “FELCC shift” or borrowing of water from later years of the critical period to the first year. The third case allows such bor- rowing and increases both the first year and annual generation limits by the amount of the firm deficit. This strategy acts as though there actually is more water in the river than before and goes the furthest in treating firm deficits as though they are balanced situations. The first of the major variables summarized in this section is firm load curtailment. Firm load generally includes three quartiles of the direct service industry load. If the top quartile was previously served by borrowed water in the fall, there could be a restriction right against the third quartile, in which case it would not be considered a firm load, but rather a proxy for the top quartile. (Whether this occurs or not depends on whether the reservoirs started full and on subsequent water conditions.) The second major variable is the direct ser- vice industry top quartile service. The top quartile is served by borrowed nonfirm in the fall and by priority access to nonfirm in the spring. Itis particularly vulnerable to changes in the reliability level, since firm load deficits have a higher priority than nonfirm and because the borrowing for fall service is con- tingent on prior reservoir refill. The third major variable is system refill. This is particularly of interest to the Corps of Engineers and the Bureau of Reclamation, the owners of the major U.S. storage proj- ects. They are concerned that some of the multiple purposes for which reservoirs are operated, such as summer recreation, would be jeopardized by more frequent failures of the reservoirs to refill. The fourth major variable of interest is the water budget flows for fish in the spring. The results for various indicators of interest are plotted in Figures 7-3 through 7-14. Each figure is described below. The distribution of the results from 500 simulations is presented in a duration plot, a description of which fol- lows (refer to Figure 7-3, “Firm Load Service’). A plot of ideal operation, showing no failures to meet load, would go straight across the top to 100 percent on the horizontal axis and would not drop at the right-hand end. The area above and to the right of the lines repre- sents failure to meet load; the larger this area is, the more frequent and severe are the failures. Point A, the intersection of 0.9 on the vertical axis, 0.95 on the horizontal axis and the line labeled “Deficit: 2,000” in the legend, shows several perspectives on operating with a 2,000 megawatt firm deficit. First, at least 5 percent of the time, the average sea- sonal curtailment is greater than 10 percent of firm load (about 2,600 megawatts in the relatively high load used for the analysis). Viewed another way, 95 percent of the time the average seasonal curtailment is less than 10 percent of firm load and the load service is greater than 90 percent of firm load. Curtailments are calculated as averages over a four-month season. It is important to note how this averaging affects the results. Aver- aging the original monthly results tends to treat curtailments caused by failures of ther- mal plants more realistically, since some of the results could be mitigated in the short term by drawing more water from the reser- voirs. However, the averaging process masks potential large curtailments caused by emp- tying the reservoirs. The lower the reservoirs are, the more the system depends on natural flow, i.e., rainfall and snowmelt, to meet load. In these cases, large month-to-month curtail- ments are entirely realistic. This latter situa- tion becomes more likely as the region departs further from critical water. Refill plots tally July 31 reservoir elevations, and fish flow plots tally average May flows. Note that the scales in the various graphs differ from each other; most scales are trun- cated above zero. Figure 7-7 shows some perspective on the firm curtailment plots by using a 0-100 percent scale rather than the smaller scale employed for clarity in the ear- lier figures. 7-5 Chapter 7 Percent Service 1.9 |Curtailment (MW) 0.9 0.8 0.7 0.6 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 Percent Of Time Exceeded 7-6 —o “047% ""1:500 —=—"=500 owmmmm= 2,000 “mmm 400 Figure 7-3 Firm Load Service—Current Rules Percent Service 1.0 0.9 0.8 0.7 0.6 = 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 Percent Of Time Exceeded —po °*""* 1500 ae: 000 swmms 2,000 1,000 Figure 7-4 Firm Load Service—Shift for Deficit The balanced case is not a guarantee against failure to meet load, as can be seen from the plot. This is a function of several things. First, some load uncertainty around a forecast load is realistic and is modeled in the System Analysis Model. Second, in plan- ning, an expected availability for thermal plants is used. The model treats plant avail- ability realistically as a random variable, with a distribution of possible states each month. Third, and importantly, critical water is defined to be the worst historical water sequence, not the worst possible water sequence.4* The System Analysis Model allows repeated and out-of-sequence selec- tion of water years, which simulate this effect. Figure 7-3 has been described above. It shows that the curtailments under a 500 megawatt firm deficit are not very different from those under a balanced situation. The differences start getting more significant above 1,000 megawatts. Figure 7-4, plotted on the same axes as Fig- ure 7-3, shows little difference using a slightly more flexible operating strategy. Figure 7-5, again on the same axes, shows the results of a considerably more flexible operating strategy. With this strategy, which attempts to operate to a firm deficit as though the system were in balance, more nonfirm sales are made in the fall than in the previous two examples. Both the frequency and the magnitude of the subsequent failures to meet load increase. This result shows up when the lines shift down toward the lower left corner of the plot. In general, the annual nonfirm sales are not larger because the increase in fall sales in good years is completely offset by the decrease in winter sales in bad years. This does not appear to be a good operating strategy for dealing with the deficits. Under repeated poor water conditions it can lead to empty reservoirs at the time of the region's winter loads, which are the highest of the year. Figure 7-6 directly compares two operating strategies for a 500 megawatt case. The first strategy is that shown in Figure 7-3, charac- terized as “Current Rules,” while the second is a hybrid between the second and third strategies described above. It allows the bor- rowing of water for deficits and an increase in the maximum first-year hydropower genera- tion without attempting to operate the system completely to a larger amount of hydropower generation than can be met under critical water. This plot shows this operating flexibility decreases the magnitude and duration of the small curtailments but increases that of the large curtailments. By drafting reservoirs deeper in the first year of the critical period, early curtailments are avoided. However, this occurs at the expense of larger curtailments later, in those cases when reservoirs fail to refill followed by poor water conditions. Chapter 7 Percent Service 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 Percent Of Time Exceeded re2227" 4500 somn=" 2,000 Figure 7-5 Firm Load Service—Shift for Deficit Full Adjustment to Annual and 1st Year FELCC Percent Service 1.0 0.9 0.8 0.7 0.6 ‘0.80 0.85 0.90 0.95 1.00 Percent Of Time Exceeded — | vseres 9 (1) Shift for Deficit + 500 MW 1st Year FELCC, (2) 500 MW Firm Deficit Figure 7-6 Firm Load Service—Current Rules 7-7 Chapter 7 Percent Service 1.0 0.8 0.6 0.4 0.2 0.0 0.0 0.1 0.2 0.3 04 05 06 0.7 08 0.9 Percent Of Time Exceeded 1.0 —-O om 500 1,000 279°°*- 1500 ===== 2000 Figure 7-7 Firm Load Service, Current Rules (Full Scale) Percent Service 1.0 0.8 0.6 0.4 0.2 0.0 i 0.0 0.10.2 0.3 0.4 05 06 0.7 0.8 0.9 Percent Of Time Exceeded 1.0 some 4 000 7-8 0 +7777°4 500 =e 5090 =" 2,000 Figure 7-8 Top Quartile Service Figure 7-7 puts the previous figures in per- spective. It is simply a replotting of the data from Figure 7-3 onto axes that run from 0-100 percent rather than the close-up views in the previous figures. Figure 7-8 shows annual average service to the direct service industry top quartile on a full scale of 0-100 percent under the base operating strategy. The step characteristics of the data are probably due to the model's logic, which serves either all or none of the top quartile for a season, depending on water conditions. This plot clearly shows how top quartile nonfirm service degrades with increasing levels of firm deficit. Chapter 7 Figure 7-9 shows another indicator of inter- est, system refill. This is a plot of July 31 reservoir contents under the base operating strategy. It suggests there is not much impact on system reservoirs when deficits remain under 1,000 megawatts. Figure 7-10 shows the impact on system res- ervoirs under the third operating strategy, complete adaptation to firm deficits. These results begin to show significant changes from the balanced case even at the 500 megawatt firm deficit level. Percent Full 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 ] 0.0 0.1 0.2 03 04 0.5 06 0.7 08 0.9 1.0 Percent Of Time Exceeded 0 rererr"” 1500 =omew= 500 sammm=m= 2,000 Lali Figure 7-9 System Refill Percent Full 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.0 0.1 0.2 03 04 05 06 0.7 0.8 0.9 1.0 Percent Of Time Exceeded ne ——"= 500 eee 1000 «*r*""" 1 500 aoamme 2,000 Figure 7-10 System Refill, Shift for Deficit Full Adjustment to Annual and 1st Year FELCC Chapter 7 Flows KCFS 500 400 300 200 100 0 0.0 0.1 0.2 0.3 04 05 06 0.7 0.8 0.9 1.0 Percent Of Time Exceeded a 0 eeepeneneneer 1,000 sown 2,000 See 000 St ta -* 1,500 Figure 7-11 Priest Rapids Flow Flows KCFS 200 175 150 125 100 75 50 0.0 0.1 0.2.0.3 0.4 05 06 0.7 0.8 0.9 1.0 Percent Of Time Exceeded 0 umes 4000 "== 2,000 wm 500 "7" 1,500 Figure 7-12 Lower Granite Flow 7-10 Figures 7-11 and 7-12 show the fish flow at the two check points during May under the base or “Current Rules” strategy. The hori- zontal lines indicate the level of the water budget. The increased firm deficits appear to have little effect at Priest Rapids and some effect at Lower Granite only above the 1,000 megawatt deficit level. It should be noted that the lower limit of the flows in these studies is the natural flow. If the water were used to refill reservoirs for future firm loads, as is likely, the natural flow would not be simply passed and the effects on fish migration could be much more severe. While the water budget has a higher priority than refill in the Fish and Wild- life Program, the program did not contem- plate large firm deficits at the time it was adopted. Chapter 7 Figures 7-13 and 7-14 show the same thing for the fully adaptive third strategy. While the degradation remains small for Priest Rapids, ‘it becomes more severe even at the 1,000 megawatt level for Lower Granite. Flows KCFS 500 400 300 200 100 0 — 0.0 0.1 0.2 03 04 05 0.6 0.7 0.8 0.9 1.0 Percent Of Time Exceeded 0 wen 4 QQ “== 2,000 —— 500 +7774 500 Figure 7-13 Priest Rapids Flow, Shift for Deficit Full Adjustment to Annual and 1st Year FELCC Flow KCFS 200 175 150 125 100 75 50 0.0 0.1 0.2 03 04 05 06 0.7 08 0.9 1.0 Percent Of Time Exceeded 0 wenn 4000 =-====" 9 000 m= 500 77777" 1,500 Figure 7-14 Lower Granite Flow, Shift for Deficit Full Adjustment to Annual and 1st Year FELCC 7-11 Chapter 7 Conclusions Several conclusions can be drawn from the plots. For a 1,000 megawatt firm deficit, the firm curtailments differ substantially from the balanced case. With increasing levels of firm deficit, the direct service industry top quartile service degrades seriously. There would be a major incompatibility between any plans to increase the interruptibility of the direct ser- vice industries and go off critical water at the same time. System refill begins to be affected between the 500 and 1,000 megawatt level, especially if steps are taken to operate the system with the deficits. The ability to meet fish flows is apparently not seriously affected by the increased firm defi- cits, unless a complete adjustment of operat- ing procedures is made, although this result needs further examination. The analysis does not include the effect of imposing addi- tional firm requirements for fish passage spill. Such potential requirements are assumed to be interim measures only. Finally, although additional operating flexibility (the third strat- egy described above) would seem to be an advantage given operation to firm deficits, it probably is not. A complete adaptation increases curtailments, decreases the prob- ability of refill and probably adds little nonfirm revenue, although there may be other operat- ing strategies that would be more effective. Even a partial adaptation appears to in- crease the magnitude and frequency of large curtailments while decreasing that of the smaller and less serious curtailments. Institutional Issues Formal departures from the critical water standard, to the extent they demand different operating procedures, would involve changes to the Coordination Agreement and probably the Canadian Treaty. Informal departures may not, although the extent of departure would depend on mutual accept- ability to all parties, including the reservoir owners. 7-2 The analysis has generally assumed current flow requirements for downstream fish migra- tion. There is currently a rough working agreement among the entities responsible for fish, power and reservoir conditions over the use of the water. Major policy shifts toward operating with firm deficits could require renegotiation of that rough agree- ment. The distribution of the water in the region would make the situation more complex. Bonneville and the generating public utilities have approximately 75 percent of the nonfirm energy in the region, but the investor-owned utilities are expected to be deficit first. The match of the resource to the load is not partic- ularly good. This issue is discussed further in Volume |, Chapter 1. In addition, the Corps of Engineers and the Bureau of Reclamation have different per- spectives on the costs and benefits than the utilities and the region's ratepayers. They are generally concerned with the refill of the sys- tem reservoirs for non-power reasons. While the deficit operating levels proposed in this section are not severe in their impacts on system refill, the Corps and the Bureau have not been willing to make many concessions on earlier issues involving refill of their reservoirs. Strategies for the Increased Use of Nonfirm in the Region There are two major kinds of strategies to achieve increased use of nonfirm energy in the region. The first uses generating resources with low capital cost and relatively high variable costs that can be displaced by nonfirm energy whenever it is available. In this case, the net cost to the region would be a relatively low fixed cost and some weighted average of the high variable cost and the lost nonfirm sales revenue. The example of this strategy examined in detail by the Council is the use of additional combustion turbines in the region. Other examples would be high- cost purchases from out of region—for example, British Columbia or California—to back up nonfirm energy. The second approach involves reductions in demand for electricity whenever the nonfirm energy is not available. This kind of approach could involve either temporary rate increases to reduce demand or a contractual right to reduce service in exchange for a payment. Increasing the interruptibility of the direct service industries is an example of this approach. The Council has done its resource portfolio studies for the draft plan using combustion turbines as the only nonfirm strategy. How- ever, the Council believes that any combina- tion of several uses of nonfirm to meet or reduce firm loads would be economical up to a maximum of about 700 megawatts. Backup Generation: Combustion Turbines and Extra-Regional Purchases The Council has analyzed the use of com- bustion turbines in detail during the prepara- tion of the 1986 plan. This analysis was originally presented in an issue paper on “Combustion Turbine Cost Effectiveness.” Since the original issue paper was com- pleted, some of the cost and financing assumptions for coal plants and combustion turbines have been revised. Also, the System Analysis Model, used for this analysis, has been modified to assure a more accurate simulation of the power system. Because of these changes in assumptions and refine- ments in the model, the cost effectiveness of combustion turbines has been re-examined. Background One of the major issues of the 1983 plan was the use of combustion turbines in the region's resource portfolio. The benefits of combus- tion turbines can be evaluated in two ways: @ their cost effectiveness based on system operation; and @ their value in planning as a hedge against long-term load uncertainty. Chapter 7 In the 1983 plan, the Council recommended that combustion turbines be included in the resource portfolio only as a hedge against higher-than-expected rates of load growth. At that time it was unclear whether combustion turbines were cost effective on an operational basis. The Council did recognize, however, that short lead-time resources have a signifi- cant value in planning for unexpectedly high load growth. They recommended, therefore, that 1,050 megawatts of combustion turbine energy be included in the medium-high and the high load forecast scenarios. Since the 1983 plan, fuel prices and their assumed real escalation rates have fallen, as have capital cost assumptions and their escalation rates. Because of these changes in assumptions and changes to the simula- tion model used for the analysis, the cost effectiveness of combustion turbines on an operational basis has been re-examined. Analysis The cost effectiveness of individual re- sources can only be determined by consider- ing how they integrate with the entire system. Cost effectiveness is a relative quantity— that is, a resource is cost effective if it pro- duces power at an “incremental system cost” less than another resource. As was done for the 1983 plan, the cost effectiveness of com- bustion turbines was determined by com- parison to coal plants. The System Analysis Model, used for the analysis, probabilistically simulates the oper- ation of the region’s power system to meet loads. For this analysis a comparison was made between two systems, one which met load growth with coal plants and the other which met load growth with combustion tur- bines. Total system costs were compared to compute net benefits. The comparison included the benefits of current uses of non- firm power. This analysis was done for differ- ent levels of installed resource energy in order to determine the maximum amount of combustion turbine energy to include in the resource mix. Table 7-1 Assumptions ITEM Coal Life Capacity Availability Capital cost (millions) Variable fuel cost ($/MMBtu) Fuel real escalation rate Variable O&M (cents/kWh) Single Cycle Combustion Turbine Life Capacity Availability Capital cost (millions) Variable fuel cost ($/MMBtu) Fuel real escalation rate Variable O&M (cents/kWh) Other Thermal Data Assumptions Sponsorship Debt/equity ratio Capital cost real escalation rate Variable O&M real escalation rate Curtailment Costs Firm load curtailment cost Firm curtailment cost real escalation rate Nonfirm load curtailment cost Nonfirm curtailment cost escalation rate Southwest Market Standard rate for Southwest sales Maximum rate for Southwest sales (approx.) Southwest price real escalation rate Southwest intertie capacity (by October 1989) ASSUMED VALUE 40 years 603 MW 75% $757 2.0 1.0% at 30 years 210 MW 85% $53 5.1 1.8% 21 Private 80%/20% 0.4% 0.0% 5.67 cents/kWh 1.8% 2.20 cents/kWh 0.0% 2.20 cents/kWh 3.00 cents/kWh 1.0 to 1.8% 7,786 MW 7-AB Chapter 7 Table 7-2 End Effect Corrections for Combustion Turbine Studies (Millions of Dollars) 178 MW 356 MW 534 MW 712MW Capital 11.0 22.0 33.0 44.0 Fixed Costs 47 9.4 14.1 18.8 Variable Costs 31.7 75.4 163.2 257.9 Total Adjustment 47.4 106.8 210.3 320.7 Table 7-3 Net Benefits INSTALLED MW 178 MW 356 MW 534 MW 712MW Net Benefits (millions) $112 $171 $194 $ -32 Capacity Factor Coal (%) 56.5 56.3 57.0 57.6 Turbine (%) 94 10.8 15.6 18.8 Degradation of Service Top Quartile (%) -2.5 -5.4 5.5 6.8 Top Quartile (MW) -16 -35 -36 45 S.W. Sales (%) -2.4 -4.8 7.1 -8.8 S.W. Sales (MW) -68 -132 -192 -237 The existing thermal resource mix was used, along with a set of loads which yielded a 2,550 megawatt surplus in the first year that decreased to a balanced condition by 1994. From that point until the end of the study period, the load/resource balance was approximately zero (slightly surplus). An incremental load growth in September of 2000 was met by the installation of an equal amount of coal or combustion turbine energy. Four scenarios were examined: one in which 178 megawatts of generic resource were used to meet load growth, one with 356 megawatts, one with 534 megawatts, and one with 712 megawatts. Two studies were performed for each scenario, one to deter- mine system costs when coal was added and the other to determine system costs when combustion turbines were added. In order to compare equal amounts of coal and com- bustion turbine energy, a scaled-down coal plant was used. For this analysis, each coal plant had capital and operating costs based 7-14 on the standard 603 megawatt capacity unit scaled down to a capacity of 237 megawatts, with an average availability of 75 percent (to yield a net energy of 178 megawatts). Each combustion turbine had a capacity of 210 megawatts with an average availability of 85 percent (to yield a net energy of 178 megawatts). Undeclared existing combustion turbines were removed from the analysis along with the option to make out-of-region emergency power purchases. The resource mix includes 164 megawatts of existing firm combustion turbine energy. Assumptions Plant operating data and assumptions were obtained from the Thermal Resources Data Base publication (Pacific Northwest Utilities Conference Committee, October 1984). This data was updated to represent January 1985 values. Financial data assumptions are in Chapter 4 of this volume. Table 7-1 below lists other assumptions used for this analysis. End Effects In this analysis, an obvious end effect prob- lem exists due to the different assumed lives of the two resources being compared. All of our cost analysis is based on the present value life cycle net revenue requirements computed by the System Analysis Model. Life cycle costs are based on a projected resource operation beyond the study period. During that period, the assumed operation of each resource is based on its average opera- tion during the last five years of the study period. Operating costs are computed for each year that the resource is in existence. In this analysis, the combustion turbines expire ten years before the coal plants. The net revenue requirements for the coal studies, therefore, contain an additional ten years of operating costs. To compensate for the shorter combustion turbine life, it was assumed that when the turbines expire, new combustion turbines would replace them. A separate calculation was made to determine the present value capital and operating costs of the first ten years life of the replacement combustion tur- bines. It was assumed that the replacement turbines would operate at the same average capacity factor as the expired turbines. These additional costs were added to the present value life cycle net revenue require- ments for the combustion turbine studies and are reflected in the summary of net benefits in Table 7-3. Adjustment costs for the combus- tion turbine studies due to end effect errors are summarized in Table 7-2. In addition to this, the System Analysis Model only projects secondary revenues and curtailment costs 25 years beyond the study period. Since the new coal plants were installed in September of 2000, their operat- ing costs are projected 35 years beyond the end of the study period, ten years beyond the point where the secondary revenues and cur- tailment costs stop. In order to correct this end effect, the model was modified to extend the projected secondary revenues and cur- tailment costs to 35 years beyond the study horizon period. Chapter 7 Results Net benefits for combustion turbines were computed for each scenario and are summa- rized in Table 7-3 and in Figure 7-15. The optimum amount of additional combustion turbine energy to include in the resource mix is determined by the point where the benefits are greatest. That point appears to be some- where close to 500 megawatts of additional combustion turbine energy. Of course, there are also operational effects to consider. Since the variable cost of com- bustion turbines is greater than the South- west is willing to pay for nonfirm power, tur- bines are never operated to meet Southwest nonfirm loads. Northwest coal plants, on the other hand, are generally operated to meet nonfirm loads because their variable operat- ing costs are lower. Thus, aS more combus- tion turbines are used to meet firm load growth (instead of coal plants), the service to the Southwest nonfirm markets will decrease. Figure 7-16 shows how the South- west market would be affected by a combus- tion turbine scenario (based on comparisons to acoal scenario). At534 megawatts of addi- tional combustion turbine energy, Southwest sales decrease by about 7.1 percent (192 megawatts) on an expected value basis. Service to the direct service industry's top quartile load is also affected in the combus- tion turbine scenario. The impact is relatively small, however, because the top quartile is served in the fall by borrowing rather than by nonfirm in every case when the hydro system refills, whether coal or combustion turbines are used in the mix. Combustion turbines are displaced by available nonfirm in the fall. In the winter and spring, nonfirm hydropower is used to displace combustion turbines prior to using it to serve the top quartile; but since the availability of nonfirm hydropower in these periods is significantly larger than in the fall, the relative priority has a small effect. Coal plants, on the other hand, would generally be operated to meet the direct service industry's top quartile load. Under a coal scenario, therefore, the service to the top quartile should not be affected. Figure 7-16 also shows how service to the direct service industry's top quartile load is affected for a combustion turbine scenario. At 534 mega- watts of additional combustion turbine energy, top quartile service drops by about 5.5 percent (36 megawatts) on an expected Percent 250 200 150 Maximum Benefit 100 $210 Million At 480 50 MW O Gosrccccccchesscccccecepecersssceregeresrssee -50 0 175 350 534 712 Additional Combustion Turbine Megawatts Figure 7-15 Net Benefits of Combustion Turbines vs. Coal (Based on Comparison to Coal Plants) Percent 100 90 80 70 At 534 MW: Percent Megawatts Top Quartile 5.5% 38 MW 60 S.W. Sales: 7.1% 192 MW 50 _t 0 175 356 534 712 Additional Combustion Turbine Megawatts @—mm Top Quartile Service “//7 Southwest Sales Figure 7-16 Impacts to Southwest Sales and Top Quartile Service (Compared to a Coal Scenario) 7-15 Chapter 7 Percent 100 90 80 70 60 50 40 30 At 534 MW Coal-57.0 Combustion 20 Turbines-15.8 STS) 10 epererreetssssTsfs@- ALLELE 0 0 175 356 534 712 Additional Combustion Turbine Megawatts mmm Coal Plants ‘77/7 Combustion Turbines Average Capacity Factor of Coal and Combustion Turbine Plants Figure 7-17 (All Additional Units) Rate 1.3 1.5 2.3 7-16 Percent Benefit 309.0000 194.0000 62.0000 0 100 200 300 400 Millions Of Dollars Figure 7-18 Sensitivity to Fuel Escalation Rate (At 534 Megawatts of Additional Turbine Energy) value basis. Unserved top quartile load was valued at 2.04 cents per kilowatt-hour in Jan- uary 1985 dollars. As the number of combustion turbines in the resource mix increases, the average capacity factor for the turbines also increases, i.e., they are operated more often. This occurs because of the limited amount of nonfirm hydropower which can be used to displace them. Once this nonfirm hydropower is used up, the remaining turbines must be operated to meet firm loads. As their capacity factor increases, the benefits quickly drop because their operating costs are so high. Figure 7-17 depicts the change in capacity factor for both coal plants and combustion turbines as a function of installed energy. At 534 mega- watts of installed energy, the additional tur- bines are operated about 15.6 percent of the time on an expected value basis. Coal plants operate at about a 57.0 percent capacity factor. Sensitivity Analysis Coal plants have very high capital costs but moderately low operating costs, whereas combustion turbines have low capital costs but very high operating costs. Obviously, the advantage of combustion turbines is their low capital cost. Thus, turbines can be cost effec- tive compared to coal as long as they are not operated at a high capacity factor. Any change in assumptions, which affects plant operation or plant capital cost, may cause the benefits to change significantly. The cost effectiveness of combustion turbines, there- fore, should be very sensitive to coal capital cost and capital real escalation rates and to turbine fuel prices and their escalation rates. The size of the Southwest market and the price that the Southwest is willing to pay for nonfirm power will also affect the cost effec- tiveness of combustion turbines. Normally a coal plant would be operated to serve South- west nonfirm loads, whereas a combustion turbine would not because of its high operat- ing costs. Revenues received from sales of coal generation to the Southwest offset other system revenue requirements. Thus, the greater the Southwest market and/or the higher the price, the greater are the benefits of having coal in the resource mix. Chapter 7 Conclusions Based on this analysis, it appears that the addition of about 500 megawatts of combus- tion turbine energy to the existing resource mix would provide the region a net benefit of about $210 million. (Recall that 164 mega- watts of firm combustion turbine energy is already present in the existing resource mix.) This analysis examines the value of combus- tion turbines on an operational basis only. A similar analysis was done using the Deci- sion Model, to show the value of combustion turbines in the portfolio. The conclusion was that 700 megawatts of new combustion tur- bine energy would be cost effective, with a benefit of $175 million. Since the two models are different in structure and complexity, a test was made of the operation of combustion turbines and coal plants in the two models, while forcing the cases to be as similar as the models allowed. The results from the two models, when presented with similar cases, were very close. The results used in the final portfolio were those from the Decision Model. The Decision Model showed benefits for a larger amount of combustion turbine energy, because it valued scheduling advantages inherent in short lead times, small plant size, and low capital cost in relationship to load forecast uncertainty, which the System Anal- ysis Model did not account for. Fuel Use Act Exemptions to the Powerplant and Industrial Fuel Use Act are assumed to be available for combustion turbines. Through research of the Fuel Use Act and its regulations and through informal consultations with the U.S. Department of Energy's Economic Reg- ulatory Administration, the Council deter- mined that permanent exemptions most likely to be obtainable for the uses of com- bustion turbines envisioned under the Coun- cil’s plan are those available for: @ maintenance of reliability of service; @ lack of alternate fuel at a cost not substan- tially above that of imported oil; ® cogeneration; @ fuel mixtures; and © peaking. Exemptions are granted only for proposed plants actually designed and nearing con- struction, and each exemption requires certain showings by the applicant as pre- requisites. Should the region decide to include an addi- tional amount of combustion turbine energy in the portfolio, rather than some other strat- egy for using nonfirm, that energy could come from existing undeclared combustion turbines. Many of those turbines are “grand- fathered” under the Fuel Use Act and may not face the problems associated with obtain- ing exemptions to the Fuel Use Act. Other potential sources of back-up genera- tion are out-of-region utilities, either in Califor- nia and the Southwest or in British Columbia. There is a large amount of oil-fired generation in California which may be available. Although no energy was available from Cal- ifornia during the two times in the 1970s when the region needed it (due to the oil embargo in 1973 and overlapping droughts in 1977), there is some indication that the cor- relation between the Northwest snowpack and the Sierra snowpack is small to nonexis- tent. There appears to be a similar indication about the correlation of the Northwest with the Peace River in British Columbia. Both these areas need further investigation before firm conclusions can be drawn. Bonneville is currently looking at the British Columbia sit- uation. The Council expects to learn more about the potential during the next two years as part of its West Coast energy study. Using Nonfirm Energy without Backup Generation: Load Management An alternative approach to the increased use of nonfirm involves simply attempting to meet firm loads using nonfirm energy without backup generation. This would require some institutional mechanism for reducing loads when no nonfirm energy is available. This issue was addressed in the Council issue paper, “Critical Water Planning,” and has been expanded for the plan. Aset of studies estimated the dollar benefits of relaxing the critical water criterion and the dollar costs of curtailment alone. These monthly plant life-cycle cost studies used the same five levels of load/resource balance described above in the system reliability sec- tion in approximately the last third of the 20- year simulation. Generally, the question of critical water is only significant when the tegion faces deficits and potential major resource acquisitions. Therefore, the study simulated some more-likely load cases than the high load. These studies used only the first (current) operating strategy. They are summarized in Tables 7-5, 7-6 and 7-8, which give 1985-dollar present values of the results. The reliability plots (described in the section on critical water studies) were done because it is often difficult to establish a common denominator to compare such things as changes in refill and flows for fish. While there have been several studies of the cost to con- sumers of curtailments, including one done for the Council in 1982, there is no general agreement about that either. The three stud- ies described below use three different approaches to the cost of curtailment. It is clear from the results that the conclusions are extremely sensitive to the imputed cost of curtailment and, in particular, to the cost of curtailing the direct service industry top quar- tile. This is so because service to the top quartile in the fall using the flexibility of the hydro system is dependent upon load/ resource balance. With firm deficits, their fall service is dependent on secondary availabil- ity only. TAZ Chapter 7 Based on the previous analysis, it was Table 7-4 observed that, of all the assumptions which End Effect Adjustments for Sensitivity Study could affect combustion turbine cost effec- (For 534 Megawatts of Additional Combustion Turbines) tiveness, coal capital costs and combustion (Millions of Dollars) turbine fuel escalation rate assumptions a aa GAS ESCALATION RATE DEBT/EQUITY RATIO. were the most sensitive. Two sensitivity stud- ies were designed to determine the effects of 2.3% 1.3% 50% changing these two assumptions. Both sen- Capital 33.0 33.0 “39.9 sitivity studies were performed for the 534 . megawatt scenario. Fixed Costs 14.4 14.4 14.4 Variable Costs 181.2 148.2 163.2 The first parameter examined was the com- bustion turbine fuel real escalation rate. Two Total Adjustment 228.3 195.3 217.2 studies were performed: one with an escala- tion rate of 2.3 percent (an increase over the base case of almost 30 percent) and the other with an escalation rate of 1.3 percent. Ratio Debt/Equity Ratio Results from these studies are depicted in — Figure 7-18. Net benefits drop from $194 mil- lion to $62 million when the fuel escalation rate is increased to 2.3 percent. Inthe second study, when the escalation rate was dropped to 1.3 percent, the net benefits jumped to $309 million. Benefit 438.0000 The second parameter examined was the capital finance assumption for the debt/ equity ratio. For this study, this ratio was changed from 80/20 to 50/50 (which has the 50 | 194.0000 net effect of increasing capital costs). Results G4, for this sensitivity study are shown in Figure 7-19. Since combustion turbine capital costs are low, this change in assumptions had little effect on the cost of the turbines. A more significant change in capital costs was observed in the coal studies. Net benefits of ° 100 200 eee a = combustion turbines increased from $194 Millions Of Dollars million to $458 million (a change of 126 percent). Figure 7-19 As evident in Figures 7-18 and 7-19, the ben- Sensitivity to Debt/Equity Ratio efits of combustion turbines are very sen- (At 534 Megawatts of Additional Turbine Energy) sitive to coal capital costs and to combustion turbine fuel prices and their escalation rates. Any change to these assumptions may alter the results significantly. As in the base case, an end effect problem exists in these sensitivity studies. To correct this, it was assumed that when the combus- tion turbines expire, new turbines would replace them. Adjustments to the capital and operating costs were computed and added to the present value net revenue requirements for the combustion turbine studies. Table 7-4 below summarizes the corrections used for the sensitivity analysis. 