Loading...
HomeMy WebLinkAboutFeasibility report Vol 1 Engineering and economic aspects 1982I I I I I I I I • I I I I I I I I I I I, ALASKA POWER AUTHORITY SUSITNA HYDROELECTRIC PROJECT FEASIBILITY REPORT VOLUME 1 -ENGINEERING AND ECONOMIC ASPECTS FIRST DRAFT FEBRUARY,~ 1982 ~-< -.. --• '·,1 ~ ' ' I I I I I ;I I I I I I I I •• I I. I I fl r SUSITNA HYDROELECTRIC PROJECT FEASIBILITY<REPORT PRELIMINARY OUTLINE VOLUME 1 -ENGINEERING AND ECONOMIC ASPECTS 9 -SELECTION OF WATANA GENERAL ARRANGEMENT 9.1 -Site Topography 9.2 -Site Geology -0 c-·c--- 9.3 -Geotechnical Design Considerations 9.4·-Seismic Conside~~tions 9.5 -Selection of Reservoir Levels 9.6 -Selection of Installeci Capacity 9.7-Selection of Spillway Capacity 9.8 -Main 2am Alternatives 9.9 -Diversion Scheme Alternatives 9.10 -Spillway facilities Alternatives 9.11-Power Facilities Alternatives 9.12 -Selection of Watana General Arrangement 9.13 -Preliminary Review 9.14-Intermediate Review 9.15-Final Review 10 -SELECTION OF DEVIL CANYON GENERAL ARRANGEMENT 10.1 -Site Topography 10.2 -Site Geology 10.3 -Geotechnical Considerations 10.4 -Seismic Considerations 10.5 -Selection of Reservoir Level 10.6 -Selection of Installed Capzc-ity 10.7 -Se1ection of Spillway Capacity 10.8 -Main Dam Alternatives 10.9 -Diversion Scheme Alternatives 10.10-Spillway Facilities Alternatives 10.11 -Power Facilities Alternatives 10.12-General Arrangement~election 10.13 -Preliminary Review 10.14 -Final Review 11 -SELECTION OF MAIN ACCESS PLANS 11~1 -Background 11.2-Objectives 11.3 -Appt"oach 11.4-Corridor Selection and Evaluation 11.5 -Route Selection an.d Evaluation 11.6-Description of Basic Plans 11.7 -Additional Plans Page 9-1 9-'1 9-1 9-9 9-12 9-13 9-16 9-19 9-20 9-22 9-27 9-28 9-31 9-33 9-38 9-41 10-1 10-1 1 0.-l 10-7 10-10 10-11 l0-12 10-13 10-13 1:0•16 10-18 10-19 10-21 10-21 10-26 11-l 11-1 11-2 11-2 il-3 11-4 11-7 11-8 r• .,, •• I I -1 I I I I I I I I I I I I I I ~\··, I VOLUME 1 -TECHNICAL AND ECONOMIC ASPECTS (Continued) J2 - 13 - 11.8-Evaluation Criteria 11.9-Evaluation of Access Plans 11.10-Identification of Conflicts 11.11 -Comparison of Access Plans 11.12-Recommended Access Plan 11.13 ... Mitigation Recommendations 11.14-Tradeoffs Made in the Selection Process WATANA DEVELOPMENT 12.1-General Arrangement 12.2 -Site Access 12.3 -Site Facilities 12.4 -Diversion 12.5 -Emergency Release Facilities 12.6 -Main Dam 1'2.7 -Relict Channel Treatment 12.8-Primary Outlet Facilities 12.9 -Main Spillway 12.10-Emergency Spillway 12.11 -Intake 12.12 -Penstocks 12.13 -Powerhouse 12 .. 14 -Reservoir 12.15-Tailrace 12.16 -.Turbines and Generators .. 12.17 -Miscellaneous Mechanical Equipment 12 .. 18-Accessory Electrical Equipment 12.19-Switchyard Structures and Equipment 12.20 -Project lands DEVIL CANYON DEVELOPMENT 13.1 -General Arrangement 13.2 -Site Access 13.3 -Site Facilities 13.4 -Diversion 13.5 -Arch Dam 13.6 -Saddle Dam 13.7 -Primary Outlet Facilities 13.8 -Main Spillway 13.9 -Emergency Spillway 13.10 -Devil Canyon Power Facilities 13.11 -Penstocks 13.12-Powerhouse 13.13 -Reservoir 13.14-Tailrace 13.15 -Turbines and Generators Page 11-9 11-15 11-21 11-22 11-25 11-26 11-27 12-1 12-1 12-2 12-3 12~8 12-10 12-11 12-31. 12-37 12-41 12-43 12-44 12-49 12-50 12-54 12-55 12-55 12-62 12-70 12-84 12.,.84 13-1 13-1 13-2 13-3 13-6 13-8 13-10 13-14 13-16 13-18 13-19 . 13-20 13-21 13-24 13-25 13-26 ,,, I I I I I I I I I I I I I I I I ··I I VOLUME 1 TECHNICP1L AND ECONOMIC ASPECTS (Continued) 13 .. 16 -Miscellaneous Mechanical Equipment 13.17-Accessory Electrical Equipment 13.18-Switchyard Structured and Equipment 13.19 -Project Lands 14 -TRANSMISSION FACILITIES 14.1 -Electric Systems Studies 14.2 -Corridor S~le~tion . 14.3 -Route Selection 14 .. 4 -Towers, Foundations and Conductor·s 14.5 -Substations 14.6 -Dispatch Center and Communications 15 PROJECT OPERATION 15.1 -Plant and System Operation Requirements 15.2 -General Power Plant and System Railbelt Criteria 15.3 -Economic ·operation of Units 15.4 ... Unit Operation Security Criteria 15.5 w Dispatch Control Centers 15.6 -Susitna Project Operation . 15.7 -Performance Monitoring 15.8 -Plant Operation and Maintenance 16 -ESTIMATES OF COST 16.1'-Construction Costs 16~2 -Mitigation Costs . 16'. 3 -Operation, ~ia i ntenance and Rep 1 acement Costs 16.4 -Engineering and Administration Costs 16.5 -Allowance for Funds Used During Construction 16.6 -Escala.tion 16.7 -Cash Flow Requirements 17 -DEVELOPMENT SCHEDULE 17.1 -·watana Development 17.2 -Devil Canyon Development :7.3-System Development Schedule 18 -ECONOMIC AND FINANCIAL EVALUATION . 18.1 -Economic Evaluation 18.2 -Risk Analysis 18.3.-Marketing 18',,4 ~ Financial Evaluation 18.5 -Financial Ri~k ·' >' 0 Page 13-28 13-30 13-35 13-35 14-1 14-1 14-7 14-14 14-19 14-25 14-25 15-1 15-1 15-2 15-4 15-6 15-7 15-8 15-15 15-17 16-l 16-1 16-5 16-6 16-6 16-8 16-8 16-8 To fol1o\'l 18-1 18.-l 18-13 18-25 18-27 18-31 I I I I I I I I I I I I I I I •• I I l·t Plate 1 2 3 4 5 6 7 8 9 10 11 12 13 - 14 15. List of Plates Title Railbelt Area Devil Canyon Hydro Development Fill Dam Watana Hydro Development Fill Dam Watana Stages Fi 11 Dam High Devil Canyon l-lydro Development S-usi tna I I I Hydro Development Vee . Hydro Development Denali & Maclaren Hydro Developments Preferred Tunnel Scheme 3 Plan View Preferred Tunnel Scheme 3 Sections Watana Arch Dam Alternative Watana Alternative Dam Axes Watana Preliminary Schemes Watana Scheme ~JPl Plan Watana Scheme WP3 Sect tons " I List.of Plates (cont'd} -. I P1ate Title -. 16 Wa:tana I Schemes WP2 & WP3 Plan and Section 17 Watana I Scheme WP2 Sections •• 18 Watana Scheme WP4 Plan I 19 Watana Scheme WP4 Sections I 20 Watana Scheme WP3A I 21 Watana Scheme WP4A I 22 Watana Simulated Reservoir Operation I 22A Devil Canyon Simulated Reservoir Operation I 23 Devil Canyon Scheme DCl I 24 Devil Canyon Scheme DC2 1- 25 Devil Canyon Scheme DC3 26 Devil Canyon I Scheme DC4 27 Devil Canyon I Selected Scheme 28 Alternative Access Corridors I 29 Alternative Access Routes ' 30 Access .. Plan I I Recommended Route ~ I :J ·' •• ~ I ·, .... . . . ----------- I I I I I I I I I I I I .I I I I· I I ,I I. Plate - 31 32 32A 33 34 35 36 37 38 39 40 41 42 ·43 List of ..Plates (cant' ~l Title Watana Reservoir Plan Watana Site Layout Watana General Arrangment Layout of Structures Plan Watana · Hydrological Data Sheet 1 Watana Hydrological Data Sheet 2 Watana General Layout Site Facilities Watana Village and Townsite Watana Main Construction Camp Site Watana and Devil Canyon Construction Camp Details Watana Diversion General Arrangement Watana Diversion Scheme Sections Watana Diversion Intake Structures Watana Downstream Portals Plan and Section Watana Emergency Release Sections I I I I I I I I I I I I I I I I I I I Plate 44 45 46 47 48 49 50 51 52 53 54 55 List of Plates !cont'd) Title Watana Main Dam Plan vJatana Main Dam Sections Watana Main Dam Grouting and Drainage Watana Outlet Facilities Gate Structure Watana Outlet Facilities General Arrangement Watana Main Spillway General Arrangement Plan and Profile Watana ~1a in Spillway Control Structure Watana Main Spillway Chute Sections Watana Main Spillway Flip Bucket Discharge Structure Watana· Emergency Spillway Watana Power Facilities General Arrangement Watana Power Facilities Plan and Sections 0 ... I I I I I •• I I I I I I I I I I I I I • ~.~ ..... ~-•'<''---' <' h·'"''-"•" .... _. ·~·"' "' j'> '<"N" Plate 55 A 56 57 58 59 60 60A 61 62 63 64 64A 65 66 List of Plates (cont 1d} Title '\ t wa ana Power Facilities Plan, Sections and Elevations Watana Power Facilities Access Watana Powerhouse Plans · Watana Powerhouse Sections Watana Transformer Gallery Plan anG Sections Electrical Legend Watana Powerhouse Single Line Diagram \~a tan a Swi tchya rd Single Line Diagram Block Schematic Computer-Aided Coutrol System Oevil Canyon Reservoir f'lan Devil Canyon Site Layout Devil Canyon General Arrangement Layollt of Structures [levi 1 Canyon Hydrologic Data Sheet 1 Devii Canyon Hydrologic Data Sheet 2 -'. >t ~ ~ ., ·~. . ·-· .... I I •• I I I I I I I I I I I I I •• I I Pl. ate 66A 67 68 69 J 70 71 72 73 74 75 76 76A 77 List of Plates (cont'd) Title Devil Canyon Genera 1 Layout Site Facil itie~~ Devil Canyon Temporary V<:llage Devil Canyon Construction Camp Plan Devil Canyon Diversion General Arrangements Devil Canyon Di ve\"S ion Sections Devil Canyon Dams Plan and Profile Devil Canyon Main Dam Geometry Devil Canyon Main Dam Geometry Crown Section Devil Canyon Main Dam Thrust Blocks Devil Canyon Main Dam Grouting and Drainage Devil Canyon Saddle Dam General Arrangement Sections Devil Canyon Outlet Facilities Devil Canyon Main Spillway General Arrangement Plan and Profile I List of Plates (cant • d) I Plate Title 78 Devil Canyon •• Main Spillway Control Structure 79 Devil Canyon I Main Spillway Chute I 80* Devil Canyon Main SpiYlway Flip Bucket I 81 Devil Canyon Emergency Spillway I General Arrangement 82 Devil Canyon Emergenc~ Spillway I Sections 83 Devil Canyon I Power Intake Structures Plan and Sections 84 Devil Canyon I Power Facilities 85 Devil Canyon I Powerhouse Plans I 86 Devil Canyon Powerhouse Plan and Sections I 87 Devil Canyon Powerhouse Sections I 87A Devil Cdnyon Trans forme•" Ga 11 ery I General Arrangement Plan and Sections I 88 Devil Canyon Powerhouse Single Line Diagram 89 Qevil Canyon Switchyard I Single Line Diagram I *Not Included f I 0 I I I I I I I I I I I I I I I I I I' I Plate 90 91 ,• List of Plates (cont'd) Title Watana Construction Schedule Devil Canyon Construction Schedule • I I I I I I I I I I I I I I .I I I I I 9 -SELECTION OF WATANA GENERA~ ARRANGEMENT This section describes -the ~volut·ion of the general arrangement of the Watana project, which, together with the Devil Canyon project, comprises the develop- ment plan selected as part of Section 8. This section also describes the site topography, geology, and seismicity of the Watana site relative to the design and arrangement of the various site facilities. The process by which reservoit operatiq~} levels and the installed generating capacity of the pm>~er facilities were estaDlished is also presented, together with the means of handling floods expected dJring construction and subsequent project operation. The main components of the Wat ana deve 1 opment are as follows: -Main dam; Diversion facilities; -Spillway facilities; -Outlet facilities; -Emergency release facilities; and -Power facilities. A number of alternatives are available for each of these components and they can obviously be combined in a number of ways. The following paragraphs describe the various components and methodology for the preliminary, intermediate, and fi na 1 screening and revi e'fl of a 1 ternati ve genera 1 arrangement of the components, together with a brief description of the se 1 ected scheme. A ~detailed descri p- tion of the various project components is given in Section 12. 9.1 -Site Topography The project site is located in a broad U-shaped valley at river mile 183~ approximately 2-1/2 miles upstream of the confluence of Tsusena Creek with the Susitna River'. The river at the site is relatively wide, although turbulent (Figure 7.11). On the right bank, the valley rises at an approximate slope .of 2H:lV from river level at elevation 1450 for approximately 600 feet, then gradually flattens to a maximum elevation of 2350 between the Susitna River and Tsusena Creek. The left bank rises more steeply from the river for about 450 feet at a slope of 1.4H:1V, then flattens to 3H:lV or less to approximate elevation 2600. 9.2 -Site Geology General This section s·ummarizes the geological and the geotechnical investigations con- ducted to date and the geologic conditions present at the Watana ~i~e. A detailed description of the geology and site investigations is presented in the 1_980-1981 Geotechnical Report (1). 9-1 (a) §eologic Setting The Watana site is located on the western side of a Tertiary age (2 to 70 m.y.b.p) intrusive body. The rock is primarily a gr~y to green medium grained crystalline igneous rock of diorite-quartz diorite composition. Associated-with the pluton are andesitic dikes and volcanic flows, which are generally the fine grained equivalents of the intrusive rocks, and volcaniclastic·sedimentary rocks. The underlying rock is hard, fresh, and of good quality. It is slightly weathered at the surf ace and a 1 ong joint surfaces~ The overburden is generally thin on the valley walls, thickening to the north of the damsite, and consists of glacially derived silts~ sands, and gravels~ Shear, frac- ture, and alteration zones have been delineated highlighting two major structural features to be considered in the design of the dam. No evidence of major structural deformation or faulting has been found. Permafrost conditions exist on the north facing slopes with temperatures near the freezing point. '} (b) Geo loai ca 1 and Geotechni ca 1 Investigations Surface and subsurface investigations for the site have been conducted by several organizations at different times. Preliminary reconnaissance work was done by the USSR in the 1950s. During the years of 1975 and 1978 the COE conducted site specific investigations to determine the suitability of the site. As part of the. current study program, more detailed investigations were undertaken in 1980 and 1981 to establish the technical feasibilitoy of the project. The investigations have included air . reconnaissance, air photo interpretation, geologic mapping of rock and surficial materia 1, dri 11 i ng of both rock and overburden, including 1 n-ho le geophysical tests and seismic refraction surveys. Both insitu and laboratory tests have been performed to determine the engineering ~ characteristic of soils and rocks. The location of drill holes and other investigations is shown in Figures 9.1 and 9.2. Geo 1ogi c mapping was concentrated in the immediate proposed damsite area between Tsusena and Deadman Creeks. All accessible areas were mapped for the rock and overburden exposures. The lithology or type of material~ bed- ding, jointing, weathering, degree of consolidation, exposure size, and elevation were noted and plotted on maps for use in the interpretations. Seismic refraction surveys were performed throughout the investigations on both banks, the river channe 1, the borrow areas, and the upper slopes of t~e damsi te area to determine the bedrock depth and other si gni fie ant features. A total of feet of seismic refraction traverse were run at the Watana site. Information gathered was integrated and used with the other subsurface information for correlation and development of a thre.e dimensional representation of site charactel·istics. 0 9-.2 I ••••• I I I I I I I I I I I I I I I I I -----~--~-'---~~~--~-----~~-~--- I ~· 0 . I I I I I (c) I I I. I I I I I I I I I Di amend core and rotary dri 111 ng were performed in the foundations and abutments of the proposed project structures. A total of . feet was drilled in 28 holes with feet of core recovered. In-haTe permeability tests were conducted in each hole upon completion and geophysical logging and borehole photographs were performed in selected borings. To monitor the ground water and ground temperature conditions at the site, piezometers and thermistor strings were installed in selected drill holes both in 1978 and 1980. A regular monitoring program for those instruments has been established to collect data. A series of tests were performed on the rock recovered from coring to determine the engineering char acteri st i cs of the rock mass. The resu 1 ts of these investigations were used to develop the geologic picture of the dam- site and the adjacent areas .. Figures 9.3 and 9.4 present the rock outcrop map and the stl"uctural geology map of the site. The results of the labora- tory rock tests are summarized in Table 9.1. Construction Material Investigations Extensive investigations have been conducted both prior to and during the current studies to identify sufficient quantities of suitable materials for the construction of an embankment dam and for concrete aggregates. The investigation methods have included geologic mapping, auger and rotary drilling, excavation of test pits, seismic refraction surveys, and laboratory tests. A total of rotary holes, auger holes, test pits, and linear feet of srumic refraction surveys have been- performed.---:rhe locations of the various potential sources of embankment material and concrete aggregates together with the locations of boreholes and test pits are shown in Figure 9.5. ( i) Rock F111 Material Two sources for the rockfill material, designated Quarry A and Quarry B, have been identified. The rock in Quarry A is located on the south bank and primarily contains a relatively thin layer of andesite (an extensive igneous rock) overlying diorite. The diorite is generally hard, durable and fresh, and suitable for use as rockfill in the dam. Quarry B is an alternative source of rockfill, however, it is estimated that sufficient quantities of rock are available in Quarry A to meet the project requirements. ( i i) Impervious Core Material Two sources have been identified for the impervious core material and are dt~si gnated as Borrow Area D and Borrow Area H. Borrow Area Dis located within 1.5 miles upstream of the damsite on the north bank. The upper few feet of materia 1 comprises tundra, topsoi 1 ~ and boulderss and is underlain by glacial tills composed of dense gravelly silty sands with some clay. The tills range from 15 to 25 feet in thickness and overlie sandy gravelly clay. A composite grain size curve for these soils is presented in Figure 9.6. The I mat$rl a 1 is well graded with natural water content at about the plastic limit. Figure 9.7 presents the results of Atterberg limits : •. on the finer portion of the material.. More than adequate quantities of materi aJ are present in Borrow Area D for the impervious fi 11 requirements at Watana. .J An alt.ernative source of core material, designated Area H, is locat- ed approximately 5 to 7 miles downstream from the damsite on the south bank of the river in the Fog Lakes a~ea. This area contains a relatively thick layer of till composed of silt, sand, and gravel with some cobbles. A composite grain size curve for the till materia 1 is prese!1ted in Figure 9.8 and the Atterberg 1 i mit test results in Figure 9. 9. The material is quite similar to that in Borrow Area D except that the natural water content is higher. Sufficient quantities of impervious fill material are also available at this location if required. (iii) Fi 1 ter Materia 1 Borrow Area E has been identified as a primary source of material for filter and transition zones of the embankment. This area is located at the confluence of the Tsusena Creek and the Susitna River approximately 2.5 miles downstream from the damsite. The area is covered by about 2 feet of organics and silt and is underlain by a few feet thick layer of silty sand to clean sand. Below that a .thick layer of sandy and gravelly material exists. A composite grain-size di stri but ion curve for Borrow Area E materia 1 is presented in Figure 9-10. Sufficient quantities are available in this borrow area to meet the project requirements for wilter materi a 1 s. The material wi 1] have to be processed to meet the gradation requirements of specific zones. Additional alternate sources of material i dent i fi ed inc 1 ude Borrow Area ,C and Borrow Area F, at greater distances from the damsite. Also, some material from Borr0\'1 Area D and riverbed alluvium (as described later) is suitable with processing to be used as filter and/or transition materia 1. {iv) Gravels and Gobbles for Shells Seismic refraction survey investigations were performed within miles upstream and miles downstream from the damsite in the- Susitna River valley:-These investigations confirmed that suffi- cient quantities of granual material are available for use in the supporting she 11 zones of the dam. In the upstream she 11 of the embankment, to meet design requirements, the material will require processing to remove fines and \tast age of materia 1 3/8 inch or sma 11 er in size, and ov~rsi ze materia 1 1 arger than 18 inches. Available data indicates that the grain size distribution of these materials will be similar to that from Borrow Area E, with probably a higher percentage of coarser material. 9~4 I I I I I I .I I I I I I I I I I I I I I I I I I I I I I I I I I I I I (v) Concrete Aggregate The material avai 1 able from Borrow Areas E, C, F, and the riverbed all uvi urn is suitable for use as coarse and fine aggregate for con- crete. Processing wi 11 be required to produce desired gradations .. The coarser particles are rounded and petrographic analyses have indicated the material to be of good quality. Sufficient quantities are available within the identified sources. (d) Geologic Conditions A summary of overburden and bedrock conditions is presented in· the follow- ing paragraphs: (i) Overburden Overbur·den thickness is generally thin on the 1alley walls and thickens away from the proposed damsite to the north (Figure 9.3). On the lower slopes, the overburden consists primarily of talus. Above the break in slope where the topography becomes more gentle~ glacial silts, sands, gravels, and boulders are encountered. Sub- surface investigations have indicated the contact between the overburden and bedrock to be relatively unweathered. The depth of the river alluvium beneath the proposed dam averages about 80 feet 1 up to a maximum of about 100 feet, and consists of sand, silt, coarse gravels, and boulders. Very little is known at this time about the denseness and character of this alluvium. A deep bedrock depression has been delineated on the north side of the river and is discussed under Section 9ol3. (ii) Bedrock Lithology The Watana site is underlain by a ·serie of sedimentary, volcanic, and plutonic rocks. The damsite is primarily underlain by an intrusive dioritic body which varies in composition from granodio- rite to quartzdiorite to diorite. The volcanic rocks (dior1te and andesite) are generally finer grained equivalents of these intrusive rocks. The sedimentary rocks consist of tuffaceous siltstones anti graywackes. The geologic map of the·site is shown on Figure 9.3. The quartz diorite is light gray and is found primarily upstream . from the damsite axis. The texture is massive and the rock is hard, competent, and fresh except within the shear zones~ which are dis- cussed later. The diorite is a dark grayish green rock with massive appearance. It is hard, competent, and generally-fresh. These two grades of di.orite occur in alternating zones on the order of several hundred feet wide. Weathering is limited to a very thin zone on the exposed surface, and along the joints to feet _depth. These rocks have been intruded by mafic and felsic dikes which are generally only a few feet thick with tight contacts. These. dikes generally trend parallel to major joint sets. The rock, downstream from these diorites, is a series of extrusive rocks ranging from rhyolite to andesite and basalt. Andesite porphyry is the more prominent of these rocks. The rock is a medium to dark gray to 9-5 ...... ' . . ~,..,'! - green and contains quartz diorite inclusions~ The contact of the andesite with the underlying diorite is generally slightly weathered and fractured. On the north bank, the andesite-diorite contact at the downstr•eam is coincident wi-th the 11 Fingerbustern shear zone. A sequence of sedimentary rocks composed of essent i a 11 y vc71 cani c debris is exposed downstream from the damsite and is comprised of generally snund sandstones and siltstones. This sequence is overlain by andesite. (iii) Bedrock Structures -Joints There are two major joint sets and two minor joint sets at the site (Figures 9 .4 and 9 .12). These joint sets are described in Table 9.2. Set I~ which is the most prominent set strikes 320° (N40W) and dips 80° NE to vertical. This set is found throughout the damsite and parallels the general structural trend in the region. Joint Set I has a subset, which strikes 29o= to 300° (N60 -70W) with a dip of 75° NE.. This subset is localized in the downstream area near where the diversion tunne 1 porta 1 s are planned. This subset also parallels the shear zones in the down- stream area of the site.. Set I I trends northeast to east and dips vertically. This set is best developed in the upstream portion of the damsi te area, although it is prominent in the downstream areas with a more easterly strike. No other structural features were found with orientations in this set. Sets III and IV are minor sets but can be locally strong. Set IIi forms numerous open joints on the c 1 iff faces near the 11 Fi ngerbuster", and sever a 1 shear zones parallel this orientation. Set IV appears to result from stress relief from glacial unloading and/or valley erosion. The average spacing, of Joint Sets I and I I is 1 to 2 feet and 6 to 12 inches, respectively;'····The spacing of Joint Sets III and IV is quite variable and can range from a few inches to several feet. -Shears and Fracture Zones Several shears, fracture zones, and alteration zones are present at the site. For the most part, these are small and discontinu- ous. During mappinglt all zones greater than 10 feet in width were mapped and delineated on the geologic may (Figure 9.4). Shears are defined as having breccia~ gouge, and/ or s·l i ckens l ides indicating relative movement and are found in two forms. The first type is found only in the diorite and are characterized by breccia of sheared rock that has been healed by a matrix or very fine grained andesite/diorite. The contacts, although irregular, are tight and unfractured. The zones were found to have high RQDs and to be fresh and hard. The second type is common to all the rock types and consists of breccia and/or gouge of fine grained t"ock types and consists of breccia and/or gouge of fine grained rock material in a silt/clay matrix. These are soft, friable, and often have secondary mineralization of carbonate and chlorite showing slickensides. These are genera11y less than 1 foot wide. I I I I I I I '& I I •• I I I I I I I I 'I I I I I I I I' I ··- •• I I I I I I I I Fracture zones are a 1 so common to a 11 rock types ranging from a 6-inch to 30-foot wide (generally less than 10 feet) zone of closely spaced joints that are often iron oxide stained or carbon- ate coated. Where exposed, the zones tend to form topographic lows. In the a 1 terati on zones, the felspars and mafic minerals of the rock have been chemically altered by hydrothermal solutions to clay and chloriteQ The degree of alteration encountered is highly variable across the site. These zones are rarely seen in outcrop as they are easily eroded tnto gullies but were encountered to some degree in all the boreholes. The transition between fr.esh and altered rock is gradational and the zones may range to 20 feet thick although are usually less than 5 feet. Core recoveries are generally very good and the rock quality is dependent on the degree of alteration. (iv) Significant Structural Features The Watana site has several significant geologic features consisting of broad areas of the shears, fractures and alteration zones described previously. The two most prominent areas have been named 11 The Fins" and the 11 Fingerbuster". "The Fins 11 is located on the north bank of tht.