7-18 Chapter 7 The key result of Table 7-5 is that the addi- Table 7-5 tional system revenue requirement declines Curtailment: Firm, 6.2 Cents and Top Quartile, 2.2 Cents continuously and significantly to the limit of Present Values—Plant Life Cycle the study, 2,000 megawatts of firm deficit. System Analysis Model: Millions of September 1985 Dollars The sensitivity to the curtailment cost is CASE shown in the lower part of the table. The 500 MW 7,000 MW 7,500 MW 2,000 MW underlined number identifies the lowest-cost COST Balance Deficit Deficit Deficit Deficit point. This table uses a standard secondary rate of approximately 2.2 cents, the rate as of Capital $ 4,959 $ 3,743 $ 2,529 $ 1,345 $ 192 July 1985, and models the market structure + Production 22,348 21,487 20,429 19,386 18,031 that the region is currently seeing under Bon- ; neville's near-term intertie access policy. It + Curtailment 1,167 1,853 2,787 3,992 5,506 also assumes the full expansion of the inter- —Nonfirm Revenue 12,427 11,803 11,215 10,596 9,807 tie to 7,786 megawatts. : Revenue Requirement $16,046 $15,281 $14,530 $14,109 $13,832 The base case curtailment cost (and real SENSITIVITY OF REVENUE REQUIREMENT TO CURTAILMENT COST escalation) rate used for Table 7-5 is approx- imately the same as the operating cost of Revenue Requirements if Curtailment Cost Multiplied generic combustion turbines, 6.2 — in times 1.5 $16,630 $16,028 = $15,924 $16,105 $16,585 panier 126-2 GOlats. Thi ate = appeen 2 times 2: 17,213 17,134 17,313 18,101 19,338 firm curtailments (including the firm direct . : service industry load) and escalates at 1.8 times 3: 18,380 18,987 20,104 22,093 24,844 percent per year in real terms, as does the combustion turbine fuel cost. The rate applied to top quartile and third quartile cur- ; Table 7-6 tailments is 2.2 cents in January 1985 dollars, Curtailment: All Loads, 6.2 Cents approximately the same as the direct service Present Values— Plant Life Cycle industry base rate as of July 1985. This System Analysis Model: Millions of September 1985 Dollars remains constant in real terms, a reasonable CASE expectation for the direct service industry 500 MW 1,000 MW 1,500 MW 2,000 MW rate. If the costs of curtailment are twice as cost Balance Deficit Deficit Deficit Deficit high as these, the lowest cost point shifts toa | 500 megawatt deficit, and if three times as Capital $ 4,959 $ 3,743 $ 2,529 $ 1,345 $ 192 high, it shifts to balance. + Production 22,348 21,487 20,429 19,386 18,031 . i + Curtailment 3,257 5,298 7,098 8,838 10,651 However, these results are quite sensitive to the imputed curtailment cost of the direct ser- —Nonfirm Revenue 12,427 11,803 11,215 10,596 9,807 vice industry top quartile. Table 7-6 shows Revenue Requirement $18,136 $18,725 $18,841 $18,955 $18,977 the results for the same studies as reported in Table 7-5, except that the top quartile and third quartile curtailments were valued at the same rate, 6.2 cents escalating at 1.8 per- cent, as the firm load. The difference in results occurs because the primary impact of moving away from critical water is on the direct service industry top quartile load. With firm deficits, the top quartile service in the fall comes only from nonfirm energy rather than from borrowing energy against the third quar- tile of their load. There are several perspectives that can be applied to evaluate these curtailment costs. The first perspective is that of short-run elas- ticity. This poses the question of what increase in price is needed to reduce demand in the short run by the required amount? With an elasticity of -0.1 and an assumed current average retail rate of 5 cents, a5 percent decrease in firm load could be achieved by raising prices 50 percent and a 10 percent decrease by raising prices 100 percent. The distribution of firm curtailments shows that only about 6 percent of the curtail- ments are greater than 10 percent of load at the 1,000 megawatt deficit level and only 20 percent are above 10 percent of load at the 2,000 megawatt level. This result suggests that two times the original curtailment cost (two times 6.2 cents) is not out of line with what firm customers are willing to pay for service for the low end of the curtailment distribution. This could also be construed as the amount the power system could pay con- sumers to reduce service, although this total amount would consist of both payment and foregone revenue at a level equal to the assumed rate. 7-19 Chapter 7 Table 7-7 The second perspective focuses on the Variable Cost of Direct Service Industries direct service industries. Both short-term elasticity and the required potential payment COST a . comronent are functions of the state of the aluminum Labor: 0.005 manhr/lb x 2,300 cents/manhr x 0.7 variable = 8.1 cents/lb market. In recent years, the price of alurmi- ; ; num has been as high as $1 per pound, Alumina: 20 cents/Ib x 0.85 variable = 17.0 though forecasts of long-term average prices Other costs: 22 cents/Ib x 0.55 variable =| 12:4 tend to be in the 75-80 cent per pound range. oe At 75 cents per pound for aluminum, the cost 22.) Ih I = 17.6 . ‘ ; Electricity: 2.2 cents/kWh x 8 kWh/Ib 17.6 to the power system of making the direct 54.8 cents/Ib service industries and their employees indif- ; ferent between operating and not operating 's/Ib profit = 2! I = 25 its/kWh : a ee a eae ie would be about 5.7 cents per kilowatt-hour. variable wages = 8.1 cents/Ib = 1.0 This calculation is shown in Table 7-7, using lost revenue = 22 roughly average data for the ten smelters i from the Bonneville Direct Service Industry 5.7 cents/kWh Table 7-8 Curtailment: Firm, 10.0 Cents and Top Quartile, 5.7 Cents Present Values—Plant Life Cycle System Analysis Model: Millions of September 1985 Dollars CASE 500 MW 1,000 MW 1,500 MW 2,000 MW cOsT Balance Deficit Deficit Deficit Deficit Capital $ 4,959 $ 3,743 $ 2,529 $ 1,345 $ 192 + Production 22,348 21,487 20,429 19,386 18,031 + Curtailment 2,624 4,185 5,747 7,429 9,329 —Nonfirm Revenue 12,427 11,803 11,215 10,596 9,807 Revenue Requirement $17,504 $17,613 $17,490 $17,546 $17,655 SENSITIVITY OF REVENUE REQUIREMENT TO CURTAILMENT COST Revenue Requirements if Curtailment Cost Multiplied times 1.5: $18,816 $19,706 $20,364 $21,261 $22,320 7-20 Options Study. At $1 per pound of aluminum, the cost to the power system of making the direct service industries and their employees even would be on the order of 9 cents per kilowatt-hour for about 2,600 megawatts. The cost to the power system would include paying wages and profits, as well as the lost revenue from the foregone power sales. The variable por- tion of alumina and other costs would not be incurred, so they are not included in the cal- culation. Again, this indicates that at a max- imum less than twice the original estimate of the curtailment cost (two times 6.2 cents) estimate is not unreasonable through most of the range of the curtailments, and as an expected value the original estimate is appropriate. Because of this sensitivity, and taking into account the conclusions above, the studies were run a third time with firm curtailment costs set at 10 cents and the top quartile and third quartile curtailment costs set at 6.2 cents. The costs are in January 1985 dollars and, in this case, are held constant in real terms. Note in comparing the tables that, because the base numbers are higher, the multiplied values for the same multiplier are not comparable across tables. The results are shown in Table 7-8. In addition, this table is relatively conservative, because it values firm direct service industry curtailments (half the direct service industry load) at 10 cents per kilowatt-hour rather than 6.2 cents. The System Analysis Model does not distinguish firm direct service industry loads from non- direct service industry loads in this context. Chapter 7 The cost at the bad tail of the distribution is harder to estimate. The appropriate cost curve for curtailment for very large, but infre- quent, events is probably severely non-linear. A 20 percent restriction of demand based on price alone could require a 200 percent increase in rates, with an elasticity of -0.1. The 20 percent curtailment level for firm load was exceeded only about 1 percent of the time even with a 2,000 megawatt firm deficit. A 200 percent increase in rates corresponds 1./ When the Council reconsidered its interim spill objectives in early 1986, it did not change the interim mainstem fish passage objective at mainstem federal projects but did extend the objective to cover 80 percent of the fish runs up to August 15. As a result of the change, spill will be used to meet the objec- tive even when only firm power is available. roughly to the multiplier of 3 in Table 7-5. However, this is probably an area in which the seasonal averaging distorts the results of the study, since some monthly failures to meet load could be much larger than shown on the plots, although correspondingly less fre- quent. Note on the other hand that the multi- plier overstates the cost in the table, because it is applied to all curtailments, not just the very large ones. 2./ The water budget is a means of increasing survival of downstream mein juvenile fish by increasing flow during the spring migration period. The Council proposed this practice and oversees it in conjunction with the U.S. Army Corps of Engineers, the fishery agen- cies and tribes, Bonneville, and the Bureau of Reclamation. The water budget is discussed in Section 304 of the Council's Columbia River Basin Fish and Wildlife Program. Conclusions This chapter has explored the characteristic variability of the Northwest hydropower sys- tem along with some uses of the nonfirm energy available because of that variability. The Council believes that its studies have demonstrated several strategies for increas- ing the value of the nonfirm energy to the region. The most promising strategy for the region at this point appears to be increased use of combustion turbines or extra-regional purchases to back up nonfirm energy to meet firm loads. Because of this, the Council has included about 700 megawatts of combus- tion turbine energy in its resource portfolio. Other strategies, such as load management or load buyback coupled with planning to somewhat better than critical water, should be explored further. 3./ The Pacific Northwest Coordination Agree- ment is a contract among the U.S. Army Corps of Engineers and all the Northwest enerating utilities (except the Idaho Power ompany) that governs the operations of the region's hydroelectric system. 4./ “A Synthesis Flow Model for The Dalles Flows,” unpublished letter from Dennis Let- tenmaier (UW) to Ron Hicks (BPA). Also see L.A. Dean and J.A. Polos, “Frequency of Failure to Meet Firm Loads for the Pacific Northwest Hydroelectric System,” unpublished paper, Dec. 6, 1983. 7-21 This chapter of the plan describes in detail the Councils resource portfolio. Section A describes the analysis that led to the Coun- cil's choice of the portfolio, and gives a brief overview of the computer models employed. Section B contains a description of sensitivity analyses performed on the portfolio. Section C gives details of the Council's analysis of the cost effectiveness of the two Washington Public Power Supply System's nuclear plants WNP-1 and WNP-3. Section D presents a more detailed description of the Decision Model. Finally, Section E discusses generat- ing resource lost opportunities. Section A: Resource Portfolio Analysis Introduction In the Council's 1983 Power Plan, the resource portfolio was presented as four dif- ferent regional resource schedules, one for each of the four different load forecasts. In the 1986 plan, the portfolio is presented not just as a set of particular resource development schedules, but also as a set of resource pri- orities and decision rules. These can be used in conjunction with resource availabilities and evolving load forecasts, to guide the deci- sion-making process toward the most eco- nomic resource decisions as the region's energy future unfolds. In developing this resource portfolio, the Council's primary objective was to achieve the lowest present value expected cost across the wide range of uncertainty faced by the region. In addition, because future events are not likely to turn out as forecast, the Councils portfolio continues to exhibit a high degree of flexibility, allowing opportune responses to unforeseen changes in need and thereby maintaining a reliable, economic power system. The Council believes the con- cept of risk management should play an important role in the resource decision-mak- ing process. The flexible planning strategy that evolved out of the Council's 1983 plan is emphasized again in the 1986 Power Plan. Generating resource characteristics which lead to enhanced flexibility and reduced risk are, primarily, short lead times and small unit size. Shorter lead times reduce the period over which the need for new resources must be forecast, and allow resource sponsors to move closer to the point of actual need before committing large amounts of capital for resource construction. Shorter lead times produce a greater likelihood that resources will actually be useful once they are ready for service. Resources with small plant sizes would allow the region to make many smaller decisions rather than a few large ones, and provide the ability to match resource devel- opment and load growth more closely. The concept of resource options, developed and emphasized in the Council's 1983 plan, has as an important objective the reduction of resource lead times. The option concept per- mits the region to enter into the preliminary stages of resource development, siting, licensing and design, based on a relatively high projection of future load growth. This strategy is expected to prove cost effective, because the cost of acquiring options is low compared to the cost of actual resource con- struction. The options concept leads to a second decision point regarding the appro- priate time to begin constructing a resource. After option acquisition, load forecasts would continue to be updated and the projected need for the resource reevaluated. If loads have not grown sufficiently to justify entering construction, the option would be held until it was either appropriate to construct the resource or the option was lost. The options concept enhances the flexibility of the Coun- cil’s resource portfolio and continues to war- rant additional analysis and policy develop- ment. Over the planning horizon, the ability to option resources will improve the ability to match the rate of resource development with resource need and reduce the cost of the resource portfolio. Chapter 8 Resource Portfolio Analytical Tools For the resource portfolio analysis the Coun- cil relied primarily on two computer models. The first of these, the System Analysis Model, is a large, very detailed model of the Pacific Northwest generation system. It was developed principally by the Bonneville Power Administration, the Intercompany Pool, and the Pacific Northwest Utilities Con- ference Committee (PNUCC). It uses com- plex models of Northwest hydro/thermal operation, and sophisticated techniques to capture the physical and economic effects of uncertain variables, such as hydro condi- tions, thermal plant availability, thermal plant arrival, and short-term fluctuations in load. It also uses detailed accounting methods to model the capital cost recovery streams required by the various types of utilities and resource sponsors in the Pacific Northwest. The System Analysis Model is an excellent tool for evaluating questions concerning sys- tem reliability, or to isolate the operation and cost of a particular resource and its impact on the system as a whole. In development of the plan, the System Analysis Model was used for analysis in the areas of combustion tur- bine cost effectiveness, relaxation of the criti- cal water standard, value of additional inter- ruptible load, and long-run marginal costs. A wide range of documentation for the model is available upon request from PNUCC. 8-1 Chapter 8 Load Forecasts Decision Analysis Decision Resource Portfolio Conservation & Generating Resource Availability Economic Assumptions, Figure 8-1 Northwest Power Planning Council Resource Portfolio Development Process The Decision Model was also used exten- sively by the Council for development of the resource portfolio. This relatively new model was developed over the last year by indi- viduals from the Council staff, the Intercom- pany Pool, PNUCC and Bonneville. This model grew out of the Council's recognition of a need for the analytical capability to assess the impact of long-term load uncertainty on resource cost effectiveness. Related to this was a need for the ability to value charac- teristics which would enhance resource flexi- bility, such as shortened resource lead times and resource options, or small unit sizes. The Decision Model provides the capability to assess the load-related risk associated with a particular decision to option or build a resource, and the consequences of errors likely to occur in the resource planning pro- cess. It helps planners determine what types of resource strategies over the long term bet- ter enable the region to manage the risks imposed by load uncertainty. The model enhances strategic planning capability, and provides information flow to the decision- making process in an area which previously had to rely largely on intuition and judgment. 8-2 Section D of this chapter provides additional background, an overview of the model, and a brief description of the model's algorithms. More detailed documentation should be available sometime in the spring of 1986. Together, the highly detailed nature of the System Analysis Model and the ability of the Decision Model to deal with load uncertainty provide the Council with a capacity for analy- sis over a wide array of resource planning issues. Portfolio Development Process The Council's resource portfolio develop- ment process consisted of four major interre- lated activities. These are depicted graphi- cally in Figure 8-1 and summarized below. 1. Load Forecasts. The process began with development of electricity demand fore- casts for the region. Four forecasts were developed, each representing a possible regional future. A probability distribution for future loads was also developed. In order to focus on the obligations of the Bonneville Administrator, the forecasts were also broken down into demands of the public and investor-owned utilities. Vol- ume Il, Chapters 2 and 3, provide a detailed description of the forecasting pro- cess and its results. . Avoided Cost Studies. Next, long-run marginal cost studies for the region were performed. This analysis was performed with the System Analysis Model and used coal units with arrival dates near the year 2000 as the avoided resource. These studies estimated avoided costs to be 4 to 4.5 cents per kilowatt-hour. These studies are discussed in detail in Volume |, Chap- ter 8. The System Analysis Model was also used to derive levelized cost esti- mates for initial ranking of the generating resources for the portfolio analysis. . Determination of Resource Availability. Information from the load forecasts and the avoided cost estimates were used to screen resources for the portfolio analysis. Initial estimates of the amounts of cost- effective resources were developed for generating resources and conservation programs. For many conservation pro- grams, the amount of efficiency improve- ment available depends on the level of economic activity modeled for that sector in the load forecast. This correlation between conservation availability and load level is used in the portfolio analysis. For a full discussion of the conservation and generating resource potential see, respec- tively, Chapters 5 and 6 of this volume. . Portfolio Analysis. The load forecast range, its probability distribution, and the conservation and resource availabilities and costs were used with the Decision Model to develop the Council's resource portfolio. The Decision Model is used here because it incorporates the effect of long- term load uncertainty, resource option and construction lead time, conservation pro- gram ramp rates, seasonality and system operating impacts into the cost-effective- ness analysis. The process involved sev- eral iterations back through the forecast- ing and resource screening activities to ensure consistency among the portfolio, loads and electricity prices, and conserva- tion energy potentials. After the resource portfolio had stabilized, scheduling stud- ies focused on the Administrator's obliga- tions, to determine what actions might be required in the Action Plan for Bonneville. Chapter 8 Load Treatment The third chapter of Volume I! describes the development of the four load forecasts in detail. The forecasts provide the starting point for the portfolio analysis and obviously are a critical piece of information. However, these four specific forecasts are not used directly in the analytical process. Rather, they are incorporated into the analysis through definition of the probability distribution for regional loads. As for any specific forecast, the likelihood is extremely small that future regional load will evolve exactly along any one of the four spe- cific forecast paths. However, because of the philosophy underlying their development, the forecasts can readily be used to define a probability distribution for future electricity demand. The forecasts were developed in such a way that future load outcomes either below the low forecast or above the high were felt to have probabilities so low as to justify exclusion for planning purposes. In addition, the two medium level forecasts define the range of most likely load outcomes. These characteristics can be represented with the trapezoidal probability distribution shown in Figure 8-2. This distribution, expressed in terms of 20-year compound growth rates, has a uniform probability of occurrence for loads between the medium-low and medium-high, with probabilities dropping off linearly to zero at both the low and the high. This is a continuous distribution, implying that any load outcome across the entire range would be possible. The probability of a load occurrence between the low and medium-low is 31 percent; between the medium-low and medium-high, 42 percent; and between the medium-high and high, 27 percent. Another component of load uncertainty included in the portfolio analysis is that of the direct service industries. The economic con- ditions driving this uncertainty are discussed in Volume II, Chapter 2. For analytical pur- poses, the Council has assumed at least half the load from aluminum producers will be present across the planning horizon; thatis, a minimum of 50 percent of aluminum direct service industry firm load is included in all load cases. The remaining 50 percent is regarded as uncertain and is represented by Probability 1.3 2.0 3.0 Compound Load Growth Rate (%) Figure 8-2 Load Growth Probability Distribution Probability 0 400 800 1,200 Average Megawatts Figure 8-3 Distribution for Uncertain DSI Load 8-3 Chapter 8 Average MW 12,500 10,000 7,500 5,000 2,500 0 -2,500 -5,000 1 1986 1988 1990 1992 1994 1996 1998 2000 2004 2006 Year Figure 8-4 Regional Resource Requirements Average MW 12,500 10,000 7,500 5,000 2,500 0 -2,500 -5,000 1986 1988 1990 1992 1994 1996 1998 2000 2004 2006 Year 8-4 Public Utility Figure 8-5 Resource Requirements the probability distribution shown in Figure 8-3. This is a discrete distribution, with only four possible outcomes: 0, 400, 800, and 1,200 average megawatts of firm load above the 50 percent base, with a constant 25 per- cent probability for each outcome. The dis- crete nature of the distribution is intended to reflect the nature of plant and potline size. This uncertain portion of direct service indus- try load is treated as independent of other regional load; the probability of observing a given direct service industry load is not affected by the level of other regional loads. This reflects the idea that Northwest alumi- num industry activity will to a large extent be driven by factors outside of the Pacific North- west economy. Resource Requirements Subtracting the capability of existing system resources over time from the Council's range of forecasts yields an estimate of the resource energy additions required to main- tain the load/resource balance under each of the load scenarios. The loads used in this calculation were the frozen efficiency loads referred to in Volume II, Chapter 3, without adjustment for any conservation program energy savings. The estimates for capability of existing resources were based on the 1985 Northwest Regional Forecast, published by PNUCC in March 1985. (See Volume I, Chapter 5.) Under the assumption that the investor- owned utilities place all of their resource needs above their current surplus on Bon- neville, these regional values represent an upper bound to the potential range of Admin- istrator obligations. Figure 8-4 depicts regional resource requirements and shows the current surplus lasting anywhere from four to more than 20 years, depending on demand growth and on the load path fol- lowed by the region. Under the high load forecast the first need for new resources occurs in about 1990. In the low load case there is no additional resource requirement. The total amount of resource additions that might be required over the 20-year planning horizon ranges from zero to almost 12,000 average megawatts. In addition to regional requirements, the Council also estimated the resource needs of only Bonneville’s public utility and direct ser- vice industry customers. In the event that no investor-owned utility loads are placed on Bonneville, these values provide a lower bound on the potential range of the Admin- istrator’s obligations. These public utility requirements are shown in Figure 8-5. Com- parison of Figures 8-4 and 8-5 shows the Bonneville/public utility system to own the lion's share of the current surplus, with pro- jections of deficits occurring at much later dates across the forecast range. In fact, no resource additions are required by the pub- lics in either the low or the higher probability medium-low forecasts. While not shown here graphically, the investor-owned utilities are currently far less surplus than the publics and are forecast to have a higher proportion of regional load growth occur in their service territories. Most of the early resource devel- opment in the region is likely to be driven by investor-owned utility needs. Resource Availability and Cost-Effectiveness Studies The Council has undertaken a detailed anal- ysis of the conservation program actions and generating resource development alter- natives available to meet the region's energy needs over the planning horizon. These analyses were described in detail in Volume ll, Chapters 5 and 6. A summary of the results is shown in Table 8-1. This table shows the amounts of cost-effective energy estimated to be available for each resource across the load forecast range. Except for cogeneration, the amount of energy available from generating resources does not vary with the load forecast. The amount of cogenera- tion available is dependent on the level of economic activity in the industrial sector and has an availability correlated to load growth. Likewise, many of the conservation program potentials are driven by the level of economic activity in their sectors; for example the rate of new building starts affects the energy avail- able from the model conservation standards. Potential savings from many of the Council's conservation programs are directly corre- lated to the assumptions used in develop- ment of each of the load forecasts. Resource Availability (Average Megawatts) Conservation Program MCS Residential MCS Commercial Refrigerators & Freezers Water Heat Manufactured Homes Existing Residential Existing Commercial Existing Industrial (with DSIs @ 100%) Agriculture Transmission & Distribution Efficiency Improvements Generating Resource Hydropower Efficiency Improvements New Hydropower Nonfirm Strategies Cogeneration Licensed Coal (2 Units) Chapter 8 Table 8-1 LOAD SCENARIO Medium- Medium- High High Low Low 792 468 405 129 398 195 109 51 352 293 224 206 396 324 266 219 35 36 32 13 455 455 455 455 802 614 475 345 538 538 538 538 124 105 105 105 34 34 34 34 110 110 110 110 200 200 200 200 714 714 714 714 320 190 190 130 905 905 905 905 4,520 4,520 4,520 4,520 Unlicensed Coal (10 units) For the discussion in this chapter, conserva- tion programs are described as either “dis- cretionary” or “nondiscretionary.” Nondiscre- tionary programs are used in the portfolio analysis to model implementation of building and appliance codes, or the forced acquisi- tion of cost-effective lost opportunity resources. The development rates for the nondiscretionary programs are not subject to program management in response to resource need. These programs produce energy savings regardless of need. For example, once incorporated into building codes, the level of activity of the model con- servation standards (MCS) would be driven primarily by the number of building starts. The MCS would automatically produce energy savings across the entire load range. It would produce more energy in the high than in the low, but would produce energy in the low even though no additional savings are required for the region in low load condi- tions. Many of the nondiscretionary pro- grams automatically produce more energy savings as load levels increase because of the higher economic activity at those load levels. This automatic correlation of savings to load can add to the value of a resource and is captured in the portfolio analysis. Addi- tionally, all nondiscretionary programs have equal and top priority in the resource devel- opment order in the portfolio analysis. For modeling purposes in the portfolio develop- ment, the residential MCS, commercial MCS, manufactured homes, refrigerator/ freezers, and water heaters are all treated as nondiscretionary resources. 8-5 Chapter 8 Table 8-2 Resource Priority Order NONDISCRETIONARY RESOURCES Residential MCS Commercial MCS Refrigerators & Freezers Water Heat Manufactured Homes DISCRETIONARY RESOURCES Hydropower Efficiency Improvements Agriculture Conservation Existing Commercial Conservation Transmission & Distribution (T&D) Efficiency Improvements Existing Residential Conservation Existing Industrial Conservation Combustion Turbines Small Hydropower Cogeneration Licensed Coal Unlicensed Coal Discretionary programs are those programs whose development is managed in response to need. These programs are targeted pri- marily at the existing sectors (e.g., existing residential weatherization or existing indus- trial) where a savings potential already exists and can be developed as needed. Delay on implementation of these programs is not likely to produce large lost opportunity impacts. These are programs which are likely to be subject to direct program management and whose energy contributions can be developed in response to need. A large portion of the industrial conservation potential comes from direct service industry (DSI) load. Because the portfolio assump- tions regard 50 percent of DSI load as uncer- tain, the level of industrial conservation is uncertain as well. In load outcomes where all DSI loads remain throughout the entire plan- ning horizon, there is more industrial conser- vation potential than in load cases where only half the DSIs remain. This correlation between DSI loads and industrial conserva- tion potential is captured in the Decision Model. The estimates of resource availability in Table 8-1 can be thought of as individual invest- ment opportunities to be used in developing the regional resource portfolio. A number of cost-effectiveness studies were performed using the Decision Model to determine the best priority order for resource development. 8-6 These studies were conducted by making pairwise comparisons of programs and gen- erating resources until the order was found that led to lowest expected value system cost. This priority order analysis involved only the discretionary conservation pro- grams and generating resources. The non- discretionary programs were excluded from the priority order tests; however, they were included in the model runs to ensure that their system effects and impact on the cost effectiveness of other resources would be included. The initial priority order was based on levelized cost estimates for the programs and resources, and the process allowed the generating resources to compete with con- servation programs for priority order. A limit of at least a $10 million improvement in system cost was judgmentally imposed as the mini- mum improvement to justify a switch in pri- ority order between two competing programs and/or resources. Except for the amount of energy available for several of the resources, the conservation program assumptions for this analysis were consistent with the data described in Volume ll, Chapter 5, and generating resource assumptions were consistent with Volume II, Chapter 6. For programs and generating resources in which the energy available was less than 200 average megawatts, the energy availability for these studies was raised to 200 average megawatts to ensure that the system effects of the resource would be captured in the present values. This increase in energy availability pertains only to these priority order studies. After the priority order was determined, the energy limits were again set back to those in Table 8-1 for further portfolio analysis. All sponsorship and finan- cing assumptions were consistent with those described in Volume II, Chapter 4. The results of this analysis are shown in Table 8-2. This is the priority order that was found to produce the lowest expected pres- ent value system cost across the entire load range, under the Council’s base data assumptions and given the constraints men- tioned above. This order was used as the basis for developing the resource portfolio, conducting sensitivity analysis, and develop- ment of Action Plan items. As stated earlier, the nondiscretionary programs are all given equal and top priority in resource develop- ment, and are only shown in the table for the sake of completeness. The results of the last pass through the pair- wise comparisons are shown in Table 8-3. This table shows the impact of switching the priority order of all contiguous pairs of discre- tionary resources in the portfolio. For exam- ple, moving existing industrial conservation ahead of existing residential conservation in the priority order, with all other resources in their original positions, causes a present value $35 million increase in the expected value of system costs. Note that cost dif- ferences due to a switch of the order for any pair of resources are generally quite small. This results primarily because the resources are already ordered according to cost effec- tiveness. Moving unlicensed coal to the top of the discretionary resource list would have a very large cost impact. Other factors which would tend to produce small cost changes are similarity of resource costs, relatively small amounts of energy for some programs and generating resources, and parallel resource development schedules. Because development on many of the resources in the portfolio occurs simul- taneously, a switch in priority order may lead only to small timing differences in resource development over most of the load range. Given that the same total amounts of two resources that have similar costs are devel- oped, small changes in the timing of develop- ment will generally produce only small pres- ent value impacts. The resource portfolio priority order shown in Table 8-2 represents a general order for development of resources during periods of acquisition. It does not mean that all of the potential of one type of conservation program or generating resource should be exhausted before moving to the next. As mentioned above, constraints on program development rates and resource lead times are likely to require parallel development paths for many of the resources in the portfolio. Additionally, the methodology used in this analysis necessarily treats programs and resources as generic blocks. For instance, all of the potential cogeneration units have the same physical characteristics, capital costs, operating costs, lead times, seasonal dis- tributions, etc. In reality, there are likely to be significant differences between individual cogeneration installations competing for resource acquisition. In the actual acquisition decision, all projects should be evaluated on their own merits, taking their own unique characteristics into account. Option and Build Decision Rules In addition to the order of resource priorities, two other decision rules are required to define the resource portfolio. These are referred to as the option and build levels. The option level governs the amount of resource for which options would be acquired and held in inventory. The build level governs the amount of resource moved out of inventory and into actual construction. The option and build levels represent levels within the range of load uncertainty to use as guides for making resource decisions. A hypothetical example is shown in Figure 8-6. In this example, the region has moved out along a somewhat random load path and finds itself at load level L in time period T. The future load path is still unknown and deci- sions must be made in the face of this uncer- tainty. To do this, a range forecast is first made from period T and a probability dis- tribution is applied to the forecast range. Within this range forecast, two additional forecasts are made, one corresponding to the option level and the other to the build level. In this example, the option level of 90 percent would mean that, of all the possible Table 8-3 Priority Order Studies RESOURCES SWITCHES INPRIORTY ORDER Chapter 8 CHANGE IN SYSTEM COST (millions of present value $) Agriculture ahead of Existing Commercial ahead of T&D Efficiency ahead of Existing Residential ahead of Existing Industrial ahead of Combustion Turbines —_ ahead of Small Hydropower ahead of Cogeneration ahead of Licensed Coal ahead of Unlicensed Coal ahead of Hydropower Efficiency Agriculture Existing Commercial T&D Efficiency Existing Residential Existing Industrial Combustion Turbines Small Hydropower Cogeneration Licensed Coal 15 3 -2 "1 35 19 12 4 2 32 Load Range Forcast Current Year. Forcast Original High Ve darerererceserrees .4 | 90% Option Level eanpesrerecesrreres sy 50% Build Level uses Current Year Longest Resource Lead Time Figure 8-6 Option and Build Level 8-7 Chapter 8 P.V. Million $ 2,000 1,500 1,000 500 Oo Build 50 10 20 30 40 50 $ Percent Load Range 60 70 80 90 100 Figure 8-7 Cost of Option/Build Level Combinations load paths from T forward, 90 percent would fall below the option level forecast and 10 percent above it. Similarly, a build level of 50 percent implies that there would be an equal chance of observing a load path either above or below the build level forecast. Once these forecasts have been made, the resource priorities, resource availabilities, and option and construction lead times are used to make resource decisions. In the example, enough resources would be optioned to ensure that if the future loads did not exceed the 90 percent option level, there would be enough resources in inventory to meet the region's needs. Construction deci- sions, however, would be made only to cover the more conservative 50 percent build level, leaving an equal risk of being either surplus or deficit at some future time period. The Council conducted a number of studies at various combinations of build and option levels to determine which combination would result in the lowest present value cost on an expected value basis. The results are shown in Figure 8-7. The solid line shows the sys- tem cost impact of holding the option level constant at 90 percent and changing the 8-8 build level from 0 to 100 percent in 10 percent increments. The dashed line shows the cost impact of holding the build level constant at 50 percent and changing the option level in 10 percent increments. The graph illustrates that, generally, option levels toward the higher end of the load range and build levels toward the middle of the load range produce lower system costs. This result makes intu- itive sense because the option cost of the resources in the portfolio is much less than the cost of their actual construction. Options can be thought of as a relatively cheap form of insurance that reduce resource lead time and allow the region to guard against unanticipated periods of rapid load growth. It appears cost effective to “over option” resources and build an inventory that exceeds expected need in order to assure flexibility in the resource acquisition process. However, because of the much higher costs associated with build decisions, they should be guided by using more conservative load level targets, near the expected value of load, to produce the most cost-effective portfolio on an expected value basis across the wide range of possible load outcomes. Figure 8-7 also shows that the expected value of system costs is quite stable across build levels from 30 to 60 percent and for option levels from 70 to 100 percent, for the set of data used in this analysis. The shapes of these curves will be driven by the data that influence the relative costs of underbuilding and overbuilding, such as option costs, the structure and price of the secondary market, availability of extra-regional purchases, cost of curtailment of interruptible and firm load, and the fixed/variable cost ratios of resources in the portfolio. System costs will also be influenced by characteristics of resources in the portfolio that affect the ability to correct for errors in the planning process, such as gen- erating resource lead time and conservation program ramp rate constraints. For example a portfolio comprised totally of ten-year lead time resources will show a much higher vari- ance in the load/resource balance than a portfolio comprised primarily of two- or three- year lead time resources. For a 50 percent build level, the ten-year lead time portfolio will show much higher levels of overbuilding in low load conditions and much higher levels of underbuilding in high load conditions. This occurs because short-term forecasts are likely to be much more accurate than long- term forecasts, and the degree of expected forecast error will diminish rapidly with shorter lead times. Insufficient time was available in develop- ment of the 1986 plan to perform the exten- sive sensitivity analysis required to investi- gate all of these issues more fully. For purposes of portfolio analysis in this plan the Council has assumed the use of a 90 percent option level and a 50 percent build level. The level of the current surplus allows time for further study of the appropriate levels within the load forecast range to use as guides in resource decision making. Description of the Resource Portfolio Because the resource portfolio is defined through the availability of resources, the pri- ority order for resource development, and the option and build decision rules, resource activity contained in the portfolio can be described in a number of ways. Perhaps the most straightforward description is to present the implied resource schedules required to meet load under several different load sce- narios. Figure 8-8 illustrates the regional Chapter 8 Hydroelectric Efficiency Improvements Hydroelectric Nonfirm Conservation] Medium-high High MW MW 12,000 12,000 10,000 10,000 8,000 Cogeneration} 8,000 Coal 6,000 Small 6,000 3 Hydroelectric , 4,000 Nonfirm 4,000 Hydroelectric Efficiency 2,000 | Improvements 2,000 Conservation} 0 0 1985 1990 1995 2000 2005 1985 1990 2000 2005 1995 Medium-low MW Low MW 12,000 10,000 8,000 6,000 4,000 Hydroelectric Efficiency Improvements 2,000 0 12,000 10,000 Small 8,000 Hydroelectric| 6,000 Nonfirm 4,000 2,000 Conservation} 1985 1995 1990 2000 2005 1985 Figure 8-8 1990 Conservation 1995 2000 2005 Regional Resource Schedules: High, Medium-High, Medium-Low, Low resource development that would be required to attain load/resource balance under each of the four regional load forecasts. A very wide range of resource activity is apparent, mov- ing from only nondiscretionary conservation programs such as the model conservation standards in the low forecast, to full develop- ment of all conservation programs, small hydropower, additional nonfirm energy, and cogeneration, along with 12 new coal units, in the high forecast. These data, showing the annual loads, year by year development for each of the resources, and the resulting load/ resource balance, are presented in tabular form in Appendix 8-A. In Figure 8-9, this same set of resource schedules is shown for Bonneville with obli- gations for only public utility loads. Because of the large current surplus on the federal system and the lower levels of load increases for the the public utilities, the only resources developed through the medium-high forecast are public utility conservation and a small amount of hydropower efficiency improve- ment. In the high forecast, some generating resources are developed, but this develop- mentis much later than in the regional cases. The tabular data supporting Figure 8-9 are also contained in Appendix 8-A. Note that the resource schedules as shown in Figure 8-8 and 8-9 are not based on the 90/50 option/build decision rule. The sched- ules shown here produce load/resource bal- ance in all but the low case, where the region is surplus for the entire planning horizon with- out addition of new resources. These sched- ules contain an implicit assumption of perfect knowledge of where long-term loads will eventually lead before the resource deci- sions are made. Given the uncertainty in long-term load, this is an unrealistic assump- tion. The schedules are represented this way because public comment received on the 8-9 Chapter 8 Average Average MW MW 4,000 4,000 Cogeneration Small 3,000 Hydro. 3,000 Combustion Turbines: 2,000 ~ 2,000 Hydro Efficiency 1,000 1,000 Conservation ° o Hydro Efficiency Conservation 1985 1990 1995 2000 2005 1985 1990 1995 2000 2005 Public High Public Medium High Average Average MW MW 4,000 4,000 3,000 3,000 2,000 2,000 1,000 1,000 O Conservation Conservation 1985 1990 1995 2000 2005 1985 1990 1995 2000 2005 Public Medium Low Public Low Figure 8-9 Public Utility Resource Schedules: High, Medium-High, Medium-Low, Low draft plan urged the Council to indicate the amount of resource development needed to meet the high forecast and argued that show- ing resource schedules which did not meet the high understated the risk inherent in the resource portfolio. The Council agrees that there is certain infor- mation value in providing illustrations of the amounts and types of resources needed to meet load in the high and has done so here. However, the reader should be aware that, given imperfect knowledge of load, a non- zero probability of loads at or near the high, and resource lead times approaching ten years for some programs and generating 8-10 resources, the only way the high forecast can be met with certainty is to commit to resource build decisions that would cover a high load event, well in advance of any indication that high loads would actually occur. The Council would argue that a strategy that committed resource decisions to cover the very low probability event of a high load outcome is a much riskier strategy than one that builds toa dynamic expected value of load (a 50 percent build level). Building to the high would be likely to lead to the kind of overbuilding that led the region to the condition in which it finds itself today: a 2,500 megawatt surplus, two nuclear units on hold, and several others ter- minated. The Council's analysis, as illus- trated in Figure 8-7, shows that the policy of building to the high (100 percent build level) has an expected value penalty approaching $1.7 billion. For purposes of illustrating resource sched- ules, the Council has used an assumption of perfect knowledge of load. However, all of the portfolio analysis and any sensitivity studies use the more realistic assumption of imper- fect knowledge of load and employ the 90 percent option and 50 percent build decision rules. Figure 8-10 Nondiscretionary Conservation Program Energy as a Function of Load Path Figure 8-12 Hydropower Efficiency Energy as a Function of Load Path The four specific scenarios just presented are indications of resource development actions should a particular load scenario occur. In fact, the likelihood is extremely small that any of these specific regional load paths, and the associated resource actions, will occur. The actual portfolio analysis is con- ducted across a large number of load paths and the resource schedules vary continu- ously across the entire load range. A more complete illustration of the portfolio's impact is illustrated in Figures 8-10 through 8-16. These three-dimensional surfaces show the timing and amount of energy additions for each resource as a function of load. Figure 8-10 shows the effect of the nondiscre- tionary conservation programs through time. This graph includes the effects of both the residential and commercial model conserva- tion standards as well as energy savings from the water heater, refrigerator/freezer, and manufactured home programs. The axis going across the page is time, moving from Chapter 8 Figure 8-11 Discretionary Conservation Program Energy as a Function of Load Path Figure 8-13 Combustion Turbine (Nonfirm) Energy as a Function of Load Path 1985 to 2005. The axis going into the page represents the cumulative probability for load and ranges from 0 to 1.0. A value on this axis of 0 would represent the low load forecast, and value of 1.0 would represent the high. Finally, the vertical axis shows the average megawatts of resource developed in the par- ticular combination of future year and load level. For readability, the data for these graphics were developed based on only 20 load scenarios. Each line moving across the 8-11 Chapter 8 Figure 8-14 Figure 8-15 Small Hydropower Energy as a Function of Load Path Cogeneration Energy as a Function of Load Path study period represents one of these sce- narios. However, the actual portfolio analysis is conducted across hundreds of different load paths. Because of the continuity of savings across the load range, the graph of the model con- servation standards is a relatively smooth surface. The graph also illustrates the ability of the standards to respond to load, providing increased savings at higher loads and lesser amounts of savings at the lower load levels. Other resources, such as combustion tur- bines and small hydropower (Figures 8-13 and 8-14), show different characteristics. Because of their discretionary nature, they do not automatically respond to load, but are brought on line as triggered by forecast need. Because of their lower position in the priority list, they are generally scheduled either later or only in the higher forecasts, as shown by their position in the back right-hand corner of the figures. This is especially pronounced in Figure 8-16 the case of coal, as shown in Figure 8-16. Coal Energy as a Function of Load Path While the three-dimensional surfaces imply that the resource priorities regarded to be most cost effective are generally followed, this does not necessarily mean that all of one resource is exhausted before moving to the next. For instance, limitations on conserva- tion program development rates may mean that small hydropower or nonfirm strategies 8-12 have to be developed in parallel. Additionally, to maintain an approximate load/resource balance in periods of rapid load growth, lead time considerations may require low priority resources with short lead times to be devel- oped ahead of higher priority resources with longer lead times. The uneven surfaces for small hydropower and cogeneration are due partly to this, plus the relatively small amounts of energy available for these resources and the exaggerated scales on the graphs. Up to this point, the description of the port- folio has focused primarily on the timing of firm energy contributions that could be expected from the various resources at differ- ent load levels. Equally important are the decision schedules required to achieve these contributions. Figure 8-17 illustrates the timing of discretion- ary conservation program start-ups required for both the region as a whole and for the public utilities only. The start-ups are shown as a function of the load path followed, with high loads represented by a probability of 1.0 and low loads by a probability of 0. If regional loads were to follow the high forecast, some conservation programs would start up or increase over current activity levels as early as 1987. This start-up date slides to 1999 at about a 10 percent load probability level. This is the lowest point in the load range for which any new conservation is required during the planning horizon. The most probable time period for regional need to increase activity in conservation programs is the early 1990s. For public utility needs only, the conservation start-up dates are much later. If high loads were to develop, the earliest start-up dates would be around 1990. This date slides out rapidly to 1995 at the 90 percent load proba- bility level, indicating only a 10 percent chance of a need for new conservation activity in the public sector before 1995. No new activity at all is required in the lower half of the load range to meet public utility needs within the planning horizon. Information regarding the timing of initial decision points for all the generating resources is shown in Figures 8-18 through 8-23. These figures depict the first new option and build decisions needed for both the region as a whole and also for just the Chapter 8 Lead Probability High1.0 0.5 Low 0.0 1985 1995 2000 2005 Year Of Program Start-up 11177. Bublic wee Region Figure 8-17 Conservation Program Start-ups Lead Probability 1 “ny eo =a Po “Um, My, “my, ‘a, 0 1985 1990 1995 2000 2005 Year Of First Decision ™@™ Region: Builds 'Region: Options v0077? Public: Builds '™ asm! Public: Options ™ Figure 8-18 Initial Decisions, Hydropower Efficiency Improvements 8-13 Chapter 8 Lead Probability 1 ut . “um % | we iZiten, “Ym, ee “on, ey “Ny —-—_—---—— ii — eee He ee ee eH ‘tty, ey “Wn, ea, lim, “m, T&e, 0 1985 1990 1995 2000 2005 Year Of First Decision werrr" Public: Builds = mms Region: Builds Ml Public: Options Region: Options Figure 8-19 Initial Decisions, Combustion Turbines (Nonfirm) Lead Probability Ny a ve, 1 Yn, Eee: “om, sey, Wining eee Uy, ey “ny 7 as oe ee es, meses ng ei TMT “Um, %, m, %e, i lls “Ye, “im, | Ye, “lly, “ey teen ne ae oem Vy — az - Oo 1985 1990 1995 2000 2005 Year Of First Decision 8-14 “errs Public: Builds" Public: Options m=immim Regions: Builds ! Region: Options Figure 8-20 Initial Decisions, Small Hydropower public utilities. The information is again dis- played as a function of load path, with low loads represented by a probability of 0 and high loads by a probability of 1. Figure 8-18 shows the timing of decisions on the highest priority discretionary resource in the portfolio, hydropower efficiency improvements. In a high load case, the first option decisions needed to meet regional load are made in 1987 and the first build decisions occur two years later in 1989. The initial option and build decision dates move out to 1998 and 2000 respectively in the lowest load condition in which hydropower efficiency improvements are needed, a load probability of about 8 percent. Because hydropower efficiency is a higher priority resource than any of the dis- cretionary conservation programs, activity extends slightly further down into the load range than for conservation programs. For the public utilities, the first option decisions made under high load conditions occur as early as 1990, with build decisions following in 1992. These dates move out ten years to 2000 and 2002 at about a 50 percent load probability level. Chapter 8 Most of the other generating resources show decision activity occurring several years later than that on hydropower efficiency, and across a narrower portion of the load range, primarily because of their lower priority. Note also that, for coal, if the region follows a load path in the upper end of the load range, option activity may be called for as early as 1989. Even though coal is the lowest priority resource in the portfolio, the length of its lead time may require decisions in the earlier years of the planning horizon. Load Probability 1 le 1985 1990 1995 2000 2005 Year Of First Decision Decision %777777%4 Public: Builds mata! Region: Builds ™®= Public: Options MMM Region: Options Figure 8-21 Initial Decisions, Cogeneration Load Probability 1 1985 1990 1995 2000 2005 Year Of First Decision Decision '%777777% Public: Builds '""""®" Public: Options Iam Region: Builds "NN Region: Options Figure 8-22 Initial Decisions, Licensed Coal 8-15 Chapter 8 Load Probability 1 1985 1990 1995 2000 2005 Year Of First Decision Decision 7777777 Public: Builds ameeeBe Public: Options =I Region: Builds =! Regions: Options Figure 8-23 Initial Decisions, Unlicensed Coal Percentage 25 0 12,000 24,000 36,000 48,000 60,000 72,000 System Cost Present Value (Millions) 8-16 Figure 8-24 System Cost Distribution Section B: Portfolio Uncertainty As has been stated previously, the Council believes that recognition of the large uncer- tainties inherent in long range resource plan- ning is imperative to producing an effective and adaptable power plan. Most of the uncer- tainty directly included in the analysis leading to the final portfolio concerns future load and the large impact it has on the types and amounts of resources that might be needed. Resource uncertainty has been included in the analysis to the extent that conservation and generating resource supply can vary with the economics and demographics across changing load forecasts (e.g., more energy available from the model conserva- tion standards in the high than in the low). However, the planning models are determin- istic for resource availability on a specific load path. While the Council feels that the data development process has produced reason- able and balanced estimates of future resource supply, there's no question that a range of uncertainty exists around these val- ues as well. Based on the public comment received on the draft plan, the Council performed a number of sensitivity analyses on the resource portfolio. These studies were designed primarily to investigate the impact of differing levels of conservation supply than projected in the final portfolio, or of having less flexibility in development of resources. Additional studies were performed to esti- mate the impact of having more uncertainty on direct service industry load than is assumed in the base portfolio, the impact of not being able to secure resource options, and the consequences of failing to attain regional cooperation on resource develop- ment. These studies were all performed using the Decision Model, and, except as noted, used the same data, resource pri- orities, and decision rules used in the final resource portfolio. All studies were per- formed using 100 load paths, with the same set of load paths in each case. The param- eters of interest in each study were the changes in cost from the base portfolio and the changes in the timing of resource decisions. Figure 8-24 is a frequency distribution for system cost present values under the base portfolio assumptions, using 100 load paths. It has has an expected value of nearly $30 billion, with a range from $0 to $72 billion. The variation in system cost is driven pri- marily by variation in load. Loads in the low end of the load range will require very little additional resource development, will have relatively low production and purchase costs, and will exhibit high levels of secondary reve- nue. High loads require intensive resource development and high levels of capital expenditure, have high production costs, and generate lesser amounts of secondary reve- nue because of the shorter duration of the surplus. The impact on the frequency distribution of initial unlicensed coal options was isolated and used as an indicator of the impact on timing of resource decisions. This is because an option on a coal unit may be required in the relatively near future (due to the length of its lead time, as early or before option deci- sions on most other generating resources — see Figure 8-23), and because of the amount of energy represented in the unit size of this resource. Figure 8-25 shows a histogram for the timing of the first options taken on unlicensed coal units in the base portfolio across the various load paths. It is not based on all the coal option decisions made in the simulation; those histograms would have more density toward the mid 1990s. It is based only on the timing of the first coal option taken, if any, in the load paths experi- enced by the model. The last period in which coal option decisions occur is 1995 because, with a ten-year total lead time, decisions made after this point would be targeted for dates outside the planning horizon. The size of the bar to the far right represents the prob- ability that no coal options are taken in the planning horizon, and shows that, in the base portfolio, no options are needed on unlicensed coal about one-third of the time. The sum of the probabilities between 1986 and 1990 yield an estimate of the probability that at least one coal option is needed by 1990. For the base case portfolio, that value is about 30 percent. Chapter 8 Percentage 40 30 20 10 wd 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Year Of First Option Figure 8-25 First Coal Options, Base Portfolio Percentage 25 PV Cost Change (Millions) Figure 8-26 Cost Impact of One-Third Less Conservation 8-17 Chapter 8 Percentage 30 20 WWW WW YW Waa 10 NN \\ YW YW YWda \ \ SS SS 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Year Of First Option Figure 8-27 First Coal Options, One-Third Less Conservation Percentage 40 30 \ 10 Y __ a AAA -300 O 300 600 900 1,200 1,500 PV Cost Change (Millions) 8-18 Figure 8-28 Cost Impact of Slower Conservation Ramp Rates Impact of Less Conservation Supply One analysis was performed under the assumption that one-third less energy would be available in all of the conservation pro- grams in the resource portfolio. The study changed only the total energy supply, it did not affect the cost effectiveness of individual conservation programs. The results of the analysis are shown in Figures 8-26 and 8-27. Figure 8-26 shows the present value cost impact of this reduction in conservation sup- ply to have an expected value system cost increase of $2 billion to the region, with the possible cost penalty ranging from $0 to over $3.6 billion. Figure 8-27 shows a frequency distribution for the first unlicensed coal options needed with one-third less conserva- tion supply. The probability here that a coal option will be needed by 1990 is about 65 percent, compared to about 30 percent for the base portfolio. Impact of Slower Conservation Ramp Rates The Council's assumptions for the maximum activity levels of conservation programs and the rates at which the programs can be accel- erated to those activity levels, yield an aver- age of about ten years total development time to capture the bulk of the energy in the existing sector conservation programs. While there is very little data available on this sub- ject, some public comment indicated that this was an optimistic assumption, and that total lead times of 15 to 20 years were more rea- sonable. A sensitivity analysis was con- ducted by reducing the existing sector pro- gram ramp rates by 50 percent, which would yield total program lead times of about 20 years. The impact is shown in Figures 8-28 and 8-29. The cost impact ranges from $0 to an increase of $1.5 billion, with a mean increase of about $340 million. The distribu- tion for first coal options is shown in Figure 8-29 and indicates a probability of about 50 percent that a coal option would be needed by 1990. Impact of Less Conservation Combined with Slower Ramp Rates The assumptions in the two previous sen- sitivity analyses were combined to determine the impact of having both one-third less con- servation supply available and 20-year ramp rates on the remaining conservation supply. Impacts of the previous two sensitivities are not directly additive, because in this case the reduced ramp rates act on a reduced conser- vation supply. The cost and schedule impacts are shown in Figures 8-30 and 8-31. The cost impact shows an expected value increase of about $2.25 billion, with a potential range from $0 to over $4.2 billion. Probability of need for a coal option by 1990 increases to 85 percent. Chapter 8 Percentage 25 20 15 NA AW ., \ AK WW AY WK WY 10 WKY UY WW WW CK 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Year Of First Option Figure 8-29 First Coal Options, Slower Conservation Ramp Rates Percentage 25 20 Ws VA, 15 10 4 WU WU Yl WM TH / VM WM UMW Wd YUU Ys % 0 © 9 Ss Ss PV Cost Change (Millions) 6 % = Za 29, ° %, wv =) % z Figure 8-30 Cost Impact of Less Conservation and Slower Ramp Rates 8-19 Chapter 8 40 NK NN NN 10 f - IN \\ NW N 3 \ «\ wy K \ SA & @ LS Mm © S PP Po or SF oh oP PP PP PP PP IP IP S Year Of First Option Figure 8-31 First Coal Options, Less Conservation and Slower Ramp Rates Percentage 25 20 Wd 1, Z Z 15 10 Ys MWh WWW, YWWdu«:sLs YY Ys Wa Wa Wa WMA Y/ Wa dda PV Cost Change (Millions) Figure 8-32 Cost Impact of One-Third More Conservation 8-20 Impact of Higher Conservation Supply Itis also possible that the Council has under- estimated the amounts of cost-effective con- servation supply available over the study horizon. A study was performed under the assumption that one-third more energy was available in each of the programs in the port- folio, at their current average cost. The results are summarized in Figures 8-32 and 8-33. The cost impact ranges from $0 to a cost savings of $3.6 billion, with an expected value savings of $1.71 billion. The probability of need for a coal option by 1990 falls to under 5 percent. Impact of Delay in Implementation of the MCS The impact of delayed implementation of the model conservation standards (MCS) was investigated by assuming that no energy sav- ings would be produced by either the resi- dential or commercial MCS before Sep- tember 1991. The interim energy savings were treated as a lost opportunity resource, and would be replaced by resources lower in the portfolio priority list as needed across the forecast range. The cost and schedule impacts are represented in Figures 8-34 and 8-35. The cost impact ranges from a cost reduction of $120 million to increased costs of $600 million, with an expected value cost increase of $175 million. Because the MCS is a nondiscretionary resource, with energy savings and costs accruing regardless of need, delay of the MCS produces present value cost reductions at the lower end of the load range, where the regional surplus con- tinues throughout the planning horizon and there is no need for any MCS savings. How- ever, the cost savings in low load conditions are more than offset by the cost increases occurring in medium and high load cases, where the MCS energy lost during the delay is replaced with more expensive resources. Under the MCS delay assumptions, the probability of need for a coal option by 1990 is about the same as in the base case, approx- imately 30 percent. Chapter 8 : c 20 ~ 10 SS | — wWNN oN Figure 8-33 First Coal Options, One-Third More Conservation Percentage 40 120 240 360 480 600 PV Cost Change (Millions) Figure 8-34 Cost Impact of MCS Delay 8-21 Chapter 8 Percentage 30 20 10 ) MUU, LL WW WY), WAM, Y; WY, WY) Uf WWW LY, WA, Wa VY o> ok (© 2” & PD DM WM NM K Year Of First Option Oo A > Oo DO Oo »@ & 2 9, 9, a 9, So a 9, % So. Figure 8-35 First Coal Options, MCS Delay Percentage 25 0 300 600 900 1,200 1,500 1,800 PV Cost Change (Millions) Figure 8-36 Cost Impact of Losing the MCS 8-22 Impact of Losing the MCS Another sensitivity was performed by com- pletely eliminating both the residential and commercial MCS from the portfolio. This case is represented in Figures 8-36 and 8-37. The cost impact ranges from $0 to a cost increase of $1.8 billion, with an expected value increase of about $620 million. This gives an indication of the cost effectiveness of the package of MCS measures to the region. The likelihood of need for a coal option by 1990 without the MCS is about 40 percent. Impact of Not Being Able to Option Generating Resources One of the important attributes of the resource portfolio is its reliance on the ability to obtain resource options. The option pro- cess provides the opportunity for two-stage decision making on resources, enhances flexibility and improves the ability to match capital intensive generating resource con- struction decisions to load growth. The ability to option resources reliably should reduce the probability and magnitude of errors likely to occur in the planning process. The Council believes the optioning process can be a workable and reliable one; however, the option concept is still largely an unproven one. Asensitivity study was performed to evaluate the impact of not being able to option resources. This was done by setting the con- struction lead times for the generating resources in the portfolio equal to the sum of their option and construction lead times in the base case, and eliminating the option lead time. The effect is a commitment to build decisions significantly earlier than would be required in an option environment, resulting in systems that show a higher variance in the load/resource balance. The cost impact is depicted in Figure 8-38. It ranges from $0 in the cases where no generating resources are required to a maximum cost increase of $3.6 billion. The expected value is a cost increase of about $710 million over the base portfolio. The impact on the schedule of decisions can be shown by comparing the initial resource build decisions instead of the option deci- sions as in the previous sensitivities. Figure 8-39 shows the differences in the distribution of first coal build decisions between the no options case and the base portfolio. The bars moving from above the axis in the early 1990s to below the axis in the late 1990s reflect movement in time of the build deci- sions in the no options case. Simply because of the longer lead times involved, the no options case moves most of the probability for builds into the early years of the study horizon, even though the information about where future loads will eventually lead is of much poorer quality in that time period. The analysis above is directed essentially at the system cost impact of longer resource lead times. It assumes that, if the option pro- cess does not work, only one decision point will exist for a resource, and that it will move forward in time to the point where the option decision would otherwise have been made. The premise is that, without guaranteed compensation for siting, licensing and design, resource developers will move directly into construction after completion of these activities. Chapter 8 Percentage 25 Year Of First Option Figure 8-37 First Coal Options, No MCS . 