~ river upstream from the diversion t unne 1 intake. It is approximately a 400 foot wide area characterized by three major northwest trending zones of shearing and alteration that have eroded into steep gullies. These alteration zones are separated by intact rock bands (ribs} 5 to 50 feet wide. The 20-foot wide upstream zone of the series coincides with the diorite/andesite porphry contact. The other two zones, approximately 55 and 30 feet wide, are fi 11 ed with severely altered talus. This area is characterized by a 300 - 310° orientation (NSO to 60W) and near vertical dips of its component shears and by low seismic refraction velocity of the rock. The extension of the zone is extrapolated by topographic and seismic lows northwest to a sheared/altered outcrop on Tsusena Creek. The 11 Fi ngerbuster 11 is 1 oc a ted downstream from the d amsi te and is exposed in a 40 foot wide deep talus filled gully along the andesite porphyr-y/diorite contact. The rock is severely weathered with closely spaced joints trending parallel to Set I (330°) and Set III (0°). Slickensides indicate vertical displacement. The extension of this zone to the south is ba~ed on a strong north-south topo- graphic lineament, although no outcrops are exposed. Becapse of the 1 ack of exposure and defi nab 1 e continuity of this feature~ its . location orientation is extrapolated. A prominent alteration zone wa~ encountered in BH-12 on the south bank. The·ho1e encountered aprroximately 200 feet of hydrothermally altered rock. Although core recovery in thin boring was very good, the quality of rock was relatively poor and a zone of significant artesian pressure was encountered. The south eastern extension of this zone passes under. the core of the dam. 9-7 (e) Ground Water Conditions The gr1'lund water regime in the bedrock is confined to movement a 1 ong fractur-es and joints. Measured water 1 eve ls have ranged from to feet below surface. It is assumed that the ground water in the non-pe_r_m-a-- frost areas on the north side .of the river is a subdued rep 1 i ca of the topography with the gradient towards the Susitna River and its tributaries. Artesian condit7ons are present in isolated fr.acture isolation zones. (f) _Permafrost Conditions Permafrost conditions exist on th~ north facing slopes (left bank) of the damsite area. Measurements in the borings indicate that it penetrates to a depth of feet and show marginal temperatures within loC of freezing. Only sporad1 c areas of permafrost have been encountered on the right bank~ {g) Reservoir Geology The topography of the \~atana Reservoir and adjacent slopes is characterized by a narrow V-shaped stream-cut valley superimposed on broad shapes U-shaped glacial valley. Overburden masks much of the bedrock especially in the 1 ower and uppermost r'1aches of the reservoir. Figure 7.11 shows the general Watana reservoir geology. · The lower portions of the Watana reservoir are predominantly covered by a veneer of glacial till with scattered outwash deposits. On the south side of the Susitna River, the Fog Lakes area is characterized by a fluted ground moraine surface. Upstream in the Watana Creek area, a broad flat plain is mantled with glacial till and semi-consolidated Tertiary sedi- ments. The·se are predominantly stratified~ poorly graded, fine sands and silts with some clays.. The river valleys contain significant amounts of alluvial deposits and reworked outwash. Ice disintegration features such as Kanes and eskers have been observed in the river valley. A non-conformable contact between argillite and the diorite pluton in the ct3ffisite area was mapped approximately three miles upstream of the damsite. Semi-consolidated, Tertiary age sedimentary rocks, and volcanics of Trias sic ag'e are present just downstream from the confluence of \~at ana Creek and the Susitna Rvier. These volcanics consist of metabasalt flows with thin interbeds of metachert, argillite, marble, and metavolcaniclastic rocks. Metamorphosed from Watana Creek to Jay Creek. Tife rocks between ·Jay Creek and Oshetna Creek are metamorphic amphibolite and minor amounts of greenschist and foliated diorite. The main structural feature of the Watana Reservoir is the Talkeetna Thrust Fault which trends northeast -southwest~ The Talkeetna Thrust Fault crosses the Susitna River approximately eight miles upstream from the damsite,. This fault has been studied in detail as part of the seismic studies, and has been determined to be inactive in relation to the project design. 9-8 I I I I I I I I I :I I I I I I I I I I I 1: I I I I I I I ·- .-. I I I I I 'I I I 9~3 -Ge9technical Design Considerations This section deals with the geotechnical aspects of design of the dam and other major stt"uctures at the Watana site. (a) Main Dam Excavat.i on and Foundation Treatment As discussed previously, the riverbed alluvium ranges up to approximately 100 feet in depth. The character of the this material is difficult to define~ however~ its stability during a strong earthquake event is quect.ionable. Considering the nature of the materi,al, and the height of the Jam, the riverbed material will be removed entirely within the limits of the dam. th~ overburden materia 1 on the abutments is relatively thin, except for gu 11 i es and pockets. Most of this materia 1 is-frozen and wi 11 become unstable when thawed and is therefore unsuitable for the dam foundation. Accordingly, the overburden on the abutments will also be required. Details of foundation treatment are preseflted in Section 12. The presence of numerous shear zones, containing gouge material, indicates the need to remove ali weathered rock under the impervious core and upstream and downstream filter zones. Excavation will include shaping of valley walls along the abutments to provide a proper contat surface in accordance with good modern design practice. Excavation under the outer shells will include removal of loose rock blocks, extensively weathered rock and local reshaping as necessary. The strength of the rock foundation is otherwise adequate to support the embankment and associated reservoir 1 oads. The two major geologic structures at the site, "The Fins" on the upstream side and the HFi ngerbuster" on the downstream side, have a major influence on the overall project layout., however, they do not directly affect the darn at its proposed location. · Extensive permafrost is present on the Jouth bank (as deep as several hundred feet) and sporadic pet"'mafrost has been not~d on the north bank, This permafrost is within loC of freezing and is protected in most part by the thick ·tundra and heavy vegetation. During the foundation excavation, the 1 ass of i nsu l at i ng tundra may cause th aw1 ng of overburden and may result in unstable slopes and/or mud flows. Therefore, proper care and planning will be required during removal of this material. (b) Contra 1 of Under seep age and Up 1 i ft , A grout curtain wi 11 be provided under the core of the dam and extending several hundred feet into the abutment beyond the dam. The initial phases of the grout curtain wi 11 be exploratory in nature, in order to identify areas in the dam foundation rock that require more extensive treatment and provide detailed information of the foundation. A series of drain holes wi 11 be drilled downstream from the grout curtain to form a continuous drainage curtain. These holes wi 11 drain into underground grouting/drain- age g a 11 eri es, which wi 11 permit moni taring of seepages and water pressures and access for necessary remedial work if required at a later date. Si nee 9-9 -"': .. ~~ the rock mass is permafro.st ... affected, ground thawing will bEl required prior to grouting. Consolidation grouting is all planned under the core and the filters to provide a good contact surface free from open joints and fractures. (c) .Cofferdams and Dewatering Becauso. of the pervious nature of the thick riverbed desposits and the relatively high pool behind the upstream cofferdan, major dewatering opera- tions wi 11 be required during the foundation excavation and unti 1 the dam construction reaches above the diversion stage pool level. A slurry trench cutoff is currently proposed beneath the upstream cofferdam to control water flows during diversion. Further exploration is necessary in the riverbed to better define the extent and condition of the a 11 uvi a 1 mate- rials at the cofferdam site prior to construction. (d) Undergro~nd Structures The rock conditions at the Watana site are suitable for the construction of tunnels and underground caverns. From the geological and geotechnical viewpoint, the location and the orientation of these structures is influ- enced by the orientation and location of ·rock discontinuities. Permafrost conditi-ons will not have any major adverse impact except where thawing may be required for grouting. The RQD values indicate that 85 percent of the rock is of a good to exct=1- lent category. The remaining 15 percent represents poor quality rock associated with rock discontinuities. The major joint sets at the t~atana site are oriented at N40W (Set I) and N45E (Set II). Other four joint sets are minor. The major shear and fracture zones a 1 so parallel these general trends. The most favorable orientations for the tunnels and the large underground caverns are those with their long axes perpendicular to the major joint sets. These factors have been a major factor in selection of the alignments of the tunnels and major caverns to achieve maximum stabi 1- ity and minimum support requirement. Although little is known at this time about the insitu stress regime at the site, the general tectonic stress regime within the region is in a compres- sion mode.. The unconfined compressive strength of the rock ranges from to and suggests that overstressing problems such as spalling or slabing are not likely. Conventional rock bolt support using 1 inch diameter' bo 1 t s is genera 11 y considered adequate in most areas with spans less than 40 feet. For larger spans and in areas of poor quality rock, the support requirements have been determined on a case-by-case basis. In the case of large span openings, intersection of n~arby vertical and subhori- zontal joints can create unstable blocks in the crown. Allowances have · been made for the use of support measures such as shotcrete, welded wire fabric, and concrete lining in areas of potentially poor rock quality and water carrying tunnels under high head (such as penstocks). 9-10 I I I I 1 a· I 'I ~ I :I I I 'I I I I I I -.• ~ I I I I I I I I I I I' I I I I I I I I " (e) Although the rock mass by itself is fairly impervious, intersection of rock di scontinui ties may cause ground water problems during construction and act as path of seepage and high pore pressures during operation. Provisions have, therefore, been made for consolidation/ring grouting and suitably placed drain holes to reduce the risk of a build up of high pore pressures. Tunnel excavation can be performed using conventional drill and blast techniques or higD production mechanical excavations. Sufficient informa- tion is not available at this time to make this decision, and for feasibil~ ity assessment purposes, conventional drill and blast methods have been assumed.. The excavation of powerhouse caverns wi 11 be performed by dri 11 and blast using a primary heading, side slash and bench excavations approach. The spacing between long tunnels has been set at 2 .. 5 times the diameter of the 1 argest tunne 1. The spacing between the major caverns has been set such that a pillar thickness of 1.5 times the span of the larger cavern is mai nt ai ned. Stabi lity_of Soi 1 and Rock Slopes In most areas the excavation slopeS? wi 11 be in the rock. The slopes in the overburden, where necessary, have been based on the nature of soi 1, ground water table, and the height of the slope. In general, slopes in overburden will not be steeper than 2H:lV below the water table and 1.5:1V above the water tab 1 e. A bench of adequate width wi 11 be provided at the over~burden­ rock contact to accommodate any local slumping or slope failure and to intercept and d 1 spose of ground/seepage \vater. Fl atter s 1 opes may be required where frozen ground may be become unstable because of high pore pressures oduring thawing. The slopes in rock are controlled by the joint dips and orientations. Since major joint set dips are almost vertical, lH:lOV slopes are con- sidered reasonable up to 40 feet in height.. Where the height of slope exceeds 40 feet, a minimum of 10 root wide bench have been provided every 40 feet depth to facilitate construction and to provide access for future maintenance .. These berms will also intercept falling loose rock pieces and surface/ground water drainage. Locally, rock bolting or similar support techniques and drain holes have been provided in appropriate areas to maintain stable rock slopes. Excavation of tunnel portals will be accomplished by liberal use of pattern rock bolting and some provision for concrete/shotcrete to reduce the risk of unstab 1 e s 1 opes~ Speci a 1 det ai 1 s have been incorporated .; n areas where slopes-intersect or cross larger shear zones or otherwise unstable rock. 9-11 (f) Use of Excavated Rock in Dam Construction Since most of the rock excavation wi 11-be within the diorite and andesite ·rocks, the quality of rock will be acceptable for use in the rockfill portions of the dam. The exception to this could be excavation in poor quality or weathered rock which will-result in unacceptable fill. The use of the rockfill in the dam will be be limited to portions of the downstream shell, and in zones of rip rap material.· Proper quality control wi11 be exercised in selecting this material. (g) Relict Channel A deep bedrock depression exists on the north bank of the river extending from about 2,500 feet west of Deadman Creek northwest-toward Tsusena Creek. The depth to.bedrock is as much as 400 feet below the surface and the reservoir level. The overburden consists of several sequences of glacial deposits, lake sediments, and a 11 uvi urn varying in thickness and character both laterally and with depth. Some of these granular deposits exhibit high permeability, and ice inclusions were noted at a depth of several hundred feet suggesting the possibility of permafrost. The ground water surface has not been well defined and a perched water table has been encountered in at least one boring indicating artesian pressure, and is also evidenced by the presence of several surface lakes. With the proposed range of reservoir levelss these overburden deposits will become saturated. A bedrock contour map of the Re 1 i ct Channe 1 area is presented in Figure 9.13. A saddle dam of relatively low height is planned across the topographic low of this Re 1 i ct Ch anne 1 . Det ai 1 s of the potentia 1 design prob 1 ems to be dealt with in the Relict Channel and the proposed methods of treatment are discussed further in Section 12. Additional investigation will be necessary to properly characterize the subsurface _condition and the final detai 1 s. of foundation treatment in the area prior to construction. 9.4 -Seismic Considerations The seismicity of the Susitna Basin and the sources of earthquakes are discussed in Section 7 of this report. This section presents the implications of the sei smi city on the design of the Watana project. (a) Seismic Design Approach For earthquake engineering and design considerations, the project struc- tures have been classified as either critical structures or non-crttical structures. Critical structures inc 1 ude the dam and simi 1 ar major --struc;;ii· tures whose failure may result in sudden and uncontrolled release of large volumes of water which may endanger property and lives downstream. The non-critical structures are those structures whose failure can be assessed as an econpmic or financial loss to the projec_t in terms of lost revenue, repair~ and/or replacement cost .. Critical structures will be designed to safely withstand the effect of the ''Safety Evaluation Earthquake" (SEE) for the site. No si gni fi cant damage to these structures wi 11 be accept_ed under these conditions. The design of non-critical structures for earthquake conditions is undertaken using conventional Uniform Building Code recommen- dations .. 9-12 I I I I -1 I I I I I 1 I I I I I I' I I 1 I I I I I I I I I I I I I I I I I I I I For design of critical structures the effective acceleration for the SEE has been determined as 0.8 x actual SEE acceleration. In the case of the earthfill structures, designs are basad on analyses using a projected time hi star y for the se lee ted e,arthquake event. For other structures a corres- pondingly scaled response spectrum is used. The selected SEE for Watana was based on a consideration of two of the most severe events which might occur. The first of these is the "terrain'1 or "detection level 11 earthquake which has been characterized as follows: -Magnitude: 6-1/4 to 6-1/2 -- -- -Location: Approximately 3 km from structure -Maximum Acceleration·: ~1ean 0.55 g to 0.60 g 84th percentile 0.70 g -Peak Spectral Acceleration: Mean 1.37 g to 1.50 g 84th percentile 1.77 g The duration of this event is relatively short and the time history approach to design was not developed. The response spectra for this event are shown on Figure 10.7A. The effective peak acceleration for design of structures is then: -Design a max= 0.8 x 0.70 g = 0.56 g (Sa) max : 0.56 g x 2.5 = 1.40 g The second earthquake source to be considered for design at Watana is the Benioff Zone. (b) Safety Evaluation Earthquake for Watana Although the "terrain .. earthquake would result in more severe ground motions, the duration of these motions is relatively short and the likeli- hood of occurrence of such an event is extremely small. A more likely source of strong ground shaking at the Watana site is the Benioff Zone. The estimated mean peak response spectrum for the SEE for this event is presented in Figure 7.14, along with the 84th percentile response spectrum. A maximum horizontal acceleration level for the 84th percentile response spectrum for the Benioff event is approximately . The design of the Watana Dam has been based in the projected time history for this event as shown in Figure ___ , and as discussed in Section 12. 9.5 -Selection of Reservoir Levels This section describes the approach used and the results of the evaluations made in the determintion of optimum Watana reservoir 1evel. The selected elevation of the Watana dam crest is based on considerations of the value of the hydro- e 1 ectri c energy produced from the associ a ted reserv'oi r~, and geotechni ca 1 con- straints on reservoir levels. Firm energy, average annu~l energy, construction costs and operation and maintenance costs were determined for the Watana deve.lopment with dam crest elevations of 2240, 2190 and 2140 feet. The relative 9-13 value of energy produced for ·each of these three dam elevations was then deter- mined bymeans of the OGP generation planning model as discussed in Section 6, to-determine the long term present worth cost of meeting the Railbelt system energy demand. Finally the physical constraints imposed on dam height and reservoir elevation by geotechnical considerations were reviewed and incor- porated into the crest elevation selection process. (a) Methodology Firm and average annual energy produced by the Susitna development are base(' on 32 years of hydrolog;:cal records. The energy produced was deter- mined by using a mult-reservoir simulation of the operation of the Watan·a and Devil Canyon reservoirs. A variety of reservoir drawdowns were ex ami ned, and drawdowns producing the maxi mum firm energy consistent with engineering feasibility and cost of the intake structure were selected (see Section 9.11). Minimum flow requirements were established at both project sites based on mechani ca 1 p 1 ant-re 1 ated restrictions and dm-Jnstream fisheries considerations. As discussed in Section 9.6, to meet system demand the required maximum generating capability at Watana in the period 1993 and 2010 ranges from 665 MW to 908 M~~. For the reservoir level determinations, energy estimates were made on the basis of assumed average annua 1 capacity requi rememts of 650 MW at Watana in 1993, increasing to 1020 MW at ~~atana in 2007, \t:'ith an additional 600 MW at Devi 1 Canyon coming on line in the year 2002. Yearly ~ystem demand and monthly and daily load patterns within the Railbe1t over a 29 year period·were based on forecasts developed as described in Section 5 and 6. The long term present worth costs of the generation system required to meet the Railbelt energy demand were then determined for each of the three· crest elevations of the Watana dam using the OGP V model. As discussed in Section 6, these present worth costs are based on economic parameters, not of inflation. The construction cost estimates used in the OGP V modeling process for the Watana and Devil Canyon projects were based on preliminary conceptual layouts and construction schedules. Further refinement of these 1 ayout s has taken p 1 ace in the process of deve 1 opment of the estimates presented in Section 16. These ref''\nements have no signi- ficant impact on the reservoir 1 eve 1 selection.. The oasis of assumed costs for construction and operation of alternative energy generation facilities is also discussed in Section 6. (b) Optimization Optimization of the Watana reservoir level was based on an evaluation of three dam crest elevations of 2240, 2190, and 2140 feet. These crest elevations apply to the central portion of the embankment with appropriate allowances for freeboard and seismic slumping, and correspond to maximum operating levels of the reservoir of 2215, 2165 and 2115 feet, respec- tively. Average annual energy calculated for each case using the reservoir simulation model are given in Table 9 .. 3, together with corresponding project construction costs. 9-14 I I ;I I I I I I I I I I I I I I I I I' I I I I I I I ••• I I I I ·I I I I I I I In the determination of long term present worth of production costs, the Susitna capital costs were adjusted to include an allowance for ·interest duri n9 .construct 1 on and then used as input to the OGP V mode 1. Si mu 1 a ted annual energy yields were distributed on a monthly basis by the reservoir operation model to match as closely as possible the project energy demand of the Railbelt and then input to the OGP V model. The long term present worth production costs of meeting the Railbelt energy demand using the Susitna development as the primary source of energy, were thus determined for each of the three reservoir levels. The results of these evaluat1ons are shown in Table 9 .. 4, and plots showing the variation of the long term present worth with dam crest elevation are shown in Figure 9 .17. This figure indicates that on the basis of the assumptions used, the minimum long term present worth of production costs occurs at a Watan.a dam crest elevation ranging from approximately 2160 to 2200 feet (reservoir levels 2140 to 2180). A higher dam crest will still result in a development which has an overall net economic benefit relative to displaced energy sources. However, it is also clear· that as the height of the Watana dam is increased, the unit costs of additional energy produced at Watana is somewhat greater than fo·r the displaced alternative energy sources~ Hence, the long-term present worth of the overall system increases. Conversely, as the height of the dam is 1 owe red, and thus Watana produces less energy, the unit cost of the energy produced by alternative generation sources to replace the lost Susitna energy, is more expensive than Susitna energy. In thise case also, the long-term present worth again increases. (c) Conclusions It is important to clearly establish the overall objective of setting the Watana reservoir level. An objective which is to minimize the long term present worth of energy cost will lead to selection of a lower reservoir level than an objective vJhich is to maximize the amount of energy which can be obtained from the available resource, while still doing so economically and within accepted technical and environmental constr~nts. The three values of long term present worth developed by the OGP V computer runs defined a relationship between long term present worth cost and Watana dam height which is relatively insensitive to dam height. There is an indication that a small difference in system present worth occurs as the Watana dam crest is raised or lowered ovesr the range considered. However these di.fferences are of the same orde~ as the inaccuracies which a~e inherent in capital cost estimates for the development of such major generation facilities within the Railbelt. Little value would thus be gained from analyzing intermediate dam heights to further define the curve. The insensitivity is highlighted by the graph of present worth against dam height in Figure 9.17. This figur€ shows these slight variations in context within the total long term present worth cost of the system. 