7 15 10 AI AWW \ DW \ WY -10,000 -8,000 -6,000 -4,000 -2,000 PV Cost Change (Millions) Figure 8-38 Cost Impact of Inability to Option 8-23 Chapter 8 Percentage 60 50 40 30 20 10 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Year Of First Option Figure 8-39 Build Decision Impact of Inability to Option : 15 -12,000 -10,000 -8,000 -6,000 -4,000 -2,000 0 Midpoint PV Cost Change (Million) 8-24 Figure 8-40 Cost Impact of 100 Percent DSI Uncertainty Another potential difficulty with the option process is that siting agencies may not allow resource options to be held idly in inventory for long periods of time, and then moved rapidly into construction upon indication of need. Changes in technology or social or political climate may require extensive review of the resource before moving ahead with construction. A possible scenario may be that two-stage decisions still take place, but that build decisions must occur immediately after completion of siting, licensing, and design or the options are lost. In effect, the options would have a very short shelf life. This would probably result in the loss of many options, but would retain the lead time and better load knowledge advantage for those options which actually moved into construc- tion. An additional study was performed under the assumption that there would only be one opportunity for a build decision on all resource options. This single build oppor- tunity would come at the end of the option process, with failure to build immediately resulting in loss of the option. The expected value cost impact of this sen- sitivity was an increase of approximately $100 million over the base portfolio. This $100 million increase for zero shelf life, ver- sus the $710 million increase for longer lead times described above, implies that most of the benefit in the optioning process results from reductions in resource lead time, and that any resource development process that captures the advantages of reduced resource lead time is likely to prove valuable to the region. Chapter 8 Impact of Increased Direct Service Industry Uncertainty As discussed earlier in this chapter, the Council’s base planning assumptions for direct service industry (DSI) load are that at least 50 percent of the DSls will remain as firm energy customers in the region through- out the planning horizon, with a uniform prob- ability of occurrence applied to loads above the 50 percent level. A sensitivity was con- ducted with 100 percent of DSI load uncer- tain, rather than only 50 percent uncertain, with a uniform probability distribution applied to the entire DSI load range. This has the effect of changing the expected value of DSI load remaining at the end of the study hori- zon from 75 percent of maximum to 50 per- cent of maximum and produces a significant number of load cases where DSI loads are much lower than in the base portfolio. The system cost impact is depicted in Figures 8-40. The expected value impact is a reduc- tion in system cost of about $5.8 billion, with a range from $0 to about $14 billion. The values at the upper end of the range result from cases where regional non-DSI load is quite high and DSI load is very low, allowing the region to use the drop-off in DSI load to avoid the need for new resources. Note that these values are from a regional generating system perspective. They do not include any effects of short-term lost revenue when DSI loads fall off and before other regional loads can grow to replace them, or any of the primary and secondary economic effects due to the loss of jobs these industries represent. The impact on timing of a coal option is depicted in Figure 8-41. Under this scenario, the probability of needing an option by 1990 is about 10 percent. Lack of Regional Cooperation The portfolio development process employed by the Council treats the region as homogeneous, with no differentiation between public and investor-owned utility loads and resources. The methodology makes the implicit assumption that the Bon- neville acquisition process will allow full development of the region's least expensive conservation programs and generating resources before having to turn to the more N s0 ‘N \N 20 \ N 10 : 1 N NK VN wT awww vy \ SV Init-Opt Year Of Iniy Initial Option Figure 8-41 First Coal Options, DSIs 100 Percent Uncertain expensive thermal resources, regardless of which customer groups lay.claim to the resources or which groups exhibit needs. Because the investor-owned utilities as a group exhibit an earlier need for resources, the Council's regional perspective results in public conservation program energy and nonfirm potential being developed to meet load growth in the investor-owned utility sec- tors. This assumption of full cooperation in resource development between the region's customer groups is an optimistic one, but is important because it leads to the lowest cost energy future for the region. Studies were performed to determine the value of regional cooperation. Using public sector scheduling studies to determine which resources the public utilities would develop to meet only their own load growth across the load range, new conservation and generating resource data bases were devel- oped that would not allow development of public resources for investor-owned utility needs. This causes the deferral of some con- servation and nonfirm energy in a significant portion of the load range, with subsequent earlier and higher development of coal. The assumption was also made that the investor- owned utilities would not use the acquisition process and would maintain a capital struc- ture of 50/50 debt to equity rather than the Council's base assumption of 80/20. Portfolio studies were then rerun with the new data assumptions to determine the cost and schedule impact on the portfolio. 8-25 Chapter 8 PV & Renewables Figure 8-42 Expected Value of Regional Cooperation Percentage 30 S \ 20 N X NNNA VN NV \ il VY NNNNA NNNNYAV NY~ VNYVNVNea NNN. NNN VNYV YS Ss SSS NNN g 0 ASSAANAA SNS SSS J 9 SS Ff x ee tee se Value Of Cooperation 8-26 Figure 8-43 Distribution of Benefits Due to Regional Cooperation The cost impacts are summarized in Figure 8-42. Two components of the value of regional cooperation were isolated. Cooper- ation on conservation and renewable devel- opment has an expected value benefit of $1.28 billion. Because the cost of renewables in the portfolio is close to that of coal, most of this value derives from the conservation pro- grams. The value of nonfirm cooperation shows an expected value benefit of about $280 million. This value is higher than the $175 million benefit attributed to nonfirm in Chapter 7, becausé here it competes with new coal under a 50/50 debt-equity ratio for the investor-owned utilities. The analysis described in Chapter 7 used the Council's base assumption of an 80/20 debt-equity ratio for new coal financing. Figure 8-43 is a histogram of the combined benefits of cooperation on conservation, renewables, and nonfirm. This distribution has a mean of $1.56 billion and a range from $0 to over $3 billion. An important point, although not discernible from the histogram, is that a significant portion of the value to cooperation comes from the relatively high probability middle load range outcomes. This results because very little resource develop- ment is needed in the low load cases, and in the higher load cases all of the utility groups develop all of their resources with or without cooperation. Its in the middle of the load range, where the public utilities have surplus conservation and the investor-owned utilities can use it, that most of the benefits of cooper- ation are derived. Figure 8-44 is a histogram of the first coal option decisions without regional coopera- tion. The probability of need for an option by 1990 is about 65 percent. Chapter 8 Section C: WNP-1 and WNP-3 Cost Effectiveness The Council devoted a significant amount of effort to study of the two unfinished Wash- ington Public Power Supply System nuclear plants, WNP-1 and WNP-3. A number of issues regarding the economic and physical characteristics of the units were examined and numerous sensitivity analyses were per- formed. This section provides the detailed analysis of cost effectiveness supporting Vol- ume |, Chapter 7. Generally, both WNP-1 and WNP-3 appear to be cost-effective resources for the Pacific Northwest as a region. Under the Council's base set of assumptions, maintaining both units as options until and if they are needed to meet regional load has an estimated present value benefit of $630 million. Most of the sensitivity studies performed continued to show present value benefits from these plants; however, the range of potential out- comes is very wide. The sensitivity analyses show expected value outcomes ranging from benefits of over $1.5 billion to losses of nearly $1.3 billion. WNP-1 and WNP-3 are not included as firm resources in the draft resource portfolio. Instead, they are included in the plan as potential options due to their potential value to the region. As discussed in Volume |, Chapter 7, their exclusion from the portfolio is based on significant barriers to their development, not on their cost effective- ness. The rest of this section will be devoted to description of the cost-effectiveness analy- sis and results. Methodology All of the WNP-1 and 3 studies used the Council's Decision Model. This model pro- vides the capability to assess the impact of a specific resource strategy or decision over a wide range of load futures. It is particularly well-suited for studies of this nature, where issues such as plant shelf life or forced restart are of interest. The Decision Model attempts to simulate the resource decision process and representative errors in the resource planning process, estimating the conse- quences of being wrong, and incorporating Percentage Percentage Bar Chart 40 \ 35 ‘N 30 \ 25 \ N 20 l N NX N s;-—_\\ N SN i \ \ eae s NaNINININIGIN ws NV VN VV S\N Ss se Ss F Ss - s s Init-Opt Year Of Initial Option Figure 8-44 First Coal Options, No Regional Cooperation the impact of load uncertainty, resource lead time, and unit size on the cost effectiveness of a resource strategy. (See Section D for a more complete description.) The criterion used to compare alternatives in all cases was the expected value of the pre- sent value of incremental system cost. Under the accounting methods used in the Decision Model, this quantity consists of the produc- tion costs associated with all existing and new resources, revenue from secondary sales to the Pacific Southwest, and the cost of imports required to meet regional needs when deficits and poor water conditions occur simultaneously. Also included are the capital costs for the optioning and construc- tion of all new generating resources and con- servation programs. To deal with end-effects of the study, the method included the costs associated with replacement resources to a time beyond which all systems would be identical. Assumptions The economic and physical assumptions used for conservation programs and gener- ating resources were consistent with those described in Volume II, Chapters 5 and 6, respectively. The base case cost assump- tions for WNP-1 and 3 were as described in Chapter 6, except as noted in the sensitivity cases. The load assumptions and probability distribution for load was consistent with that discussed in Volume II, Chapter 3. In addition there were a number of other assumptions needed to perform the studies. These included (any costs specified are in January 1985 dollars): 1. Option Level: 90 percent 2. Build Level: 50 percent 3. Resource Priorities: 8-27 Chapter 8 Percentage 80r- 60 40 x N X S 20 Na S S NINN \ S SSS Sx N S S 0 sXSS$ysF8EEyuu rc SVS LX PHF PPS OH PMP AT AS AT AS AS AT NS OY YD QB a YW a Y Arrival Figure 8-45 Arrival Distribution of First WNP Unit Percentage 80 60+ Z Z Z 40 4 g Z 20 Z Z Z Z. ZZ,3 a wn» t24G4GG4G 4 3 a 9? © 5h (9? 0? PO SY OP PP APPA APRS No aS a a? ae a? a AS Arrival Figure 8-46 Arrival Distribution of Second WNP Unit 8-28 Analysis was performed to find the loca- tion for WNP-1 and WNP-3 in the priority order which would minimize system cost (and maximize their benefit). These stud- ies were conducted in the same manner as those used to determine the priority order for the resource portfolio. WNP-1 and WNP-3 were treated as separate units, and allowed to compete with each other as well as all of the other conserva- tion programs and generating resources for priority order. The priority order for the discretionary programs and resources which produced the lowest expected value portfolio cost was: hydropower efficiency improvements, agriculture, existing com- mercial, transmission and distribution effi- ciency improvements, existing residential, existing industrial, WNP-1, WNP-3, com- bustion turbines, small hydropower, cogeneration, licensed coal, and unlicensed coal. 4. Cost of Out-of-Region Purchases: 15 cents per kilowatt-hour to meet firm load. 2.5 cents per kilowatt-hour to meet non- firm load. 5. Replacement Resource: A 4.0 cent per kilowatt-hour load reduction resource. 6. Number of Simulations: 300 7. Inflation: 5 percent 8. Real Discount Rate: 3 percent Probability of Need for WNP-1 and WNP-3 Figures 8-45 and 8-46 show the arrival dis- tributions for WNP-1 and WNP-3 for a study with 300 simulations. These arrival distribu- tions for the plants would be the same for all studies, with the exception of the forced restart and limited shelf life studies. It is important to point out that, while for modeling purposes the studies assume WNP-1 would be restarted ahead of WNP-3, the levelized cost estimates are nearly equal for both units. The Council has adopted no position on which of the units should be completed first. The results of these studies would change very little if the order were reversed. Chapter 8 The height of the vertical bars in Figures 8-45 and 8-46 represents the percentage of time that the unit arrives in-service in a particular year. The bar at the far right represents the probability that the unit is not needed within the 20-year study horizon. The arrivals for the first unit range from 1994 to 2005, with a most likely arrival near 2000. With five-year lead times, this would imply a most likely construction restart in the mid 1990s. The second unit's arrival tends to lag the first unit by two or three years. Its earliest arrival is 1995 or 1996, with a most likely arrival around 2003. This implies a late 1990s restart of construction for the second unit. The probability that neither unit is needed before 2006 is the same as the probability that the first unit is not needed—about 35 percent. Results Option Value of WNP-1 and WNP-3 The value of WNP-1 and 3 depends in part on how long they can be preserved and still be restarted. The shorter their shelf life, the less likely that they will be available to meet regional load when needed, and the more likely that no return will be realized on pay- ment of additional hold costs. Three studies analyzed this issue, at five, ten and 15-year shelf lives for each unit. (With a construction lead time of five years, a 15-year shelf life ensures that the plants will be available for service anytime within the 20-year planning horizon.) Additional studies were performed including only one unit in the portfolio, to isolate the relative value of each unit. The results are summarized in Figure 8-47. The ability to hold the plants for five years results in a benefit of $330 million, a ten-year shelf life yields benefits of $570 million, and a 15- year shelf life benefits of $630 million. Of the $630 million benefit produced by both plants, $440 million is derived from the first unit and $190 million from the second. This higher marginal value for the first unit arises from the fact that it is needed in more of the load paths, and also that its total hold costs will generally be less. The same type of pattern is seen in the five and ten-year shelf life single unit studies. P.V. Benefit Million $ 2,000 630 1,000 0 330 5 s 1st 200 WY 420 unit Oo [5 Year 10 Year 15 Year Option Option Option -1,000 -2,000 P.V, Cost Figure 8-47 WNP-1 and WNP-3 Option Value Percentage 60 40 ‘ NI NY S S S S S 20 + s > S S S < eS S Ss S$s8$ X x < < S < < \ Nay 0 sSaSSSSSESESESESS SHO OO TO OOO COOL A MAO GOO DOOD GO AOD Orn’ XX e S Oo OY no oO . Ca vw 4 4 Figure 8-48 Value of WNP-1 and WNP-3 8-29 Chapter 8 Percentage 60 40 20 VIIIIIIITLILITLTLYLTITST Ld W447 \III/// V“IIIIIIIIT11. WYIIITITITTTTTTLRE fa VWIIIITTTT) oO a S SHH OO THO KP FH HP Shc ci DOPE SS oO POP ASS SS? ¥ A Poa} S © OD ,O“O as Figure 8-49 Value of First Unit Percentage 60 Z Z 5 Z 40 % 4, 4 Z 4 Z 4 Z 4 5 4 20 % 4 Z Z Z Z 44 Z a3 $4.34 4 44433 ol. « 444445 OAH AAP LH CA AHA HSH OAH So ce) DED GOGO. OGO™,.O" OHO SELLS POM MOAB yar qh Figure 8-50 Value of Second Unit 8-30 The range of potential costs and benefits for the individual units is shown in Figures 8-49 and 8-50. The means of these distributions are $440 and $190 million for the first and second units respectively, as shown in Figure 8-47. The first unit is built more frequently than the second, shows a higher probability of producing some benefit for the region, and a lower probability that the investment in its hold costs will be wasted. The $630 million benefit of both units under a 15-year shelf life scenario is the Council's estimate of the expected value of the benefit across the entire load range. In reality, there is a very wide range of potential outcomes for the value of these units. In low load cases, itis likely that the units may never be needed and may be held for a long period of time with eventual termination. This would result in a net loss to the region consisting primarily of the hold costs. On the other hand, in the event of higher load growth scenarios the units may be held for a relatively short period of time; successful restart and construction would avoid the need for more expensive coal units and could yield large benefits to the region. Figure 8-48 shows a frequency distribution for the value to the region of being able to hold WNP-1 and 3 for 15 years. Its mean is the $630 million benefit mentioned above, and it shows a range anywhere from a loss of $1.5 billion to benefits of $2.7 billion. The spike at -$300 million represents the cases where the units are not needed and the hold costs become a wasted investment. The larger negative values of $-1.2 to $-1.5 billion are occurrences where loads begin to come up, the plants are restarted and constructed, loads fall back off and the plants are not needed. These cases represent occurrences of the kinds of resource planning errors the Decision Model was designed to evaluate. The higher load outcomes are represented at the upper end of the benefits distribution. Both WNP-1 and WNP-3 are built early, hold costs are kept to a minimum, the units ulti- mately displace high cost coal and produce large regional benefits of approximately $2.7 billion. Impact of the Future Status of the Direct Service Industries Another issue which has considerable impact on the value of WNP-1 and 3 is the future of the aluminum industry in the North- west. The aluminum industry uses large amounts of electricity, and the industry's needs must be taken into account in long- range resource planning. However, there is currently significant uncertainty regarding the long-term operating viability of a portion of the Northwest aluminum industry. (See Volume II, Chapter 2, for more discussion.) This uncertainty should be taken into account in the planning process. It would be a poor economic outcome to embark on con- struction of long lead time, large thermal units such as WNP-1 and 3, only to find out as they near completion that a portion of the load which justified their completion is gone. Eventually, load growth may again produce a need for the plants, but in the meantime the region would have incurred significant costs with little benefit. The Council performed two studies to esti- mate the impact of the future of the direct service industries (the majority of which are the aluminum producers) on WNP-1 and 3. Both assumed all direct service industry loads would fall to zero after 2001, the year the current industry contracts expire. In the first study, WNP-1 and 3 were used as resources in the portfolio, with the arrival schedules depicted in Figures 8-45 and 8-46, even though frequently there would be little need for the plants once they had arrived. In the second study, the availability of a short- term purchase was substituted for WNP-1 and 3. It was modeled as a revolving account limited to a maximum of 1,600 average megawatts (the energy capability of WNP-1 and 3), with a two-year negotiation lead time, two-year contract duration, reservation costs equal to the annual capital costs of combus- tion turbines, and displaceable energy costs of 6.5 cents per kilowatt-hour. This shart-term purchase was more expensive than WNP-1 and 3 on an annual basis, but in most cases would have been needed for only a short period of time. Because of the two-year con- tract life, such a resource would represent a much more flexible strategy and would allow the region to move back toward load resource balance much more quickly after the direct service industry load had fallen off. Chapter 8 P.V. Benefit Million $ 2,000 880 G 630 Yy G 1,000 + DSIs Leave & Loads Grow [ After 2005 DSIs DSIs Stay Wp Stay (75%) ~ 1260 (100%) -1,000 -2,000 P.V. Cost Figure 8-51 Impact of DSIs on WNP-1 and WNP-3 P.V. Benefit Million $ 2000 910 630 Y 1000 260 \ WW \ 30 YR 40 YR 50 YR -1000 + -2000 P.V. Cost Figure 8-52 Impact of Plant Lives 8-31 Chapter 8 P.V. Benefit Million $ 2000 1490 630 SS S 220 1000 N \N 390 \N X SS 0 Base Cost $24 Million 25% Capital Assumptions Hold Costs Cost Increase -1000 2.5% Increase In Cost Of Coal -2000 P.V. Cost Figure 8-53 Impact of Changing Cost Assumptions P.V. Benefit Million $ 3000 2000 1000 SN Availability -110 -1000 P.V. Cost Figure 8-54 Impact of Equivalent Availability 8-32 The results of these studies are shown in Figure 8-51. The $630 million benefit is the benefit of WNP-1 and 3 in the portfolio, if the direct service industry loads remain after 2002. The remaining DSI load is a random variable, but would average 75 percent remaining across all load cases. The middle bar represents the differences in costs between using the short-term purchase strat- egy and using WNP-1 and 3 in cases where the direct service industries leave. In this case, the short-term purchase strategy costs the region about $1.3 billion less than using WNP-1 and 3. As one additional comparison, the bar to the right represents the value of WNP-1 and 3 as options with 100 percent of the DSI load remaining at the end of the planning horizon, and shows a benefit to the units of $880 million. Impact of Plant Operating Life The Councils base assumption for WNP-1 and 3 operating life is 40 years. With shorter operating lives the plants would produce less benefit because the fuel savings would be limited, and other, more expensive, resources would be needed to replace them more quickly. Conversely, with longer operat- ing lives the plants would have more value. Sensitivity studies were performed using 30 and 50-year operating lives for the plants. The results are depicted in Figure 8-52. With a 30-year life the plants have a benefit of $260 million, and a 50-year life raises the value to $910 million. Sensitivity to Cost Assumptions The value of WNP-1 and 3 will be influenced not only by their cost to complete but also by the costs of competing resources in the resource portfolio. Three studies were per- formed here. The first was a sensitivity on the construction costs required to complete the units. Arguments have been made that the current Supply System budgets may under- estimate costs to complete by as much as 25 percent. As shown in Figure 8-53, if remain- ing construction costs were actually 25 per- cent higher than current estimates, this would reduce the value of the plants to $220 million. Chapter 8 The second sensitivity was on hold costs for the units. The analysis to this point has been based on hold costs of $12 million per year per plant. This also assumes that the hold costs result in no earned value credit; that is, future costs to complete are not reduced through expenditure of the hold costs. Cur- rent Supply System and Bonneville budgets call for expenditures of approximately $24 million per year per plant, and it's estimated that this will result in earned value credit. A pessimistic outcome would be that the $24 million would result in no earned value credit. This case was examined and resulted in lowering the benefit of the units to $390 million. The third sensitivity investigated the value of WNP-1 and WNP-3 if the construction and operating costs of new coal units turned out to be 25 percent higher than the the Council's current estimates. This assumption increases the combined value of both units to $1.49 billion. Impact of Equivalent Availability Because of the large unit size of WNP-1 and 3, the amount of time the plants are actually available for operation once in service will have a significant impact on their cost effec- tiveness. The Council's base assumption for equivalent availability for both WNP-1 and 3 is 65 percent. Lower equivalent availabilities would reduce benefits from the plants through more frequent operation of higher variable cost resources or from losses in sec- ondary revenues. Additionally, lower avail- abilities may require the construction of other resources to maintain system reliability. Higher availabilities would have the opposite effect. The Council performed two sensitivity studies on equivalent availability for the plants, one at 55 percent and the other at 75 percent. The results are portrayed in Figure 8-54, and show a loss to the units of $110 million at 55 percent availability, and a benefit of $1.26 billion at 75 percent. Value of Forced Restart Aset of studies concerning the economics of forced restart was also performed. The term “forced restart” as used here implies an unconditional construction restart of a unit at a specific date, regardless of load path or anticipated need for the unit. These studies evaluated a series of forced restart dates for P.V. Millions 500 400 + 300 Nf 200 100 MAAN un 3 3 =) WN \\ x oa AY WN \\ WN NV WN | WK -300 Vv Vs © D> oO Vi © SUVSES OF DD DDD OD OH MODIS DW BD DW DW GP & see os SF oS 9 Figure 8-55 Impact of Forced Restart both units, to determine both the cost impact of forced restart and to investigate the appro- priate timing, if any, for a forced restart. The first set of studies was performed by remov- ing the second WPPSS unit from the portfolio and forcing the first unit across a series of system arrival dates ranging from 1992 to 2005. The second set was performed by forc- ing the arrival of the first WPPSS unit in the year 2000, and forcing the arrival of the sec- ond unit in the years post-2000. The results of these studies are shown in Figure 8-55. The bars to the far right repre- sent the value of the units if the restart date is allowed to float; that is, the units are sched- uled and built only in anticipation of future need. These are the $440 and $190 million values shown on Figure 8-47. The bars labeled 1992 to 2005 represent the value of forcing the first WPPSS unit into the system at selected points across the study horizon. The labels here represent the resource arrival or in-service dates. Construction restart dates would occur five years earlier. Forcing an arrival of the first unit in 1992 has an expected value loss over the base port- folio of $5 million. Forcing it in 1994 produces an expected benefit of about $90 million, and this value slowly rises to a maximum of $120 million in the year 2000. However, this value is far below the $440 million benefit obtained when the first unit is scheduled in anticipation of need. This occurs because forced con- struction in the lower portion of the load range causes unnecessary overbuilding, and in the middle portion of the load ranges causes displacement of more cost-effective conservation program energy. This effect is even more pronounced in the case of the second unit. The second unit has an incre- mental value of $190 million when both units are allowed to float. However, when the first unit has a forced arrival in 2000 and the sec- ond unit is forced in 2002, its has a negative value of $270 million. 8-33 Chapter 8 Summary Using the Council's base set of assumptions, the inclusion of WNP-1 and 3 in the regional resource portfolio reduces the present value of portfolio costs by $630 million. In addition, the cost effectiveness of these plants appears to be fairly robust. While several of the sensitivities performed here show nega- tive value to maintaining the plants as options, the majority of the studies continue to show varying degrees of benefits to the plants. All of the sensitivity analyses described here were performed by changing a single param- eter at a time and comparing to the base case to isolate the impact of the change in only that parameter. Modest changes in parameters such as equivalent availability, operating life, capital costs, or costs in com- peting resources can result in large shifts in present value benefits. Due to the interre- lated nature of most of these parameters, the reader is cautioned against direct addition of the individual changes presented here to estimate the impact of simultaneously changing parameters. While not presented formally here, another important factor for the cost effectiveness of WNP-1 and 3 is the nature of future load uncertainty. These units derive most of their benefit through the displacement of coal, and to a lesser extent combustion turbines, small hydropower and cogeneration. This happens primarily in the higher load cases. If for some reason the range of load forecasts were to fall, or even simply narrow, the value of main- taining WNP-1 and 3 would fall as well. Con- versely, higher loads or a wider load range would yield higher benefits. 8-34 Section D: Decision Model Introduction One of the important attributes of the Coun- cil’s 1983 plan was the formal recognition of regional load uncertainty and the incorpora- tion of this uncertainty into the planning pro- cess. This is evidenced by the range of load forecasts used in the plan, the emphasis placed on flexible, short lead time resources, and development of the options concept. During development of the 1983 plan, ana- lytical tools were available to the Council that helped characterize the nature of the load uncertainty faced by the region. In general, these were the models contained within the demand forecasting system. However, once the analytical process moved over to the sup- ply side and began the evaluation of resource alternatives to meet future load growth, there was limited analytical capability to assess the effect of this newly defined load uncertainty on the various alternatives. The Council recognized this deficiency and, in the 1983 Two-Year Action Plan Item 29.1, directed its staff to develop a model capable of dealing with load uncertainty and its inter- action with resource decisions. The Decision Model is intended to be that tool. It has been developed to date in a joint effort by indi- viduals from Council staff, the Intercompany Pool, the Pacific Northwest Utilities Con- ference Committee, and Bonneville. The model has also been influenced by activities and discussions within the Council's Options Evaluation Task Force. Council staff has taken responsibility for coordination and oversight. The major benefit of the model lies in provid- ing planners with the ability to assess the load-related risks associated with particular resource option or acquisition decisions. It automatically evaluates the consequences of errors that are likely to occur in the resource planning process. It assists in determining what types of long-term resource strategies better enable the region to manage the risks imposed by load uncertainty. The model enhances strategic planning capability and provides an information flow to the decision- making process in an area which previously had to rely largely on intuition and judgment. The remainder of this section will outline briefly some of the load-related shortcomings of traditional analytical methods and indicate how the Decision Model contributes to the planning process. It will provide an overview of the model, discuss some of the major fea- tures within the model, and briefly describe the major algorithms used in the modeling process. Background The Council's 1983 plan included four differ- ent load forecasts and, correspondingly, four different resource schedules. The range of forecasts acknowledged the highly uncertain nature of the assumptions underlying the forecast, and began to move away from the idea of point forecasting and planning resources to a specific load level with little consideration of other possible load out- comes. It recognized the possibility of alter- native futures and the large impact those futures will have on the types and amounts of resources that will need to be developed. The 1983 plan also placed an emphasis on flexible, short lead time resources. It relied on the premise that the most efficient condition for the region to maintain is one of approxi- mate load resource balance. Shorter lead time resources reduce the period over which the need for new resources must be forecast, and allow resource sponsors to move closer to the point of actual need before committing large amounts of capital for construction. The less lead time needed for resource develop- ment, the better that development can be matched with load. However, quantitative estimates of the eco- nomic value of lead time are difficult to obtain with the analytical methods used in the 1983 plan. The analytical process stopped short of complete incorporation of future load uncer- tainty. The major resource models used in the first plan were designed to schedule or evaluate resources under one specific load condition or forecast, and load uncertainty was in large part handled outside of the plan- ning models. Chapter 8 Load Forcast Load Paths Resource > Strategy Scheduling Module Option & Resource Data + Resource Secondary Purchases Operating Sales $ Mw Capital Costing $ Output Reports End $ "Decisions & Timing Effects "Cost "Reliability Figure 8-56 Decision Model Overview 8-35 Chapter 8 In developing the resource schedules for the first plan, it became necessary to make the assumption that after the first few years the region would know which of the four load paths it was on and would not deviate from that path. Essentially, all of the load uncer- tainty was resolved in the near term and had little impact over the remainder of the plan- ning horizon. This perfect knowledge of load led to resource schedules which provided virtually perfect load/resource balance in each load case, once the surplus was exhausted. This type of study structure reflects none of the benefits inherent in short lead time resources. A study that assumes perfect information on load will show no economic difference between two resources that have the same total cost, regardless of any dif- ferences in lead time. Its very difficult to evaluate the effects of load uncertainty and its impact on cost effective- ness with single load path models. The important effects to capture are the conse- quences of being wrong. It would be possible to manually set up studies which reflect errors in the resource planning process, resulting in systems that are out of load/ resource balance. However, it would be very time consuming to set up and run enough studies to be sure of a representative set of wrong outcomes. Most of the planning stud- ies performed in the region were done under an assumption of perfect knowledge. It is possible to model the single way of being right. It is virtually impossible to model all the different ways of being wrong. However, there is little doubt that the prediction of future con- ditions used to justify today's planning deci- sions will turn out to have some degree of error. 8-36 Decision Model Overview An overview of the Decision Model and the general modeling process is shown sche- matically in Figure 8-56. The process starts with the input of a load forecast range and the probability distribution for that range. Analo- gous to the Councils planning assumption for the region, the actual load experienced within the model might be anywhere in the forecast range. Because it is now possible in the model for the load to have wide variations in outcomes, it is no longer possible to spec- ify a fixed resource schedule to be imple- mented regardless of load outcome. So, instead of a fixed schedule, the user spec- ifies a “resource strategy” that, in general terms, defines the types of resources that should be scheduled as a function of time and load level. The model then moves through the future along asomewhat random load path, making decisions as consistently as possible with the resource strategy. It is essentially blind to the future within the limits of the load forecast tange, and the predictions it uses for deci- sions will generally turn out to have some degree of error. How well the model can match resources to load will depend in large part on the size and lead time of resources combined with the potential variation in load. Costing routines are used to keep track of the capital and production costs associated with the particular load/resource configuration, as well as secondary sales and need for pur- chases. When the model has completed one pass through the planning time horizon, it will have simulated the effect of the resource strategy under one set of future conditions. Because of the large number of possible alternative futures, it is necessary to make many passes through the future to ensure statistical reliability for the results. A model of this nature is useful in answering questions such as the following: * How are today’s resource decisions affected by load uncertainty? * What is the value of reduced resource lead time? ¢ What types of options should the region pursue? ¢ What level of options inventory should the region hold? ¢ Given the uncertainty in long-term load, to what level of load should the region be pre- pared to commit resources? The overall modeling approach is one that combines features of decision analysis and simulation. Decision analysis is a branch of operations research involving the evaluation of a decision in light of the uncertainty that confronts the decision maker. It allows estimation of the consequences of a decision across a range of outcomes for uncertain variables and, given the probabilities for those outcomes, allows calculation of the expected value of the decision. This is essen- tially the problem to be solved here. What are the expected cost consequences of a partic- ular resource decision or set of decisions in light of future load uncertainty? It should be pointed out that the model is not intended to be an optimizer. It does not attempt independently to find the best resource decision or decision strategy. The decisions or strategies are user-defined inputs to the model, and the model is simply a tool to allow the evaluation of the actions represented in the input. By comparing the results produced by one set of decisions ver- sus another, it is possible to discern the advantages of one over another. Major Features Load Uncertainty The uncertainty represented here is the one inherent in the long-term load trend. Alter- native load paths all start at the current load level but may end up at any point between the low and high forecasts. The user has control over the size of the load range, the shape of the distribution of ending load values, and the amount of variation present in the individual load paths. However, the model has little knowledge about where a load path will eventually lead. It has limited forecasting abil- ity and continually updates forecasts as it moves through time, but it is blind to the future load within the limits of the forecast range. Two-Stage Resource Decisions For any particular resource, decisions are made in two steps: a decision to initiate an option, and a decision to start construction or build. Once an option decision is made, the resource passes through an option period before it moves into the option inventory. Once in inventory, it becomes available to build. If it is not built before the end of its shelf life, it expires and is no longer available as a regional resource. Conservation Program Management Conservation is generally thought of as a very flexible, short lead time resource. How- ever, in periods of rapid load growth and high need, its flexibility will be influenced by pro- gram acceleration characteristics and the maximum rates for program development. Conversely, during periods of surplus, con- servation flexibility is dependent on how quickly programs can be decelerated, and the minimum levels at which they can be run. The model applies user-defined maximum and minimum program development rates, accelerations, and decelerations as con- straints to manage program activity. In this way the flexibility or limitations of program scheduling characteristics can be valued in cost-effectiveness analysis. Major Decision Variables The following are the major inputs available to the user to control definition of studies. ¢ Option Level: The level of load within the forecast range for which options should be secured. Chapter 8 Load Selected End Point 0 5 10 Years 15 20 Figure 8-57 Decision Model Load Selection ¢ Build Level: The level of load within the forecast range for which resources should be built. « Resource Strategy: Specification of the preferred conservation programs and gen- erating resources as a function of time and load level. Resource supply limits as a func- tion of load can be used to differentiate resource preferences across the load range; e.g., build 1,000 megawatts of com- bustion turbines in a high load case, but build none in the low. ¢ Forced Option and Build Decisions: Specific option and build decisions to be made regardless of load level or need. A Typical Model Simulation This section will briefly describe a typical Decision Model simulation, giving more detail than the sections above. It describes six general steps: load selection, option and build requirements, resource choice, capital costing, production costing and treatment of end effects. Load Selection The first step the model takes is the selection of a load. This process is shown on Figure 8-57. The model will choose a load end point consisting of two components. It chooses values separately for loads, exclusive of the direct service industries, and direct service industry loads. The model assumes inde- pendence between non-direct service indus- try and direct service industry loads. The model then determines four five-year trends to reflect the general time structure of the forecast, which does not have constant load growth rates over the entire planning horizon, and the time pattern of the industry's activity reflected in the forecasts. Finally, the model applies a load shape from one of three sets to the five-year trends to give the load actually observed by the model for planning and costing. The three sets of load shapes have low, medium and high volatility in their deviations from the load trends. The user selects the set from which to draw. The Council's studies have been done using medium volatility. Figure 8-58 illustrates some examples of observed load paths gen- erated by the model. 8-37 Chapter 8 8-38 30,000 27,500 25,000 22,500 20,000 17,500 15,000 OS & So Vv b © S 9 v ’ © PDP DW DW VP WMO OY OS PD DD DD WM PF & LF Figure 8-58 Example Decision Model Load Paths Load Original High Forcast Range Forcast Current Year. 90% Option Level | 50% Build Level | | I uses Current Year Longest Resource Lead Time Figure 8-59 Decision Model Option and Build Level Option and Build Requirements The selection of option and build require- ments is depicted in Figures 8-59 and 8-60. Figure 8-59 shows the use of the option and build levels at a point part way through the model's simulation. The model has followed a varying load path from 1985 to the current year. It still sees a forecast range as it looks forward the length of the longest lead time of the resources it has available to it. However, the megawatt range is narrower than the orig- inal range. The high growth rate still is achievable, but since the model is now at a middle point in the range it can never reach the Council's high load itself. The option level and build level are selected by the user, and the current Council values are shown. Figure 8-60 repeats part of Figure 8-59 and shows an example of the actual resource and build decisions, given a set of previous deci- sions. This diagram shows a set of existing resources plus a set of build decisions that were made in previous years. Since option decisions were also made in previous years, there are some resources now in the option inventory from which to choose in making the build decision. As shown in this example, while there are sufficient options available to build to the forecast 50 percent build level, additional options will need to be acquired to maintain the option level at 90 percent. The model makes all these decisions, calculates production costs and capital costs for the current year at the observed load, then steps forward another year, discovers a new load, and repeats the process. Resource Choice Figure 8-61 is a simplified illustration of the process of resource choice. Resources are ranked in priority order in an input file. The model only sees the priority order for its option and build choices; these choices are not made by the model on the basis of rela- tive cost. The resource priority order is deter- mined externally to the model by the user. The Council determined the order through a process using simple screening of levelized cost, more complex comparison of resources with the System Analysis Model, and multi- ple trials of priority orders using the Decision Model. Once this priority order is established, the model attempts to choose the resources in this order. In the example shown in Figure 8-61, energy from nondiscretionary conser- vation programs or forced resource deci- sions, represented by block A, would be scheduled automatically. Energy from discre- tionary conservation programs, represented as block B, would be managed to meet energy targets for the individual conservation programs, subject to program penetration constraints. Resource C, a generating resource with a three-year lead time, has its first point of need beyond its lead time, and would require no decision other than to con- tinue to hold in inventory. Resource D, how- ever, is projected to be needed at its lead time of six years, and a decision to initiate con- struction would be made. There can be occurrences where the resource priority is not followed explicitly. Events such as sudden spurts in load growth may require scheduling resources with lower priority, but shorter lead time, in order to maintain balance with respect to the option and build levels specified. It is also possible that reductions in observed load growth may cause options to expire before they can be used, and may lead to resource choice out of order. Capital Costing Figure 8-62 is a rough illustration of the pro- cess of calculating capital costs in the Deci- sion Model. The top portion of the figure shows various important time points for the capital costing of a resource: option decision, option arrival, build decision, build arrival or in-service date, and retirement. The lower portion of the figure identifies the nominal dollar capital revenue requirements that are observed by the model in each year from the option decision to retirement of the resource. The figure is only to scale in a very general sense. Chapter 8 Load Additional Option Requirements ‘T In Current Year Build Decision In Current Year Previously Committed Builds Existing Lo + Resources Current Se Year Longest Resource Options Lead Time Available From Inventory 1985 Figure 8-60 Decision Model Option and Build Requirements Net Resource Requirements Projected Requirements Surplus Resource Additions , A. Non-discretionary conservation programs or forced resource decision B. Discretionary conservation programs C. Generating resource - 3 year lead time D. Generating resource - 6 year lead time Figure 8-61 Decision Model Build Decisions 8-39 Chapter 8 Timing Milestones Option Option m™ Build Decision Study 3 Build Physical Decision Arrival (In Service) Retirement 1 2 4 Start —— —— Option In Service Process Held In Under 5 Construction Inventory Revenue Economic Life Requirements Hold Costs Option Revenue Requirements 4 5 Economic Lite (May Not Equal Physical Life) Construction Revenue Requirements Figure 8-62 Decision Model Process of Calculating Capital Costs Figure 8-62 shows that revenue require- ments for options begin at the end of the option lead time; payments during that time are assumed capitalized until then. The option costs are put into revenue require- ments over a period equal to the economic life of the resource. If the resource is never built, the remaining unrecovered option costs are placed directly into revenue requirements in the year the option is lost. Option revenue requirements are calculated using a nominal levelized fixed charge rate. (Chapter 4 of this volume gives background on the concepts of “nominal,” “real,” and “levelized.”) Hold costs are put directly into revenue requirements each year as they are incurred and are shown as increasing in nominal terms because of assumed inflation. Finally, construction costs are capitalized to the in-service date of the resource, and then converted to annual reve- nue requirements over the economic life of the resource using a nominal levelized fixed charge rate, similar to the treatment of option costs. 8-40 The model will eventually have the ability to convert levelized fixed charge rates to the uneven pattern of actual nominal revenue requirements (see Volume II, Chapter 4), but this capability has not been completed yet. Production Costing Production costing is based on a composite system model similar to that used for sea- sonal studies in the System Analysis Model. Because of the dominance of energy issues in Northwest power planning, it is an energy model only; there is currently no treatment of capacity. Simulation of hydropower system operation is based on a one dam model in which total hydro energy capability, natural streamflow energy, reservoir draft, and limits on draft and refill for the entire system are specified as single values for the various sea- sons and water conditions. Data for the hydropower model are based on the result of critical period studies and the 40-year hydro regulation studies performed as part of the Northwest Regional Forecast. To capture the impact of streamflow variability, the model uses complete enumeration of ten represen- tative water conditions, and weights the results in accordance with the 102-year water record. Four discrete time periods are used for evaluation within each operating year: September-December, January-April, May, and June-August. May is modeled sepa- rately to provide better resolution on the sys- tem impact of the spring fish flows. Thermal units are modeled with deration for equivalent availability and are shaped sea- sonally according to specified maintenance schedules. Nuclear units are treated as must run; all other thermal operation is modeled with economic dispatch against firm, inter- ruptible, and secondary market load blocks, as needed under the various hydro condi- tions. The secondary market is modeled as a four-tiered market with prices and seasonally shaped demand blocks changing through time. Conservation programs and renewable generating resources are typically treated as seasonally shaped load reduction resources. Any firm load not met with regional resources is assumed to be met with an out-of-region purchase at a specifiable price. (The Council currently uses 150 mills, or 15 cents, per kilo- watt-hour.) Curtailments of interruptible load are priced near interruptible rates. Treatment of End Effects End effects are incurred in any model because resources have different lives and, in addition, many of them last beyond the study horizon of the model. A resource that costs the same amount but lasts twice as long as another will be more valuable. But if both resources retire outside the study hori- zon of a model, the model will not be able to tell. One means of dealing with this problem, used in the end effects treatment for the Sys- tem Analysis Model and in the Decision Model, is to extend the simulation period in a simplified way until all resources constructed during the study horizon have retired. These resources are all replaced by the same kind of resource and the study is then truncated after the only remaining resources are the replacement resources. The use of constant real levelized capital costs for these replace- ment resources ensures that studies with resources of different lifetimes are comparable. There is an additional end effects problem to be dealt with in the Decision Model. Since the model options and builds resources under load uncertainty, some simulations will end up surplus at the end of the study horizon and some simulations will end up deficit. This distribution of ending load/resource balances can be a function of the resource strategy employed, i.e., the option and build levels, the resource priorities, forced resource deci- sions, and the amount of load variation present. While production and capital costing is car- tied out beyond the study horizon (normally 20 years), the forecasting and option and build steps stop at the study horizon. To the extent that strategies being tested have con- sequences like persistent overbuilding or underbuilding, the surpluses or deficits need to be carried beyond the study horizon to the end of the terminal horizon. (This can be as long as an additional 70 years to deal with the issues mentioned in the previous para- graph.) In some uses of the model, however, the user may wish to ensure that a certain level of load/resource balance is attained for the post-study horizon period. Because of these varying requirements, there are three methods available to calculate the terminal horizon load/resource balance. The three methods are illustrated in Figure 8-63 for a simulation that varies from deficit to surplus over the last five years of the study horizon, but that ends in surplus. The first method simply extends each simu- lation’s observed twentieth year surplus or deficit to the terminal horizon for that simula- tion. This is illustrated in the top diagram in Figure 8-63. The second method adjusts each simula- tion's twentieth year surplus or deficit to an input target load/resource balance. In this case, the target was zero surplus or deficit. It does this by building additional resources in the twenty-first year if more resources are needed, or by not replacing resources as they retire if fewer resources are needed. The latter process usually reaches the target within ten years after the study horizon. Chapter 8 Method 1 Resource Load Load ‘Surplus Load 1 60% Deficit Resource: Load 2 [40% Surplus Figure 8-63 Decision Model End Effects Treatments The third method is an elaboration of the first. Itis illustrated in the bottom diagram in Figure 8-63. This method calculates the average surplus and the average deficit over the last five years of the study horizon. (The number of years is a user input; the Council uses five.) It then conducts its production costing twice for the terminal horizon period: once using the average surplus and once the aver- age deficit value. Finally, the model weights the two results by the percent of time in the last five years of the study. horizon that the simulation was surplus and was deficit. While this is a more precise calculation than the first method, it has the disadvantage of requiring substantially more computer time because of the doubled terminal horizon production costing for each simulation. Section E: Lost Opportunity Resources A lost opportunity resource is a potential electric power generating resource or a potential electric power conservation mea- sure which is currently available to the region and which, if not acquired or otherwise secured now, will no longer be available and cost-effective to the region. If a lost oppor- tunity resource is not secured, it will have to be replaced in the future by a less cost-effec- tive resource. A lost opportunity resource is cost effective and should be secured if the present value system cost of the investment to secure and maintain the resource by the region, as determined by the Council, is less than the present value system cost of all other resources included in the Council's resource portfolio that might have to replace it. Avoided cost studies, regarding the eco- nomics of lost opportunity resources and their value during the current surplus, were discussed in Volume |, Chapter 8. This sec- tion presents a general description of the various types of lost opportunity generating resources. 8-41 Chapter 8 Table 8-4 Inventory of Potential Lost Opportunity Resources ENERGY TYPE OF RESOURCE (average megawatts) Loss of Generation Potential> Municipal water systems (28 projects) 80.0 Biomass incineration (2 projects) 10.0¢ Solid waste disposal (4 projects) 62.0 Cogeneration and misc. (5 projects) 45.7¢ Out-of-Region Sale Coal (2 projects) 138.7¢ Hydropower (2 projects) 21.3¢ Loss of Development Rights Licensed thermal sites (3 projects) 1,406.0 Loss of Development Incentives (3 hydropower projects) 53.4¢ Generation in Lieu of Transmission (1 project) — TOTAL 1,817.1 aThe projects and energy listed in this table are taken from the Bonneville preliminary inventory of lost opportunity resources, and do not necessarily agree with current Council inventories of these resources. Not included is 65 megawatts from the proposed Fast Flux Test Facility (FFTF) power addition project, earlier evaluated by Bonneville as a potential lost opportunity resource. Uncertainties regard- ing long-term Congressional funding of the FFTF project are considered by Bonneville to be to great to justify acquisition of this resource. °The energy production of one or more projects within this category was not estimated; thus the actual total would be greater than indicated. Availability of Potential Lost Opportunity Resources The availability and cost effectiveness of potential lost opportunity conservation resources in the residential and commercial sectors were well understood at the time of the 1983 Power Plan, leading the Council to call for the acquisition of these resources, where cost effective, through implementation of conservation standards. In contrast, the extent and cost effectiveness of potential lost opportunity generating resources remain less well understood. In order to gain additional information on potential lost opportunity resources, the Council included Action Item 13.3 in the 1983 Power Plan. This action item called upon Bonneville to: 8-42 Identify, by project, specific resources which may be lost to the region if decisions to acquire an option or to acquire the resources are not made. This inventory should recognize each resource sponsor's requirements for keeping the resource avail- able to the region. In response to this action item, Bonneville has compiled a preliminary inventory of potential lost opportunity generation resources. This inventory, summarized into five classes of potential lost opportunities, is shown in Table 8-4. Loss of Generation Potential The projects representing loss of generation potential are related to scheduled non-power developments that could be modified to pro- duce electric power as a byproduct. These projects include: 1) municipal and hatchery water supply systems with an available water head that could be used for hydropower gen- eration; 2) proposed solid waste incinerators that could be modified to recover energy for power generation; 3) landfills that could be provided with methane collection systems for use in powering generation equipment; and 4) industrial facilities that could be modified to accommodate cogeneration. Not included in the present inventory are planned irrigation projects with the potential for associated hydropower development, building cogeneration potential or planned transmis- sion and generation projects with additional system efficiency improvement potential. Because the basic power source exists for other purposes, the incremental lead time, cost and environmental impact are poten- tially less than for facilities constructed spe- cifically for power generation. This gives proj- ects in this category desirable planning qualities, including short development lead times, small increments of capacity, low cost and modest incremental environmental impact. The power generation capability of these projects can be secured by incorporating design features during initial construction to facilitate later addition of power generation equipment. For example, taps could be pro- vided in a new municipal water supply sys- tem to accommodate later addition of tur- bines. The power generation equipment can be added when need-for-power dictates. Out-of-Region Sales Potential lost opportunities for out-of-region sales include two types of projects. One type is existing regional power generation resources currently offered for sale as excess to the needs of the current owners. All proj- ects of this type on the current Bonneville inventory are coal-fired power plants. The capability of these plants may be sold out- side, and potentially lost to the region. Regional acquisition of the capability of these projects would likely be cost effective if a sale of power outside the region, incorporating Chapter 8 callback provisions, could be arranged after acquisition. With such an arrangement, these resources would appear to be quite valuable. Power could be made available to the region with short lead time (the time period negotiated in the callback provisions) and in appropriate increments. Incremental environmental impact within the region would be negligible. Costs would be representative of existing thermal plants. A second type of out-of-region sale project is proposed projects potentially qualifying for sale under the provisions of the Public Utility Regulatory Policies Act (PURPA). These projects could be lost to the region if sales to out-of-region purchasers were negotiated. All such projects in the current inventory are hydropower projects, although other qualify- ing facilities, such as cogeneration, might materialize. Acquisition of these projects is advantageous to the region to the extent that they are cost effective. Loss of Development Rights These opportunities consist of currently undeveloped sites for which land rights, pre- liminary engineering design, baseline environmental data and licenses have been partly or fully obtained. Currently there are three thermal sites in the inventory: Creston, Washington; Wyodak, Wyoming; and Boardman, Oregon. Not included are the Salem, Montana, thermal site or numerous partially or fully licensed hydropower sites in the region. If the present value cost of acquisition and maintenance of the development rights is found to be less than the present value cost of reacquisition of these assets, if and when needed, the development rights should be acquired and maintained as an option by the region. In assessing the value of develop- ment rights, consideration should be given to the suitability of thermal sites for siting sec- ondary hydropower firming resources such as combustion turbines. Loss of Development Incentives These opportunities consist of several hydro- power projects for which special incentives may expire unless exercised. Securing this type of lost opportunity would likely require construction of the project. Because of this, acquisition would likely be cost effective under the present surplus only for very low- cost projects. Generation in Lieu of Transmission This opportunity presently includes one pro- spective cogeneration project located in an area needing transmission upgrade to serve increased load. Construction of the project, offsetting the transmission load, may be more cost effective than the planned trans- mission expansion. The reliability of the proj- ect must be considered in assessing the cost effectiveness of the project in comparison with upgraded transmission. Additional Resource Information The current inventory, while adequate for ini- tial identification of potential lost opportunity generation resources, is not adequate to determine if specific resources should be acquired. Additional information required includes: * The timing and duration of the present “win- dow of opportunity” for each resource. The nature and cost of actions that might be taken to secure the resource. ¢ The cost, availability and shelf life of the resource if actions are taken to secure the resource or to extend the window of opportunity. * The cost, availability and shelf life of the resource if actions are not taken to secure the resource. Bonneville should expand its efforts to include the above information in the lost opportunity data base. The data base should be expanded to include the following resource types: « Planned irrigation development with power potential. ¢ Planned generation, transmission and dis- tribution system upgrades with additional system efficiency improvement potential. ¢ Hydropower development rights. ¢ Building cogeneration potential. Lost opportunity resources are not, by defini- tion, static. For this reason it is desirable to periodically update the lost opportunity resource data base. Resource Evaluation and Acquisition Acquisition of certain lost opportunity resources may be cost effective even during the current period of surplus. Resources would likely be cost effective if their acquisi- tion results in a present value system cost less than the forecast present value system cost without acquisition. This determination can be made using available system plan- ning models. Certain resources may have energy costs less than the value of surplus energy. Immediate development of such a resource may be cost effective. For example, preliminary information on the cost of system efficiency improvements indicates that the cost of certain improvements of this type is extremely low, with resulting costs of energy less than the value of surplus. Because near-term acquisition of certain lost opportunity resources may be cost effective, actions should be taken to develop the institutional mechanisms to acquire lost opportunity resources. This will require a methodology for the evaluation of lost oppor- tunity resources and adoption of a policy for lost opportunity resource acquisition. The policy should include consistent criteria for determining when a lost opportunity resource should be acquired. These activities are called for in Volume |, Chapter 9. 