9-15. Thus~ from an economic standpoint, the optimum crest elevation could be considered as varying over a range of elevations of as much as 50 to 100 feet. The governing factors is establishing the upper limit of dam height were consequently physical and geotechni ca 1· considerations, a 11 owing the objective of maximizing the economic use of the Susitna resource still to be. sati sfi ect~ · The normal maximum operating level of the reservoir was therefore set at elevation 2185 feet. At this level, for up to the 1:10,000 year flood occurrence, there will be no danger of over-topping the low lying portion of the relict channel on the right side of the river. In the· unlikely event of floods of greater severity, a freeboard dike in the low area of up to 10 feet in height has been incorporated in the design" ·With this approach, the Watana project will develop the maximum energy reasonably available without incurring the need for costly water retaining structures in the relict channel area. 9.6 -Selection of Insta 11 ed Capacity The generating capacity to be installed at both Watana and Devil Canyon was determined on the basis of generation planning studies described in Sections 6 and 8, together ~ith appropriate consideration of the following: -Available firm and average energy from Watana and Devi 1 Canyon; -The forecast energy demand and peak load demand of the system; -Available firm and average energy from other existing and committed plant; -Capital cost and annual operating costs for Watana and Devi 1 Canyon; -Capital cost a.nd annual operating costs for alternative sources of energy and capacity; -Environmental constraints on reservoir operatioh; and -Turbine and generator operating characteristics. {a) Methodology The following procedure was used to select the installed capacity at Watana: -The firm and average energy available at both Watana and Devil Canyon was determined using a reservoir operation computer si mui ati on program based on the 32 years of hydrological record developed as described in Section 7 (see Plate 22). -An assessment was made of the a 1 tern at i ve therma 1 energy required to meet the predicted 1 cad forecast, using a computer ~; mul at ion of economic 1 oad 'dispatch from av ai 1 ab 1 e p 1 ant in the firm year. This determined the optimum schedu 1 i ng and capacity of new therma 1 p 1 ant required to meet the . minimum Loss-of-Load Probability (LOLP) criterion for system security. - A determination was then made of the generating capacity required to utilize the available energy from the Susitna Project in the hydrological years of record, based on the following assumptions: 9-16 c "I' s_"' . I I I I I I I I I I I I I I I ' •'• I I I I I I I tl I I I I I I I I ,, I I' I I . In a wet year, hydro energy in excess of system demand) displaces thermal energy (from coal,· gas turbine, combined cycle, or diesel p 1 ant). . In an average year, where thermal energy is required to meet system energy demand, hydro energy is used either to satisfy peak demand with thermal energy supplying base load (Option 1); or hydro energy is used to supply base load requirements with thermal energy at peak demand (Option 2). The actual choice is based on made on economic load dispatch criterla. . Devil Canyon energy is used predominantly as base load energy because of environmental constraints on downstream flow variations. • The maximum installed capacity required was determined on the basis of the est ab 1 ished peak generating capacity described above plus any hydro standby or spinning ~eserve requirement~ (b) Total Installed Capacity .. The required total capacity at Watana in a wet year (determined as descri be·d above), exc 1 udi ng st-andby and spinning reserve capacity, is summarized below. The capacities are based on the Battelle medium load forecast. Capacity (MW) Demand Year Option 1 Option 2 1993 801 801 1995 839 839 2000 862 742 2002 (In~ 1 • Devi 1 Canyon) 660 655 2005 (Incl. Devil Canyon) 750 740 2010 (Incl. Devi 1 Canyon) 908 900 On the basis of this evaluation, the ultimate power generation capability at Watana was selected as 1000 MW for preliminary design purposes, to allow a margin for hydro spinning reserve and standby for forced outage. This installation also provides a low cost margin in the event than an accelerated growth of demand occurs. (c) Unit Capacity Selection of the unit size for a given total capacity is a compromise betw~en the initial least cost solution, generally inv9lving a scheme with a smaller number of large capacity units, and the improved plant efficiency and security of operation provided by a larger number of smaller capacity 9-17 units. _Other factors include the size of each unit .as a proportion of the total system load and the minimum anticipated load on the station_. Any requirement Jor a mlnimum downstream flow would also affect the selecticm, since, for example, Francis turbines will. not operate effectively at less than about 50 percent full output. Growth of the actual load demand is also a significant factor~ ·since the unit installation may be phased to ~atch the actual load growth~ The number of units and their i ndi vi du a 1 ratings were determined by the requirement to deliver the design peak capacity in the critical demand month of December, at minimum December reservoir level, with turbine wicket gates fully open. In addition, unit selection was based on consideration of the following: -Rate of load growth with time;- -Load following capability at part station operation; -Efficiency variation with load and head; p -Minimum acceptable load on each machine; -Minimum downstream compensation flow; -Standby capac1ty and spinning reserve; and -Sensitivity to change in forecasted load growth. An examination was made of the economioc impact on power plant production costs for various combinations of unit numbers and rated capaci ty:s which would provide the selected capability of 1000 MW and satisfy the considera- tions outlined above. As discussed above, for any given installed capa- city, plant efficiency increases as the number of units increases. This is illustrated in Figure 9.18. The assumed capitalized value of the resulting additional annual energy used for this evaluation was 1000 mills per kWh; based on economic parameters developed in previously described system studies. Variations in unit numbers and capacity will affect the cost of the power intakes, penstocks, powerhouse, and tailrace; the differences in these capital costs were estimated and included in the evaluation. The results of this analysis are presented below. Number of Units 4 6 8 Rated Capacity of Unit (MW) 250 170 125 Capitalized Value of Additional Energy ($ Millions) 40 50 Additi anal Capital Cost ($ Millions) 31 58 It is apparent from th1s analysis that a six-unit scheme is the most economic alternative. This scheme also offers a high degree of flexibility and security of operation compared to the four-unit alternative, as well as advantages if unit i nsta 11 ati on is required to be phased to match actua 1 load growth. The net economic benefits of the six-unit scheme are greater than those of the eight-unit scheme, whi 1 e at the same time, no significant operational or scheduling advantages are associated with the eight-unit scheme. Accordingly, a scheme incorporating six units each with a rated capacity of 170 MW, for a total of 1020 MW~ has been adopted for all alternatives. - :_., 9-18 I I I I J ll I I I I I I I 'I I I I I I I I I ' ·. •• 1: I I I •• I I I I I .I 'I I I I I For proj,ect .design and cost estimating purposes in the current studys the installed capacity of 1020 MW has been assumed. From generation planning and financial analyses, certain advantages may be gained from staging the installation of generating equipment over a somewhat longer period. These aspects wi 11 be addressed further during detai 1 ed design of the project .. The power f aci l i ties and associated equipment are described in detail in Section 12. 9.7 -Selection of Spillway Capacity Normal design practice for projects of this magnitude, together with applicable design regulations, require that the project be capab 1 e of passing the probab 1 e, "c .•. maxi mum flood (PMF) routed through the re·servoi r without overtopping the damo In addition to this requirement, the project should have sufficient spillway capacity to safely pass a major f1 ood of 1 esser magnitude than the probab 1 e max- imum flood without endangering the main dam or ancillary structures, in a manner which will avoid injury or loss of life, or damage to the project itself. The frequency of occurrence of this flood, known as the spi ·1 lway design flood or Standard Project Flood (SPF), is generally selected on the basis of an evalua- tion of the risks of the project if the spillway flood is exceeded, compared to the costs of the structures required to safely discharge the flood. A list of spillway design flood frequencies and magnitudes for several major projects is presented below. Spi 11 way Design Flood Frequencies and Magnitudes Spi~lwax Design Flooa Basin Spillway Capacity Peak PMF After Routing (cfs) Project ~------~----~--~--~----~~--~~~------------~------ Frequency Inflow {cfs) (cfs) Design* Mica, Canada Churchill Falls, Canada New Bullards, USA Oro vi 11 e, USA Guri, Venezuela (final stage) Itaipu, Brazi 1 Sayano, USSR · PMF 1:10,000 PMF 1:10,000 PMF PMF 1:10,000 250,000 600,000 226,000 440,500 1,000,000 2,195,000 480!t000 250,000 1,000,000 226,000 711,400 1,000,000 2s195,000 N/A 150,000 230\JOOO 170~000 440,500 1,000,000 2,195,000 680s000 *All spillways except Sayano have capacity to pass PMF with surcharge On the basis of the foregoing, a spillway design flood with a return frequency of 1:10,000 years was selected for Watana. 9-19 . - ></ ;:; ~ ,~,..., ~ '''• -.<"-•4 ''' ,, ,,.,,.. . ..,,".' ' '•"~ ..... ~;;. "''""' • <;,••-."''•·•••!;~ ... ,_ -~~~--·J "'" "•••"'>~>-'""~•:...,._ 0 h ••' ''"'"" "'""' 0 .,,,)>,...,., •• , •. -' •• • • • < '• ., "••' • • •••• •-~ oo-<o.---"'"•'"w --~~ "' •'""'" '"' •• ,,., • 0 • .... The flood frequency analysis undertaken as described in Section 7.2 indicates the following values: Flood Probable Maximum Spillway Design Frequency --1:10,000 Inflow Peak 326~000 cfs 156~000 cfs (0.47 PMF) -AdditioP.al capacity required to pass the Ptt1F discharge will be provided by an emergency spillway consisting of a fuse plug and rock channel cut on the right bank. 9.8 -Main Dam Alternatives This section describes the a 1 ternati ve types of dam considered at the Wat ana site and the basis for the selected. alternative. (a) Comparison of Embankment and Concrete Type Dams The selection between an embankment type or a concrete type dam is usually made based on the configuration of the ·~·alley, the condition of the founda- tion rock and depth of the overburden, and the relative availability of construction materials. Previous studies by the COE envisaged an embank- ment dam at Watana. Initial studies as part of these current evaluation included comparison of a earthfi 11 dam with a concrete arch dam at the site. An arrangement for a concrete arch dam alternati ;e at Watana is presented in Plate 11. The results of this analysis indicated that the cost of the embankment dam was somewhat lower than the arch dam, based on the use of concrete costs significantly lower than comparable costs used for Devil Canyon. This preliminary evaluation did not indicate any signi- ficant advantages for the concrete arch relative to the arrangement of other structures, or the construction schedu 1 e relative to the use of a concrete arch. Based on the overall cost differences described above, and the likelihood that the cost of the arch dam wou 1 d increase re 1 at i ve to that of the embankment dam, the arch dam a·lternative was eliminated from further consideration. (b) Selection of Dam Type The deve 1 opment of the design of the main dam, togethP.r with a description of the various features of the dam, is given in Section 12. The dam is, of course, the central and most costly component of the project, and a brief discussion of the development of the finally selected design, together with some of the factors which influenced deve 1 opment of the genera 1 arrangement are presented in this section. Selection of the configuration of the embankment dq.m cross-section \~as un- dertaken within the context of the following basic considerations: -The avai 1 ability of suitable construction materials within economic haul distance, particularly impervious core material; '9-20 . : .. I I I I I I I I I I I I I I I I I I I I I I I I I 'I. ! I I I ·~ I I' •• I I I I -The requirement that the dam be capable of withstanding the effects of a significant earthquake shock, as well as the static loads imposed by the reservoir and its own weight; -The relatively limita· constrtiction season available for placement of compacted fi 11 materia 1 s. The exp lor at ion program undertaken during 1980 and 1981 indicated that ade- quate quantities of materia 1 s sui tab 1 e for dam construction were located within reasonable distance from the site. The locations of potential borrow materials for the dam are shown on Figure ·9.6. The well graded silty sand material from Borrow Area D is considered the most promising source of impervious filt. Compaction tests indicate a natural moisture constant slightly on the wet side of optimum moisture content, so that control of moisture content will ·be critical in achieving a dense imper- vious core \'lith high shear s"trength. Potential sources for the upstream and downstream shells included either river gravel from borrow areas along the Susitna Riyer~ or compacted rock fill from structural excavation of quarries. The main dam wi 11 consist of a compacted impervious core protected by fine and coarse filter and transition zones on both the upstream and downstream slopes of the core. The upstream and downstream outer supporting f"ill zones wi 11 comprise t"'e 1 ati ve ly free draining materials such as compacted gravel or rockfill, providing stability to the overall embankment struc- ture. The location. and inclination of the impervious core is fundamental to the design of the embankment. Two basic alter·natives exist in this regard.: - A vertical core located centrally within the dam; and -An inclined core with both faces sloping upstream. The advantages and disadvantages of these two alternatives are discussed in Section 12. A central vertical core was chosen for the embankment based on a review of precedent design and the nature of the avai 1 able potential i m- pervious material. In order to evaluate the relative sensitivity of the project arrangement.to changes in.exterior dam slopes, two alternatives were used in the prelim- inary review: -2.4H: lV upstream and 2H:1V· downstream -2.25H: lV upstream and 2H:lV downstream As part of the i ntermed1 ate review, the volume of the dam with an upstream slope of 2o4H:lV was.computed for four alternative dam axes. The location of these alternative axes are shown on Plate 12. The results of this comparison are described below: 9-21 ! . .. .J Alternative 1 2 3 4 Total Volume (millions c.y.) 69.2 71.7 69.3 71.9 During the intermediate review, the upstream slope of the dam was flattened to 2.75H:1V. This slope was based on a conservative estimate of the effective shear strength parameters of the available construction materia 1 s, as well as a conservative a 11 owance in the design for the effects of earthquake shock on the dam. During the fi na 1 review stage, the exterior upstream slope of the dam was steepened from 2.75H:lV to 2.4H:lV, reflecting the results of the static and dynamic design analyses being undertaken at the same time as the general arrangement studies. This section was used for the final review of alternative schemes. Further refinements to the design were subsequent 1 y incorporated in the final design presented in Section 12, but these did not influence the selection of the final scheme. 9.9 -Diversion Scheme Alternatives The topography of the site essentially dictated that diversion of the river during construction be accomplished using one or two diversion tunnels with upstream and downstream cofferdams protecting the main construction area. The configuration of the river in th~e vicinity of the site favors location of the diversion tunnel or tunnels on the right bank, since the tunnel length for a comparable scheme on the left bank could be approximately 2,000 feet greater. In addition, rock conditions on the right bank are more favorable for tunneling and excavation of intake and outlet portals. Notwithstanding these considera- tions, the selection process for establishing the final general arrangement included examination of tunnel locations on both banks. (a) Design Flood for Diversion The recurrence interval of the design flood for diversion is generally established based on the characteristics of the flow regime of the river, the length of the construction period for which diversion is required and the probable consequences of overtopping of the cofferdams. These last two considerations are usually evaluated as part of an economic risk analysis in which the cost of the diversion scheme, and the risks involved in exceeding the capabilities of the scheme.. This type of analysis wi.ll be undertaken as part of the detailed design phase of the project, but for the purposes of the feasibility analysis, design criteria and experience from other projects similar in scope and nature have been used. 9-22 ., I I 'I I I I 'I I I I I 'I I I 'I I I I I I I I :a I I ' I I I I I. I I I I) I I '" I .• ''-· At Watana, damage to the parfia1ly completed project tr~ould be significant, or more importantly, would probably result in at least a one-year delay in the completion schedule •. A ·preliminary evaluation of the construction schedule indicates that the diversion scheme would be required to operate for 4 or 5 years until the dam was completed sufficiently to permit initial filling of the reservoi~. A design flood with a return frequency of 1:50 years was selected based on experience and practice with other major hydroelectric projects. This approximates a 90 percent probability that the cofferdam 1Ni 11 not be overtopped over the critical 5 year constructi·on period. The equivalent inflow for the design_ flood together with average flow char acteri st i c s of the river s i gni fi cant to diversion are presented below: Average annual flow Maximum average monthly flow Minimum average monthly flow Design flood inflow (1:50 years) (b) Goff erd ams 7,860 cfs 23,100 cfs (June) 890 cfs (March) 81,:100 cfs The character and considerable depth of riverbed alluvium at both cofferdam sites indicate that embankment type cofferdam structures waul d be the on 1 y technically and economically feasible alternative at Watana. For the purposes of establishing the overall general arrangement of the project and for subsequent diversion optimization studies, the upstream cofferdam section adopted comprised an initial closure section approximately 20 feet high constructed in the wet, with a zoned embankment constructed in the . dry. The downstream cofferdam comprises a closure dam structure approxi- mately 30 feet high placed in the wet. Control of underseepage through the relatively pervious underlying alluvium mat·erial will be achieved by means of a soil/ bentonite slurry wall. The selected cofferdam sections are described in more detail in Section 12. (c) Diversion Tunnels A basic consideration in evaluation of any diversion tunnel'scheme is an ex ami nation of the advantages and disadvantages of concrete-lined tunne 1 s compared to unlined tunnels. Preliminary hydraulic studies indicated that the design flood routed through the diversion scheme would result in a de- sign discharge of approximately 80,500 cfs. For concrete-lined tunnels, design velocities of the order of 50-feet per second have been used in sev~ eral projects. For unlined tunnels~ maximum design velocities ranging from 10 fps in good quality rock to 4 fps in less competent material are typical. Using a maximum permissible velocity of 10 fps!! four unlined tunne 1 s each with an equivalent diameter of 50 feet waul d be required to pass the design flow. Alternatively~ a design velocity of 50 fps would theoretically permit the use of one concrete~lined tunnel with an equivalent finished d1qmeter of 44 feet. The unlined tunnels would require 4.5 times as much excavation as the lined alternative, together with at least four times as much tunnel support cost. This would only be partially 9-23 __ (d) offset by the cost of the concrete lining~ Apart from co'St, the most important single factor relates to the security and reliability of the diversion scheme. The tunne 1 s wi 11 undoubted 1 y traverse numerous unfavorable geological conditions and structures, as yet undefined, during construction. The reliability of an unlined tunnel is more dependent on rock conditions than is a lined tunnel, particularly given the extended period during which the diversion scheme is required to operate. These considerations, together with cost and the somewhat questionable feasibi 1- i ty of a tunne 1 with a diameter approaching 50 feet in this type of rock,. are considered sufficient to eliminate consideration of unlined tunnels for the diversion scheme. · The following alternative lined tunnel schemes were examin~d as part of this analysis: -Pressure tunnel with a free outlet; -Pressure tunnel with a submerged outlet; and -Free flow tunnel. Pressure tunnels are designed to flow full and accordingly must withstand internal pressure. The most widely used type of pressure tunnel for diver- sion has the crown of the outlet portal submerged during all flow conditions .. Emergency Release Facilities While not an integral part of the diversion scheme itself, the emergency release facilities greately influenced the number, type, and arrangement of the diversion tunnels selected for the final scheme. At an early stage of the study, it was established that, in accordance with current design practice, some form of low level release facility was roequired to permit lowering o~ the reservoir in the event of an extreme emergency. Since the primary discharge facilities wil1 be located near the crest of the dam, they would be ineffective if the reservoir level had to be reduced below approximate elevation 1950. The most economical alternative available would involve converting an existing diversion tunnel to permanent use as a low level outlet facility. Since it obviously would be necessary to maintain the diversion scheme in effective service during construction of the low level outlet works, two or more tunnels would be required if this a 1 ternati ve was adopted. The use of two diversion tunnels, while not contributing to the overall economy of the project, prov1des an additional measure of security to the diversion scheme in case of the loss of service of one tunnel during an emergency. The use of two· tunnels also provides greater flexibility in construction scheduling:> particularly since concrete-lined tunnels are required. Additionally, potential problems with stability of two smaller openings are likely to be 1 ess severe than for the 1 arger spans ,g.ssoci ated with a single tunnel .. 9-24 I I I I I I I I I I I I I I. I I I I "I J. I I I. I If operation of the emergency low level release facilities is required, it \t~i ll extend over a considerable period of time. Discharge of the faci li- ties at the heads required could r<esult in serious erosion downstream .. This requirement necessitated some form of energy dissipation prior to returning the reservoir water to the river. Given the space restrictions imposed by the size of the diversion tunne 1, it was decided to uti 1 i ze a double expansion system with concrete plugs within the tunnel. The operation of the expansion chamber is described in Section 12. The use of this arrangement requires that the chamber be located above tai lwater to prevent ~cavitation in the area of the emerging jets from the downstream plug. The implications of this restraint require that if a diversion tunnel is to be used as part of the emergency low level release facilities, it must act as a free flow tunnel. II (e) Optimization of Diversion Scheme I ' ., I I ,, I \. ~--· I ' I I I Given the considerations described above relative to design flows, coffer- dam configuration and alternative types of tunnels, an economic study was undertaken to determine the optimum combination of upstream cofferdam height and tunne 1 type and diameter .. Capital costs were developed for three heights for an upstream cofferdam embankment with a 30 foot wide crest and exterior slopes of 2H:1V. A freeboard allowance of 5 feet for settlement and wave runup and 10 feet for the effects of downstream ice jamming on tailwater was adopted. Capital costs for the a::·,;)ociated tunnel alternatives included allowances for excavation, conc\·ete liner, rock bolts, and stee'l supports. Casts were a 1 so deve 1 oped for the upstream and downstream por·t a 1 s. inc 1 udi ng excavation and support. The cost of intake and out 1 et gate structures and associated gates was determined not to vary significantly with tunnel diameter and was excluded from the analysis. A right bank configuration was selected· and the corresp?nding tunnel length in all cases was assumed to be 4,700 feet. Curves of headwater elevation versus tunnel diameter for the various tunnel alternatives with submerged and free outlets are presented in Figure 9.13. The relationship between capital cost and crest elevation for the upstream cofferdam is shown in Figure 9.14. The capital cost for various tunnel diameters with free and submerged outlets is given in Figure 9.15. The results of the optimization study are presented in Figure 9.16" and in- dicate the following optimum solutions for each altern-ative. Cofferdam TypP of Tunnel Diameter (ft) Elevation (ft} Total Cost ($) 2-Pressure tunnels 30 1595 66,000,000 2-Free flow tunnels 32.5 1570 68,000,000 2-Free flow tunnels 35 1545 69,000,000 9-25 I The foregoing indicates that a relatively small cost differential (4 to 5 J.~."·. · .. percent) separates the various alternatives in range of tunnel diameter : from 30 to 35 feet. (f) Selected Diversion Scheme J Although a scheme incorporating two 30 foot diameter pressure tunnels with submerged outlets is marginally the most economical soiution as discussed in (d), at le.ast one tunnel must have a free outlet if it is to be con- verted into a low level outlet tunnel. An important consideration at this point is cofferdam closure. For the pressure tunnel scheme, the invert of the tunnel entrance is below riverbed elevation, and once the tunnel is complete diversion can be accomplished with a closure dam section approximately 10 feet high. The free flew tunnel scheme however requires a tunnel invert approximately 30 feet above riverbed level, and diversion will involve an end-dumped closure section 50 feet high.. Two basic problems are associated with closure embankments of this height -velocities during final closure would be quite high, requir- ing 1 arge size stone to remai~n in p 1 ace, and subsequent sea 1 i ng of the closure embankment-in the wet must be done at significant depth, with relatively less control than for lower embankments. In consideration of these problems and restraints, a combination of one pressure tunnel and one free flow tunnel (or pressure tunnel with free out- 1 et) was adopted. This wi 11 permit i ni ti a l diversion to be made using the lower pressure tunnel, thereby simplifying this critical operation and avoiding potentially serious delays in the schedule. Two alternatives were re-evaluated as follows: Tunne 1 Diameter (feet) 30 35 Upstream Cofferdam Crest Elevation Approximate Height (feet) (feet) 1595 1545 150 <: 100 More detailed layout studies indicated that the higher cofferdam associated with the 30 foot diameter tunnel alternative \vould require locating the in- let portal further upstream into "The Fins" shear feature. Since good rock conditions for portal construction are essential, and the 35 foot diameter tunnel alternative would permit a portal location downstream of 11 The Fins 11, this latter alternative was adopted. As noted in {e), the·overall cost difference was not significant in the range of tunnel diameters considered, and the scheme incorporating two 35 foot diameter tunnels with an upstream cofferdam crest elevation 1545 was incorporated as part of the selected general arrangement. The various components of the selected diversion scheme are described in Section 12. 