8-43 The tables in this appendix contain the data supporting the resource portfolio graphics, figures 8-8 and 8-9. Eight tables are included, one for each of the four load forecasts for the region as a whole, and for just the public utility and direct service industry customers of the Bonneville Power Administration. The loads shown are firm loads only and have been adjusted for transmission and distribu- tion losses. The “existing resource” category pendix 8-A Regional and Public Utility Resource Schedules contains hydro Firm Energy Load Carrying Capability (FELCC), existing thermal, mis- cellaneous resources, and imports net of exports. Values for existing resources were derived based on the 1985 Northwest Regional Forecast, compiled and published by PNUCC. The resource schedules shown here are based on the assumption of perfect knowl- Table 8-A-1 Regional High (1985-1995) edge of load, and they attain load/resource balance in all load conditions within the con- straints of the current surplus and generating resource unit size. Line item entries for con- servation programs show cumulative energy developed through time for each program. Each line item entry for the generating resource represents the energy associated with a set of new additions. System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 16,258 16,634 17,262 17,915 18592 19,117 19,614 20,157 20,726 21,380 Observed Rate 0.00% 2.31% 3.78% 3.78% 3.78% 2.82% 2.60% 2.77% 2.82% 3.16% Resources: Existing 18,824 18,834 18,605 18,555 18,531 18,522 18,461 18,415 18,006 18,011 Conservation Programs: MCS Single Family 5 22 41 64 91 120 151 183 217 254 MCS Multifamily 1 3 6 8 12 15 19 23 27 31 MCS Commercial 0 15 36 57 78 98 118 139 161 183 Refrigerators/Freezers 0 0 0 0 0 0 10 26 49 72 Water Heat 0 0 0 0 0 0 1 29 53 79 Manufactured Homes 0 0 1 3 4 5 io 9 10 12 Agricultural 0 2 10 20 30 40 50 60 70 80 Existing Commercial 0 15 60 125 195 265 335 405 475 545 Trans & Distr Efficiency 0 1 4 8 13 18 23 28 33 34 Existing Space Heat 8 16 34 64 109 159 209 259 309 359 Existing Industrial 0 0 20 80 170 270 370 450 450 450 Subtotal 14 74 212 429 702 990 1,303 1,611 1,854 2,099 Generating Resources: Hydropower Efficiency 0 0 0 0 0 60 60 60 60 60 Hydropower Efficiency 0 0 0 0 0 0 15 15 15 15 Hydropower Efficiency 0 0 0 0 0 0 0 10 10 10 Hydropower Efficiency 0 0 0 0 0 0 0 0 15 15 Hydropower Efficiency 0 0 0 Oo oO Oo 0 0 0 10 Subtotal 0 0 0 Oo 0 60 75 85 100 110 Combustion Turbines 0 0 0 0 0 0 0 0 714 714 Subtotal 0 0 0 0 0 0 0 0 714 714 Small Hydropower 0 0 0 0 0 0 0 52 52 52 Small Hydropower 0 0 0 0 0 0 0 0 7 7 Small Hydropower 0 0 0 0 0 0 0 0 0 140 Subtotal 0 0 0 0 0 0 0 52 59 199 (table continued on next page) 8-A-1 Appendix 8-A Table 8-A-1 (continued) Regional High (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Cogeneration 0 0 0 0 0 0 0 0 0 255 Cogeneration 0 0 0 0 0 0 0 0 0 0 Cogeneration 0 0 0 0 0 0 0 0 0 0 Cogeneration 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 255 Licensed Coal 0 0 0 0 0 0 0 0 0 0 Licensed Coal 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 Oo 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Unlicensed Coal 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 Total Firm Resources 18,838 18,908 18,817 18,984 19,233 19,572 19,839 20,163 20,733 21,388 Load/Resource Balance 2,580 2,274 1,555 1,069 641 455 225 6 7 8 Regional High (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 22,027 22,738 23,388 24,033 24,849 25,375 26,045 26,746 27,485 28,260 Observed Rate 3.038% 3.238% 2.86% 2.76% 3.40% 2.12% 2.64% 2.69% 2.76% 2.82% Resources: Existing 17,991 18,012 17,948 17,916 17,784 17,796 17,780 17,740 17,643 17,790 Conservation Programs: MCS Single Family 295 337 381 425 470 515 560 607 656 705 MCS Multifamily 36 42 47 53 58 64 69 75 81 87 MCS Commercial 206 229 253 277 303 328 354 380 392 398 Refrigerators/Freezers 97 122 148 176 204 233 261 291 321 352 Water Heat 106 135 165 196 227 259 292 326 361 396 Manufactured Homes 14 16 19 21 23 26 28 30 33 35 Agricultural 90 100 110 120 123 123 124 124 124 124 Existing Commercial 615 685 755 801 801 801 802 802 802 802 Trans & Distr Efficiency 34 34 34 34 34 34 34 34 34 34 Existing Space Heat 409 455 455 455 455 455 455 455 455 455 Existing Industrial 450 450 450 450 450 450 450 450 450 450 Subtotal 2,352 2,605 2,817 3,008 3,148 3,288 3,429 3,574 3,709 3,838 (table continued on next page) 8-A-2 Appendix 8-A Table 8-A-1 (continued) Regional High (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Generating Resources: Hydropower Efficiency 60 60 60 60 60 60 60 60 60 60 Hydropower Efficiency 15 15 15 15 15 15 15 15 15 15 Hydropower Efficiency 10 10 10 10 10 10 10 10 10 10 Hydropower Efficiency 15 15 15 15 15 16 15 15 15 15 Hydropower Efficiency 10 10 10 10 10 10 10 10 10 10 Subtotal 110 110 110 110 110 110 110 110 110 110 Combustion Turbines 714 714 714 714 714 714 714 714 714 714 Subtotal 714 714 714 714 714 714 714 714 714 714 Small Hydropower 52 52 52 52 52 52 52 52 52 52 Small Hydropower it 7 7 7 7 7 7 7 7 7 Small Hydropower 140 140 140 140 140 140 140 140 140 140 Subtotal 199 199 199 199 199 199 199 199 199 199 Cogeneration 255 255 255 255 255 255 255 255 255 255 Cogeneration 0 0 5 5 5 5 5 5 5 5 Cogeneration 0 0 0 40 40 40 40 40 40 40 Cogeneration 0 0 0 0 0 0 0 20 20 20 Subtotal 255 255 260 300 300 300 300 320 320 320 Licensed Coal 452 452 452 452 452 452 452 452 452 452 Licensed Coal 0 452 452 452 452 452 452 452 452 452 Subtotal 452 904 904 904 904 904 904 904 904 904 Unlicensed Coal 0 0 452 452 452 452 452 452 452 452 Unlicensed Coal 0 0 0 452 452 452 452 452 452 452 Unlicensed Coal 0 0 0 0 904 904 904 904 904 904 Unlicensed Coal 0 0 0 0 0 452 452 452 452 452 Unlicensed Coal 0 0 0 0 0 0 452 452 452 452 Unlicensed Coal 0 0 0 0 0 0 0 452 452 452 Unlicensed Coal 0 0 0 0 0 0 0 0 904 904 Unlicensed Coal 0 0 0 0 0 0 0 0 0 452 Subtotal 0 0 452 904 1,808 2,260 2,712 3,164 4,068 4,520 Total Firm Resources 22,073 22,799 23,404 24,055 24,967 25,571 26,148 26,725 27,667 28,395 Load/Resource Balance 46 61 16 22 118 196 103 -21 182 135 8-A-3 Appendix 8-A Table 8-A-2 Regional Medium-High (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 16,258 16,466 16,833 17,208 17,591 17,969 18,345 18,727 19,108 19,580 Observed Rate 0.00% 1.28% 2.23% 2.23% 2.23% 2.15% 2.09% 2.08% 2.038% 2.47% Resources: Existing 18,824 18,834 18,605 18,555 18,531 18,522 18,461 18,415 18,006 18,011 Conservation Programs: MCS Single Family MCS Multifamily MCS Commercial Refrigerators/Freezers Water Heat Manufactured Homes Agricultural Existing Commercial Trans & Distr Efficiency Existing Space Heat Existing Industrial Subtotal Generating Resources: Hydropower Efficiency Hydropower Efficiency Hydropower Efficiency Hydropower Efficiency Hydropower Efficiency Hydropower Efficiency Hydropower Efficiency Subtotal Combustion Turbines Combustion Turbines Combustion Turbines Subtotal Small Hydropower Small Hydropower Small Hydropower Small Hydropower Small Hydropower Small Hydropower Subtotal Cogeneration Subtotal Licensed Coal Licensed Coal Subtotal Unlicensed Coal Subtotal Total Firm Resources Load/Resource Balance 32 48 68 6 10 14 a7 28 39 109 131 153 176 22 26 31 35 61 73 85 96 10 25 47 69 10 27 49 72 9 11 13 15 30 40 50 60 195 265 335 405 13 18 23 28 163 213 263 313 80 170 270 370 702 999 1,319 1,639 ocooos-0ONWON a =< Booonoo =u ° > o eohSnooSaB 69 0 269 elomoooocoeoeo-ua = Blos o ° ea 8a & 5 5 55 on a ooooo°$o a a cloolooolo coloooaoaa oloacaaoaolaaoaaGa a0 Oo colooloooloocolcooooooolcno0ocol!tconcnaoeo0o o clooloo colo oloooeoae ao oOloaoaaoaolaaoaaaaG oo clooloooclo colo oooooco colo oaooltconoooo & cloocoloo colo coloacoaaoaaaaoOloaconoolaoaaoaaoaaaa eo clocloo clo cloc oc GC CoO Cloc co Bloco oon clooloooloolooooooco!ccecnd Oo oO 0 0 Oo 0 0 Oo 0 0 0 0 0 0 0 oO 0 Oo Oo 0 0 Oo 0 0 0 0 4 2 Pr © = 6 18,971 19,163 19,41 19,380 1,002 818 687 272 = 18,878 18,690 18,703 2,412 1,857 1,495 elslalorolororcilarelsretsiorsorclsais ciloroelsreilcrenre SHE Bolboobaoboooopobooobouoeeoc nN 8-A-4 (table continued on next page) Appendix 8-A Regional Medium-High (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 20,006 20,471 20,863 21,240 21,766 21,990 22,342 22,708 23,094 23,498 Observed Rate 2.18% 2.32% 1.91% 1.81% 2.48% 1.03% 1.60% 1.64% 1.70% 1.75% Resources: Existing 17,991 18,012 17,948 17,916 17,784 17,796 17,780 17,740 17,643 17,790 Conservation Programs: MCS Single Family 199 223 246 269 290 310 329 348 368 387 MCS Multifamily 40 45 49 54 59 63 67 72 76 81 MCS Commercial 108 120 131 143 155 167 178 189 194 195 Refrigerators/Freezers 91 114 137 161 184 207 228 250 272 293 Water Heat 95 120 145 171 196 222 247 273 299 324 Manufactured Homes 17 19 22 24 26 28 30 32 34 36 Agricultural 70 80 90 100 105 105 105 105 105 105 Existing Commercial 475 545 611 611 611 611 611 611 611 611 Trans & Distr Efficiency 33 34 34 34 34 34 34 34 34 34 Existing Space Heat 363 413 455 455 455 455 455 455 455 455 Existing Industrial 450 450 450 450 450 450 450 450 450 450 Subtotal 1,941 2,163 2,370 2,472 2,565 2,652 2,734 2,819 2,898 2,971 Generating Resources: Hydropower Efficiency 55 55 55 55 55 55 55 55 55 55 Hydropower Efficiency 5 5 5 5 5 5 5 5 5 5 Hydropower Efficiency 10 10 10 10 10 10 10 10 10 10 Hydropower Efficiency 0 10 10 10 10 10 10 10 10 10 Hydropower Efficiency 0 0 10 10 10 10 10 10 10 10 Hydropower Efficiency 0 0 0 10 10 10 10 10 10 10 Hydropower Efficiency 0 0 0 0 10 10 10 10 10 10 Subtotal 70 80 90 100 110 110 110 110 110 110 Combustion Turbines 0 178 178 178 178 178 178 178 178 178 Combustion Turbines 0 0 178 178 178 178 178 178 178 178 Combustion Turbines 0 0 0 357 357 357 357 357 357 357 Subtotal 0 178 356 713 713 713 713 713 713 713 Small Hydropower 7 7 7 7 7 7 7 7 7 7 Small Hydropower 0 37 37 37 37 37 37 37 37 37 Small Hydropower 0 0 55 55 55 55 55 55 55 55 Small Hydropower 0 0 0 0 95 95 95 95 95 95 Small Hydropower 0 0 0 0 0 0 0 0 2 2 Small Hydropower 0 0 0 0 0 0 0 0 0 Z Subtotal 7c 44 99 99 194 194 194 194 196 198 Cogeneration 0 0 0 0 210 210 210 210 210 210 Subtotal 0 0 0 0 210 210 210 210 210 210 Licensed Coal 0 0 0 0 0 452 452 452 452 452 Licensed Coal } 0 0 0 0 0 0 452 452 452 Subtotal 0 0 0 0 0 452 452 904 904 904 Unlicensed Coal 0 0 0 0 0 0 0 } 452 452 Subtotal 0 0 0 0 0 0 0 0 452 452 Total Firm Resources 20,009 20,477 20,863 21,300 21,576 22,127 22,193 22,690 23,126 23,348 Load/Resource Balance 3 6 0 60 -190 137-149 “18 32 = -150 8-A-5 Appendix 8-A Table 8-A-3 Regional Medium-Low (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 16,258 16,282 16,367 16,454 16,540 16,721 16,956 17,213 17,346 17,798 Observed Rate 0.00% 0.15% 0.52% 0.53% 0.52% 1.09% 1.41% 1.52% 0.77% 2.61% Resources: Existing 18,824 18,834 18,605 18,555 18,531 18,522 18,461 18,415 18,006 18,011 Conservation Programs: MCS Single Family 5 10 16 24 34 44 56 68 80 92 MCS Multifamily 1 3 5 8 12 16 20 24 28 33 MCS Commercial 0 2 5 10 14 20 25 31 37 44 Refrigerators/Freezers 0 0 0 0 0 0 7 19 35 51 Water Heat 0 0 0 0 0 0 8 21 39 57 Manufactured Homes 0 0 1 2 4 6 7 9 1 13 Agricultural 0 0 0 0 0 0 0 2 10 20 Existing Commercial 0 0 0 0 0 0 0 0 15 60 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 1 4 Existing Space Heat 8 16 28 36 44 52 60 68 76 94 Existing Industrial 0 0 0 0 0 0 0 0 0 0 Subtotal 14 31 55 80 108 138 183 242 332 468 Generating Resources: Hydropower Efficiency 0 0 oO 0 0 0 0 0 0 0 Hydropower Efficiency 0 0 0 0 0 0 0 0 0 0 Hydropower Efficiency 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 Oo 0 Combustion Turbines 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 0 Small Hydropower 0 0 0 0 0 0 0 0 0 0 Small Hydropower 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 0 Total Firm Resources 18,838 18,865 18660 18635 18639 18,660 18,644 18,657 18338 18,479 Load/Resource Balance 2,580 2,583 2,293 2,181 2,099 1,939 1,688 1,444 992 681 oe nn, (table continued on next page) 8-A-6 Appendix 8-A Regional Medium-Low (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 18,158 18,489 18,773 19,047 19,465 19,612 19,899 20,197 20,503 20,811 Observed Rate 2.02% 1.82% 1.54% 1.46% 2.19% 0.76% 1.46% 1.50% 1.52% 1.50% Resources: Existing 17,991 18,012 17,948 17,916 17,784 17,796 17,780 17,740 17,643 17,790 Conservation Programs: MCS Single Family 105 118 131 144 157 171 184 197 211 225 MCS Multifamily 37 41 46 50 55 60 65 70 75 80 MCS Commercial 50 57 64 71 78 85 92 99 104 109 Refrigerators/Freezers 68 84 101 118 135 153 170 188 206 224 Water Heat 76 96 116 136 157 179 200 222 244 266 Manufactured Homes 14 16 18 20 22 24 26 28 30 32 Agricultural 30 40 50 60 70 80 90 100 105 105 Existing Commercial 125 195 265 335 405 475 475 475 475 475 Trans & Distr Efficiency 8 13 18 23 28 33 34 34 34 34 Existing Space Heat 132 181 231 281 331 381 431 455 455 455 Existing Industrial 0 0 20 80 170 270 370 450 450 450 Subtotal 645 841 1,060 1,318 1,608 1,911 2,137 2,318 2,389 2,455 Generating Resources: Hydropower Efficiency 0 0 80 80 80 80 80 80 80 80 Hydropower Efficiency 0 0 0 15 15 15 15 15 15 15 Hydropower Efficiency 0 0 0 0 15 15 15 15 15 15 Subtotal 0 0 80 95 110 110 110 110 110 110 Combustion Turbines 0 0 0 0 0 0 0 0 357 357 Subtotal 0 0 0 0 0 0 0 0 357 357 Small Hydropower 0 0 0 0 0 0 0 22 22 22 Small Hydropower 0 0 0 0 0 0 0 0 0 87 Subtotal 0 0 0 0 0 0 0 22 22 109 Total Firm Resources 18,636 18,853 19,088 19,329 19,502 19,817 20,027 20,190 20,521 20,821 Load/Resource Balance 478 364 315 282 37 205 128 7 18 10 8-A-7 Appendix 8-A Table 8-A-4 Regional Low (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 16,258 16,064 15,826 15,591 15,360 15,295 15,322 15,367 15,436 15,587 Observed Rate 0.00% -1.19% -1.48% -1.48% -1.48% -0.42% 0.18% 0.29% 0.45% 0.98% Resources: Existing 18,824 18,834 18,605 18,555 18531 18,522 18,461 18,415 18,006 18,011 Conservation Programs: MCS Single Family 5 6 6 7 10 13 18 22 27 32 MCS Multifamily 1 1 1 2 3 4 6 9 11 MCS Commercial 0 1 2 4 6 8 10 13 15 18 Refrigerators/Freezers 0 0 0 0 0 0 iv 18 34 49 Water Heat 0 0 0 0 0 0 6 17 31 45 Manufactured Homes 0 0 0 0 0 1 2 2 3 4 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 0 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 8 16 28 36 44 52 60 68 76 84 Existing Industrial 0 0 0 0 0 0 0 0 0 0 Subtotal 14 24 37 49 63 78 109 148 195 243 Total Firm Resources 18,838 18,858 18,642 18,604 18,594 18,600 18,570 18,563 18,201 18,254 Load/Resource Balance 2,580 2,794 2,816 3,013 3,234 3,305 3,248 3,196 2,765 2,667 Regional Low (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) “PERIOD —s—~<“‘“CO™CSCS™CSC™*C*;*;*;*#«SO6~#«#«G-S7~«=«#97-88 © 98-99 «99-00 00-01 -—=«T-02,-=S«02-03«=«03-04 4-05 Observed Load 15,604 15,699 15,792 15,894 16,153 16,149 16,290 16,443 16,601 16,775 Observed Rate 0.11% 0.61% 0.59% 0.65% 1.63% -0.02% 0.87% 0.94% 0.96% 1.05% Resources: Existing 17,991 18,012 17,948 17,916 17,784 17,796 17,780 17,740 17,643 17,790 Conservation Programs: MCS Single Family 38 43 49 55 61 68 74 81 87 94 MCS Multifamily 13 16 18 20 22 25 27 30 32 35 MCS Commercial 21 24 27 30 34 38 41 45 48 51 Refrigerators/Freezers 65 81 98 114 130 146 161 176 191 206 Water Heat 60 76 92 110 127 145 163 182 200 219 Manufactured Homes 4 5 6 Z 8 9 10 1 12 13 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 0 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 92 100 108 116 124 132 140 148 156 164 Existing Industrial 0 0 0 0 0 0 0 0 0 0 Subtotal 293 345 398 452 506 563 616 673 726 782 Total Firm Resources 18,284 18,357 18,346 18,368 18,290 18,359 18,396 18,413 18,369 18,572 Load/Resource Balance 2,680 2,658 2,554 2,474 2,137 2,210 2,106 1,970 1,768 1,797 8-A-8 Appendix 8-A Table 8-A-5 Public High (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 8,444 8,548 8,809 9,079 9,356 9,576 9,784 10,011 10,252 10,568 Observed Rate 0.00% 1.23% 3.05% 3.07% 3.05% 2.35% 2.17% 2.32% 2.41% 3.08% Resources: Existing 10,303 10,345 10,190 10,212 10,231 10,238 10,244 10,233 9,957 9,982 Conservation Programs: MCS Single Family 2 9 18 28 39 52 64 78 92 108 MCS Multifamily 0 1 3 4 6 7 9 11 13 15 MCS Commercial 0 6 15 23 31 39 47 56 64 73 Refrigerators/Freezers 0 0 0 0 0 2 6 12 21 31 Water Heat 0 0 0 0 0 2 6 14 23 34 Manufactured Homes 0 1 1 2 3 4 5 6 7 8 Agricultural 0 0 0 Oo 1 4 8 12 16 20 Existing Commercial 0 0 0 0 6 24 50 78 106 134 Trans & Distr Efficiency 0 0 0 0 0 2 4 6 8 10 Existing Space Heat 4 8 13 16 19 26 41 61 81 101 Existing Industrial 0 0 0 0 0 0 0 0 0 15 Subtotal 6 25 50 73 105 162 240 334 431 549 Generating Resources: Hydropower Efficiency 0 0 0 0 Oo 0 0 0 30 30 Hydropower Efficiency 0 0 Oo 0 O 0 0 0 0 40 Subtotal 0 0 0 0 0 0 0 0 30 70 Combustion Turbines 0 0 0 0 0 0 0 0 0 0 Combustion Turbines 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 0 Small Hydropower 0 0 0 0 0 0 0 0 0 0 Small Hydropower 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 0 Cogeneration 0 0 0 0 0 0 0 0 0 0 Cogeneration 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 0 Licensed Coal 0 0 0 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 Oo Oo 0 Total Firm Resources 10,309 10,370 10,240 10,285 10,336 10,400 10,484 10,567 10,418 10,601 Load/Resource Balance 1,865 1,822 1,431 1,206 980 824 700 556 166 33 (table continued on next page) 8-A-9 Appendix 8-A Public High (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 10,811 11,116 11,387 11,652 12,070 12,207 12,479 12,761 13,057 13,364 Observed Rate 2.30% 2.82% 2.44% 2.33% 3.59% 1.14% 2.23% 2.26% 2.32% 2.35% Resources: Existing 10,063 10,367 10,364 10,351 10,315 10,340 10,364 10,351 10,290 10,296 Conservation Programs: MCS Single Family 125 143 161 180 198 217 236 256 276 297 MCS Multifamily 18 20 23 25 28 31 33 36 39 42 MCS Commercial 82 91 101 111 121 131 141 152 156 158 Refrigerators/Freezers 40 51 61 72 84 95 107 119 131 143 Water Heat 45 57 69 82 95 108 121 135 149 164 Manufactured Homes 9 1 12 13 15 16 18 19 21 22 Agricultural 24 28 32 36 40 44 48 49 49 50 Existing Commercial 162 190 218 246 274 302 320 320 320 321 Trans & Distr Efficiency 12 14 14 14 14 14 14 14 14 14 Existing Space Heat 121 141 161 181 182 182 182 182 182 182 Existing Industrial 60 130 210 290 337 337 337 337 337 337 Subtotal 698 876 1,062 1,250 1,388 1,477 1,557 1,619 1,674 1,730 Generating Resources: Hydropower Efficiency 30 30 30 30 30 30 30 30 30 30 Hydropower Efficiency 40 40 40 40 40 40 40 40 40 40 Subtotal 70 70 70 70 70 70 70 70 70 70 Combustion Turbines 0 0 0 0 357 357 357 357 357 357 Combustion Turbines 0 0 0 0 0 0 178 178 178 178 Subtotal 0 0 0 0 357 357 535 535 535 535 Small Hydropower 0 0 0 5 5 5 5 5 5 5 Small Hydropower 0 0 0 0 0 0 0 155 155 155 Subtotal 0 0 0 5 5 5 5 160 160 160 Cogeneration 0 0 0 0 0 0 0 60 60 60 Cogeneration 0 0 0 0 0 0 0 0 0 50 Subtotal 0 0 0 0 0 0 0 60 60 110 Licensed Coal 0 0 0 0 0 0 0 0 452 452 Subtotal 0 0 0 0 0 0 0 0 452 452 Total Firm Resources 10,831 11,313 11,496 11,676 12,135 12,249 12,531 12,795 13,241 13,353 Load/Resource Balance 20 197 109 24 65 42 52 34 184 “11 Appendix 8-A Table 8-A-6 Public Medium-High (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 8,444 8,473 8,618 8,766 8,916 9,071 9,230 9,391 9,554 9,795 Observed Rate 0.00% 0.34% 1.71% 1.72% 1.71% 1.74% 1.75% 1.74% 1.74% 2.52% Resources: Existing 10,303 10,345 10,190 10,212 10,231 10,238 10,244 10,233 9,957 9,982 Conservation Programs: MCS Single Family 2 7 13 21 29 38 47 56 66 75 MCS Multifamily 0 1 3 5 i” 8 10 12 15 17 MCS Commercial 0 3 ie 12 16 21 25 30 35 40 Refrigerators/Freezers 0 0 0 0 0 2 6 12 21 29 Water Heat 0 0 0 0 0 2 6 13 22 31 Manufactured Homes 0 1 2 3 4 6 7 8 10 11 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 0 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 4 8 13 16 19 22 25 28 31 34 Existing Industrial 0 0 0 0 0 0 0 0 0 0 Subtotal 6 20 38 57 75 99 126 159 200 237 Generating Resources: Hydropower Efficiency 0 0 0 0 0 oO Oo 0 0 0 Hydropower Efficiency 0 0 oO 0 0 0 0 0 0 0 Subtotal 0 0 0 0 0 0 0 0 0 0 Total Firm Resources 10,309 10,365 10,228 10,269 10,306 10,337 10,370 10,392 10,157 10,219 Load/Resource Balance 1,865 1,892 1,610 1,503 1,390 1,266 1,140 1,001 603 424 (table continued on next page) 8-A-11 Appendix 8-A Public Medium-High (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 9,948 10,154 10,322 10,480 10,784 10,803 10,955 11,113 11,279 11,450 Observed Rate 1.56% 2.07% 1.65% 1.53% 2.90% 0.18% 1.41% 1.44% 1.49% 1.52% Resources: Existing 10,063 10,367 10,364 10,351 10,315 10,340 10,364 10,351 10,290 10,296 Conservation Programs: MCS Single Family 85 95 105 114 123 132 140 148 156 164 MCS Multifamily 19 21 24 26 28 30 33 35 37 or) MCS Commercial 44 49 54 58 63 68 72 ea 78 79 Refrigerators/Freezers 38 48 57 67 76 85 94 103 112 121 Water Heat 41 51 61 72 83 93 104 114 125 136 Manufactured Homes 13 15 16 18 20 21 23 24 26 er Agricultural 0 1 4 8 12 16 20 24 28 32 Existing Commercial 0 0 6 24 50 78 106 134 162 190 Trans & Distr Efficiency 0 0 0 2 4 6 8 10 12 14 Existing Space Heat 37 40 43 50 65 85 105 125 145 165 Existing Industrial 0 0 0 } 0 0 3 26 78 152 Subtotal 277 320 370 439 524 614 708 820 959 1,119 (table continued on next page) Public Medium-High (1995-2005) (continued) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Generating Resources: Hydropower Efficiency 0 0 0 0 40 40 40 40 40 40 Hydropower Efficiency 0 0 0 0 0 0 30 30 30 30 Subtotal 0 0 0 0 40 40 70 70 70 70 Total Firm Resources 10,340 10,687 10,734 10,790 10,879 10,994 11,142 11,241 11,319 11,485 Load/Resource Balance 392 533 412 310 95 191 187 128 40 35 8-A-12 Appendix 8-A Table 8-A-7 Public Medium-Low (1985-1995) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 8,444 8,394 8,418 8,443 8,468 8,545 8,647 8,757 8,728 9,012 Observed Rate 0.00% -0.59% 0.29% 0.30% 0.30% 0.91% 1.19% 1.27% -0.338% 3.25% Resources: Existing 10,303 10,345 10,190 10,212 10,231 10,238 10,244 10,233 9,957 9,982 Conservation Programs: MCS Single Family 2 4 7 11 16 22 27 32 37 42 MCS Multifamily Oo 1 2 4 6 8 10 12 14 16 MCS Commercial 0 1 2 4 7 9 12 14 16 19 Refrigerators/Freezers 0 0 0 0 0 1 4 9 15 22 Water Heat 0 0 0 0 0 1 5 10 17 25 Manufactured Homes 0 1 1 2 3 5 6 we 8 10 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 0 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 4 8 13 16 19 22 25 28 31 34 Existing Industrial 0 0 0 0 0 0 0 ) 0 0 Subtotal 6 15 25 37 51 68 89 112 138 168 Total Firm Resources 10,309 10,360 10,215 10,249 10,282 10,306 10,333 10,345 10,095 10,150 Load/Resource Balance 1,865 1,966 1,797 1,806 1,814 1,761 1,686 1,588 1,367 1,138 Public Medium-Low (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 9,195 9,352 9,475 9,590 9,847 9,833 9,955 10,082 10,210 10,337 Observed Rate 2.08% 1.71% 1.32% 1.21% 2.68% -0.14% 1.24% 1.28% 1.27% 1.24% Resources: Existing 10,063 10,367 10,364 10,351 10,315 10,340 10,364 10,351 10,290 10,296 Conservation Programs: MCS Single Family 48 54 60 65 71 77 83 89 95 101 MCS Multifamily 18 21 23 25 28 30 33 35 38 40 MCS Commercial 22 24 27 30 33 36 39 42 43 45 Refrigerators/Freezers 29 36 43 50 57 64 71 79 86 94 Water Heat 33 41 50 58 67 76 85 95 104 113 Manufactured Homes 1 12 14 15 17 18 20 21 23 24 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 Oo 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 37 40 43 46 49 52 55 58 61 64 Existing Industrial 0 i} 0 0 0 0 0 0 0 0 Subtotal 198 228 260 289 322 353 386 419 450 481 Total Firm Resources 10,261 10,595 10,624 10,640 10,637 10,693 10,750 10,770 10,740 10,777 Load/Resource Balance 1,066 1,243 1,149 1,050 790 860 795 688 530 440 8-A-13 Appendix 8-A Public Low (1985-1995) Table 8-A-8 System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Observed Load 8,444 8,283 8,142 8,005 7,869 7,832 7,849 7,876 7,917 8,037 Observed Rate 0.00% -1.91% -1.70% -1.68% -1.70% -0.47% 0.22% 0.34% 0.52% 1.52% Resources: Existing 10,303 10,345 10,190 10,212 10,231 10,238 10,244 10,233 9,957 9,982 Conservation Programs: MCS Single Family 2 2 2 3 4 5 7 9 11 14 MCS Multifamily 0 0 0 1 1 2 3 4 4 5 MCS Commercial 0 0 1 1 2 3 4 5 6 iw Refrigerators/Freezers 0 0 0 0 0 1 4 9 15 21 Water Heat O 0 0 0 0 1 4 8 14 19 Manufactured Homes 0 0 0 0 0 1 1 2 2 3 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 0 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 4 8 13 16 19 22 25 28 31 34 Existing Industrial 0 0 0 0 0 0 0 0 0 0 Subtotal 6 10 16 21 26 35 48 65 83 103 Total Firm Resources 10,309 10,355 10,206 10,233 10,257 10,273 10,292 10,298 10,040 10,085 Load/Resource Balance 1,865 2,072 2,064 2,228 2,388 2,441 2,443 2,422 2,123 2,048 Public Low (1995-2005) System Summary: Observed Loads and Resources (Average Megawatts) PERIOD 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 Observed Load 8,047 8,120 8,173 8,223 8,421 8,347 8,412 8,481 8,550 8,624 Observed Rate 0.12% 0.91% 0.65% 0.61% 2.41% -0.88% 0.78% 0.82% 0.81% 0.87% Resources: Existing 10,063 10,367 10,364 10,351 10,315 10,340 10,364 10,351 10,290 10,296 Conservation Programs: MCS Single Family 16 19 21 24 26 29 32 34 37 40 MCS Multifamily i 8 9 10 1 12 13 14 16 17 MCS Commercial 8 9 10 12 13 15 16 18 19 20 Refrigerators/Freezers 27 34 40 47 53 60 66 72 78 84 Water Heat 25 32 39 46 53 60 68 75 83 90 Manufactured Homes 4 4 5 5 6 7 7 8 9 10 Agricultural 0 0 0 0 0 0 0 0 0 0 Existing Commercial 0 0 0 0 0 0 0 0 0 0 Trans & Distr Efficiency 0 0 0 0 0 0 0 0 0 0 Existing Space Heat 37 40 43 46 49 52 55 58 61 64 Existing Industrial 0 0 0 oO 0 0 0 0 0 } Subtotal 124 146 167 190 211 235 257 279 303 325 Total Firm Resources 10,187 10,513 10,531 10,541 10,526 10,575 10,621 10,630 10,593 10,621 Load/Resource Balance 2,140 2,393 2,358 2,318 2,105 2,228 2,209 2,149 2,043 1,997 Chapter 9 Consideration of Environmental Quality and Fish and Wildlife Anessential element of the Northwest Power Act is the careful balance between electrical power planning and environmental and fish and wildlife protection. The Act requires that the Council give due consideration in its power plan to environmental quality and the protection, mitigation and enhancement of fish and wildlife. The Council complied with this mandate throughout development of its 1983 plan and this 1986 plan. The Act also requires the Council to consider the com- patibility of the planned resources with the existing regional power system, to choose the most cost-effective resources, and to fol- low certain priorities in selecting those resources. For this reason, selection of the resource portfolio involved not only choosing those resources that were most environmen- tally sound or most protective of fish and wildlife, but also balancing these concerns with the other requirements. This balancing means that, because of new overriding fac- tors such as lost resource opportunities or relative ease of project construction, some resources may be chosen even if they lead to some adverse environmental effects. In addition, the Act requires that all resource cost-effectiveness evaluations must include quantification of environmental costs and benefits. Costs for pollution abatement equipment and fish and wildlife mitigation required under state and federal regulations were included in the Council's estimates of resource costs. The Act further specifies that the Council must develop a method to be used by Bonneville to quantify these environ- mental costs and benefits in measuring the cost effectiveness of specific resource acquisition decisions. This method, devel- oped by the Council, is presented as Appen- dix Il-A. The Council expects Bonneville to use this method in evaluating each resource and resource site prior to acquisition. This chapter describes the process the Council has used in giving due consideration to environmental quality and fish and wildlife in its selection of resources. Environmental Quality Due Consideration Process When the Council drafted its first plan in 1983, it performed studies in support of the plan to identify the potential environmental and fish and wildlife effects of particular types of resources. These studies and important issues arising from them were subjected to public review and comment and guided the Council as it drafted its resource portfolio for the 1983 plan. Additional public comment was sought as the Council revised its resource portfolio for this 1986 plan. During the public comment period on the 1983 draft plan, many comments and consid- erable data were received regarding the environmental effects of the various resources discussed in the plan. In particular, many public commentors offered data docu- menting the environmental effects of hydro- power dams, coal-fired power plants, and high-voltage transmission lines. In addition, a public consultation attended by represent- atives of environmental groups, Indian tribes, utilities, and an agricultural organization pre- sented views and data which assisted the Council in furthering its consideration of environmental quality and fish and wildlife concerns. All this information was carefully considered by the Council in forming its origi- nal plan and was reconsidered by the Council in addition to comments and data which were submitted during the public comment period on the draft version of this power plan. No resource is without its potential adverse effects. In giving due consideration to environmental quality, the Council examined the relative magnitudes of various effects and the practicality of mitigation. Analysis and Resource Alternatives While selecting the individual components of its resource portfolio, the Council assessed all available energy technologies, including their environmental benefits and impacts. The Council also considered the amounts of power to be expected from each resource type, how effects on environmental quality and fish and wildlife could be mitigated, and how mitigation measures may affect energy production. Although not included as major components of the Council's plan at this time, the environmental costs and benefits of alter- native resources such as geothermal, solar- electric generation and wind resources were considered. These alternative resources will be closely monitored and assessed in the future for their environmental effects as well as for their increased cost effectiveness and feasibility. As they become eligible for inclu- sion in the Council's resource portfolio, these resources again will be subject to environ- mental considerations. This section discusses some of the mitiga- tion measures that the Council expects Bon- neville to consider in any resource acquisi- tion or other actions that are required by the Act to be consistent with the plan. While the Council has adopted specific standards only for protection of fish and wildlife in hydro- power development (see Appendix II-B), it is expected that the implementing agencies will be guided by all the considerations set forth in this chapter. During the course of developing this 1986 Power Plan, the Council considered estab- lishing a general set of resource acquisition criteria for nonhydropower resources. (See December 12, 1984, staff issue paper, “Environmental Criteria for Resource Acquisition.”) However, the Council decided to rely on existing federal, state and local regulation of the development of non- hydropower electrical generation resources and to take no specific action relative to addi- tional environmental controls other than the evaluation of environmental effects inherent in the development of the resource portfolio. Among the reasons for the Council's decision were concerns about the Council’ role in the possible acquisition of nonmajor resources and of resources not in the Council's portfolio. 