9-26 I I I I I I I I I I 'I I ' I I I I .I I I: I I, I I I I I I. I I ·~- 1 '·' . !I I ,---~--,-, -.·.~----,.,-, .. ' l 9.10-Spillway Facilities Alternatives As discussed in Section 9.7, the project has been designed to safely pass the floods with the following return frequencies: Flood Design Flood Probable Maximum Flood Frequency 1:10,000 years Spillway Discharge ( cfs) 120,000 235,000 Discharge of the spillway design flood wi 11 require a gated service spillway on either the 1 eft or right bank. Three basic alternative spi 11way types were ex ami ned: / -Chute spillway with flip bucket; -Chute spillway with stilling basin; and -Cascade spillway. Consideration was also given to combinations of these alternatives with or without supplemental facilities such as valved tunnels and an emergency spillway fuse plug for handling flood con~itions. Clearly the selected spillway alternatives will greatly influence and be influenced by the project general arrangement. A discussion of the development of the general arrangement is presented in Section 9.12. (a) Energy_ Di ssi pati on The two chute spillway alternatives considered effect energy dissipation either by means of a flip bucket which directs the spillway discharge in a free-fall jet into a plunge pool in the river well downstream from the structure, or a stilling basin at the end of the chute which dissipates energy in a hydraulic jump. The cascade type spillway limits the fr,ee fall height of the discharge by uti 1 i zing a series of excavated steps do\'Jn to river level, with energy dissipation at each step and reduction of the velocity heads. All spillway alternatives were assumed to incorporate a concrete agee type control section controlled by fi.xed roller vertical lift gates. Chute spillway sections were assumed to be concrete lined, with ample pr 'lision for air entrainment in the chute to prevent cavitation and pres sur::.:. ~eli ef drai~s and rock anchors in the foundation. A detailed description 6i the selected spillway alternative is given in Section 12. (b) Environmental Mitigation During development of the general arrangements for both Watana and [}evi 1 Canyon~ a restriction was imposed on the allowance of excess dissolved nitrogen in the spillway discharges. Supersaturation occurs when aerated flows are subjected to pressure increases~ forcing excess nitrogen into 9-27 1 "f I I :· ,,;.;,.S· ... , .. ~,. 'S •• ~.,.1 .. ~ ,,,:"·v~· "·"·· ·, . ,m.~· .,.,,_,. ~,. ~'*'"'''!'" ~.·'··~-~.~·''-"I"''""'~·~;.,_~~.!~~!..!C._-'~~tb_.:....•_!_>::::_,·~-·':_._~~~:.c0~~,,.u, •. _,<!•'~.·~·"',.,"'""'·-·'"''"''i'';.,..._~cc~"'~\·<"'"'·'""' .. '' ,,..~-. ... ,H-.• ...... ...-fll.•:.·-..._ .. ,.1--,..·~"·"'"""'-•~•-H•""-•~~-.""'' '~·~'-'"~""'····~·-'>-"-·,·· w ·-.~·#-~--.-~.t .. .-.c• solution~ This occurs when water is subjected to pressures approaching two atmospheres and would occur in deep plunge pools or at large hydraulic jumps.. The excess nitrogen would not be dissipated within the downstream Devil Canyon reservoir and a buildup of nitrogen concentration could occur throughout the body of water. It would eventually-be discharged downstream from Devil Canyon with extreme··Jy harmful effects on the fish population. On the basis of an evaluation of the related impacts, and discussions with interested federal and state environmental agencies, spillway facilities were designed to limit discharges of nitrogen supersaturated water from Watana to a recurrence period of not less than 1:50 years. 9.11 -Power Facilities Alternative Selection of the optimum power plant development involved consider~ion of the fol!owing: -Location, type and size of the power plant; -Geotechnical considerations; -Number, type, size and setting of generating units; -Arrangement of 1 ntake and water passages; and -Environmental constraints. The se 1 ecti on of the i nsta 11 ed capacity of 1020 MW at Watana is described in . detail in Sect1on 9.6. The detailed comparison of power facilities alternatives is described in Appendix D. A summary of the general conclusions is described below., (a) Comparison of Surface and Underground Powerhouse Preliminary studies were carried out to compare the construction casts of a surface powerhouse and an underground powerhouse at Watana. These studies were undertaken on the basis of preliminary conceptual layouts assuming units and an installed capacity of 840 MW. The comparative cost --.---estimates for powerhouse civil works and electrical and mechanical equip- ment (excluding common items) indicated an advantage in favor of the under- ground powerhouse of $16,300,000. The additional cost for the surface powerhouse arrangement is predominantly associated with the longer pen- stocks and the steel linings required. Although construction cost esti- mates for a 1020 MvJ plant would be somewhat higher, the overall conclusion favoring the underground location would not change. The underground powerhouse arrangement is also better suited to the severe winder conditions in Alaska, is less affecteci ~Y river. flood floes in summer, and is aesthetically less obtrusive~ l)is arrangement has therefore.been adopted for further development. (b} Comparison of Alternative Locations Preliminary studies were undertaken during the development of conceptual project layouts at Watana to investigate both right and. left bank loc·ations for power facilities. The configuration of the site is such that left bank locations generally required longer penstock and/or tailrace tunnels and were therefore strictly more expensive. 9-28 I I I I I ,I I I I ' I I I I :I .. ill ~ I I I I I J I "'· ....... '- ' I -,.; I I· ....... I 1 I. I I I ·. ' t I ~-- 'I I~. The location of the left bank was also not favored because of indications that the underground facilities would be located in relatively poor quality rock. The underground powerhouse was therefore located on the right bank such that the major openings lay between the two major shear features (*'The Fi ns 11 and the "Fi ngerbuster••) .. (c) Underground Openings Cost estimates have been based on assumptions of full concrete lining of the penstocks and tai 1 race tunnels. The 1 atter is a conservative assump- tion for preliminary design; in practice, a large proportion of the tail- race tunnels could be unlined, depending on the· actual rock quality encountered. The mini mum center-to-center spacing of rock tunne 1 s and caverns has been assumed for layout studies to be 2.5 times the size of the larger excavation. (d) Selection of Turbines The sel~ction of unit type is governed by the available head and flow, an~ economic considerations. For the design head and specific speed, Francis type turbines have been selected; these have a reasonably flat load-efficiency curve over a range from about 50 percent to 115 percent rated output, and a peak efficiency of about 92 percent. The number and rating of 0 individual units is discussed in detail in Section 9.6. The final arrangement selected is six units of 170 MW rated at minimum reservoir level (from reservoir simulation studies) in the peak deman9 month (December) at fu 11 gate. Th.e unit best efficiency output at rated head {680 feet) is 181 MW. (e) Transformers The selection of transformer type, size, location and step-up rating is descr1bed in Section 12.18 and summarized below: -Single phase tran·sformers are required because of transport limitations on Alaskan roads and railways; -Direct transformation from 15 kV to 345 kV is preferred for overall system transient stability; -An underground transformer gallery has been selected for minimum total cost of transformers, cables, but, and transformer losses; and - A grouped arrangement of three single ~hase transformers for each two units has been selected to rer~\;ce the physical size of the transformer gallery and to provide a tran.iformer spacing comparable with the unit spacing. 9-29 (f) Power Intake and Water Passages The power intake and approach channel are significant items in the cost of the overall power facilities arrangement. The size of the intake is controlled by the number and minimum spacing between the penstocks!~ which in turn is dictated by geotechnical considerations (Sections 9.2 and 9.3) .. The preferred penstock arrangement comprises six inrlividual penstocks, one for each turbine~ With this arrangement, no inlet valve is required in the powerhouse since penstock dewatering can be performed by using the control gate at the intake. An alternative arrangement with three penstocks was considered in detai 1 to assess any possible advantages. This scheme would require a bifurcation and two inlet valves on each penstock and extra space in the powerhouse to accommodate the inlet valves. Estimates of relative cost differences are summarized below: Item Intake (increment) Penstocks (increments) Bi fur·c at ions Valves Power·house Capitalized Value of Ext~a Head Loss Total Cost Difference 6 Penstocks 3 0 ($000) Penstocks -20.0 -3.0 + 3.0 + 4.0 + 8.0 + 6.0 -2.0 Despite a marginal saving of $2 million (or less than 2%) in favor of three penstocks, in a total estimated cost of $120 million, the arrangement of six individual penstocks has been retained. This arrangement provides improved flexibility and security of operation. The preliminary design of the power facilities involves two tailrace tunne 1 s 1 eadi ng from a common surge chamber. An alternative arrangement with a single tailrace tunnel was also considered, but no significant cost saving was apparent. Optimization studies on all water passages were carried out to determine the minimum total cost of initial construction plus the capitalized value of anticipated energy losses caused by conduit friction~ bends and changes of section. For the penstock optimization, the construction costs of the intake and approach channe 1 were inc 1 uded, as a function of the penstock diameter and spacing. Similarly, in the optimization studies for the tailrace tunne 1 s, the costs of the surge chamber were included~ as a function of tailrace tunne 1 diameter. (g) Environmental Constraints Apart from_the potential nitrogen supersaturation problem discussed in Section 9.10, the major environmental constraints on the design of the power facilities are: 9-30 I ·'1;. ~ - I I .j I I """' • "';;'_ .. ' I .... - I I "'-• I I I I ... -~ ' a I ' I I I I I ·- 1 I ' ' t. I t I I I ' t I ' I I .. Contra l of downstream river temperature's; and .. Control of downstream' flows. The intake design has -been modified to enable power plant flows to be drawn from ~he reservoir at four different levels through the anticipated range of dra\'idown in order to control the downstream river temperatures within acceptable limits. Guaranteed minimum flows at Gold Creek during the critical summer months have been studied to mitigate the project impacts on salmon spawning downstream of Devil Canyon. These minimum flows represent a constraint on the reservoir operation, and influence the computation of average and firm energy from the Susi tna development. These studies are discussed in detai 1 in Section 15. In average to wet years, the Watana development will be capable of operating as a daily peaking plant for load following. The actual extent of daily peaking will be dictated by unit availability, system demand, unit generating costs, system stability, etc., (as described in Section 15). Predicted downstream water level fluctuation caused 'By daily peaking at Watan& is within acceptable limits .. 9.12 -S"election of Watana General Arrangement Preliminary alternative arrangements of the Watana Project were developed and subjected to a series of review and screening processes. The layouts selected from each screening process were developed in greater detail prior to the next review, and where necessary, additional layouts were prepared combining the features of two or more of the alternatives. Assumptions and criteria were evaluated at each stage and additional data incorporated as ne~essary. The selection process followed the general selection methodology established for the Susitna project, and is outlined below. (a) Selection Methodology The determination of the project genera 1 arrangement at ~4atana was undertaken in three distinct review stages: preliminary, intermediate., and final. · (i) Preliminary Review This comprised four steps: -Step 1: Assemble avai 1 able data; Determine design criteria; and Establish evaluation criteria. -Step 2: Develop preliminary layouts based on the above data and design criteria including all plausible alternatives for the constituent facilities and structures. -Step-3: Review all layouts on the basis bf technical feasibility, readily apparent cost differences, safety, and environ- ment a 1 imp act. ,, ' ,-;.'; (i i) (iii) ~ Step 4: Select those layouts that can be identified as most favoriible~~Pased on the evaluation criteria determined under Step lc, taking into account the preliminary nature af the work at this stage. · Intermediate Review This involved a series of 5 steps: -Step 1: Review all data, incorporating additional data from other work tasks. Review and expand design criteria to a greater 1 eve1 of detail. Revie~ evaluation criteria and modify, if necessary~ -Step 2: Revise selected layouts on basis of the revised criteria and additional data. Prepare plans and principal sections of layouts. -Step 3: Prepare quantity estimates for major structures based on drawings prepared under Step 2. -Step 4: -Step 5: Develop a preliminary construction schedule to evaluate whether or not the selected layout will allow completion of the project within the required time frame. Prepare a preliminary contractor's type estimate to determine the overall cost of each scheme. · Review all layouts on the basis of technical feasibility, cost impact of possible unknown condi ti ens and uncertain- ty of assumptions, safety, and environmental impact. Se 1 ect the two most f avorab 1 e 1 ayouts based on the evaluation criteria determined under Step 1~ Fi na 1 Review -Step 1: -Step 2: Assemble and review any additional data from other work tasks. Revise design criteria in accordance with additional available data. Finalize overall evaluation criteria. Revise or further develop the two layouts on the basis of conclusions from Step 1. Determine over a 11 dimensions of structures' water passages' gates' and other key ; terns ·• ' 9-32 I I I J I I I ' ' I t I I I J; -~ f ' ' I "-* II I - I ~..--.. ·t ~ .. I I I ' ' I ~ t I I I I J• ' t I I. I -Step 3: Prepare quantity take-offs for all major structures. Review cost components within a preliminary contractor's type estimate using the most recent data-and criteria, and develop a construction.schedule. Determine overall direct cost of schemes. -Step 4: Review all layouts on th~ basis of practicability, technical feasibility, cost, impa:ct of possible unknown conditions, safety, and environm~ntal impact. -Step 5: Select the final layout on the basis of the evaluation criteria developed under Step 1. (b) Design Data and Criteria As discussed above, the review process included assembling of relevant de- sign data, establishing preliminary design Criteria, and expanding andre- fining these data during the intermediate and final revie\vs of the project arrangement. The design data and design cri ter i a which eva 1 ved through the final review is presented in Table 9.5. Data and criteria developed during the preliminary and intermediate review stages are given in Appendix D for reference. (c) Evaluation Criteria The various layouts were evaluated at each stage of the review process on the basis-of the criteria summarized in Table 9.6. The criteria listed in Table 9.6 illustrate the progressively more detailed evaluation process 1 eadi ng to the final se 1 ected arrangement. 9.13 -Preliminary Review The development selection studies described in Section 8 involved comp~risons of hydroelectric schemes at a number of sites on the Susitna River. For these comparisons a preliminary conceptual design was·developed for the Watana project known as the "DSR Sci1emes 11 • ·· Eight further 1 ayouts were subsequently prepared and ex ami ned for the Watana project during this preliminary review process, in addition to the DSR scheme. These eight layouts are shown in schematic form on Plate 13. Alternative 1 of these·layouts was that recommended for further study in the Development Selection Report. This section describes the preliminary review Ur1dertaken of alternative Watana 1 aym·?~s. (a) Basis of Comparison of Alternatives Although it was recognized that provision would have to be made for downstream r·e leases of water during filling of the reservoir and for emergency reservoir drawdown, these features were not incorporated in these 9-33 preliminary layouts. These facilities would either be inter-connected with the diversion tunnels or be provided for separately. Since the system selected would be similar for all layouts with minimal cost differences and little impact on other structures, it was decided to exclude these faci 1 i ties from over a 11 assessment at this ear 1 y stage. Ongoing geotechnical explorations had identified the two major shear zones c~ossing the Susitna River and running roughly parallel in the northwest direction. These zones enc'lose a stretch of watercourse approximately 4500 feet in length (see Section 9.2). Preliminary evaluation of the existing ge.ological data ·indicated that the fracture materials and infill within the actual shear zones would be unable to support standard tunneling methJds and would be inadequate for founding cf massive concrete structures. The originally proposed dam axis was located between these shear zones, and as no apparent major advantage cppeared to bn gained from large changes in the dam location, layouts generally were kept within the confines of these bounding zones. An earth and rockfi 11 dam as described in Section 9.8 was used as the basis for all layouts. The downstream slope of the dam was assumed as 2H~1V in all alternatives, upstream slopes varying between 2.5H:1V and 2.25H:lV were examined ·;n order to determine the ifluence of variance in the dam slope on the congestion of the layout. In all these preliminary arrangements, except that prepared for the DSR, cofferdams were incorporated within the body of the main dam. Floods greater than the routed 1;10,000 year spillway flood and up to the probable maximum flood were assumed to be passed by surcharging the spillways except in cases where an unlined cascade or stilling basin type spillway served as the sole discharge facility. In such instances, under 1 arge surcharges, these spi 11 ways wou 1 d not act as efficient energy di.ssipators but would be drowned out, acting as steep open channels \'lith the possibility of their total destruction. In order to avoid such an occurrence the design flood was considered as the routed probable maximum flood. On the basis of information existing at the time of the preliminary review, it appeared that an undergrpund powerhouse cou 1 d be located on either side of the rive~. A surface powerhouse on the right bank appeared feasible but was precluded from the left bank by the close proximity Of the downstream toe of the dam and the adjacent broad shear zone. Locating the powerhouse- further downstream would require tunneling across the shear zone, whtch would be expensive, and excavating a talus slope. Furthermore~ it was found that a left bank surface powerhouse would either interfere with a left bank spillway or would be directly impacted by discharges from a right bank spillway. (b) Description of Alternative (i) Preliminary DSR Scheme The preliminary I}DSR scheme as shown on Plate 3 has a dam axis location similar to that originally proposed by the COE, and a right. bank double stilling basin spillway. The spillway follows the · 9-34. t I -.. -' I _ .. I I ' ' ' I .... t I I I ' I I ' I I I I ~-- , 'I {1_ ' I ' ' I ' f I I I t ' I I I . , I shortest line to the river avoiding interference with the dam and -discharges downstream, almost parallel to the flow~ into the center of the river. A substantial amount of excavation is required for the chute and stilling 'basins, although most of this material could probably be used in the dam~ A large volume of-concrete is also required for this type of spillway, however, and the system would be very costly. The maximum head dissipated within each stilling basin is approximately 450 feet, within world experience, and cavitation and erosion of the chute and basins should not be a problem if the structures are properly designed. Extensive erosion downstream would not be expected. The diver·sion fo11o~rs the shortest route, cutting the bend of the river on the right bank, and has-inlet portals as far upstream as possible-without having to tunnel through 11 The Fi nsu. It is possi b 1 e that the underground powerhouse is in the area of "The Fi ngerbuster", but it caul d be located upstr.eam almost as far as the system of drain holes and galleries just downstream of the main dam grout curtain. (ii) Alternative 1 This alternative is that recommended for further study in the Development Selection Report and is similar to the preliminary DSR 1 ayout, except that the right side of the dam has been rotated clockwise, the axis relocated upstream and the spillway changed to a chute and flip bucket. The revised dam al~gnment resulted in a slight reduction in total dam volume compared to the DSR alterna- tive. A localized downstream curve was introduced in the dam close to the right abutment in order to reduce the length of the spillway. The alignment of the spillway is almost parallel to the downstream sec~ion of the river and it discharges into a pre-excavated plunge pool. in the river approximately 800 feet downstream from the flip bucket~ This type of spillway should be considerably less costly than one incorporating a stilling basin, provided that excessive excavation of bedrock within the plunge pool area is not required. Careful design of the bucket will be required however, to prevent excessive erosion downstream causing undermining of the valley sides and/or build up of materia 1 downstream which could cause elevation of the t ai lwater 1 eve 1 s ~ (iii) Alternatives 2 through 20 Alternative 2 consists of a left bank cascade spillway with the main dam axis curving downstream at the abutments. The cascade spillway would require an extremely large volume of excavation but it is probable that most of this material, with carefule scheduling~ could be used in the dam. The excavation would cross "The Fingerbuster" and extensive dental concrete would be required. In the upstream portion of the spillway~ velocities would be relatively high because of the narrow configuration of the channel and erosion could take place in this area in proximity to the dam. This discharge from the spillway enters the river perpendicular to the general flow but velocities would be rei atively low and should not cause substanti a1 erosion prol1lt:p1s. The powerhouse is in the most suitable location for a surtace ~alternative where the bedrock is close to the surface and the overall slope is approximately 2H:lV • " Alternative 2A is similar to Alternative 2 except that the upper end of the channel is divided and separate control structures are. provitied ... This division would allow the use of one structur·e or upstream channe 1 whi 1 e maintenance or remedi a 1 ~1ork is being performed on the other. Alternative. 2B is similar to Alternative 2 except that the cascade spillway is replaced by a ~ouble stilling basin type structure. This spillway is somewhat longer than the sinli 1 ar type of structure on the right bank in A 1 tern at ive 1. However, the s 1 ope of the ground is less than the· rather steep right bank and it may be easier to construct, a factor which may partly mitigate the cost of the longer· structure. The discharge is at a sharp angle to the river and being more concentrated than the cascade could cause erosion of the opposite bank. · • Alternative 2C is a derivative of 28 with a similar arrangement, except that the double stilling basin spillway is reduced in size and augmented by an addi ti ona1 emergency spillway in the form of an inclined~ unlined rock channel~ Under this arrangement the concrete spillway acts as the main spillway, passing the 1:10,000 year design flood with greater flows passed down the un 1 i ned channe 1 which is closed.at its upstream end by an erodable fuse plug. The problems of erosion of the opposite bank still remain, although these could be overcome by excavation and/or slope protection. Erosion of the chute would be extreme for significant flm'ls,. although it is highly unlikely that this emergency spillway would ever be used. Alternative 20 replaces the cascade of Alternative. 2 with a lined chute and flip bucket. The comments relative to the flip bucket are the same as for Alternative 1 except that the left bank location in this instance requires a lange~ chute, partly offset by lower construction costs bee ause of the flatter slope, and the flip bucket discharges into the river at an angle which may.cause erosion of the opposite bank. The underg~·ound powerhouse is located on the right bank, an arrangement which provides an ov~~rall reduction ·of the length of the water passages. (iv) Alternative 3 This arrangement h~s a dam axis location slightly upstream from Alternative 2, but retains the downstream curve at the abutments. The main spillway is an unlined rock cascade on the left bank which passes the design flood. Discharges beyond the 1:10,000 year flood would be discharged through the auxiliary concrete-1 i ned chute and flip bucket spillway on the right bank. A gated contra 1 structure is provided for this auxiliary spillway which gives it the flexibi 1- ity to be used as a backup if maintenance should b·e required on the main spillway.. Erosion of the cascade may be a prob 1 em, as mentioned previously, but erosion downstream should be a less important consideration because of the low unit discharge and the infrequent · operat 1 on of the spi 11way. The diversion tunne 1 s are situated in the right abutment~ as with p·revi ous arrangements, and are of simi 1 ar cost for all these alternatives. 9-36 I a ~ t) ' ' I t ' i "'" ·a < ~ I I I .I t I ' .,.--:;,. ' I I ;. W> ,.. ., ' ' ' t I ' I .. I I ...... I I I ' t . t •• (v) Alternative 4 -This alternative involves rotating the axis of the main dam so that the left abutment is relocated approximate1y 1000 feet downstream from its Alternative 2 location. The relocation results in a -reduction in the overall dam quantities but would require siting the impervious core of the dam directly over the "Fi ngerbuster" shear zone at maxi mum dam height. The 1 eft bank spillway, consisting of chute and flip bucket, is reduced in length compared to other left bank locations, as are the power facility water passages. The diversion tunnels are situated on the left bank; there is no advan- tage to a right bank location~ since the tunnels are of similar length owing to the overall downstream relocation of the dam. Spillways __ and power facilities would also be lengthened by a right bank location with this dam configuration. (c) Selection of Schemes for Further Study A basic consideration ·during design developm-ent was that the main dam core should not cross the major shear zones because of the obvious problems ~Jith treatment of the foundation~ Accordingly, there is very little scope for realigning the main dam apart from a slight rotation to place it more at right angles to the river. Location of the spillway on the right bank results in a shorter distance to the river anc allows discharges almost parallel to the general direction of river flow. The double stilling basin arrangement of the preliminary DSR scheme would be extremely expensive, particularly if it must be designed to pass the probable maximum flood. An alternative such as 2C wou.ld reduce the magnitude of design flood to be passed by the spi 11way but would only be acceptable if an emergency spillway with a high degree of operational predictability could be constructed. A flip bucket spillway on the right bank, discharging directly down the river, would appear to be an economic arrangement~· although some scour might occur in the plunge pool area.. A cascade spillway on the 1 eft bank caul d be an acceptable so 1 ut:i~11 providing most of the excavated materia 1 could be used in the dam, and adequate rock conditions exist. The length of diversion tunne 1 s can be decreased if they are 1 ocated on the right bank. In addition, the tunnels would be accessible by a preliminary access road from the north, which is the most likely route. This location would also avoid the area of 11 The Fingerbuster" and the steep cli.ffs which would be encountered on the left side close to the downstream dam toe .. The underground configuration assumed for the powerhouse in the.se prelimi- nary studies allows for location on either side of the river with a minimum of interference with the surface structures. Four of the preceding layouts., or variations of them, were selected for further· study: 0 9-37 (i) A variation of the preliminary DSR scheme, but with a single stilling basin main spillway on the right bank, a rock channe 1 and fuse plug emergency spillway, a left bank underground powarhouse and a right bank diversion scheme; (ii) Alternative 2 with a right bank flip bucket spillway, an underground powerhouse on the left bank, and right bank diversion; · (iii) A variation of Alternative 2 with a reduced capacity main spillway and a right bank rock channel with fuse plug serving as an emergency spillway; and (iv) Alternative 4 with a :left bank rock cascade spillway, a right bank underground powerhouse, and a right bank diversion. 9.14 -Intermediate Review For the intermediate review process, the four schemes selected as a result of the preliminary review were examined in more detail and modified. A description of each of the schemes is given below and shown on Plates 14 through 19. The general locatio~s of the upstream and downstream shear zones shown on these plates are approximate, and have been refined onthe basts· of subsequent field investigations for the design_studies described in Section 12. (a) Description of Alternative Schemes The four schemes at"e snown on Plates 14 through 19: (i) Scheme WPl (Plates 14 and 15) This scheme is a refinement of Alternative 1. The upstream slope of the dam is flattened from 2.5:1 to 2.75:1. This conservative approach was adopted to provide an assessment of the possible impacts on project layout of conceivable meansures which prove necessary in dealing with severe earthquake design conditions. Uncerta1nty with regard to the nature of river alluvium also led to the location of the cofferdams outside the limits of the main dam embankment. As a result of these conditio~s, the intake portals of the diversion tunnels on the right bank are also moved upstream from "The Fins". A chute spillway with a flip bucket is located on the right bank together with the underground powerhouse. (ii) Scheme WP2 (Plates 16 and 17) This scheme is derived from the DSR layout. The main dam and diver- sion facilities are similar to Scheme WPl except that the downstream cofferdam is relocated further downstream from the spillway outlet, and the diversion tunnels are correspondingly extended. The main spillway is located on the right bank, but the two stilling basins of the preliminary DSR scheme are combined into a single stilling basin at the river 1 evel. .An emergency spillway is a 1 so located on the right bank, and consists of a. channe 1 excavated in rock, dis- charging downstream from the area of the relict channel. The 9-38 . ) I ' ,, I ' I t ' ' 'I .... I :t a· I ' ' ,, ' ,, I ' ' ' I t I I ' t tl\ -· ' ' ' I I t ' I (b) channel is closed at its upstream end by a compacted earthfi fl fuse plug and is capable of discharging the flow differential between the probable maximum flood and the 1:10,000-year design flood of the main spillway 8 The underground powerhouse is 1 oc ated on the left bank. (iii) Scheme WP3 This scheme is similar to Scheme WPl in all respects, except that an emergency spillway is added, consisting of right bank rock "channe 1 and fuse plug (see Plate 16). {iv) Scheme WP4 (Plates 18 and 19) The dam location and geometry for Schemo WP4 are similar to that for the other schemes. The diversion is on the right bank and discharges downstream from the powerhouse tailrace outlet. A rock cascade spi 11way is 1 oc ated on the 1 eft bank and is served by two separate contra 1 structures with downstream sti 11 i ng basins. The underground powerhouse is located on the right bank. Comparison of Schemes The main dam is in the same location and has the same configuration for each of the four layouts considered. The cofferdams have been locat!=d outside the limits of the main dam in order to allow more extensive excavation of the alluvial material and to ensure a sound rock foundation beneath the complete area of the dam. The overall design of the dam is conservative, and it was recognized during the evaluation that savings in both fill and excavation costs can probably be made after more detailed study. · The diversion t unnE! 1 s are l oc at ed on the right bank. The upstream flatten- i ng of the dam slope necessitates the location of the diversion inlets up- stream from "The Fins" shear zone which wi 11 require extensi v·e excava);i on and support where the tunnels pass through this extrem1.~ ly poor rock zone and could cause delays in the construction schedule. A low-lying area exists on the right bank above the area of the relict channel, and t~is is closed by an approximately 50-foot high saddle dam. A slurry trench cutoff will be combined with grouting to seal the .200-f,oot depth of pervious material infilling this channel. A summary of capita 1 cost estimates for the four alternative schemes is given in Table 9.7. The results of this intermediate--analysis indicate that the chute spillway with flip bucket of Scheme WPl is the least costly spillway alternative. 9-39 (c) The scheme has the additional advantage of relative1y simple operating characteristics. The control structure has provision for surcharging to pass the design flood. The probable maximum flood can be passed by additional surcharging up to the crest level of the dam. In Scheme WP3 a similar spillway is provided, except tha the control structure is reduced in size and discharges above the routed design flood are passed through the rock channel emergency spillway. The arrangement in Scheme WPl does not provide a backup faci] i ty to the main spillway, so that if repairs caused by excessive plunge pool erosion or damage to the structm·e itself require remova 1 of the sp,i llway from service for any 1 ength of time, no alternative discharge facility would be available. The additional spillway of Seheme WP3 would permit emergency discharge if it were absolutely required under extreme circumstances. · The stilling basin spillway (Scheme WP2) would reduce the potential for ex- tensive erosion downstream, but high velocities in the lower part of the chute could cause cavitation even with the provision for aeration of the discharge. This type of spillway wou'l d be very costly, as can be seen from Table 9.7. The feasibility of the,rock cascade spillway is entirely dependent on the qua 1 i ty of the rock, which dictates the amount of treatment required for the rock surface and cdso the proportion of the excavated material which can be used in the d,:tm. For determining the capital cost of Scheme \4P4, conservative assumr:cions were made regarding surface treatment and the portion of materi a1 that woL·ld have to be wasted. The diversion tunnels are located on the right bank for all alternatives ex ami ned in the intermediate review. For Scheme WP2, the downstream portals must be located downstream from the stilling oasin, resulting in an increase of approximately 800 feet in the length of the tunnels. The left bank location of the powerhouse requires its placement close to a suspected shear zone, with the tailrace tunnels-passing through this shear zone to reach the river. A 1 anger access tunne 1 is a 1 so requ·i red, together with an additional 1,000 feet in the length of the tailrace. The left-side location is remote from the main access road, which will probably be on the. north side of the river, as will the transmission corridor. Selection of Schemes for Further Study Examination of the technical and economic aspects of Scheme WPl through WP4 indicates there is 1 itt l e scope for adjustment of the dam axis owing to the confinement imposed by the upstream and downstream shear zones. In addition, passage of the diversion tunnels through the upstream shear zone could result in significant delays in construction and additional cost. From a comparison of costs in Table 9.7, it can ba seen that the flip bucket type spillway is the most economi ca 1, but because of the potential for erosion under extensive operation it is undesirable to use it as the on·ly discharge facility. A mid-level release will be required for emer--- gency drawdown of the reservoir, and use of this release as the first-stage service spillway with the flip bucket as a backup facility would combine flexibility and safety of oper~tion with reasonable cost. The emergency rock channel spillway would be retained for di schiirge of flows above the route 1:10,000 year flood. 9-40 I ' ' ' ' ' ! ' ' I ' I • I ., i I I t I t I I I ' I • , .• I ' t ' ' ' I t I ' I ' I t The stilling basin spi1lway is very costly and the operating head of 800 feet is beyond precedent experi.ence. Erosion downstrceam should not be a pt"Oblem but cavitation of the chute could occur. Scheme \vP2 was therefote elimi·nated from further ·consideration .. The cascade spillway was also not favored for technical and economic reasons. However, this arrangement does have an advantage in that it pro- vides a ·means. of preventing nitrogen supersaturation in the downstream discharges from the project which could be harmful to the fish population, as discussed in Section 9.10. A cascade configuration would reduce the di sso 1 ved nitrogen content ... and hence~ this a 1 ternati ve was ~et ai ned for further evaluation. The c~tJacity of the cascade was r;educed and the emergency rock channe 1 spi'llway was included to take the extreme floods. The results of the intermediate review indicated that the following compon- ents should be incorpo~"ated into any scheme carried forward for f~nal re- view: ... TvJo diversion tunnels located on the r·ight bank of the r~iver; ·· An underground powerhouse also 1 oc ated on the right bank; An emergency spillway~ comprising a rock channel excavated on the right bank and discharging well downstream from the right abutment. The channel is sea1ed by an erodible fuse plug of impervious material designed to fail i.f overtopped by_the reservoir-; and - A compacted earthfi 11 and rockfi 11 dam situated b-etween the two major shear zones which tiaverse the project site. As discussed above, two specific alternative methods exist with respect to routing of the spillway design flood and mi nimi zing the adverse effects of nitrogen supersaturation on the downstream fish populatior~. These alterna- tives are: - A chute spillway with flip bucket on the right bank to pass the spi 1lway design flood, with a mid-level release system designed to operate for floods with a frequency of up to about 1:50 years; or - A cascade spillway on the 1 eft bank. Accordingly, two schemes were developed for fruther evaluation as part of the fi na i review process. These schemes are described separate 1 y in the paragraphs be 1 ow. ~e15 -Final Review The two schemes considered in the final review process were essentially deviations of Schemes WP3 and WP4. (a) Scheme WP 3 A (Plate 20) This scheme is a modified version of Scheme WP 1 described in Section 9.14 with an emergency spillway as included in Scheme WP 3. Because of 9~41 schedu 1 i ng and cost considerations, 1 t is extreme 1 y import ant to rnaint ai n the diversion tunnels downstream from "The Fins .. n It ,is also important to keep the dam axis as far upstream as possible to avoid congestion of the downstr,eam structures. Far these reasons, _the in 1 et portals to the diversion tunn~ls were located in the sound bedrock forming the downstream boundar,y of the '~Fingerbuster. u The upstream cofferdam and main dam are maintained in the upstr-eam locations as shown on Plate 14. As mentioned previously, additional criteria. have necessitated modifications in the spillway configuratton, and low-level and emergency drawdown outlets have been introduced. The main modifications to the scheme are ~s follows: ( i) Main Dam Further investigation of preliminary design studies and review of world practice sugg.ests that an upstream slope of 2.4H :IV-would be acceptable for the rock shell. Adoption of this slope results not only in a reduction in dam fill volume but also in a reduction in the base width of the dam which permits the main project components to be located between the major shear zones. The downstream slope of the dam is retained .as 2H:1V. The coffer- dams remain outside the limits of the dam in order to allow complete excavation of the riverbed alluvium. (i i) Diversion In the intermediate revlew arrangements, diversion tunnels passed through the broad structure of 11 The Fins,11 an intensely sheared area of breccia, gouge, and i nfill s. Tunne 1 i ng of this materia 1 wou 1 d be difficult, and might even require excavation in open cut from the surface. High cost would be involved, but more important would be the time taken for construction in this area and the possibility of unexpected delays. For this reason, the inlet portals have been relocated downstream from this zone with the tunnels located closer to the river and crossing the. main system of joi nt.i ng at approxi ... mately 45°. This arrangement allows for shorter tunnels with a more favorable orientation of the in let and out 1 et porta 1 s with respect to the directions of river flow and di'version inflow and outflow at the portals. A separate low-1 eve 1 in 1 et and concrete-1 i ned tunnel is provided leading from the reservoir at approximate elevation 1550 feet to downstream from the diversion plug where it merges with the .diversion tunnel closest to the river. This low-level tunnel is designed to pi::\S flows in excess of 2000 cfs acting as a low-level release durin~ reservoir filling. It will also pass up to 10~000 cfs under 500-foot head to allow emergency draining of the reservoir as discussed in Section 9 .. 9. 9-42 I I ' I I I I I I I t I t t - t t ~ t I ' ., I t I I ' ' I I ,. ' I ., I t I I (iii) {iv) Initial closure is made by lowering the gates to the tunnel located c 1 osest to the river and constructing a concr:ete c 1 a sure p 1 ug in the tunnel at the location of the grout curtain Ulnderlying the core of the main dam. On completion of the plug~ thEi 1ow-leve1 release is opened and controlled d1 scharges are passed downstream. The gates within the second port a 1 are lowered and a mass concrete c 1 osure p 1 ug constructed on a 1 i ne with the grout cur·tai n. Afi>-=r c 1 osure of the gates, filling of the reservoir can commence. Emergency and Out let Release Faci 11~;; es As a pro vision for drawing down the reservoir in case of emergency, a mid-level release is provtded~ The intake to these facilities is located at deptch adjacent to the power facilities intake structures. Flows wi 11 then be passed downstr·eam thr"ough a hi red shaft and tunne 1, existing beneath the downstream end of the main spi 11 way flip bucket, as ·described be 1 ow. In order to overcome potential nitrogen supersaturation problems, Scheme WP 3 A also incorporates a system of fixed core valves into the downstream end of the emergency release facility. The valves were sized to discharge in conjunction with the powerhouse operating at 7S percent capacity, flows up to the equivalent routed 50-year flood. Six valves are required, located on branches off a steel manifold and protected by i ndi vi dua 1 upstream c 1 osure gates. The v a 1 ves are partly incorporated into the mass concrete block forming the flip bucket of the main spillway. The rock downstream is protected from erosion by a concrete facing slab anchored back to the sound bedrock. Spillways As discussed in Section 9.10 above, the designed operation of the main spillway facilities was arranged to limit discharges of potentially nitrogen supersaturated water from Watana to flows having an equivalent return period of not less than 1:50 years .. The main chute spillway and flip bucket discharge into an excavated plunge pool in the downstream river bed. Releases are controlled by a three-gated agee structure 1 oc ated adjacent to the out 1 et release and power intake structures just upstream from the dam centerline. The design discharge is approximately 80,000 cfs corresponding to the routed 1:10,000-year flood (120,000 cfs) reduced by the 40,000 cfs flet11.s attributable to outlet release and power faciliti.es discharges. The plunge pool is formed by excavating the alluvial river deposits to bedrock; and as this approaches the limits of the calculated maximum scour hole, it is not anticipated that~ given the infrequent discharges, significant downstream erosion wi'll occur. The emergency spillway is provided by means of a channel excavated in rock on th~ right bank~ discharging \'/ell downstream from the right abutment in the direction of Tsusena Creek. The channe 1 is sealed by an erodible fuse plug of impervious material designed to 9-43 fail if overtopped by the reservoir; although some preliminary excavation may be. necessary. The crest level of the plug wi 11 be set at elevation 2230 fe~t, well below that of the main dam. Th~ channel will be capable of passing the exce·ss discharg~ of floods greater than the l:IO,OOQ .... year flood up to the probable maximum .flood·of 235,000 cfs. · (v) Power Facflities The power intake is set slightly upstream from the dam ·centerline. deep within·sound bedrock. at the downstream end of the approach channeL. The intake consists of six units with provi sian in each unit for drawing flows from a. variety of depths covering the complete drawdown range Of the reservoiro This facility also . provides for drawing water from the different temperature strata within ·the upper part of t.he reservoir~ and thus regulating the temperature of the downstream di scharg~s close to the natural temperatures of the river. For this p:~el imi nary conceptual arrangement, flow withdrawals from difterent levels is affected by a series of upstream verti.ca1 shutters moving in a single set of guides and operated to form openings at the r~qui red 1 eve 1. Downstream from these·shutters each unit has a pa.ir of wheel-mounted closure gates which will i-solate the individual penstocks. The six penstocks are 18-foot-di ameter, concrete-lined tunnels inclined at 55° immediately downstream from the intake to a nearby .horizontal portion leading to the powerhouse. This horizontal por- tion is steel-lined for 150 feet upstream from the turbine units to extend the seepage path to the powerhouse and contain the flow with- in the fractured rock area caused by b 1 asti ng in the adjacent power- house cavern. The six 170 MW turbine/generator units are housed within the major powerhouse cavern and are setviced by an overhead crane which runs the 1 ength of the· powerhouse and into the service area adjacent to the units. Switchgear, area maintenance room and offices are 1ocat~ ed within the main cavern, with the transformers situated downstream in a separate gallery excavated above the tailrace tunnels. Six inclined tunnels carry the connecting bus ducts from the main power hall to the transformer gallery upstream. A vertical elevator and vent shaft run from the power cavern to the main office. bui 1d1ng and control room located at the surface.. Vertical cable shafts., one for each pair of transformers, connect the transformer gallery to the switchyard directly overhead. Downstream from the transformer · ga.llery$ the unde~lying draft tube tunnels merge fnto two sur.ge chambers, one chamber for three draft tubes, which also house th'e draft tube gates for isolating the units from the tailr.ace .. The gates are operated by an overhead tt"aveling gantry located in the--- upper part of each of the surge chambers. Emerging· from the ends of the chambers, two concrete-lined, low-pressure tailra<:e tunne.ls carry the discharges to the river. Because of space restrictions at the river, one of these tunnels has been merged with the downstream end of th.e diversion tunnel. The other tunnel emerges in a separate portal with prQvi si on for the i nsta llati on of bulkhead ga.tes. I~ .. ' : i ' ' I I I I I I I I .I t. t I I I i t ·~ 'I I t I I I I I I. t ' I ' I I I I I t < .-:- (b) The qrientation of water passages and underground caverns ts such as to avoid as far as possible the ma.i n, alignment of the excavations running paralle-l to _the major Joint sets as described in Section 9.3~ . (vi) Access Access is assumed to be from the north (right) side of the river~ Permanent access to_ structures close to the river is by a -road along the right downstream river ba.nk and then· vi'a'··-a tunnel passing through the concrete formin~ the flip bucket. A· tunnel fr6m this point to the power cavern provides for vehicular access.. A secondary access road across the crest of the dam passes dawn the 1 eft b an_k of the v a11 ey and across the 1 ower part of the dam. Scheme WP 4A (Plate 21) This scheme is similar in most respects to Scheme WP 3A previously dis~ cussed~ except for the spillway arrangements. (i) Main Dam The main dam axis is similar to that of Scheme -wP 3A, except for a s 1 i ght downstream rotation at the 1 eft abutment at the ·spillway con- trol structures. (ii) Diversion The diversion and l0\'1 head releases are exactly similar for the two schemes. (iii) Emergency and Outlet ~eleas~. Facilities The emergency drawdown re 1 ease faci 1 i ty is separated from the main sp'i llway for this scheme. The emerging re 1 ase consists of a low-level gated outlet structure located upstream, discharging up to 30,000 cfs well into the river through a concrete-lined~ free-flow tunnel with a ski jump flip bucket. This facility may also be· operated as an auxiliary outlet to augment the main left bank spi 1lway. ( i v) S pi 11 ways The main left bank spillway is capable of passing a design flow equivalent to the 1:10,000-year flood through a series of 50-foot drops into sha 11 ow pre-excavated p 1 unge poo 1 s.. The emergency spillway is designed to operate during floods of greater magnitude up to and including the PMF. Main spillway discharges are coQtrolled by a broad multi-gated control structure discharging into a shallow stilling basin. The feasibility of this arrangement is governed by the qua·lity of the r·ock i.n the area.~ requiring both durabi l).ty to withstand erosion caused by spillway flows, and a high percentage of sound rockfill material that can be used f~om the ex{:avation directly in the rna1n dam. 9-4b '.) I I I' I I l {c) !.; On the: basis -Of the .sit·e information deve.foped concurrent Ty-with the gen~ral arr·angement studies, it became app.arent that the major shear zone::_ known to exist in the left bank area extended ·further - dO\·Jnsl're.am than initial st·udies have indicated. The cascade spillway channe 1 was therer:ore 1 engthened to avoid the shear area at the lower end of the cascade. The arrangement shown on Plate 21 for Scheme WP 4A does not reflect this relocati'on, which would increase the overall cost of the scheme. The_ emergency spillway consisting of rock channe 1 and fuse plug is simi 1 ar to that o.f the right bank spillway scheme. (v} Power Facilities The power facilities are similar to those in Scheme WP3A. -- Evaluation of Final Alternative Schemes An evaluation of the dissimilar features for each arrangement (the main spillways and the discharge. arrangements at the downstream end of the outlets) indicates a saving in capital cost of $197,000,000, excluding conti ngenci_es and indirect cost, in favor of Scheme WP 3A. · If this difference is adjusted for the savings associ atad with using an appropriate proportion of sxcavated material as rockfi 11 in the main dam, this repre- sents a net overall cost difference of approximately $110,000,000 including contingencies, engineering, and administration costs. As discussed above, although limited information exists regarding tne qual- ity of the rock in the downstream area of the left bank, it is known that a major shear .zone runs through and is adjacent to the area presently allocated to the spillway. This waul d require re locating the· left bank cascade spillway several hundred feet farther downstream into an area where the rock quality is unknown and the topography less suited to the gentle overall slope of the cascade. The cost of the excavation would substan- tially increase compared to previ ou.s ass\lmpt 1 ens, irrespective of the rock quality. -In addition, the resistance of the rock to erosion and the suit- ability for use as excavated material in the main dam would become less certain. The economic feasibility of this scheme is largely predicated on this 1 ast factor, si nee the abi 1 i ty to use the materi a1 as a source .of rockfi 11 for the main dam represents a major cost saving. The prabl em of the occurrence of nitrogen supersaturation can be overcome by t,he use of a regularly operated di-spersion type valve outlet facility in conjunction with the main chute spillway. As this scheme presents a more econcmic solution with fewer potential problems concerning the· geotechnical aspects of its design, the right bank chllt-~ ~;'~·\angement has been adopted as the final selected scheme. 9-46 . ·~ -~ ~ 1 t t I I I t I t I I I I I I ·-,,_.