9-1 Chapter 9 The Council was also concerned about a possible duplication of effort in regulatory matters. The analysis that follows first discusses the resources that are included in the 1986 Action Plan and then discusses the re- sources identified in the Council's portfolio for acquisition in later years if higher growth occurs. Conservation The Council expects that conservation will contribute the largest share of energy to the resource portfolio. To that end, the Action Plan includes measures in the residential sector to weatherize existing homes and to build new homes to the model conservation standards. The Action Plan calls for weath- erization of existing homes at a reduced rate because of the current energy surplus and because efficiency standards for both new homes and homes converting to electrical space heating will save more energy than weatherizing existing houses. In both the res- idential and commercial sector, the Council has emphasized the model conservation standards for new buildings. The plan pro- vides that programs for existing commercial buildings should be implemented only to build the capability to acquire this resource when it is needed by the region. The Action Plan also calls for building capability in the industrial and agricultural sectors to achieve conservation savings. The Council recog- nizes that the model conservation standards and programs to build capability to acquire conservation in the various sectors represent important lost opportunity resources that if not acquired now, may no longer be available and cost-effective to the region. These con- servation actions were developed by the Council with full consideration of their poten- tial environmental costs and benefits. The environmental benefits of conservation are substantial. First, reduction of electrical demand due to conservation measures can help the region avoid construction and opera- tion of new energy resources with their accompanying environmental impacts. Con- servation “generates’ electricity without requiring transmission lines; without creating significant air or water pollution, noise, solid waste, or land use impacts; and without 9-2 creating the array of adverse impacts imposed on fish and wildlife by hydropower development and generation. In addition, buildings containing conservation measures tend to be more comfortable. The environmental costs of conservation can be negligible if appropriate provisions are made for acceptable indoor air quality and adequate ventilation in energy efficient build- ings that have less air leakage than ordinary buildings. In buildings with less natural air leakage, the potential exists that there will be higher concentrations of normally occurring indoor air pollutants than would be the case in buildings with ordinary levels of air leakage. Formaldehyde, radon, and combustion byproducts such as benzo(a)pyrene are the indoor air pollutants considered the major potential health risks. Health effects of inhal- ing higher than average concentrations of these chemicals can range from headaches and sore throats to increased chances of incurring lung cancer. Moisture (i.e., humid- ity) is also perceived as an indoor air pollutant when it becomes excessive, contributing to the growth of molds, mildews and fungi. Pollutants can enter a home from a variety of sources. These include the materials used to build the home, the appliances and furnish- ings within it, materials smoked in the home, chemicals brought into the home, cooking and even the taking of showers. In general, new energy efficient homes and new con- ventional homes do not differ significantly in their sources of pollutants. The amount of pollution within a building depends on three factors: the strength of the source, the ventilation rate of the building and the rate at which the pollutant is removed from the air by chemical reaction or physical processes. The source of the pollutant is a very important factor. If there is no source in the home to start with, there is no need to remove it. Although some pollutant sources are unavoidable, many pollutant sources can be avoided or minimized at the time a build- ing is constructed to meet the model conser- vation standards. For example, formalde- hyde off-gassing can be reduced through the use of “10W fuming” formaldehyde wood products rather than the use of ordinary plywood and particle board. Many studies have been undertaken during the past five years, both in the United States and Canada, to better understand the rela- tionship between indoor air quality and energy conservation. These studies are showing that energy-efficient homes with whole-house mechanical ventilation are no more prone to indoor air quality problems than non-energy-efficient homes. Further- more, the studies are showing that very leaky houses, with hourly air change rates (ACH) of two can have indoor air pollution problems, while relatively tight homes with .5 ACH can have very low levels of pollutants. These find- ings indicate that strong pollutant sources can overwhelm ventilation. However, at lower pollutant levels, ventilation is one important means for pollution control. To date, there are no widely accepted stan- dards that establish a “bad” or unhealthy level of indoor air pollutants. Proposed guide- lines from the American Society of Heating, Refrigerating and Air-Conditioning Engi- neers (ASHRAE) and the Environmental Protection Agency (EPA) standards are fre- quently mentioned in discussions on indoor air quality. Although the ASHRAE guidelines were developed for assuring good indoor air quality, they proved to be highly controversial because provisions were lacking on how to implement these guidelines. It is one thing to adopt a guideline of two pico curies per liter for radon, but another matter to use it. (The curie is the unit of measure for radioactivity; a pico curie is one trillionth of a curie.) Currently, the only way of knowing that the guideline is being met is to monitor a building for radon after it is built. New houses constructed under the Uniform Building Code are required to provide mechanical ventilation in bathrooms and kitchens, if windows are not present or oper- ative. These fans are not sized to ventilate the entire house, and, under current building practice, whole-house ventilation depends primarily on unreliable factors such as wind. Current codes provide no assurance of either sufficient ventilation or sound indoor air quality. To guard against worsening indoor air quality, the Council has recommended that mechan- ical ventilation be used in houses with tight construction. Model conservation standard Chapter 9 houses with this mechanical ventilation that were built in the Residential Standards Demonstration Program (RSDP) are being monitored and compared to a control group of houses built to current practice. The early data from both the RSDP and from the Cana- dian R-2000 program show no significant dif- ferences in indoor air quality between houses built to the level of energy efficient standards and houses built to current practice. The Council has designed a research program in the Action Plan to address remaining con- cerns about indoor air quality. On June 20 and 21, 1985, Council staff and some Council members were briefed on pos- sible health impacts of the model conserva- tion standards by five recognized indoor air quality experts. Discussions at this meeting supported the conclusion that pollutant source strength is a primary determinant of indoor air quality. However, both source strength and ventilation rate are important, and strategies to control pollutant levels should focus on both factors. The group also emphasized the need to design programs to minimize source entry of pollutants and to improve the reliability and performance of heat recovery ventilators. Though the production of conservation devices (insulation, storm windows, etc.) may have some environmental impacts, the Council recognizes that the amount of elec- tricity “produced” by conservation is more environmentally acceptable than, for exam- ple, the equivalent amount of energy gener- ated by a coal-fired power plant or hydro- power dam. The 3,920 average megawatts of energy expected to be contributed by conser- vation under the Council's high growth fore- cast is equivalent to the output of more than eight coal-fired power plants that produce 452 average megawatts each (the size of plants assumed by the Council if increments of coal-fired generation are required). (See Volume II, Chapter 6.) On balance, conservation can be an environ- mentally acceptable resource. The potential for indoor air quality problems can be reduced or mitigated through planning for mechanical ventilation to control such pollu- tants as carbon dioxide and moisture, and through source control strategies for such pollutants as radon and formaldehyde. The energy conserved through energy-efficient building design means that the need for addi- tional generating facilities and transmission lines will be reduced, thereby reducing the effects on land, air, water, and fish and wildlife resources. Better Uses of the Hydropower System The Council's plan includes considerations involving improved uses of the hydroelectric system. Although the Council has not deline- ated specific strategies at this time, some strategies for increasing the region's reliance on nonfirm energy, such as using combus- tion turbines to back up nonfirm energy, may have effects on the environment. Developing new uses for nonfirm energy could affect the willingness of hydropower system managers to provide flows and spill for fish passage. The more the hydroelectric system is put to high-valued uses such as meeting firm loads or shutting down combus- tion turbines, the greater the potential conflict with its use for flows and spills needed for fish Passage. This is a concern that the Council will monitor to assure consistency with its plan and program. Environmental effects resulting from hydropower development and operation are discussed more fully below. The plan includes combustion turbines as one possible means of firming up nonfirm hydropower supplies, in order to make possi- ble more economical uses of the nonfirm energy. The combustion turbines are not the only means of backing up nonfirm energy, but their costs provide an upper planning limit for the costs of implementing firming strat- egies on average. Over the long term, it is expected that the combustion turbines would only be operated at most 19 percent of the year. Because of their flexibility, combustion turbines can also be used as a “planning hedge” against rapid growth. Fueled by natural gas or oil, combustion tur- bines are expected to emit certain air pollu- tants. To date, the Council's data show that emissions of natural gas-fired turbines are minimal compared to those of oil-fired tur- bines or coal plants. Combustion of natural gas releases small amounts of nitrogen oxides and about half the amount of carbon dioxide emitted by coal plants. The Council's data have suggested that nitrogen oxides from gas-fired turbines can be reduced to comply with air quality regulations by reduc- ing the temperature of combustion air, recir- culating flue gas, or injecting demineralized water. Oil-fired turbines release larger amounts of these pollutants, plus sulfur dioxide. Accord- ing to Council studies, sulfur dioxide emis- sions from oil-fired turbines can be minimized by limiting the sulfur content of fuel oil used. Noise impact may be mitigated by siting the plants away from population centers, install- ing mufflers, and developing buffer zones. Use of combustion turbines fueled with natu- ral gas or oil also raises certain environmen- tal concerns in connection with exploration, development and transportation of the fuel. The Council notes that off-shore exploration and development of fossil fuels can interfere with commercial and recreational fishing and could cause aesthetic impacts on shoreline areas. On-shore exploration and develop- ment can intrude on roadless areas and wild- life habitat and affect the aesthetics of natural areas. If reliance is placed on imports, there also may be increased risk of oil spills from tanker accidents. Transportation by pipeline involves potential spills and can disrupt exist- ing land uses and cause some aesthetic impacts. Combustion turbines are included as a potentially low-cost option for firming sec- ondary hydropower and as insurance to meet unexpected load growth. Merely pre- serving the potential for using these turbines can postpone or avoid construction and oper- ation of large-scale coal or nuclear facilities. The Council chose combustion turbines as one strategy for firming nonfirm hydropower because they can be brought on-line quickly and operated in harmony with the hydro- power system. This flexibility and avoidance of other impacts, in the Council's judgment, outweighs the effects of combustion turbines on environmental quality and fish and wildlife. 9-3 Chapter 9 Hydropower Development As with the 1983 plan, the Councils 1986 Action Plan directs Bonneville to secure options on hydropower projects at several sites, although Bonneville is not to acquire power from these sites at this time. The development process for the Council's Columbia River Basin Fish and Wildlife Pro- gram, adopted November 15, 1982, and amended October 10, 1984, provided a wealth of information on the effects of hydro- power development on fish and wildlife as well as measures for mitigating those effects. Those considerations have also been taken into account in this plan to the extent they are appropriate outside the Columbia Basin. Some measures adopted by the Council for the Columbia River Basin and the rest of the region are more fully described in the discus- sion of fish and wildlife impacts in a later section of this chapter. For a more complete description of the impacts and mitigation measures applicable to the Columbia River Basin, see the Council's Columbia River Basin Fish and Wildlife Program. The effects of hydropower generation are limited generally to the stream and fisheries affected by a dam. That is, no serious air pollution or solid waste problems are raised by hydropower projects, and they do not rely on a finite fossil fuel. Dams can alter gravel recruitment patterns, because they block downstream movement of gravel and some sediment. Loss of fish spawning and rearing habitat may occur. This effect can be miti- gated somewhat by habitat restoration proj- ects downstream. Among the adverse impacts on migrating and resident fish are turbine-related mortality, migration barriers, dewatering of streams, alteration of flows, inundation of habitat and the effects of increased travel time. Although they are not entirely effective or feasible in all locations, mitigation measures include fish screening and bypass systems, spill for pas- sage, fish ladders, establishment of mini- mum flows, and flow augmentation. 9-4 Construction of a hydropower project may also result in erosion and sedimentation near the stream, causing increased water turbidity. These effects can reduce the aesthetic qual- ity of the stream as well as harm its value for fish, wildlife, and recreational uses. Some- times, these effects are limited to the period of construction and are not considered signif- icant enough by themselves to warrant fore- going otherwise feasible hydropower sites. In addition, the transformation of a river to a deep, still reservoir can alter the temperature of the water. Because of reduced flows, increased temperatures, and the buildup of sediment, many reservoirs become exces- sively productive, sometimes turning eutrophic. The use of special structures, res- ervoir draft techniques, and control of upstream nutrient sources through better land management practices can mitigate these effects. Another impact is nitrogen supersaturation caused by excessive spilling of water over the dam. Though lethal to fish, it can be mitigated with the use of devices that deflect spilled water. Altered water temperatures and nitrogen supersaturation are generally limited to large hydropower projects involving reservoirs, while the Council expects many new hydro- power projects will be small stream diver- sions without reservoirs. These smaller proj- ects are not necessarily benign. Their effects can become cumulative when considered in combination with other projects. (See Columbia River Basin Fish and Wildlife Pro- gram, Sections 1200-1204.) Federal law prevents licensing hydropower projects on or directly affecting wild and sce- nic rivers, and special consideration is required when Indian lands, Indian fisheries, historic or archaeological sites, national wild- life refuges, national monuments, national recreation areas, endangered species hab- itat, or lands adjacent to wilderness are involved. In estimating the amount of hydro- power potential for the 1983 plan, the Council accordingly eliminated such areas from con- sideration. This estimate was reduced even further for this 1986 plan, pending completion of the Pacific Northwest Hydropower Assess- ment Study being conducted by the Council and Bonneville. The study will help rank potential hydropower sites according to impacts on fish and wildlife. With the excep- tion of the hydropower options described ear- lier, which will not be producing power under the Action Plan, only hydropower from exist- ing facilities is included in the 1986 resource portfolio. Installation of hydropower projects on a pre- viously free-flowing stream also can reduce or eliminate the stream's value for kayaking, rafting, and some types of fishing, as well as reduce the forest land base and destroy Indian religious sites through inundation. Also, although the effects of particular proj- ects may be relatively minor, the cumulative effects of several hydropower dams ona sin- gle stream or in a single basin, drainage or subbasin, can be serious. As a result, this plan includes measures to support future hydropower development only at the least sensitive locations and with minimum environmental impact. Because of these safeguards, the Council believes needed additional hydropower development can occur in an environmen- tally sound manner. The first hydropower included in the plan would not be needed until the early 1990s. This allows sufficient time to study the impacts of hydropower and to refine methods for alleviating those impacts. Industrial Cogeneration The Council expects about 80 percent of the available cogeneration to be fueled with bio- mass such as wood waste. Particulates would be emitted from combustion of wood chips or other biomass fuel, but the effects of these emissions could be controlled to a large extent by pollution control technology. Cyclone separators can remove larger parti- cles, while wet scrubbers, electrostatic pre- cipitators, and baghouses can remove smaller ones. However, control technology for cogeneration may not be as sophisticated as it is for larger central station thermal plants, and some residual effects may remain. Also, cogeneration units are more likely to be located near population centers. Use of coal as a backup fuel would entail the air quality impacts discussed below regard- ing coal. Chapter 9 Timber harvesting raises concerns regarding erosion, sedimentation, aesthetic impacts, and destruction of wildlife habitat. Because biomass fuels are usually byproducts of lumber processing, the Council believes most of these effects would not be attributa- ble to biomass electrical generation. None- theless, if and when biomass harvesting involves picking up fallen wood in forests, it may independently cause the effects described above. Use of cogeneration to generate electricity would reduce the need to construct coal-fired or nuclear plants, which, for the reasons stated below, may be less environmentally sound. Some cogeneration projects may be coal-fired and could thus have many, if not more, of the environmental effects associ- ated with coal-fired power plants, discussed below. Nevertheless, cogeneration, even coal-fired, can entail fewer environmental tisks than the separate production of elec- tricity and process steam. Because cogeneration depends largely upon existing facilities, it normally does not include the “boom town” impacts or major transmission lines associated with larger thermal plants. The Council also recognizes that, unlike fos- sil fuel-fired generators, some cogeneration has the advantage of using a renewable resource. Coal-Fired Power Plants Coal-fired generation was the most contro- versial resource included in the Council's resource portfolio. As considered by the Council, the environmental effects of coal- fired generation span the entire fuel cycle. Coal to fuel regional generators most likely will come from strip-mines in eastern Mon- tana or Wyoming. Exploration for coal can include drilling and blasting that risk con- tamination of groundwater. Strip-mining coal involves removing large amounts of soil and other materials overlying the coalbeds. Federal law requires reclamation of strip- mined lands and includes procedures for efilling and regrading, water protection, and revegetation, as well as prohibitions against mining sensitive lands, such as alluvial valley floors and prime farm land. However, there is some question whether these reclaimed lands can sustain long-term productivity or establish a diversity of species characteristic of native range. Because coalbeds often serve as aquifers, their removal by mining often disrupts groundwater and can dry up neighboring wells used for domestic or stock water uses. The resaturation of soils when mined pits are refilled can degrade water quality. The Coun- cils data indicated that acid mine runoff can contaminate local surface and groundwater, and toxic materials exposed by mining can both contaminate nearby water sources and hamper later efforts to reclaim the land. In addition, extraction of coal releases large quantities of dust into the air, hindering nearby livestock operations and decreasing local visibility. Opening mines in rural com- munities can disrupt the agricultural economy. Council studies have shown that transporta- tion of coal to the generating plant incurs various environmental effects, depending upon the location of the generators. Plants located where the coal is mined include fewer transportation effects. However, they con- centrate the effects of both mining and gener- ation in one community and increase the number of transmission lines required. Load- center generation, where the coal is trans- ported long distances from the mine for gen- eration in the area where the electricity is needed, somewhat eases the effects on the community where the coal is mined but increases transportation-related effects. Most coal is transported via railroad, and in some areas new mines require additional rail spurs. These lines can disrupt local farms and ranches by consuming valuable bottom land, hindering drainage, increasing noise, and bisecting fields and pastures. Use of unit trains consisting of up to one hundred coal cars can increase noise, coal dust pollution, and railroad crossing accidents and traffic tie-ups in the rural towns they pass through. Coal slurry pipelines have been proposed to carry crushed coal suspended in water from the Great Plains coal fields to generating plants in Washington and Oregon. Council reports indicated that such pipelines would require large quantities of water and could pose serious water pollution problems at the terminus where the water must be removed from the coal. Also, the pumping systems required for such pipelines would need large amounts of energy to transport the coal sev- eral hundred miles. Such pipelines would require rights-of-way that could disrupt local land uses and affect aesthetics. Coal generation can also have air quality impacts. Though federal and state laws require pollution control, all coal plants emit sulfur dioxide, nitrogen oxide, particulates (small particles), carbon dioxide, and trace elements. Sulfur dioxide has demonstrated detrimental effects on some crops and is known, in many instances, to be harmful to human health. Along with nitrogen oxide, sul- fur dioxide can react in the atmosphere to form sulfates and nitrates, which in turn cause acid rain downwind from coal-fired generators. Acid rain appears to be capable of harming fish, vegetation, soil, surface water and other materials. Particulates can cause respiratory ailments in humans and reduce the traditionally excellent visibility in rural areas of the Great Plains. Sulfates can also reduce visibility. Although sulfur dioxide emissions can be reduced through the use of flue gas desulfurization equipment, these devices may in turn produce large amounts of sludge as a byproduct. This sulfur-laden sludge poses a solid waste disposal problem because it must be prevented from leaching into local water supplies. Advanced combus- tion technologies such as fluidized bed com- bustion can reduce or eliminate production of sludge. Also, fly ash left over from combus- tion of coal contains various trace metals and also must be disposed of in a safe manner. Public comments from Montana suggested that water demands for power plant cooling could conflict with water needs for irrigation and other purposes such as fish and wildlife protection, and that ponds used to store cool- ing water can alter local water tables. As with coal strip-mining, construction and operation of coal-fired generators in rural communities can cause boom and bust impacts. When the plant ceases operation, it can cause rapid out-migration, unemploy- ment and declining tax base. Because coal plants are generally sited away from load centers, electricity generated at most coal-fired power plants must be trans- ported long distances to load centers using high-voltage transmission lines. Council reports have indicated that siting these lines can change local land use patterns, disrupt agricultural operations, and cause aesthetic impacts. Construction of lines through moun- tainous areas can cause erosion as well as interrupt wildlife habitat and recreational pur- 9-5 Chapter 9 suits, and clearing rights-of-way often involves use of controversial herbicides detri- mental to fish and wildiife. Transmission line corridors may interfere with migratory pat- terns of birds or big game. High-voltage transmission lines may produce noise, inter- ference with local television and radio recep- tion, and risk of electrical shock. The Council, in part because of its concern for these effects of coal-fired generation, has only included coal in the energy plan to meet loads under the high growth scenario in 1995 and under the medium-high growth scenario in the year 2000. Even in those cases, the 1986 plan calls for development of coal plants at already-licensed sites first. This would cause lower construction and mining impacts, such as boom-town problems, than starting from an undeveloped site. Nuclear Power Plants Although not included in the resource port- folio, Washington Public Power Supply Sys- tem Nuclear Plants 1 and 3 are retained as potential resources in the 20-year power plan. The environmental effects of nuclear power, described in data analyzed by the Council, also span the entire fuel cycle. Ura- nium, the fuel source for nuclear generators, is extracted by surface or open pit mining. Exploration can involve drilling, blasting and road building that may contaminate ground- water and disrupt wildlife habitat. The Coun- cils data indicated that many of the same water pollution, air pollution and reclamation problems are encountered in uranium mining as in coal mining; the scale of uranium mining is substantially smaller, however, for a given energy content in the fuel. Also, the radioac- tive nature of uranium ore poses potential health risks to miners and persons living near uranium mines. Uranium ore processing results in large amounts of tailings that con- tain radioactive waste materials. These tail- ings may raise human health concerns and must be disposed of properly to avoid con- tamination of water sources or transportation by the wind. Construction of a nuclear power plant is a major undertaking and, because of large plant sizes, can create more severe “boom and bust” social and environmental effects than coal plants. Significant local socioeconomic impacts have already been 9-6 experienced at Washington Nuclear Projects (WNP) 1 and 3. WNP-1 is located, however, in acommunity with a long-term commitment to nuclear work, and mechanisms for adjusting to economic fluctuations due to construction may be better developed there than else- where. Some central station power develop- ments (including nuclear plants) require high-voltage transmission lines and their associated effects. Operation of nuclear power plants may also require large amounts of water for cooling. Council studies have indicated that water intake structures have the potential to harm fish, and any thermal water discharges also have the potential to be detrimental to fish. Cooling systems can also discharge chemical blowdown, which may contaminate air and water. Spent fuel and other radioactive wastes from plant operations require safe disposal. Spent fuel must either be reprocessed to recover uranium and plutonium or it must be treated as waste. Transport to disposal sites or reprocessing plants raises concerns regard- ing highway accidents, accidental spillage, and theft. Some radioactive wastes must be isolated for thousands of years. Pursuant to federal stat- ute, work is now underway to choose suitable disposal sites for spent nuclear fuel and high- level wastes. One method of decommission- ing a nuclear plant requires the removal of all fuel. Next, the plant is sealed and cooled for ten years, during which time the site must be monitored and isolated. The reactor building is then covered to withstand natural forces for 200 years. Other Resources Other resource technologies, although not included in the Council's resource portfolio because of their high-cost or technical infeasibility at this time, were nonetheless considered by the Council for their potential impacts. Geothermal Energy Pursuant to the 1983 Action Plan, Bonneville has designed an assessment and acquisition program for geothermal power. The Bon- neville-sponsored assessment by the com- bined states highlights the potential of the region's geothermal resource, the paucity of verified data pertaining to this resource, and the general sequence required for geother- mal exploration, discovery and development. Federal agencies with responsibilities for characterizing and verifying regional geo- thermal resources are directing their atten- tion to various parts of the region, with an emphasis on the Cascade Mountains. Their findings, coupled with new information from other sources, will describe the geothermal environment of specific drilling locales, and also the general nature of hydrothermal res- ervoirs associated with broader geologic regimes. From this information, appropriate conversion technologies can be determined and related environmental issues will be identified. Council studies have indicated that electrical generation from geothermal sources, where either dry steam or flashed steam conversion processes are used, can cause emission of a variety of gases, including hydrogen sulfide. At low concentrations, this pollutant causes an offensive odor and can be harmful to the human respiratory system and to local wild- life. However, the Council's analysis suggests that current pollution control technology can achieve 90 percent hydrogen sulfide removal. Even with this technology, there are some residual effects from the use of geo- thermal resources. Many of these are dis- cussed above in the context of coal-fired power plants. Clearing of land and construction of roads and pipelines required to tie the numerous geothermal wells to central generators could destroy wildlife habitat and create barriers to wildlife migration. After geothermal water or steam is used to generate electricity, it is usually reinjected into the earth. Studies sug- gest that the impacts of fluid disposal are site- specific, depending largely upon the chem- ical nature of the fluids. Though reinjection is normally preferred, the Council's data noted that other disposal techniques deserve study. Venting of steam or water vapor can create noise, having a potential impact on recrea- tional areas and wildlife populations. Noise can be controlled, however, by installation of noise attenuation equipment and proper operation. Some geothermal projects may require large quantities of water for cooling, although dry cooling can be used in water- scarce areas. Extraction of geothermal steam or water may cause the earth to settle. Chapter 9 Also, geothermal development may disrupt scenic and recreational areas and expose workers to risk of injury while working near steam or hot water. Wind Power The Council estimates that wind generators would cause only minor environmental effects. Though operation of some wind tur- bines may create low-frequency noise, this effect may be minor because generators will likely be located far from population centers. Future wind power studies should examine these potential effects further, and mitigation techniques should be identified. Wind tur- bines may alter the aesthetics of shorelines, mountains, gorges and other areas with typ- ically high winds. Also, the need to avoid obstructions around wind generators may require restrictions on certain types of land use. The Council recognizes that wind gener- ators do not pollute the air, use water, create solid waste, and probably would not cause severe “boom town” effects. With proper con- trol, erosion, siltation and water pollution can be avoided. They do not affect free-flowing rivers and can probably be sited with minimal impact on wildlife habitat. When costs are reduced, the Council expects wind power to be a desired energy resource for the region. Solar Power Solar-electric generation is another resource not yet included in the Council's portfolio because of present high costs and immature technology. The Council's data indicated that this technology also would have relatively minor environmental impacts. Solar systems using fluids to exchange heat raise a pos- sibility of contamination of water and land, albeit minor. A typical large-scale, solar-elec- tric generation plant will require installation of solar reflectors or cells on large land areas, and could affect land use, wildlife habitat, and aesthetics. However, because such plants would not include major water or air pollution or solid waste disposal problems, the Council expects that the impacts of solar-electric gen- eration would be minor compared to the wide range of serious effects associated with large-scale thermal-electric generation. As this and other emerging technologies mature, the Council will gather additional, more detailed data concerning their environ- mental effects, which will receive considera- tion in all future Council decisions regarding these resources. The Council welcomes comments regarding the development of these resources. Additional Fish and Wildlife Concerns Due Consideration Process The requirement of due consideration for fish and wildlife is in addition to the Act's mandate that the Council adopt a Columbia River Basin Fish and Wildlife Program. That pro- gram was adopted by the Council on November 15, 1982, and amended on October 10, 1984. The fish and wildlife program is limited by law to the Columbia River Basin. The power plan, on the other hand, must cover the entire region. Also, the plan covers all types of gen- erating resources, while the fish and wildlife program deals only with the effects of the hydropower system. Under the Northwest Power Act and the Council's power plan, resource acquisitions by Bonneville gener- ally must be consistent with the plan's environmental and fish and wildlife provi- sions. Those acquisitions proposed within the Columbia River Basin must also be con- sistent with the provisions of the Council's fish and wildlife program. The Council's consideration of the rela- tionship between energy supply and devel- opment and the protection of fish and wildlife began with its development of the Columbia River Basin Fish and Wildlife Program. Federal hydropower project operators and regulators (i.e., Bonneville, Bureau of Recla- mation, Corps of Engineers and the Federal Energy Regulatory Commission) must take that program into account at each relevant stage of decision making to the fullest extent practicable. Also, Bonneville must use its legal and financial powers to protect, mitigate and enhance fish and wildlife consistently with the program. On December 16, 1982, the Council released an “Environmental Document for the Colum- bia River Basin Fish and Wildlife Program.” That document described consideration of the fish and wildlife and environmental impacts of the Council's Columbia River Basin Fish and Wildlife Program. It noted that, while some minor environmental impacts might result from implementation of the Council's program, its overall effect was to remedy environmental effects that had gone largely unmitigated for decades. The docu- ment noted numerous ways in which the Council's program would benefit fish and wildlife in the Columbia River Basin. The effects of the Council's fish and wildlife program were considered as the Council developed and revised its energy plan. For example, annually 250-270 average mega- watts of energy capability are estimated to be lost due to use of the Council's water budget to provide adequate flows for migrating ana- dromous fish. This was taken into account in the Council's estimate of the amount of hydropower available to meet future demands. In addition, the costs of fish and wildlife miti- gation and protection measures required in the fish and wildlife program were included as the Council estimated costs of various resources. As previously noted, included in the Councils resource cost calculations were the costs of pollution control technology required by existing law. These measures will benefit fish and wildlife by reducing or pre- venting air and water pollution. Analysis of the Fish and Wildlife Impacts of Hydropower Development Hydropower development can have serious effects on fish and wildlife. As noted in the fish and wildlife program, hydropower proj- ects can hinder migration of fish. Juvenile anadromous fish passing downstream may be slowed by the reservoirs or killed while passing through the turbines. Successive dams and reservoirs in a single drainage or basin can eliminate the natural flushing of migrating juvenile fish to the ocean during the spring months. Without adequate passage facilities, dams present barriers to upstream migration as well. Water level fluctuations above or below hydropower dams can dis- rupt fish spawning and strand wildlife popula- tions. Water impoundments caused by hydropower dams can alter water tem- peratures to the detriment of fish. Construc- tion of dams may create reservoirs that inun- date important wildlife habitat. However, as previously noted, the Council expects many of the new hydropower projects to be stream diversion projects without reservoirs. 