-- ; ·• ., . . . /t\~ .... ~,J.J;.u~::., .... ~""-"""'""',u;MI;~ ·.I ... , I 1 I I I I I I I I I ' I ,, I I I TABLE 9.3: COMBINED WATANA AND OCVIL CANYON OPERATION Watana Dam Watana* Devil. Canyon* iota! Average Co.st Crest Elevation Cost Cost Annual f;nergy (ft MSL} 2240 (2215 reservoir elevation} 2190 (2165 reservoir elevation) 2140 (2115 reservoir elevation) ($ X 106 ) ($ X 106) ($ X 10 6) - 4,076 .1' 711 5;787 3.,785 1,711 5,496 3,516 . 1' 711 5,227 Watana Project alone (prior to year 20D2) Crest Elevation (ft MSl) 2,240 2,190 2,140 Average Annual Energy (GWh) 3,542 3 1 322 3,071 * Estimated costs in January 1°82 dollars, based on preliminary conceptual designs, including relict channel drainag·e blanket a11d .20 per.cent contingencies .. (GWh) 6,809 6,586 6,264 ~I I I ... . " I: I , I I' TABLE 9.4~ LONG-TERM PRESENT WORTH I W~tana Dam Long Term I Crest Elevat.ion System-Presen~* ' (ft MSL) Worth ($ x 10 ) - ~ 2240 (reservoir elevation 2215) 7' 123 I 2190 (reservoir elevation 2165) 7,052 2140 (:r-eservoir I elevation 2115) 7,084 * January 1982 dollars. I I I I I I I I I I I "' :~~; I c:!: :·::·\', '> '"- ·-.. 0 .".J ' ; I '" I I I I .I I I I I I .:. I ; -. I I I ., ' ':,."; ' t;:-. ,.,. , .. -··--. -... ·~~~~.,...,..--,---._.,.,.n·"'"· ......... ......., ·- TABLE 9.5: OES!'GN DATA AND DESIGN CRITERIA .FOR fiNAL REVIEW OF LAYOUTS Rivel' flaws ·- Average flow (over 30 years of recotd); Prabable maximum flood. (routed): Maximum inflow with return period of 1.:10,000 years: Maxinwn 1:lo·,ooO-year routed discharge: Maximum flood with return period ·Of 1:500 years: Ma~imum flood with return period of 1:50 years: Reservoir normal maximum operating level: Reservoir minimum operating level; Dam - Type: Crest elevation at point of maximum super elevation: Height: Cutoff and foundation treatment: ~stream slope: Downstream slope: Crest width: Diversion . Cofferdam type: Cutoff and foundation:. Upstream cofferdam crest elevation: Da'ftnstream·cofferdam crest elevation: Maximum pool level during construction: Tunnels · final closure: Releases during impounding: Spil~~ Design floods: Main spillway -Capacity: -Control str,Jcture: Emergency spillway ~ Capacity: ... Type: Power Intake . . Type·: Nt.Jllber of intakes: Draw-off requirements: Drawdown: -~· ~ -: .. '. .. .. :::::::~ 7,860 r.:fs 235,000 cfs 155,000 cfs 120,000 cfs · 116,000 cfs 87 1 000 .cfs 2,215 ft 2,030 ft Rock fill 2,240 ft 890 ft above foundation Core founded em rock; grout curtain and downstream drains 1V:2.4H:1V 1V:2H:1V 50 ft Rock fill Slurry trench to bedrock 1,585 ft 1,475 ft 1,580 ft Concrete lined, Mass concrete plugs 6,000 cfs mirimum via bypass to ou.tl.et structure Pass~s PMF, preserving.integrity of dam with no loss of life Passes routed 1:10,000.;.year f'lood with no damage t;.o struntures Routed 1 :10,000-year flood with 5 ft surcharge Gated ogee crests PMF minus 1:10,000 year flood Fuse..,plug Reinforced concrete 6 Multi-level corr.esponding to temperat:ure strata 1135 feet . -. ~ 'c ··--···A<·•·"·:;L_.li -:t.. _...:~ ''- ... -F-~ .. ~~:.. ~ .L·~-!~·,2_,'.' "'-·~~~~~~ \ \\ TABLE 9.5t {Cont'd) Type: Nunber of \)enstocks: Powerhouse Type: Transformer area: Control.room and administration: Access -Vehicle: .. Personna 1: Powerplant Type of turbines: Nllllber of rating: Rated net head: · Design flow: Normal maximum gross head: Type of generator: ·Rated output: Po.wer factor: Frequency: Transformers: Water passages: Surge: . Average tailwatex· elevation (full generation):. Concrete-lined tunnels with downstre~am steel liners 6 Un~ergr.ound Separate gallery Surface Rock tunnel Elevator from surface Francis ""' 6 X 170 MW 690ft 31500 C 1 ~S per unit 745 ft Vertical s~nchronous 148 MVPi 0.9 60 HZ 148 MVA -13.8-345 kV, 3-phase 2 concret.e-lined tunnels Separate surge chambers 1,458 ft ' •• I I I I I I I I : I I I I I I ' l I I I I I I I I I I I I I I I I I PRELIMINARY REVIEW Technical feasibility Co~atibility of layout with known .geological and topographical ~·ite features Ease of constructi~~r·; Physical dimensio:--.;J of component. structures in certain loc~tions Obvious cost dl:t' ferences of comparable structures Environmental accept- ability Operating characteristics TABLE 9,6: tVALUATJON CRITIEFtt: INTERMEDIATE REVIEW Technical Feasibility Co!JfJatibilit 't of layout with known geological and topographical site featur~s Ease of construction Overall cost Environmental accept- ability Operating characteristics Impact on constructicn schedule ~1 fiNAL REVIEW Technical feasibility Compatibility of layout with known geological and topographical site features Ease of construction Overall co&t Environmental impact Mode of operation of spill- 'Nays Impact on construction schedule ,Jesign and operating limita- tions for key structures .-~-,-.--,-.'"-0'_____ -o--~-----;--.--.----------=----·~~-e-cc--e TABLE 9. 7: SUMMARY OF COMPARATIVE COST ESTir;~;:;:e­ lNTERtt;EDIATE REVIEW OF ALT~RNATIVE ARRANGEMENTS (January 1982 $000,000) WP1 WPZ WPJ Diversion 101.4 112.6 "101.4 Service Spillway· 128.2 208.3 122.4 Emergency Spillway -46.9 46.9 Tailrace Tunnel 13.1 13.1 13.1 Credit for Use of Rock in Dam (11 .. 7) (31. 2) ( 18 •. 8} Total Non-Common Items 231.tl 349.7 265.0 Common Items 1643.0 1643.0 1643 .. 0 Subtotal 1874.0 1992.7 1908.0 Lamp & Support Costs· (16%) 299oB 318 .. 8 305.3 Subtotal 2173.8 2311.5 . 2213.3 Contingency (20%) 434.8 462.3 442.7 Subtotal 2608.6 1773.8 2656.0 Engineering and (12.5%) Administration 326.1 346.7 332.0 ' lOTAL 2934.7 3120.5 29138.0 WP4 1(..3.1 ~67.2 a~o (72.4) 305.9 1643.0 194.8.9 311 .. 8 2260.7 452.1 2712.8 339 .. 1 3051.9 . '\1 I I I I I I I I I 1 .. I I I I I I I I t·' I I I I I I I I I I I I ·I I I; I. ·I l I!!MMQ •••rm _.,.m? I l I \· ··.~ 0 • -..£;$ SCALE e~5iiiiiiiiiiiiiiiiiiii! LEGEND-~ BOREHOLES AND TEST Pr!':: . f971J,COE ROTARY DRILL ~ORI~G .....---....... GEOPHYSICAL SURVEYS~ .· ·. ·~ SEISMIC AEFR~CTION SURVEy £WO Ofl. Tl.lftlnNG 57~. DAMES & MOORE .~978, SHANNON 8 Wl!.!iON 1980-111, WOODWARD•CLYDE C~SULTANTS 0 SCALE ,. I I' I I I I I I I I I I I I I I I I ·~ It~- I ! ..---_....;....J/ +~ --------- : WATANA EXPLORATION MAP LEGEND BOREHOLES AND TE$T PITS• Gl OM ... ·l t·t·n·· .. ,t· of..·.· J·. OIA~ONO CQRE aoRtN&,"OR!Zot{f.L . ~ · ~£CTICHtAt~N ..... ~AAJ .. TPIHt .-,,AA! IA~H<lE TEST PIT GEOPtiVSICAL $UR'IEYS: ' £ ... lltiiMIC MniACTtot: SUIIIVE'1 EMil Oft TUMfiMI POINT . OM·C ttta,oAYts a MOoRE s-.1 lt78,1MiAHMON a WIL~ I SL 10•1 tHO~II,W®DW~·CLYDE c::oMim.TANTS o zoo 400 FEET SCM.£ REFERENCE• BASE . .lt4AP FROM COE.l918-1••200' WAUNA TOPOGRAPH'I',SHE£T8al3 OF 26,COORDitfATES IN FEET,Al.ASI<A STATE PLANE (ZoNE4} .CONTINUES OFF MGE COE 1 lV78( } s aW.tt:?a ( ) wcc,l978 < .. ) ~· . ~ ·- FIGURE 9 .. '2·0 ~· . iil .. ,, I I I I I I I I ·I· I ·I I I I I I I _ ____:l __ _ ··~- I + l ! Ill ~···~··· .-··' \ ----------- . ·~ ----::;J _.:t•---#-·'<"' WATANA ROCK OUTCROP MAP LEGEND LITHOU>3Y: .. "~"•,..f ANDESITE .PORPkYRY,INct..UOESMINoft. • • ,. DACITE AND LATtTE CONTACTS: ----... LIMIT OF OUTCROP CONltlUR LINES: ------.... TOPOGRAPHIC CONTOVR ~TERVAL 50 FEET ..... --~--.........----.. ~ .. + + 0 200 400 FEET SCALE + ' . . i . FIGURE 9 .. 3 -~-_,-.. --.. :0----~---,--- ."Y" I I ··~- I I li I I, I I I I I I I, I I I I .. a lt I,· i i .. · .. -- ·WATANA GEOLOGIC MAP SCALE LEGEND LttHOU>GYt D :=~~..a~ltlE:$ •to.••;.. . ~ ~.iNCt.Uo£:$ .~ •• • liiiNOit DAC11£ JND Ul11'1E j~,.v ~~ OIOftlttllfOft~ CONTACTS: . J'!!. LI'I'HOLbGIC, OUtED WHERE WEfii'E)., 0Ht WH£RE. tQtOWN STRUCTUfi£: Em• .SHEAR. WIOTH. ~ l'tWC• to~ ~ ~ lMl£SS w .SHOWfil. TQ..__ ~.·~ LESS nwe.··l9 n:tt •. · ..)"""A" INCUNEO, VERTICAL.£Xmll ~. ·.JOtDWN e;;;:.:;;::t ~ z• WIDTH ~"r14111il ~ 10 f'Ef'f; veRTICAL UNI..DS .,. .. SfiOWM: To-·::..--. ~-!~WilTH.LES$ TRoUt .. • T.r 10 FEET ~u ~ .£XTENT •• ,. WHERE ·~~WM. ' 'i'~h ~ ·~~~~~·.rJr!li~~· I'JO EXCEPT FCR OPEN .. JQtNTS.) [:J A!.TERKOON ZONE, 'WI>nt AS s .... : OTMER! w··a t ,t GEOLOGIC SEC'nON LOCAll(lM A WJ•I JOINT STATtotli 0 200 .. oo fEET FIGURE. 9 •. 4 --.·· '· :t 1- I I 1: I. I I I I I I I I I 'I I I !I 1500 1000 TAILRACE TUNNEL ~ACCESS ~ 0 TUNNELS ':r.·I'OO toO ao o I J J £l.. -1710.1·.· ·. . . -Tit.· ·. ~0.~ ... 0 ~. :~_/ #' . ~,-. -15M • . .,.'I> 0--_0 JDCVERSION TUNNELS ~· AREAS OF'FitACTURE 1DNES a MINOR SHEARS (HIGH ANCILE DIPS) TRENO~IO· ·WATANA GEOLOGIC SECTION w-1 SHEET I OF 2 . OH-s 4 ~-· I 9USITNA .. I I t I L!GEND UTHOLOG'f; gr.~J.i: ~.1.......-FP!OfTIATto r---1·. •OIOitr!E. TO. GUMTZ. t:dlftt. INa 'IDES 1......-..J . ...oR~ . n,-.~ ::-.-: .. '""· · ANDDitt POR;BIMrt,INCUJIIf:IIIINO!t . .• • •. • ••. QriQ1'E: • Ul'J"JE lq"\.. ..,,I*MtTE ~ CONTACTS~ ----~· "rtP -~ RO(;I( --.~ •. I:IQD 'WMJ;RE· IHFEJ.:D -:.c,~.·Sitowrc w.at:· tiUTf1t ~~ FP.ACTIJR£. . ~ONE.l..Wimt .SI40WN ·~ ~ GIIIEATIR TtWf. ·.u FEET . . c:: ·] ALT£JW'ION lilt£, \IIIDTH M .ltfOWN· GEOPtftSICAL SURVEYS: tsw·• ~CliQM WITH SE1SMlC ~ OM•C Jt75, ~ a _,OOM SW-I 1978, SHANNON a Wl.SON Sl. 80-2 1910, WOOOWUIO·CUO£ ~SUIJ»CT$ SL 1!·21 1981 • WOOOMRD·CUDE CONSl.UJWTS ••••••oo• SEISMC VELOC!TY ·~ 1 ~ SEISIIIC. VELOCITY .. FEET POl SECQN) BOREHOLES : .Ui'HOI.OGY 8H·I · !&SHEAR All'£RIITION ZONE ~:f . COE NOTARY a I>IAMCM!lO COR£ 80Mt6S D fJH AAI ~COM . .aft~Ne OTHER• W-5. .I INTERSEC'n~ WITH GEOl..OGIC; '+' SECilON 'W • l5 ~;-:-.. GEOL061.C f't.Al"lJM OESCfttatO-' ~~ . SCALE FIGURE 9JI .. I =~I I I I I I I I I I. ·I I I 2800 2000 11500 1000 0 8aTTOM PRQ.I£CTEO IIS'W BOTTOM PRO.IECTa) II!'E WATANA GEOLOGIC SECTION W-1 SHEET 20F2 1000 o. 100 200 FEET SCALE FIGURE 9Jta ,. I I I I I I I I I I I I I I I I I ,I ..1 E II COMPOSITE JOINT· PLOT SOUTHEAST QUADRANT N•721 II COMPOSITE JOINT PLOT NORTHEAST QUADRANT· N•!525 w w E E II COMPOSITE JOINT PLOT SOUTHVIEST QUAt)RANT N•329 N COMPOSITE "OlNT PLOT NORTHWEST QUADRANT WATANA COMPOSITE JOINT PLOTS SCALE • 0 '--· NOTES l. coto!TOURS ARE PEJK:EHT OF JOINTS PER 1"-OF AREA. CONTOUR INT£R\iU;.-I ,3, a 0 "'· Z. N EOOAU NUMBEII OF ~TA POINTS. ~. COMPOSITE .PLOTS .INCORPO'MTE ALL JOINT ~~ iFflOM THEIR ft£SP£CTIVE WAOftlNTS. .f • .IOINT .t'LCtra FOtt .1011T STAfiO~l (Wol-1,2,3,._5.6.~1 all) ott 5, FOft .1010'. PLat:TM liiET'HOD .$EE FIGURE 1 FIGURE 9;12 [iil ;1. · .. ' . I •• '.1 . . . -., I I I ·- 1 I I I I I I I I I ··-- ··-- '''tm JI'WQQQ ··-- "' . .:. .. . ·• .. I SC.AL£ o 4 8 MILES ~!!!!!iiiiiiiiiiiiiiiiiil' i:: LOCATION MAP 0 YJOO 2000 FEET SCALE .. } t WATANA RELICT CHANNEL TOP OF BEDROCK . FIGURE 9.12q l I I I I I I I I I I I 1: I 1600 --------~~--~----~--~----~----~---r~--~---; \ \ \ \ \ (') \ LESS THAN 3 ' ENTRANCE. SUBMERGED -1550 ~--------+----------~--~~;-~---------r~------~ .....: u.. - 0 TYPICAL TUNNEL SECTION 1450 ~--------L---------~--------~--~~------------~ 25 30 35 40 45 TUNNEL DIAMETER (FT.) WATANA DIVERSION HEADWArER · ELEVAT1 ON I TUNNEL DIAMETER FIGURE $.13 I 'I I I I I I I I I I I I I I I I I I l .. < • < < -,_; u.. - z 0 -1- <( > lJJ ...J lJJ ~ <( 0 ' i I f I 1600~-------h~------~~----~~------~ { f 1 I • t 1550 t----f----+-----t------4-------1 1500~------~~------~------~------~ lOX 10 6 2.0X 10 6 30XJ0 6 40XI06 " CAPITAL COST S WATANA DIVERSION UPSTREAM COFFERDAM COSTS FIGURE 9.14 I .~ C' I I I I , I -CD 0 - X I ... - .... Cl) I 0 (..) ...J < 1- I -Q. < (..) I I I I I I Q I, I I 80----~--~--------~~~----~--------~--~---r~--~--~ 50 40 30 20 15 0 0 TYPICAL TUNNEL SECTION 20 *DAM c STS INCLUDE AM HEIGHT 15 1 ABOVE HelAOWAT~R I ELEVAT ON FOR FREEBOARD. 25 30 35 40 TUNNEL DIAMETER (FT.} WATANA DIVERSION TUNNEc·a COFFERDAM COST I TUNNEL DIAMETER FIGURE 9.15 45 I I I I I I I I 1: ·~ I I I '.1 I I I. I I -U) 0 - X -- (I) 1- (I) 0 0 ..J ~ a.. <t 0 ..J <t t- 0 t- 80 70 60 0 TYPIC At: TUNNEL. SECTION 50 15 20 25 30 35 TUNNEL DIAMETER (FT.) WATANA DIVERSION TOTAL COST I TUNNEL DIAMETER l 40 FIGURE 9.16 I I I I. • I I -ID 0 I -)( ... -:r; 1- I a:: 0 == 1-z IJJ I (/) IJJ a: 0.. :E IJJ I }- (/) >-(/) :E I a: UJ 1- (!) z 0 I ..J I I I I I I I 7300 72.00 7100 ~ 7000 6900 6800 6700 6600 6500 2140 0 .- ·-·-· /~ ~ ~ . I .· l f ~ ! t ' I . t -.. 2160 2180 2200 222.0 ~240 22.60 DAM CREST ELEVATION (FEET) SELECTION OF RESERVOIR LEVEL FiGURE 9.17 I I I I I I I I I I; I I I I I I I I . 0 1. 10 ... SELECTION OF DEVIL CANYON GENERAL ARRANGEMENT This section describes the development of the -general arrangement of the Devil Canyon project. The site topography, geology, and seismicity of the Devil Can- yon site are described re 1 at ive to the design and arrangement of the various site facilities, in a manner similar to that presented in Section 9 for the Watana site. The method of handling floods during construction and subsequent project 6peration is also 9utlined in this section. The reservoir level fluctuations and inflow for Devil Canyon will essentially be contra lled by operation of the upstream Watana project. This aspect is also briefly discussed in this section. A detailed descriptioo of the various proj- ect components is given in Section 13. 10.1 -Site Topography The Devil Canyon site is located at river mile 152 of the Susitna River, approx- imately 31 mi 1 es downstream from the Watana site, in a 11 V" shaped sect ion near the entrance to the canyon which is about 2 miles long. The valley wall on the left side of the river rises w~ry steeply from Elevations 900 to 1300 on the left bank at a slope of approximately 0.4H:lV to a relatively gently sloping plateau area which reaches Elevation 1600 within the general project area. On the right side, the valley is less pronounced, rising at about 1.1H:lV to Eleva- tion 1500, then much more gradually to approximate Elevation 1900. The steep left bank features overhanging cliffs and detached blocks of rock. 10.2 -Site Geology This section summarizes the gecilo~ical and geotechnical investigations and in- terpretations conducted to date and the conditions present at the proposed Devil Canyon site. The detailed description of the site investigations and the geo- logic and geotechnical conclusions are presented in the 1980-81 Geotechnical Report ( 1). (a) Geologic Setting Devil Canyon has been eroded through hard metamorphosed sedimentary rocks, argillite and graywacke of excellent quality (Figure 7 .11). The bedding strikes roughly· parallel to the river and dips to the south. Overburden is generally thin to nonexistent. Stress relief cracks and open joints paral- lel the gorge and extend more than 100 feet from the canyon walls .. On the left (south) bank, a series of small 1 akes paral1el the valley. Deep overburden up to 80 feet thick ilhas been encountered in this area which probably represents an old stream buried under glacial material. A highly sheared and fractured zone is present under this buried stream. Work per- formed during this study, however, showed this feature to be of no seismic concern (3). A large alluvial fan exists at the confluence of Cheechako Creek with the Susitna Kiver, about 1,000 feet upstream from the damsite. This area is the rna in source of materia 1 for the concrete aggregates and the fi 1 ter materials for the saddle dam. 10-1 (b) Geological and Geotechnical Investigatf~ns Sur·face and subsurface investigations have been conducted by several organ- ; zat ions at different times. During the per tod from June ·1957 to August 1958, the USBR conducted geologic mapping and subsurface investigations at this site. The subsurface investigations included drilling of rock and overburden, in-hole testing, test pits, and laboratory tests. Subse- quently, in 1978, the COE conducted seismic refraction surveys to expand on this work to assess the suitability of the site. · During the years 1980 and 1981, more detailed investigations of geologic features were performed as part of the current work program to establish the technical feasibility of the project. These investigations have included air reconnaissance, air photo.-in~erpretation, geologic mapping of rock and overburden ir.cluding in- hole geophysical tests, and seismic refraction surveys .. Both in situ and 1 aboratory tests have been performed to determine the engineering charac- teristics of soils and rocks. The location of drill holes, test pits, and other investigations is shown in Figure 10.1. · Geologic mapping was concentrated in the immediate damsite area to define the geology of the site in as much detail as possible. Under extremely difficult conditions of accessibility, ground mapping traverses ·were accom- plished by making maximum use of technical climbers in the gorge to augment mapping activities on foot along the upper slopes. At each station, the applicable 1 itho1ogy or· type of overburden, bedding, jointing, weathering, degree of consolidation, exposure size, and elevations was noted and plot- ted on maps for use in the interpretation. All accessible areas with rock outcrops were mapped. Seismic refraction surveys totalling 3~300 feet were performed on the south side of the canyon across the lake area and the alluvial fan to delineate their extent and characteristics. Diamond core dri 11 ing was performed on the upper slopes and at river level of the canyon.. The holes were cased through the overburden leaving access for future te.sting and instrumentation. A total of 4,800 1 inear feet in 29 holes have been drilled at the Devil Canyon site. Comprehensive logs have been developed for each hole. Water pressure testing using inflatable packers was conducted in the holes for permeability calculations. Geophys- ical logging and borehole photography were attempted during the 1980 season in selected holes. I I •• I I I I I I I I I I To monitor the ground water and ground temperature conditions at the site, I piezometers and thermistors were installed in selected drill holes. Peri- odic readings of these instruments, after stabilization, have been con- ducted to give more detailed informat·ion of the conditions that may be en-·•· countered during construction. A series of tests was performed on the rock recovered from coring to deter-.•... mine the engineering characteristics of the rock mass. The results of these investigations were compiled, correlated, and interpreted to develop the geologic picture of the damsite and the adjacent areas. The results of the 1 aboratory rock tests are summarized in Table lQ.L J 10-2 •• I I I I I. I I I I I I I I I I I I I I (c) Construction Material Investigations Most of the investigations for the construction materials at the Devil Canyon site were performed during the years 1957 and 1958. Additional ex- plo~ation during the years 1980 ar.d 1981 was undertaken to supplement the previous work. The investigations have included geologic mapping, auger drilling, excavation of test pits and test trenches, seismic refraction surveys, and laboratory tests. A total of 8 auger holes, 2 test pits, and 4,100 linear feet of seismic refraction survey have been completed. The location of the various sources for the concrete aggregate and the saddle embankment dam are shown in Figure 10.1. A major source of construction materials for the Devil Canyon Project is an alluvial fan of deposits, which lies near the Cheechako Creek confluence approximately 1,000 feet upstream from the arch damsite. The area contains large quantities of sands and gravels with inclusions of boulders and cobbles above the river level. Under a thin mantle of organic material, a 3-to-4-foot layer of silty sand overlies a layer of sandy gravel with traces of silt and some cobble5 and boulders. This layer of sandy gravel is about 80 feet thic~. With proper processing, this source will provide coarse and fine aggregate for the concrete, fi 1 ters for the embankment dam a'nd cobbles for the upstream shell of· the embankment dam. A composite grain-size curve for this material is presented in Figure 10.2. ( i) Rockfi 11 Materia 1 for the Saddle Dam The required quantities of rockfill can be obtained from the area designated Quarry Site K. The rock in this site is primarily dior- ite. It is hard, durable and fresh, and suitable for the embankment construction. Also, the suitable portions of the rock excavated for the found at ion of the arch dam and other project structures, such as · underground power facilities and main and emergency spillways, are adequate for use in the embankment, subject to appropriate schedul- ing of excavation and fill operations. Sufficient fill-quantities are available at the site to meet all requirements. (ii) Impervious Core Material for the Saddle Dam No suitable source for the core material for the saddle dam has been identified at this time near the site. For current feasibility assessment purposes, it is planned that the core material will be transported from Borrow Area D near the vJatana site, where suffi- cient quantities of suitable material have been identified. A dis- cussion of the engineering characteristics of those materials is presented in Section 9. Additional investigations will be performed in the future in an at tempt to 1 ocate a potentia 1 source nearer the Devil Canyon site. (iii) Filter Matet"'ial for the Saddle Dam filter and transition zone materials will be obtained from the allu- vial fan Borrow Are a, as discussed above. 10-3 (iv) Gravels and Cobbles for the Saddle Dam If needed, sufficient quantities of clean gravel and cobbles can also be obtained from the alluvial fan with proper processing." (v) Concrete Aggregate The coarse and fine aggregate for the concrete structures wi 11 also be obtained from the alluvial fan upstream. The results of the lab- oratory testing, presented in Table 10.1, indicate that the material from this source is of adequate quality. The gravel particles are generally rounded with accompanying subangular sands. Petrographic analyses indicate that the material includes quartz diorites~ gran- ites, andesites, diorites, dacites, metavolcanics, rocks, ap1ites, breccias, schists, phillites, arqi1lites, and amphtbolites. Gener- ally, the material has less than 2 percent deleterious constituents such as chert, muscovite, and argillite. (d) Geologic Conditions The overburden and bedrock conditions at the Dev i1 Canyon site are discus- sed in this section. (i) Overburden The valley walls at the Devil Canyon site are very steep and al"'e generally covered by a thin veneer of overburden consisting primari- ly of talus at the base (Figure 10.3). The flatter upland areas are covered by 5 to 35 feet of overburden of g1 aci a1 origin. The topo- graphic depression along the elongated lakes on the south bank has an overburden covering in excess of 85 feet of g1 aci al materials. The overburden on the alluvial terrace or point bar deposit at the Cheechako Creek confluence thickens from 100 feet to more than 300 feet over a distance of less than 40U feet. The river channel alluvium appears to be composed of cobbles~ bould- ers, and detached blocks of rock and is inferred to be up to 40 feet thick. A representative cross section across the valley is pre- sented in Figure 10.4. (ii) Bedrock Lithology The bedrock at the Devil Canyon site is a low-grade metamorphosed sedimentary rock consisting predominantly of argillite with inter- beds of graywacke (Figure 10.5). The argillite is a fresh, medium- to-dark gray, very thinly bedded, very fine grained argillaceous rock with moderately well ... developed fo1 iation parallel to the bed- ding. The graywacke is a fresh, light gray, mainly fine grained sandstone with an argillaceous matrix. It is locally a conglomerate with lithic. fragments up to two inches in size. The graywacke is well indurated and·exhibits poorly developed to non-existent folia- tion. The graywacke is interbedded with the argillite in beds gen• erally less than 6 inches thick. Contacts between beds are tight 10-4 I I I I ,, I I I I I I I I I ·I I I .I I I I I I I I .~ I I I I I ·I I. I I I I -.• and both rock types are fresh and hard.. Minor quartz veins and stringers have intruded the argillite. These are generally less than 1 foot wide and unfractured with tight contacts. Sulphide min- eralization is common with pyrite occurring in as much as 5 percent of the rock. The area has also been intruded by numerous felsic and mafic dikes ranging from 1 inch to 60 feet wide (averaging 20 feet). The dikes have northwest to north orientation (Figure 10.5) with steep dips. When c 1 ose ly fractured they are easily eroded and tend to form steep talus-filled gullies, some of which exhibit shearing with the host rock.. The felsic dike~; are 1 ight gray silicic varieties including aplite and rhyolite. The mafic dikes are fine grained and appear to be of diorite to diabase composition. (ii1) Bedrock Structures Bedding The arg i 1 1 ite/ graywacke. has been comp 1 ete 1 y deformed as evidenced by refolded folds and the development of multiple foliations .. The primary foliation parallels the bedding at 35° to 90~ {N35E to E), subparallel tb the river, and dips 45° to 80°SE. \\f11en two or mor·e foliations are parallel, the rock has a very slateyl phyll it1c appearance, and when oblique, the rock appears massive. The canyon at the dams i te appears to be contra 11 ed by the: bedding planes. -Joints Four joint sets have been delineated at Devil Canyon, as shown on the stereo plots (Figure 10.6). Set I (strike 320° to 355~ and dip 60NE to 70NW) and Set ti (strike 040° to 065~ and 40° to 60~S dips) are the most significant. Set I joints are the most promi- nent with spacing of 15 feet to 2 feet, and on the upper canyon ~>;a 11 s of the south bank are open as much as 6 inches (Figure 10.5). Set III is subparallel to the bedding/foliation and 3 when it intersects with Set I, can cause the formation of loose blocks. Set III joints (strikes 005° to 030° and dips 85NW to 85SE) are also often open ·an the south bank and may dip towards the river, creating potential slip planes. This set, however, has variable spacing and sporadic distribution. The fourth set is a minor set with low dip angles and variable strike orientatlon. Joint spacings measured from the borehole cores range ftom less than 1 foot to more than 10 feet. The spacing and tightness of the joints increase with depth, and the iron oxide staining and weathering extends up to ·80 feet. ~ Shears and Ftacture Zones Shears and fracture zones were encountered in localized areas ·of the site in noth outcrops and boreholes. Corre 1 at ion of the data is shown on the interpretation map (Figure 10.5). Shears are 10-5 •w ,•."' defined as areas containing. breccia, gouge, and/or slickens1 hies indicating relative movement~ These zones are soft and friable and are characteri .zed by high permeab i 1 i ty and core 1 oss during , drilling. Fracture zones, often encountered in conjunction with tht; shears, are zones of very closely spaced joints. With depth, these .zones become smaller, tighter, and more widely spaced .. Where exposed, they are eroded into deep gullies. The most common trend of these features is northwest, parallel to Joint Set I. These have vertical to steep northeast dips and are generally less than 1 foot \vide. Northw·est trending shears are also associated with .the contacts between the argillite and mafic dikes and are up to 1 foot wide, wi·ch closely spaced joints prom- inent in the dike itself~ · A second series of shears trend northeasterly, subparallel~ng the . bedding/foliation and Joint Set II, and have· high angle southeas- terly dips. These average less than 6 inches in width. (e) Structural Features Several. structural features at the Devil Canyon site were investigated dur- ing· the 1980-81 program. In summary, these included the east-west trending sheared and fractured zone beneath the proposed saddle dam area; a bedrock drop-off beneath Borrow Site G; and bedrock conditions beneath the Susitna. Seismic refraction and drilling data confirm the existence of a highly sheared ·and fractured zone on the 1 eft bank beneath the proposed saddle dam that generally trends parallel to. the.river. The dip on this feature is inferred_ to be paralle1'or subpara1tel-i~o-_,tl1e bedding/foliation at approxi- mate 1y 65 o to the south. The 1 in ear extent of the feature has been infer- red to be approximately 2~500 feet. No evidence was found during the 1980-81 program to suggest movement along this feature. This finding was also concluded by work done by Woodward-Clyde Consultants (3). Further investigation of this feature will be required to define its extent and type of foundation treatment that will be required beneath the saddle dam in subsequent phases of investigation. Upstream from the damsite, a several-hundred-foot drop-off in bedrock sur- face was detected by seismic refraction surveys under the alluvial fan .. Land access restrictions imposed during the study prohibited any further investigation of this area. Possible explanation for this apparent ·anamal- ous drop-off could be attributed to misinterpretation of the seismic data ·or else the lower velocity material could be either a highly fractured rock in lieu of soil or an offset of the rock surface caused by faulting. The 1 atter interpret at ion is unlikely in that work performed by Woodward-Clyde Consultants (3) in this area concluded that there was no compelling evi- dence for a fault.. Future work remains to be done in this area tu more clearly define this featur.e~ Uetailed examination of rock core and mapping in the river valley bottom showed no evidence for faulting in the riverbed. 