9-7 Chapter 9 Many comments from fish and wildlife agen- cies, Indian tribes, and environmental groups have expressed concern over the role of hydropower in the Councils resource port- folio. Some have suggested that the cumulative effects of many small hydropower projects on certain stream reaches could be catastrophic to both anadromous and resi- dent fish, as well as other environmental and cultural values. Within the Columbia River Basin, the Coun- cil’s fish and wildlife program includes a water budget for the Columbia and Snake rivers designed to provide adequate flows for downstream migration. The Council's pro- gram includes other specific measures to assist fish migration. These measures incor- porate provisions for flows, spill, structural bypass systems, ladders and transportation. The Councils program includes measures applicable to the Columbia Basin to minimize the harmful effects of water level fluctuations and temperature control measures for spe- cific Columbia Basin dams. The Council recently completed rulemaking concerning spill measures, which should provide interim fish protection until permanent bypass facili- ties are in place. In addition, the Electric Power Research Institute (EPRI) is currently funding projects in the areas of: 1) fish screens, 2) minimum stream flow require- ments, 3) downstream migration and 4) fish passage through turbines. 9-8 All future hydropower projects within the Columbia Basin will be subject to specific provisions in the Council's program to avoid or mitigate the above effects. The program calls for consolidated review of all applica- tions or proposals for hydropower develop- ment in a single river drainage within the Basin. The Council intends that such review will assess cumulative effects of existing and proposed hydropower development on fish and wildlife. In conformance with the pro- gram, Bonneville has funded a study, cur- rently being performed by Argonne National Laboratory, to propose criteria and methods for assessing potential cumulative effects of hydropower development. The Council and Bonneville have undertaken a study to help collect the information needed for classifying and designating certain streams and wildlife habitat in the basin for protection from future hydropower develop- ment, based upon their value for fish and wildlife and their hydropower potential. As part of its Pacific Northwest Hydropower Assessment Study, the Council will study the existing and potential productivity of stream reaches for anadromous fish. The Pacific Northwest Rivers Study, the portion of the Hydropower Assessment Study conducted by Bonneville, will study non-anadromous values, including resident fish, wildlife, natu- ral and cultural features, recreation, and institutional constraints. In addition, the Council will study Indian tribal and cultural values. These studies will enable the Council to designate stream reaches and wildlife hab- itat within the Columbia Basin to be protected from further hydropower development. Finally, the program calls on the Federal Energy Regulatory Commission to require all license applicants within the Basin to demon- strate how their proposed projects would take the Council's program into account to the fullest extent practicable at each relevant stage of decision making. The conditions for Bonneville's support of hydropower within the entire region (included in Appendix II-B) are designed to avoid or mitigate the kinds of effects described above when they occur outside the Columbia River Basin. The Councils Hydropower Assess- ment Study will also result in a ranking of sites within the entire region in terms of their relative fish and wildlife values. Although hydropower development includes serious risks to fish and wildlife, the Council believes that the provisions of this plan will minimize the effects of any future hydro- power development. One of the fundamental purposes of the Northwest Power Act of 1980 was to provide for the participation of the four Northwest states, their local governments, consumers, the Bonneville Power Administration's cus- tomers, the users of the Columbia River sys- tem (including Indian tribes and fish and wild- life agencies), and the public in the development of regional power policies. The Northwest Power Planning Council plays a crucial role in this process. The Act specifically directs the Council to inform the region's publics about major regional electrical energy issues, obtain their views concerning those issues, and consult with them. The Council fulfilled each of these obligations in the development of the 1986 Power Plan and will continue its commitment to involve and inform the public about regional energy issues in the future. From the time of its formation, the Council has dedicated itself to an active public infor- mation and involvement program. It does not wait for people to come to it, but actively seeks people out to involve them in its process. The Councils structure is built around this commitment to involve the region in its work. It meets in public every three weeks, rotating among the Northwest states. These meet- ings are announced in the Federal Register and the Council newsletters, and are pro- moted by an agenda sent to 11,000 people, including the region's media. All decisions, except those exempted under the “Govern- ment in the Sunshine” portion of the Admin- istrative Procedure Act, are made in these meetings, with public comment opportunities provided. A calendar of public meetings is presented in Table 10-1. Throughout development of the 1986 Power Plan, the Council has held consultations with the utility and industrial customers of Bon- neville, consumer and environmental groups, state and local governments, and the Bonneville Power Administration. These have involved both regionwide sessions as well as state-level meetings in each of the four Northwest States. Consultations have included Council meetings with regional pol- icymakers and other members of the public, as well as staff-to-staff meetings between the Council and other organizations. In addition, beginning in the summer of 1984, several advisory committees and task forces were formed to examine and advise the Council on specific issues related to drafting the 1986 plan. Committee members were chosen to represent a wide range of interests as well as for their expertise. These 11 com- mittees, along with several committees focusing on fish and wildlife issues, make up mittee called for in the Northwest Power Act. Approximately 140 people sit on these com- mittees. Subjects covered include conserva- tion in all sectors, economic and demand forecasting, and resource optioning. Because support at the state level is crucial for implementation of the plan, the views of the state energy and regulatory agencies must be considered in the plan's develop- ment. To ensure this involvement and to improve ongoing communication, the Coun- cil established a State Agency Advisory Committee made up of members of the Northwest states public utility commissions. In developing its electrical demand forecast, the Council requested projections of eco- nomic growth from approximately 300 busi- nesses and industries in the region. The responses were used in developing and val- idating the Council's economic and demo- graphic projections, which are the backbone of the Council's electrical demand forecast. The Council's newsletters are used to keep people informed about the Council's work. Northwest Energy News, a bimonthly 32- page magazine, provides background infor- mation to 15,000 people. It focuses on regional energy and fish and wildlife news, major issues and the Council's activities. In November 1984, the newsletter Update! was initiated to list reports and papers avail- able from the Council, public involvement opportunities, and upcoming meetings. It also accompanies a synopsis of the previous Council meeting and an agenda of the com- ing meeting. More than 11,000 people receive Update! every three weeks. Chapter 10 Public Involvement The November/December 1984 issue of Northwest Energy News included a ques- tionnaire designed to publicize the start of the power plan process and gain insight on how the public involvement program was per- ceived by the region. Most of the respondents felt the Council's information system kept them well informed, but some had sug- gestions for improvements and others were not aware of all the opportunities available for public involvement. Subsequently, every issue of Energy News has carried informa- tion about the power planning and fish and wildlife processes, including opportunities for involvement. Both Update! and Northwest Energy News were used to announce over 20 discussion papers describing various issues relating to the 1986 Power Plan. See Table 10-2 for a list of papers published. These issue papers were circulated widely in the region and to interested parties in other states (a total of over 4,000 people). Comments were solicited and used by the Council in making prelimi- nary decisions on the issues. All comments were entered in the administrative record and distributed to the appropriate staff and Coun- cil members. Those people who sent in writ- ten comment received verification that their comment was being entered in the record. Council “Backgrounders” were developed to supplement issue papers and help non- technical readers understand the issues. These Backgrounders were used as hand- outs and were available at Council meetings. The Councils preliminary decisions were incorporated in the draft plan, adopted in August and distributed for further public com- ment. The comment period closed on October 25, following hearings in each state. Comments from over 150 groups and indi- viduals were received. The final plan was adopted in January 1986. Early in 1985, an advertisement was run in 12 regional newspapers and magazines announcing the 1985-86 power planning pro- cess and opportunities for involvement in it. Both Update! and the draft plan mailing lists were expanded with 325 responses to these ads. Two other advertisements were run in the summer to publicize the availability of the draft plan and to announce public hearings. The latter was a full page ad. 10-1 Chapter 10 At the same time that the 1986 Power Plan was being developed, the Council consid- ered an amendment to the model conserva- tion standards contained in the 1983 Power Plan. This amendment was adopted at the Council's December 4, 1985, meeting and was incorporated into the 1986 plan. In March 1985 an issue paper reviewing the model conservation standards was released for public comment. Subsequently, two addenda to that paper were also released. Over 600 copies of the issue paper and addenda were distributed. On July 11, 1985, the Council voted to consider amending the standards and initiated a public comment period that ended on September 25, 1985. Hearings were held in each state with over 20 groups testifying. Announcements describ- ing the proposed amendment were sent to more than 300 people. In addition, announcements were published in the Coun- cil's magazine, Northwest Energy News, and newsletter, Update. Press releases, which covered the amendment proceedings exten- sively, were sent to the Northwest's media. Based on the responses received, the Coun- cil decided to revise the amendment and reopen the comment period. Testimony was again taken on the standards at the power plan hearings in October 1985. In total, 150 groups and individuals commented on the proposed amendments. Throughout this entire process, Council members and staff met frequently with local governments, utilities, and other interested parties to keep them informed on the Coun- cil’s decision process and to solicit their input. 10-2 The Council's mailing list has been updated and improved to better target specific groups affected by the issues. For example, a spe- cial mailing was sent to 1,600 members of the region’s homebuilding industry to inform them that the Council was reviewing the model conservation standards and to invite their comment and participation. To ensure widespread participation in the planning pro- cess, the Council hired a contractor to help identify under-represented groups and set up meetings between these groups and Council members and staff. To assure accurate media coverage of the Council's activities, key media people were briefed on the power plan process. At each meeting, a press kit was distributed to attend- ing media people and mailed to others. Council issues generated a great deal of cov- erage in the region's newspapers and televi- sion and radio stations. The Council main- tains a newspaper clipping file of this coverage, which is available for public review. The Council is continuing its ongoing efforts to work closely with the region's local govern- ments. Through its local government liaison, the Council provides timely information on the issues to the region's local governments and consults with them individually and col- lectively, working principally through the state local government associations. The Council maintains a public reading room at its central office in Portland where the pub- lic can review staff and contractors studies, as well as comments received relating to the development of the energy plan. The Council also maintains toll-free tele- phone lines (1-800-222-3355 for Idaho, Mon- tana, and Washington and 1-800-452-2324 in Oregon) to encourage public access to the Council. With the development of the 1986 Power Plan, the Council reaffirms its strong commit- ment to an active public involvement and information program. After adopting the plan in January 1986, the Council has continued to hold regular public meetings throughout the region. These meetings are a forum for the Council to discuss ideas and hear pro- posals on major energy and fish and wildlife issues from state and federal agencies, Indian tribes, Bonneville and Bonneville cus- tomers, local governments, and the public. Consultations are continuing with these inter- ested parties on major issues. Concurrently with publication of the Draft Power Plan, the Council began the amend- ment process for its Columbia Basin Fish and Wildlife Program. Equivalent public involve- ment activities are addressed in that program. The Council knows that this plan is not a static document. As conditions change or if resources do not perform as expected, revi- sions to the plan may be needed. To encour- age increased public involvement, the Coun- cil will publicize widely its process for making revisions to the power plan. The public will be informed of proposed revisions through pub- lished material and public briefing sessions. Throughout this process, comments will be solicited from the public on proposed changes to the plan prior to Council adoption. November 28-29 November 30 December 3 December 11 December 11 December 18 December 19-20 January 9-10 January 15 January 15 January 18 January 24 January 30-31 February 13 February 13 February 14 February 14 February 14 February 19 February 20-21 March 7 March 7 March 8 March 13-14 March 26 March 27 March 28 March 28 April 1 April 2 April 2 April 3-4 April 18 April 22-23 April 24-25 April 29 April 30 April 30 Table 10-1 Council and Advisory Committee Meetings 1986 Power Plan Council Meeting, Portland, Oregon Coal Options Task Force Demand Forecasting Advisory Committee Hydropower Assessment Steering Committee Council Hearing on the Pacific Northwest/Southwest Intertie and Out-of- region Sales, Portland, Oregon Economic Forecasting Advisory Committee Council Meeting, Boise, Idaho Council Meeting, Portland, Oregon Hydropower Assessment Steering Committee Economic Forecasting Advisory Committee Demand Forecasting Advisory Committee Public Utility Commissions Task Force Council Meeting, Seattle, Washington Demand Forecasting Advisory Committee Coal Options Task Force Conservation Programs Task Force Options Evaluation Task Force Public Utility Commissions Task Force Options Evaluation Task Force Council Meeting, Boise, Idaho Demand Forecasting Advisory Committee Conservation Programs Task Force Coal Options Task Force Council Meeting, Portland, Oregon Hydropower Assessment Advisory Committee Conservation Programs Task Force Public Utility Commissions Task Force Demand Forecasting Advisory Committee Model Conservation Standards Task Force Economic Forecasting Advisory Committee Coal Options Task Force Council Meeting, Missoula, Montana Coal Options Task Force Conservation Programs Task Force Council Meeting, Seattle, Washington Options Evaluation Task Force Demand Forecasting Advisory Committee Losses and Goals Advisory Committee Chapter 10 10-3 Chapter 10 10-4 May 1 May 3 May 8 May 14 May 15-16 May 29 May 30 May 31 June 5-6 June 12 June 18 June 19 June 20 June 26-27 July 10-11 July 15 July 19 July 29 August 7-8 August 13 August 14 August 20 August 22 August 28-29 September 4 September 18-19 September 20 September 24 September 26 October 1 October 2 October 3 October 3 October 3 October 3 October 9-10 October 21 October 25 October 29 October 30 October 31 Resident Fish Substitutions Advisory Committee Model Conservation Standards Task Force Production Planning Advisory Committee Hydro Assessment Steering Committee Council Meeting, Portland, Oregon Resident Fish Substitutions Advisory Committee Options Evaluation Task Force Losses and Goals Advisory Committee Council Meeting, Portland, Oregon Production Planning Advisory Committee Hydropower Assessment Steering Committee Resident Fish Substitution Advisory Committee Losses and Goals Advisory Committee Council Meeting, Seattle, Washington Council Meeting, Missoula, Montana Production Planning Advisory Committee Resident Fish Substitutions Advisory Committee Losses and Goals Advisory Committee Council Meeting, Portland, Oregon Hydropower Assessment Steering Committee Production Planning Advisory Committee Resident Fish Substitutions Advisory Committee Losses and Goals Advisory Committee Council Meeting, Coeur d'Alene, Idaho Mainstem Passage Advisory Committee Council Meeting, Portland, Oregon Mainstem Passage Advisory Committee Hydropower Assessment Steering Committee Losses and Goals Advisory Committee Resident Fish Substitutions Advisory Committee Demand Forecasting Advisory Committee Mainstem Passage Advisory Committee Economic Forecasting Advisory Committee State Agency Advisory Committee Conservation Programs Task Force Council Meeting, Missoula, Montana Losses and Goals Advisory Committee Mainstem Passage Advisory Committee Production Planning Advisory Committee Council Meeting, Boise, Idaho Mainstem Passage Advisory Committee November 5 Resident Fish Substitutions Advisory Committee November 6-7 Council Meeting, Portland, Oregon November 13-14 Council Meeting, Portland, Oregon November 18 Mainstem Passage Advisory Committee November 20-21 Council Meeting, Portland, Oregon December 2 Production Planning Advisory Committee December 3 Mainstem Passage Advisory Committee December 4-5 Council Meeting, Portland, Oregon December 11-12 Council Meeting, Portland, Oregon December 20 Losses and Goals Advisory Committee January 8-9 Council Meeting, Portland, Oregon January 23 Council Meeting, Portland, Oregon Table 10-2 Issue Paper List for Draft Power Plan * 1985 Action Plan: Conservation Resources * 1985 Action Plan: Generation Resources + Assumptions for Financial Variables * Combustion Turbine Cost Effectiveness * Conservation Supply Curves * Cost & Availability of Generation Resources + Cost of Delaying the Model Conservation Standards until 01/01/88 * Critical Water Planning + Economic, Demographic & Fuel Price Assumptions + Environmental Criteria for Resource Acquisition + Hood River, Elmhurst & ELCAP Projects + Intertie Access Policy + Long-Term Achievable Conservation Targets + Lost Opportunity Resources + Model Conservation Standards Review + Out-of-Region Imports/Exports + Preliminary Demand Forecasts + Research, Development & Demonstration of Promising Resources + Role of Power Institutions in the 1985 Power Plan + Value of Additional Direct Service Industry Interruptibility * WNP-1 & WNP-3 Planning Assumptions Chapter 10 10-5 Priority is given in the plan to resources that are cost effective. The Bonneville Power Administrator is required to estimate all direct costs of a resource or measure over its effec- tive life in order to determine if a resource or measure is cost effective. Quantifiable en- vironmental costs and benefits are among the direct costs of a resource or measure. The Act requires the Council to include “a methodology for determining quantifiable environmental costs and benefits” in the plan. This methodology will be used by the Administrator to quantify all environmental costs and benefits directly attributable to a measure or resource. Proposed Method A. Identify the characteristics (technical, economic, environmental, and other) of the resource or measure in question. Quantify each identified environmental effect in terms of the physical units involved (e.g., acres of habitat, tons of sulfur dioxide, change in water temperature). B. Identify all potential environmental costs and benefits (e.g., the economic value of the effects of changes in the environment) that will result from the resource or mea- sure. Each one of the environmental stud- ies previously completed by the Council should be regularly subjected to public review, comment, and improvement. Research to identify the environmental costs and benefits of each resource should be continued by Bonneville in light of advancing knowledge about environ- mental impacts and of technical changes in resources. C. Screen the identified environmental costs and benefits to determine whether a meaningful economic evaluation can be performed. In making this determination, reference should be made to the work products of the Council — Study Module VI, Nero and Associates, Inc., Reports to Council (Tasks 1-6) on Quantification of Environmental Costs and Benefits, Con- tract 82-020. In particular, consideration should be given to whether economic techniques are sufficiently developed to allow for a meaningful analysis of the environmental cost or benefit. Appendix Il-A Method for Determining Quantifiable Environmental Costs and Benefits D. Determine whether environmental costs and benefits which can be meaningfully evaluated in monetary terms will be so analyzed. This determination should include consideration of: 1. whether sufficient information exists or can reasonably be obtained to allow for an analysis of the environmental cost or benefit; 2. whether the relative cost effectiveness of alternative resources is such that the as yet unquantified environmental costs and benefits would likely affect the decision on resource cost effective- ness; and 3. whether significant costs or benefits remain after considering the effect state or local standards may have on the environmental cost. . For each environmental cost and benefit that can be quantified, an information base should be assembled by the Admin- istrator that analyzes the amount of infor- mation available to quantify each cost or benefit and assesses the uncertainty affecting the ultimate quantity estimates. Federal, state, and local studies of such environmental costs and benefits, schol- arly and professional quantifications, and data obtained as a result of public com- ment should be used to the extent appropriate. . A specific economic evaluation method should then be selected by the Admin- istrator based on the type of environmen- tal cost or benefit, data available to char- acterize the environmental effect and related environmental cost or benefit, experience with the method (e.g., has it been successfully used in the past), and type of uncertainties involved. It is recog- nized that the strengths and limitations of the evaluation method will vary with each environmental impact, and this should be documented. More than one evaluation method may be needed to cross check and verify results. . For those environmental costs and bene- fits where it is not possible to develop monetary values, key physical and bio- logical parameters should be described and, if possible, quantified. H. The application of the evaluation methods should then take place. A record should be compiled that describes the resource, indicates what impacts were identified and which measurement methods were selected, documents each aspect of the calculation, and supports the final result. Throughout this process, the Admin- istrator should consult with the Council, the resource sponsor, interested persons, Bonneville customers, consumers, states, and local political subdivisions. The Administrator should involve the pub- lic to the maximum extent appropriate. . All quantified environmental costs and benefits should then be included in the decision on resource cost effectiveness. Where the environmental costs or bene- fits have been quantified in other than monetary terms, the Administrator should make a decision about the cost effective- ness of each resource or measure by comparing the dollar cost of resources or measures with such costs or benefits to the dollar cost of competing resources or measures. A determination should then be made as to whether the quantifiable but unpriceable costs or benefits are suffi- cient to make an otherwise less expen- sive resource or measure, with such unpriceable environmental costs or bene- fits, more “costly” than the next most “costly” resource or measure. To the extent that no quantification on any terms is possible, the environmental costs and benefits should be identified and described and an assessment should be made on their probable magnitude in relative terms. The environmental costs and benefits of a resource should be given due consideration by the Admin- istrator before the resource is acquired. Such environmental costs and benefits will be weighed in the decision to acquire. I-A-1 Appendix II-A In 1983 and 1984, Bonneville conducted case studies on the environmental costs and benefits of four existing individual resources—a Coal plant, a combustion tur- bine, a nuclear plant and a hydroelectric dam. These studies tested the feasibility of trying to assess environmental costs, using specific estimating techniques. The studies made environmental cost and benefit esti- I-A-2 mates for each of the four facilities. Generally, the case studies showed that it should be possible to establish costs for environmental impacts. In 1985, Bonneville undertook to estimate environmental costs for various types of resources on a generic basis. Bonneville hired consultants and conducted a public involvement process to develop generic environmental costs for hydroelectric, geo- thermal, cogeneration, biomass, wind and solar resources. Draft reports have been released, and additional public input is now being sought. See BPA Issue Backgrounder, June 1985, “Counting the Costs— How BPA Performs Environmental Cost Analysis.” Appendix II-B Conditions for Bonneville Financial Assistance To Hydropower Development in the Region The Council includes the following conditions inits planin response to the Northwest Power Act, which requires due consideration for pro- tection, mitigation, and enhancement of fish and wildlife and related spawning grounds and habitat, including sufficient quantities and qualities of flows for successful migra- tion, survival, and propagation of ana- dromous fish. i Protection, mitigation, and enhancement of fish: Bonneville should not agree to acquire power from, grant billing credits for, or take any other actions under section 6 of the Act concerning any hydropower development in the region without providing for: A. Consultation with interested fish and wild- life agencies and tribes, state water man- agement agencies, and the Council throughout study, design, construction, and operation of the project; Specific plans for flows and fish facilities prior to construction; . The best available means for aiding downstream and upstream migration of salmon and steelhead; Flows and reservoir levels of sufficient quantity and quality to protect spawning, incubation, rearing, and migration; . Full compensation for unavoidable fish or fish habitat losses through habitat restora- tion or replacement, appropriate propaga- tion, or similar measures which give pref- erence to natural propagation over artificial production of fish; Assurance that the project will not inun- date the usual and accustomed fishing and hunting places of any tribe; . Assurance that the project will not degrade fish habitat or reduce numbers of fish in such a way that the exercise of treaty rights will be diminished; and . Assurance that all fish protection and miti- gation measures will be fully operational at the time the project commences. 2. Protection, mitigation, and enhancement of wildlife: Bonneville should not agree to acquire power from, grant billing credits for, or take other actions under section 6 of the Act concerning any hydropower development in the region without providing for: A. Consultation with interested wildlife agen- cies and tribes, state water management agencies, and the Council throughout study, design, construction, and operation of the project; B. Avoiding inundation of wildlife habitat, such as winter range or migration routes essential to sustain local or migratory populations of significant wildlife species, insofar as practical; C. Timing construction activities, insofar as practical, to reduce adverse effects on nesting and wintering grounds; D. Locating temporary access roads in areas to be inundated; E. Constructing subimpoundments and using all suitable excavated material to create islands, if appropriate, before the reservoir is filled; F. Avoiding all unnecessary or premature clearing of all land before filling the reservoir; G. Providing artificial nest structures when appropriate; H. Avoiding construction, insofar as prac- tical, within 250 meters of active raptor nests; |. Avoiding critical riparian habitat (as defined in consultation with the wildlife agencies and tribes) when clearing, riprapping, dredging, disposing of spoils and wastes, constructing diversions, and relocating structures and facilities; J. Replacing riparian vegetation if natural revegetation is inadequate; 3. All pro K. Creating subimpoundments by diking backwater slough areas, creating islands, level ditchings, and nesting structures and areas; . Regulating water levels to reduce adverse effects on wildlife during critical wildlife periods (as defined in consultation with the fish and wildlife agencies and tribes); . Improving the wildlife carrying capacity of undisturbed portions of new project areas (through such activities as managing veg- etation, reducing disturbance, and sup- plying food, cover, and water) as compen- sation for otherwise unmitigated harm to wildlife and habitat in other parts of the project area; . Acquiring land or management rights where necessary to compensate for lost wildlife habitat at the same time other proj- ect land is acquired and including the associated costs in project cost estimates; . Funding operation and management of the acquired wildlife land for the life of the project; . Granting management easement rights on the acquired wildlife lands to appropri- ate management entities; and . Collecting data needed to monitor and evaluate the results of the wildlife protec- tion efforts. posals for Bonneville support of hydropower development should: A. Take fully into account the results of the Council's Hydropower Assessment Study to ensure that future hydropower develop- ment occurs only at the least sensitive locations with minimum environmental impact. . Explain in detail how these provisions will be accomplished or, where exceptions are allowed, the reasons why the provisions cannot be incorporated into the project. I-B-1