10-6 I I 'I· ::, ,' I I I I I I :1 I •• I I I I I I I • I I I I I I I I !. I 1: I I I I I I I (f) Ground Water Co~ditions Ground water migration within the rock is restricted to joints and frac- tures.. It is inferred that the ground water level is a subdued replica of the surface topography with the flow towards the river and lakes. Measured water levels in the bOl·eholes range average approximately 120 feet below surface. · (g) Permafrost Preliminary temperature measurements made in the boreholes did not enccunt- er permafrost conditions on either side of the river. (h) Devil Canyon Reservoir. Geolo91_ ' The Oevi 1 Canyon reservoir will be confined to a narrow canyon where the topography is controlled by bedrock. The overburden is thin to nonexis- tent, except in the upper reaches of the reservoir where alluvial deposits cover the valley floor. Near the Watana site, 1 ight gray to pink, medium grain diorite rock is present. This rock is hard, massive, and competent except on the upland north of the Susitna River where the biotite grana-. dionite has been badly weathered. The principal rock types in the most part of the reservoir are the arg i 11 ite and graywacke which are exposed at the damsite. The,rock has been isoclinically folded into steeply dipping structures striking generally northeast-southwest~ The argillite has been intruded by massive granodiorite, and as a result, large isolated roof pend ants of the argillite and graywacke are found 1 oc ally throughout the entire reservoir and surrounding areas. The joint measurements at selected areas indicate structural trends similar to those at the damsites. 10.3 ~-Geotechnical Consideratio~s The geotechnical investigations to date have been primarily directed toward the important geological features which may have significant impact on the feasibil~ ity o·f the project. More detailed investigatinns, including exploratory adits, will be required prior to the detailed design. (a) Arch Dam Foundation and Abutments The geologic and topographic conditions are favorable for an arch dam ·at the Devil Canyon site. The rock is principally hard, competent, and fresh with weathering 1 imited to joints and shear zones. ~1trt1sive mafic and felsic dikes, where present, are hard, and the contact with the parent rock is tight. The orientation of these dikes is generally NW toN and has no important adverse effect on the stability of the abutments. The unconfined compressive strength of the intact rock ranges between l6,00U psi and 32,000 psi. The stresses imposed by the arch dam are about l,OOO psi or less under normal conditions. Even under extreme loading conditions~ the stresses will be well withln the acceptable 1 imits for bearing considera- tions. Un the right abutment, the arch dam thrust block wi 11· be seated in good sound rock. The topcof the hil1 is at approximately Elevation 1500 and no major rock discontinuity is present. However, on the left side, massive thrust blocks will be required to transfer the loads to competent rock and to form an abutment to the saddle dam. 10-7 Essentially no continuous, poorly oriented rock d·iscontinuities, which· might adversely affect abutment stability, have been found. The major joint set -:at the site, influencing the stability of the south bank aout- ment,. ·strikes approximately northwest with a near vert ica1 average dip. The st~b il ity o.f the right abutment (north bank) is controlled by the. bedding planes and foliations that strike roughly parallel to subparallel to the canyon walls and dip steeply into the canyon. The bedding planes generally appear to be tight with undulating surfaces because of the ex- tensive folding of the rock. Preliminary analyses indicate no stability problems. Additional rock investigations and in situ testing will be re- quired during final design to C6hfirm rock properties and the J·esults of stability analyses. The dam· and the thrust blocks will be. founded on sound rock. This will require complete removal of all the overburden and weathered rock.. Along some of the northwest trending shear zones, the weathering could be as deep as 200 feet~ Extensive dental excavation may be required in these areas to form an acceptable foundation. The entire dam foundation area should be consolidation grouted to fill all the openings and cavities in rock at shallow depth. The permeability of rock varies from 1 x lo-4 em/sec at shai low depths to 1 x lo-6 em/sec at depths below 175 feet. The permeability is con- trolled by the discontinuities in the rock and may var·y widely from area to area. A grout curtain will be provided under the entire dam including the abutments and an appropriate distance beyond the dam into the abutments. A system of dra.in holes and drainage galleries will be included to control uplift pressures and to safely release seepage water. A double row curtain is proposed,. There is little evidence of permafrost at this site; however, provision should be made for thawing during grouting should permafrost be encountered. (b)_ Underground Structures The rock conditions at the site ·are suitable for the construction of tun- nels and underground caverns. From the geological and geotechnical view- point, the 1 ocat ion and the orientation of these structures are influenced by the orientation and location of rock discontinuities. The RQD values indicate that about 50 percent of the rock is in the good- to-excellent category, roughly 40 percent in the fair-to-good category, and lhe remaining percentage in the poor category. The poor quality rock is generally associated with fractures and shear zones.. The major joint sets are oriented northwest (Set I) and northeast (Set II). Both sets are steeply dipping.. The bedding plane strikes roughly in a NE-E direction and dips at 45 to 80° southeast. The orientation of the tunnels and the large underground caverns have been carefully selected to minimize the potentia 1 · adverse effects of these rock d i scant inui ties~ 10-8 I I I I I I I I I ,, I I I I I I I I I I I ~'' '· '· ''.•..c' I I I I. I I I 'I I. I I I I I I I (c) . Oeterminat ion of the magnitude and the orienfat1on .of the in situ str~sses · .. will_ rrot be possible until in situ-testing is undertaken. Nevertheless, the tectonic setting suggests that the entire site region is in a compres- sional stress regime. Near valley walls,. the stresses are expected to have been relieved .and-low horizontal stresses may exist. Considering the un- confined strength of the intact rock, overstressing problems such as rock spa11 ing ~and sl abing are not anticipated~-The rock support requirements will depend on the size and orientation of the openings and the presence and character of the rock discontinuities intersected. For. the most part, conventional rock bolt support u~ing 3/4-inch to l-inch-diameter bolts has been assumed to be adequate for openings less than 40 feet in span. For larger spans, in areas of poor quality rock and where rock discontinuities are known to be adversely oriented, support requirements have been deter- mined on a case-by-case basis~ In the case of 1 arge span openings (such as powerhouse cavern), special attention has been given to the potential presence of subhorizontal.Joints where they may intersect almost vertical joints and may create unstable blocks in the crown of the excavation. The use of shotcrete, welded wire fabric, and concrete lining will be required in poor rock quality areas. For power tunne1s; provisions have also been made for concrete lining and contact/consolidation grouting. - Although rock permeabil ities are generally lo"'' to moderate, intersection of rock discontinuities may lead to ground water inflow problems during con- struction and cause high-pore water pressures <.lfter the reservoir is flood- ed. Therefore, provjsions have been made for grouting around tunnels and caverns, and suitably placed drain holes and drainage galleries have been provided upstream from the powerhouse and surge chamber. The excavat1on of tunnels may be performed using conventional drill and blast techniques or high-production mechanical excavators. Sufficient in- formation is not available at this time to select an optimum system. For cost estimating purposes, conventional methods have been conservatively assumed at this time. The excavation of the powerhouse cavern will be by drill and blast using a heading, side slash, and benching sequence. The spacing between long tunnels should be 2.5 times the diameter of the 1 argest tunnel. The spacing between the caverns should be kept so that a minimum pillar thickness of 1.5 times the span of the larger cavern is maintained. Stabi 1 ity of Soil and Rock Slopes In most areas, the permanent excavation slopes will be confined within the rock~ except on the left bank, where a deep buried stream exists. The slopes within overburden will depend on the nature of soil, ground wat·er table, and the height of the slope. In general, slop.es within the aver- burden"nave been assumed as 2H:1V or less below the water table and 1 .. 5H:1V or .less :above the water table. A bench of appropriate width will be pro- vided at the overburden-rock contact to accommodate any local slumping/ slope failure and to intercept and dispose of ground/seepage water •. Flat- ter s 1 opes are required in some areas where frozen ground may become un- stable because of high pore pressures during thawing or where slope height/ soil conditions so dictate. 10-9 D The slopes of excavations ir rock have bee_n se1ected in accordance with ~he joint dips. and orientations and the shear. strength, of rock along discontn1-. _ uities. Slopes in intact rock or where discontinuities dip into the exc·a-. vated face will usually stand steeply without any· structural support. Slopes par;alleling the discontinuity have, whenever possiole, been laid back to the same angle as the dip of the rock discontinuity or adequa.te rock support provided. Wherever possible, permanent cuts have been set .at stable slopes without the need for rock bolts. In areas where 1 arge pore pressures·could potentially develop behind the rock cuts, allowances have been made for drain holes to relieve the pore. pressures~ In general, a 4V:1H overa'll slope hi considered stable~ For slopes exceeding 40 feet in height, benches have been included every40 feet~ Excavation of tunnel portals includes pattern rock bolting and appropriate provision for concrete/shotcrate to reduce the risk of unstable slopes. Special details are required ·in areas where s 1 opes wi 11 intersect or cross 1 arger shear zones or otherwise unstable rock. (d) Saddle Dam Found at ion The saddle dam on the south bank will be constructed across the buried stream. The thickness of overburden in this area reaches up to 80 feet. The bedrock below (argillite and graywacke) area is competent. The imper- vious core, filters, and outer shells for the saddle dam wnl be founded on sound rock. The prominent shear zone or fault which was found in the saddle dam foundation, together with various she&r and fracture zones, has been treated by means of provisions for con$olid~ion and curtain grouting under the core as a continuation of foundation treatment for the arch dam. 10.4 -Seismic Considerations The seismicity of the Sus i tn a Has in and the sources of earthquakes are discussed in Section 7 of this report. This section presents the implications ofthe seismicity on the design of the Devil Canyon project. (a) Seismic Design Approach For the earthquake engineering and,design considerations, the project structures have been classified as either critical structures or noncriti- cal structures .. Critical structures include the dam and similar major structures whose failure may Y'esult in sudden and uncontrolled release of 1 arge volumes cf water which may endanger property and ·1 ives downstream .. The noncritical structures are those structures whose failure can be assessed as an economic or financial loss to the project· in terms of lost revenue, repair, and/or replacement cost. Critical structures \'Jill be de- signed to safely withstand the effect of the 11 Safety Evaluation'Earthquake 11 (SEE) for the site. No significant damage to these structures will be . accepted under these conditions. 10-10 I I I ·1·'·. ! ' I I I I I I I I I I I I I ·. I .I· I I 'I I I I '· I I I I I I I I ' I for ·ctestgn nf critical .structures, the effective acceleration fcrr. the SEE has been determ-ined as o. 8 :x actual SEE acceleration, together with a cor- responding scaled respOnse spectrum. The selected SEE corresponds to the 11 terrain 11 or 11 detection 1eve1 11 earthquake ·which has been characterized as . follows: -Magnitude: -Location! -Maximum Acce 1 er at ion! -Peak Spectral Acceleration: 6-1/4 to 6-:1/2 Approximate 1 y 3 km from structure Me·,. 0.55g to 0.60g 84th p~rcentile 0~70 g · Mean 1.37g to 1.50g · 84th percentile l.77g The response spectra for this event are shown in Figure 10.7A. The effec- tive peak ~cceleration for design of structures is then: ~ -Design a max = 0.8 x 0.70g =· 0.56g (Sa) max = 0.56g x 2.5 = 1.4Ug The design of non-critical .structures for earthquake conditions will be undertaken using conventional Uniform Building Code recommendations. (b) Safety ·Evaluation Earthquake The design of the: Devil Canyon arch dam and other critical structures has been undertaken using response spectrum analysis. Although the "terrai'n 11 -earthquake would result in more severe ground motions, the duration of these motions is relatively short. However, the method of analysis used for the Devil Canyon Dam does not ta~-e duration into account. The most likely source of strong ground shak·ing at the Uevil Canyon site is, in fact, the Benioff Zone. The estimated mean peak response spectrum for the SEE is presented in Figure 10.7, along with the 84th percentfle response spectrum .. A maximum horizontal acceleration level for the 84th percentile response. spectrum for the Benioff event is approximately 0.47g. 10.~ -Selection of Reservoir Level The selected normal maximum operating level at Devil Canyon Dam is Eleva- tion 1455. Studies by the USSR and COE on the Devil Canyon Project were essentially based on a simi 1 ar reservoir 1 eve 1, which corresponds to the tailwater level selected at the Watana site. Although the narrow configur- ation of the Devil Canyon site and the relatively low costs involved in in- creasing the dam height suggest that.it might be economic to do so, it is clear that the upper economic limit of reservoir level at Devil Canyon is the Watana tailrace level. · The detailed studies of reservoir level at Watana (Section 9) indicated 1 ittle change in benefit-cost ratio over a 100-foot range of reservoir level at the upper limit. Maximization of hydroelectric energy pr_Qd_uction at the site was found to be an important objective which weighed heavily in the selection of reservoir level at Watana. Although a detailed determina- tion has not been undertaken, the same .is likely to be true at Devil Can- yon. t:, l0-11 Although significantly lower reservoir levels at Devil Canyon would lead to lower dam casts; it is clearly evident .that the location of adequate spill- \1/ay facilities in the nartow gorge would become extremely difficult and lead to offsetting increases in cost. In the extreme case, a spillway dis-, charging over the dam would raise. concerns regarding safety from scouring at the toe of the dam, which have already led to rejection of. such schemes. · · 10.6 -Selection of Installed CapacitX The methodology used for the preliminary selection of installed capacity at- Watana and Devil Canyon is described in Section 9.6~ Thedecision to operate Devil Canyon essentially as a run-of-river plant with _ maximum utilization of available flows from Watana was governed by the following main considerations: · · -Daily peaking is more effectively performed at Watana than at Devil Canyon; and Excessive fluctuations in discharge from Devil Caryon will have a significant impact on downstream fisheries. Given this mode of operation, the required installed capacity at Devil Canyon has been determined as the maximum capacity needed to uti1ize the available energy from the hydrologit:al flows of record, as modified by the res lts of reservoir operation computer simulation analysis (Section 9 .. 6), with the station operating at 100 percent load factor. In years where the energy from Watana and Devil Canyon exceeds the system demand, the usable energy has been reduced at both stations in proportion to the average net head available, assuming that power flow releases at Watana will also be used at Devil Canyon. The total capacity required at Devil Canyon 1n a. wet year, excluding standby and spinning reserve capacity, is summarized below. As discuss€d in Section 9 .. 6, the capacity shown is based on the Battelle medium load growth forecast. Demand Year 2002 2005 2010 Capacity MW 370 410 507 The selected total installed capacity at Devil Canyon has been established as 600 MW for feasibility design purposes. This will provide some margin of stand- by for forced outage and possible accelerated growth in demand. The major factors governing the selection of the unit size at Oev.il Canyon ·are the rate of growth of system demand, the minimum stat ion outputs and the re- quirement of standby capacity under forced outage conditions. The above tab- ulation indicates that station maximun load in December will increase by about 50 percent from 2002 to 2010 (from 370 MW to507 MW). Station minimum output in July during ·the same period will vary fr:?~ _about 150 MW to 202 MW .. 10-12 I I I I I I I :1 ;• I I I I I I ---- 1 I ' I I .I I I I I I I I I I •• 1.· '.J I I ....... I I I I For feasibility design purposes, the power facilities at Devil Canyon have been developed for 4 units at 150 IVIW, This arra.ngement ;vill provide for efficient station operation at part load.. Consideration of phasing of the capacity in-· stallation of the machines may be desirable as the system demand increases. However·, the uncertainty of load forecasts this fan into the future~ and the additional contractual costs of mobi·1ization for.equipment ins_tal1atton are such that for study purposes at this t irne it has been assumed that a11 units wi 11 be commissioned by 2002,. . The Devil Canyon reservoir wi 11 normally be full in December; hence~ Q.ny forced outage will result in spilling and a loss of available energy~ The units have . been rated to deliver 150 MW at minimum December drawdown level; this means that in o.n average year, with higher reservoir levels, the full station outp:Ut can be maintained even with one unit on forced nutage. 10.7 -Selection of Spillway Capacity A flood frequency of 1:10,000 years was se 1 ected for the spi 11 way design on the same basis as described for ·watana (Section 9), An emergency spillway with an erodible fuse plug will also be provided to handle larger discharges up to the probable maximum. flood. As discussed in Section 8 and· elsewhere, the develop- ment plan envisages completion of the Watana project prior to construction at Devil c·anyon. According1y, the inflow flood peaks wi11 be significantly ·less at UeVil Canyon because of routing through the Watana reservoir. Spillway floods as calculated in Section 7.2 are: Flood Probable Maximum 1:10,000 year Inflow Peak (cfs) 366,000 165,000 . , The restrictions with respect to nitrogen supersaturation of.downstream flows discussed in Section 9 _for Watana also will apply to Devil .Canyon, and dis- charges of nitrogen-supersaturated water from Devi 1 Canyon will be 1 imited to a recurrence period of not less than 1:50 years. 10.8 -Main Dam Alternatives The location of the Devil Canyon damsite \'laS examined during previous studies by the. USBR and COE. These studies focused on the narrow entrance to the canyon and led to the recommendation of a concrete arch dam.. ~otwithstandifig this initial appraisal, a comparative analysis was undertaken as part of thes.e fe.asi ... bility studies to evaluate the relative merits of the following types of struc- tures at the same location: -Concrete thi·n arch; -Concrete gravity arch; and -Fill embankment. (a) Comparison of Embankment and Concrete Type Dams This analysis was based on the concrete. arch and concrete gravity arch schemes developed by the COE fh'-1975 and 1978, together with a rockfill dam alternative developed as part of the current study program. The results _of lO ... lSo· the aQalysis indicated a trend in favor of the concrete arch dam alterna· tive when compared to the gravity or rockfi11 dam alternatives. The assessment showed that a gravity dam in the narr.ow gorge would tend to behave similarly to an arch dam but would not have the flexibtlity of such a structure. The technical feasibility of a gravity dam was therefore questionable particularly under severe seismic shaking conditions. This type of dam also ,tended to be more expensive and was; therefore, not considered further. ConsidE:ration of a central core rockfill dam at Devil Canyon indicated rel- atively small 'Cost differences from a conservative arch dam., significantly thicker in cross section than the finally selected design. Furthermore, no information was available to indicate that impervious core material could be found for such a dam in the necessary quantities and within a reasonable distance. The rockfill dam was accordingly dropped frGtil. further consider_a- tion. Details of this evaluation are presented in Appendix 02. Neither of the concrete arch dam 1 ayouts are intended as the final site arrangement., but were sufficiently representative of the most suitable arrangement associated with each dam type to provide an adequate basis for comparison. Each type of dam was located just downstream of where the river enters Devil Canyon close to the canyon's n~rrowest point which is the optimum location_for all types of dams. A br.ef description of each dam type and configuration is given below. (i) Thick Arch Dam The main concrete dam will be a single center arch structure, acting partly as a gravity dam; w1th a vertical cylindrical upstream face _and a sloping downstream fa,ce inclined at 1V~0.4H. The maximum height of the dam will be 6.35 feet with a uniform crest width of 30 . feet, a crest 1 ength of approximately 1, 400 feet, and a maximum foundation width of 225 feet. The crest E:levation will be 1460. The center portion of the dam wi 11 be founded on a massive mass con-- crete· pad constructed in the excavated river bed. This central sec- tion will incorporate a service spillway with sidewalls anchored into solid bedrock and gated orifice spillways discharging down the steeply inclined downstream face of the dam into a single _large stilling basin set below river level and spanning the valley. The main dam will terminate in thrust blocks high on the abutments-. The 1 eft abutment thrust b 1 ock will incorporate an emergency gated control spillway structure which will discharge into a rock channel running well downstream and terminating at a high level in the river valley. Beyond the. control structure and thrust block, a low-lying saddle on the. left abutment wi 11 be closed by means of a rockfili dike founded on bedrock~ The powerhouse wi 11 house 4 x 150 MW units and will be located underground within the right abutment. The multi-level intake wiJ l be constructed integrally with the d.am and connected to the powerhouse by vertical steel-lined penstocks~ The service spillway will be· designed to pass the 1:10~-000-year routed flood with larger floods discharged downstream via the emer- gency spillway .. 10-14 ·I ,I I .I I I I I I I I 'I I I -...-· I I I I I •••• ~-. I I .,. I .... ·· I I ' I -. a I ........ I .. I """ If I I I I -.;,· I I { i i) Thin Arch Dam The main dam will be a two-center., doub1e-curv_ed arch structure of similar height to the thick arch dam, but with a 20-f.oot uniform crest and a maximum base width o..f 90 feet. The crest elevation will be 1460. The center sect ion wi 11 be founded on a concrete pad, and the extreme upper.portion of the dam wi11 terminate in concrete thrust blocks located on the abutments • ~The main service spillway will be located on the right abutment and will consist of a conventional gated control structure discharging down a concrete-1 ined chute terminating in a flip bucket. The bucket will discharg~ into an unlined plunge pool excavated in the riverbed alluvium and located sufficiently downstream to prevent undermining of the dam and associated structures. The main spillway will be supplemented by orifice type spillways located high i.n the center portion of the dam which will discharge into a concrete-lined plunge pool immediately downstream from the dam. An emergency spillway, cons·isting of a fuse plug discharging into an unlined rock channel, terminating well downstream, will be located beyond the saddle dan on the left abutment. The concrete dam will terminate in a massive thrust block on each abutment which, on the left abutment, will adjoin a rockfill saddle dam. The service and auxiliary spillways will be designed to discharge the 1 :10,000-year flood. Excess floods for storms tip to the prob- able maximum flood will be discharged through the emergency left abutment spillway . (iii) Comparison of Arch Dam Types Sand and gravel for concrete aggregates are believed to .be avai1able in sufficient quantities within economic distance from the dam as discussed in Sections 10.2 and 10.3. The gravel and sands are formed from the granitic and metamorphic rocks of the area; at this time it is anticipated that they wil_l be suitable for the production of aggregates after a moderate amount of screening and washing,. The bedrock geology of the site is discussed in .Sections 10.2 and 10.3. At this stage it appears that there are no geological or geo- technical concerns that would preclude either of the dam types from consideration. T~te thick arch dam will allow for the incorporation of a main spill- way chute on the downstream face of the dam discharging into a spillway located deep within the present riverbed. This spillway wi 11 be able to pass routed floods with a return frequency of less than 1:10,000 ye-ars. For the thin arch and rockfill alternatives, the equivalent discharge capacity will be provided separately through the abutments. . 10-15 Under hydrostatic and temperature loadings, stresses within the thick arch dam will be generally lower than for the thin arch alter- native. However, finite.' element analysis has shown that the addi- tional mass of the dam under seismic loadings will produce stresses of a greater magnitude in the thick arch dam than in the thin arch dam. If the surface stresses approach the maximum allowable at a particular section, the remaining understressed area of concrete will be greater for the thick arch, and the factor of safety. for the dam wiii be correspondingly higher. The thin arch is, however, a more efficient design and better utilizes the inherent properties of the concrete. It is designed around acc:eptabl e predetermined fac- tors of safety and requires a much smaller volume of concrete for the actual dam structure. The thick arch arrangement did not appear to have any outstanding merits compared to a thin arch dam and would be more expensive be- cause of the larger volume of concrete needed. Studies, therefore, continued, on refining the feasibility of the thin arch alternative. 10.9 -Diversion Scheme Alternatives In this se&tion the selection of general arrangement and the basis for sizing of the diversion scheme are presented. (a) General Arrangements The steep walled valley at the site essentially dictated that diversion of the river during construction be accomplished using one or two diversion tunne 1 s ~ with upstream and dO\oJnstream cofferdams protecting the main con- struct ion area. The selection process for establishing the final general arrangement in- cluded examination of tunnel locations on both banks of the river. Rock conditions for tunneling did not favor one bank over the other. Access and ease of construction strongly favored the left bank or abutment, the obvi- ous approach being via the alluvial fan. The total length of tunnel re- quired for the 1 eft bank is approximately 300 feet greater; however~ access to the right bank could not be achieved without great difficulty. (b) Design Flood for Uiversion The recurrence interval of the design flood for diversion was established in the same manner as for Watana (see Sect ion 9). Accordingly~ at Devil Canyon a risk of exceedence of 10 percent per annum has been adopted, equi- valent to a design flood with a 1:10-year return period for each year of critical construction exposure. The critical construction time is esti- mated q.t 2.5 years. The main dam could be subjected to overtopping during construction without causing serious damage, and the existence of the Watana facility upstream will offer considerable assistance in flow regula- tion in case of an emergency. These considerations led to the selection of the design flood with a return frequency of 1:25 years. The equivalent inflow) together with average flow characteristics of the river significant to diversion, is presented below: 0 '10-16 I I I I I I .:-~ ' J I I <-~ I ... 'I I I I ._ I ,, ..... I I I ·- 1 I I •..r' I " •• "- 1 ' I I ' I ...,. I ' I I ' .I I (c) Cofferdams Average annual flow: _ Maximum average monthly flow: Minimum averagfr~onthly flow: Uesign flood inflow (1:25 years routed through Watana reservoir): 7,860 cfs 23,100 cfs (June) 890 cfs (March) 37,800 cfs As at Watana, the considerable depth of riverbed alluvium at both cofferdam sites indicates that embankment type cofferdam structures would be the only technically and economically feasible alternative at Devil Canyon~ For the purposes of establishing the overall general arrangement of the project and for subsequent diversion optimization studies, the upstream coffe·rda.'ll sec- tion adopted will comprise an initial closure section approximately 20 feet high constructed in the wet, with a zoned embankment constructed in the dry.. The downstream cofferdam will comprise a closure dam structure ap- proximately 30 feet high placed in the wet. Control of underseepage through the relatively pervious alluvium material will be achieved by means of a grouted zone.· The nature df the alluvium led to the selection of a grouted zone rather than a s 1 urry wa 11. The selected cofferdam sections are described in more detail in Section 13, (d) Diversion Tunnels Although studies for the W;:::.an a project indica ted that· concrete-1 ined tunnels were the most ecof~O~llically and technically feasible so1ution~ this aspect was reexamined at Devil Canyon. Preliminary hydraul i·c studies indi- cated that the design flood routed through the diversion scheme v1ould re- sult in a design discharge of approximately 37,800 cfs. For concrete~ined tunnels, design velocities of approximately 50 ft/s have been used in sev- eral projects. For unlined tunnels, maximum design velocities ranging from 10 ft!s in good quality rock to 4 ft/s in less _competent material are typi- cal. Using a maximum permissible velocity of 10 ft/s, four unlined tunnels, each with an equivalent circular diameter of 35 feet, would be required to pass the design flow. Alternatively, a design velocity of 50 ft/s would theoretically permit the use of one concrete-lined tunnel with an equivalent diameter of 30 feet. As was the case for the Watana diver- sion scheme, considerations of reliabnity and cost were considered suffi- cient to eliminate consideration of unlined tunnels for the diversion scheme. For the purposes of optimization studies, only a pressure tunnel was con- sidered, since previous studies (Section 9) indicated that cofferdam clo- sure problems associated with free-flow tunnels would more than offset their other advantages. Pressure tunnels are designed to flow full and, accordingly, must withstand internal pr·essure. The most widely used type of pressure tunnel for diver- sion has the crown of the outlet portal submerged during all flow condi- tions. The tunnel cross-section used for this evaluation was a modified horseshoe or 11 0 11 sh-aped configuration. The area of this type of section is 10-17 ...• ~~··or·~ .. ·,1 (e) 13.7-percent greater than for a circular tunnel with the same diameter of span. The 11 0 11 -shaped tunnel offers advantages in terms of ease of con- .struction and scheduling for the spans envisaged at Devil Canyon. Optimization of Diversion Scheme Given the considerations described above relative to design flows~ coffer- darn configuration; and alternative types of tunnels, an economic study was undertaken to determine the optimum combination of upstr·eam cofferdam ele- vation (height) and tunnel diameter. _, Capital costs were developed for these single pressure tunnel diameters and corresponding upstream cofferdam embankment crest elevations with a 30-foot wide crest and exterior slopes of 2H:1V. A freeboard allowance of 5 feet was inc 1 uded for sett 1 ement and vJave run up. Capital costs for the tunnel alternatives jncluded allowances for excava- tion, concrete liner~ rock bolts, and steel supports. Costs were also developed for the upstream and downstream portals, including excavation and support. The cost of intake and outlet gate structures and associated gates was determined not to vary significantly with tunnel diameter and was excluded from the analysis .. The centerline tunnel length in all cases was assumed to be 2,000 feet, reflecting a left bank location. Rating curves for the single-pressure tunnel alternatives are presented in Figure 10.8. The relationship between capital cost and crest elevation for the upstream cofferdam is shown in Figure 10.9. The capital cost for various tunnel diameters is given in Figure 10.10: The results of the opt im izati on study are presented in Figure 10.11 and indicate the following optimum solutions for each alternative. Tunnel Diameter 25 feet 30 feet 35 feet Cofferdam Elevation 945 feet 945 feet 945 feet Total Cost $8,000,000 $6,600,000 $7,100,000 The selection of the diversion scheme was based on economics, a single, 30- fo-ot-diameter pressure tunnel being selected. An upstream cofferdam 60 feet high, with a crest elevation of 945, was carried .forward as part of the selected general arrangement. The various components of the selected diversion scheme are described in Section 13. 10.10 -Spillway Alternatives As discussed in Section 10.7, the project has been designed to safely pass floods with the following return frequencies: 10-18 I I t I -... I ;I' :.~ I I I I -,, I I ... ' I I I ,.. I I I ' I I I ·I '- 1 I I 'I ·I I I I t I t ' I 10.10 ~ Spillway Alternatives As discussed in Section 10.7, the project has been designed to safely pass floods with the following return frequencies: Flood Spillway Design Flood Probable Maximum Flood Ftequency 1:10,000 years ,, U i scharge ( cfs) 135,000 270,000 A number of ·alternatives were considered singly and in combination for Devil Canyon spillway facilities. These included gatd overflows or orifices in the main dam discharging into a plunge pool downstream, right and left bank chute or tunnel spillways in the flip buckets or stilling basins for energy dissipation, and emergency open channel spillways. As described ·for Watana in Section 9, the seler:tion of spillway facilities greatly influenced ar{d was influenced by the gene~"Jl arrangement of the major structures. In general, the main spilh'/ay facilities will discharge the design flood through a gated spillway control structure with energy dissipation either by a flip bucket which directs the spillway discharge in a free fall jet into a plunge pool in the river or by a stilling basin. which dissipates the energy in a hydrau.ic jump. In addition, similar restrictions apply with respect to limiting nitrogen supersaturation in the spillway discharges. The various spillway arrangements developed in accordance with these considerations are discussed in Sections 10.13 and 10.14. 10.11 -Power Facilities Alternatives The selection of the optimum arrangements for the power facilities involved consideration of the same factot~s as described in Section 9.11 for Wat.ana. The selection of the installed capacity of 600 MW at Devil Canyon is des- cribed in Section 10.6. (a) Comparison of Surface and Underground Powerhouses A surface powerhouse at Devi 1 Canyon would be located either at the down- stream toe of the dam or along the side of the canyon wall. As determined for Watana, costs favored an underground arrangement. In add it ion to cost, the underground powerhouse layout has been selected based on the following: -Insufficient space is available in the steep-sided canyon for a surface powerhouse at the base of the dam; -The provision of an extensive intake at the crest of the arch dam would be detrimental to stress conditions in the arch dan particularly under earthquake loading, and would require significant changes 1n the arch dam geometry; . and _, -The outlet facilities located in the arch dam are designed to discharge directly ·into the river valley; these would cause significant \'linter icing and spray problems to any surface structure below the dam. 10-19 -. (b) (c) (d) Comparison of Alternat ive,Locat-i-ons .. The underground powerhouse and related facilities have been located on the ·right bank for.the following reasons: -Generally superior rock quality at depth; The left bank area behind the main dam thrust block is unsuitable for the construction of the power intake; and -The river turns north downstrean from the dam, and hence, the right bank -power development· is more suitable for extending the tailr·ace tunnel to develop extra head. Selection of Units The turbine type selected for the Uevil Canyon development is governed by the design head and specific speed and by economic considerations. Francis turbines have been adopted for reasons similar to those discussed for Watana in Section 9.11. The selection of the number and rating of individual units is discussed in detail in Section 10.6 .. 1he four·units will each be rated to deliver 150. MW at full gate opening at minimum reservcir level in December (the peak month). The best efficiency unit output at rated head {575 feet) is 164 MW. Transformers Transformer selection is similar to the procedure for Hatana as discus.sed in Section 9.11. The arrangement of the transformr~rs at Devil Canyon ;·s described in detail in Section 13. Power Intake and Water Passages For flexibility and security of operation, individual pensto~ks are pro- vided to each of the four units. As discussed in Section 9.11 for Watana, it was found that there is no significant cost advantage in using t\«l · 1 arger diameter penstocks with bifurcation at the powerhouse._ A single tailrace tunnel has been assumed 6,80U feet in length, to develop · a further 30 feet of head downstream from the dam. Detai 1 ed des.i gn may in- dicate this should be changed to two smaller tailrace tunnels for improved station security; the extra cost involved would be insignificant. T'ne surge chamber design for two tailrace tunnels would be relatively un- changed. The overall length of the intake structure is governed by the selected pen- stock diameter and the minimum penstock spacing. Deta'iled studies were 10-20 . . . : . .. . ' . .. I .. : •. · ..... I I I I -;· I .1: ~· I I I I I .... ,, I I I I I I I f I i I ' I '"" I I I I ' I - I I I I carried out to determine the optimum diameter of the penstocks and the tailrace tunnel, in a similar manner to that described for Watana in Section 9.11. (f) Environmental Constraints In addition to potential nitrogen-saturation problems caused by spillway operation~ as discussed in Section 10.10~ the major impacts of the Devil Canyon power facilities development are: -Changes in the normal temperature regime of the river; and -Fluctuations in downstream river flows and levels. Temperature modeling has indicated that varying the intake design at Devil Canyon would not significantly affect downstream water temperatures, since these are effectively controlled by the water released from Watana. Conse- quently, the intake design at Devil Canyon incorporates a single level draw-off about 75 feet below normal reservoir operating level (El 1455}. The Devil Canyon station will be operated as a base-loaded plant throughout the year, to satisfy the requirement for no siginificant daily variation in power flow. 10.12 -General Arrangement Selection The approach to select ion of a general arrangement for Devil Canyon was a sim- ilar but simplified version of that used for Watana, as described in Section 9. (a) Selection Methodology Pre 1 iminary alternative arrangements of the Devil Canyon project were developed and selected using two rather than three review stages. Topo- graphic conditions at this site limited the development of reasonab1y feas- ible layouts, and initially, four schemes were developed and evaluat::r:L During the final review, the selected layout was refined based on techni- cal, operational and environmental considerations identified during the preliminary review. (b) Design Data and Criteria The design data and design criteria on which the alternative layouts were based is presented in Table 10.2. Subsequent to selection of the preferred Devil Canyon scheme, the information was refined and updated as part uf the on-going study program. The aescription of the Devil Canyon project pre- sented in Section 13 reflects the most recent design d-ata for the prcject. 10.13-Preliminary Review Cons·icleration of the options available for types and locations of various struc- tures 1 ed to the development of four primary layouts for ex.aminat ion at Devil Canyon in the preliminary review phase. As discussed above, pr'evious studies 10-21 had led to the selection of a thin concrete arch structure for the main dam, and indicated that the most .acceptable technical and economic locQ.tion was at the upstream entrance to the canyon. The dam axis has been fixed in this location for all alternatives. (a) Description of Alternative Schemes 0 The schemes evaluated during the pre.l iminary review are described below. In each of the alternatives· evaluated, the dam is founded on a mass con- crete plug, constructed on the sound bedrock underlying the riverbed.. The structure is 635 feet high, has a crest width uf 20 feet, and a maximum base width of 90 feet. Mass concrete thrust blocks ·are founded high on the abutments, the left block extendihg approximately 100 feet above the exist- ing bedrock surface and ·suppor-ting the upper arches of the dam. The thrust block on the right abutment makes the cross-river profile of the dam more symetrical and contributes to a more uniform stress distribution. ( i) Scheme DC 1 (See Plate 23) In this scheme, diversion facilities comprise upstream and down- stream earthfill and rockfill cofferdams and two 24-foot-diameter tunnels beneath the left abutment (Section 10.9). A rockfill saddle dam occupies the lower lying area beyond the left abutment running from the thrust block to the higher ground beyond .. The impervious f i 11 ::ut-off for the s add 1 e dam is founded an bedrock approximately 80 feet beneath the existing ground surface. The maximum height of this dam above the foundation is approxima~ely 200 feet. · The routed 1:10,000-year design flood of 135,000 cfs is passed by two spillways. The main spillway is located on the right abutment. It has a design discharge of 90,000 cfs, and flows are controlled by a three-gated ogee control structure. This discharges down a · concrete-1 ined chute and over a ski-jump flip bucket which ejects the water in a diverging jet into a pre-excavated plunge pool in the riverbed. The flip bucket is set at Elevation 925, approximate1y 35 feet above. the river level. An auxiliary spillway, discharging a total of 33,000 cfs, is located in the center of the dam, 100 feet be 1 ow the dam crest and is contra 11 ed by three whee 1-mounted gates .. The orifices are designed to direct the flow in to a concrete-lined plunge pool just downstream from the dam. An emergency spillway is located in the sound rock south of tne saddle dam. · This is designed to pass discharges in excess of the 1:10,000-year flood up to a probable maximum flood· of 270,000 cfs, if such an event should ever occur. The spill way is an un 1 ined rock channel which discharges into a valley downstream from the dam lead- ing into t.he Susitna River. 10-22 I I ,... t I I I t I I I I I I .... ,, I I I I I I ' I I I I I I I I I I I. I I I I I •• ' -----;------;:-~.---------;-;:-;------~--.~ .. -,--:--~-., The upstre.am end of the channe 1 ·;s closed by an earthfi 11 fuse plug.·. The plug is designed to be eroded if overtopped by the reservoir~ Thus, as the crest is lower. than either the mainor saddle dams, the plug would be washed out prior to overtopping of either of these structures. · The underground power facilities are located on the right bank of the -river, within the bedrock forming the dam abutment. The rock \'lithin this abutment is of better quality with fewer shear zones and t; a lesser degree of jointing than the rock on the left side of the canyon (see Section 10.3), and hence more suitable for underground excavation. - The power intake is located just upstream from the bend in the valley before it turns sharply to the right into Devil Canyon. The intake structure is set deep into the rock at the downstream end of the approach channel. Separate penstocks for each unit lead to the powerhouse. The powerhouse contains four 150 ~lW turbine/generator units. The turbines are Francis type units coupled to overhead umbrella type generators. The units are servi~ed by an overhead crane running the length of the powerhouse and into the end service bay. Offices, the control room, switchgear room, maintenance room, etc., are located beyond the service bay. The transformers are housed in a separate, upstream gallery located above the lower horizontal section of the penstocks. Two vertical cable shafts connect the gallery to the surface. The draft tube gates are housed above the draft tubes in separate annexes off the main powerhall. The draft tubes converge in two bifurcations at the tailrace.tunnels which discharge, under free-flow conditions, to the river. Access to the powerhouse is by means of an unlined tunnel leading from an access portal on the right side of the canyon. The switchyard is located on the left bank of the river just down- stream from the saddle dam, and the power cables from the trans- formers are carried to it across the top of the dam. (ii) Scheme DC 2 (See Plate 24) The layout is generally similar to Scheme DC 1 except that the chute spillway is located on the left side of the canyon. The concrete- 1 ined chute terminates in a ski,..jump flip bucket high on the left side of the canyon which drops the discharges into the river below. The design flow is 90,000 cfs, and discharges are controlled by a 3-gated, ogee-crested-contro 1 structure, simi 1 ar to tt1at for Scheme DC 1, which abuts the left side thrust block. The saddle dam axis is straight, following the shortest route be- twe~n the contra 1 structure at one end and the rising ground beyond ·the 1 ow-lying area at the other. - 10-23 (iii) Scheme DC 3 (See Plate 25) The layout is similar to Scheme DC 1 except that the right side main spil ha.Jay takes the form of a single tunnel rather than an open chut.e. A 2-gated, agee-control structure is located at the head of the tunnel and discharges into an inc1 ined shaft 45 ·feet diameter at its upper end .. The. structure will discharge up to a maximum of 90,000 cfs. The concrete-1 ined tunnel narrows to 35 feet diameter and discharges into a flip bucket which directs the flows in a jet into the river below as in Scheme DCl. An auxiliary spillway is located in the center of the dam and an emergency spillway is excavated on the left abutment. The 1 ayout of dams and power facilities are the same as for Scheme DC 1. (iv) Scheme DC 4 (See Plate 26) 9 The dam, power facilities, and saddle dam for this scheme are the same as those for Scheme DC 1. The major differ.ence is the subst i- tution of a stilling-basin type spillway on the right bank for the chute and flip bucket. A 3-gated, agee-control structure is located at the end of the dam thrust block and contra 1 s the discharges, up to a maximum of 90,000 cfs~ The concrete-1 ined chute is built into the face of the canyon and discharges into a 500-feet-long by 115-feet-wide by 100 feet high concrete stilling bijsin formed below river level and deep within the right side of the canyon. This arrangement forms the service spill- way with central orifices in the dam and the left bank rock channel and fuse plug forming the auxiliary and emergency spillways, respectively, as in the alternative schemes. The downstream cofferdam is located beyond the spillway, and the diversion tunnel outlets are located farther downstream to enable construction of the stilling basin. (b) Comparison of Alternatives As the arch dam, saddle dam, power faci1 ities, and diversi.on vary only in a minor degree among the alternatives, a comparison of schemes rests solely with a comparison of the spillway facilities. As can be seen from a comparison of the costs in Ta~le 10. , the flip bucket spillways are substantially less costly to construct than the stilling-basin type of ~:heme DC 4 .. The left side spillway of Scheme DC 2 runs at a sharp angle to the river and ejects the discharge jet from high on the canyon face toward the opposite side of the canyon. Over a 1 anger 10 .... 24 . 1 - I . ' I , ..... I I I I I I I I I I I t I I I I - I I I I I I I I I I I I I I I I I I I I (c) period of operation, scour of the heavily jointed rock could be a consider- able problem causing undermining of the canyon sides· and their subs.eque.nt. instabi1 ity, together with the possibility of a deposition of material downstream with a correspondi11g elevation of the tailrace.. Construction of a spillway on the steep left side of the river could be more difficult than on the right side because of the presence of deep fissures and large un- stable blocks of rock which are present on the left side close to the top of the canyon.- The two-right side flip bucket spillways schemes, based on e.ither an open chute or a tunnel, take advantage of a downstream bend in the river to eject discharges parallel to thu course of the river. This will reduce the effects of erosion but could still present a problem, as can be seen from the outline of the estimated maximum possible scour hole which would occur aver a period of time. The tunnel type spillway could prove difficult to construct because of the large diameter inclinedshaft and tunnel paralleling the bedding planes~ The high velocities~ encountered in all spillways, could particularly cause troubles in the tunnel with the possibility of spiraling flows and severe cavitation. The stilling bas in type spillway of Scheme DC 4 reduces downstream erosion problems within the canyon. However; cavitation could be a problem under the high-flow velocities experienced at the base of the chute. This \"lOU1d be somewhat alleviated by aeration of the flows, introducing air into the water/concrete contact area at offsets along the chute invert. There is, however, 1 ittle precedent for stilling basin operation at heads of over 550 feet; and even where floods of much less than the design capacity have been discharged, severe damage has occurred. Selection of Final Scheme The chute and flip bucket spill ways of Schemes DC 1 and DC 2 could generate downstream erosion problems which could, in the case of Scheme DC 2~ re- quire considerable maintenance costs an.d cause reduced efficiency in opera- tion of the pro.ject al a future date. Scheme DC 3 causes hydraulic prob- lems and cavitation co.uld be severe. There is no cost advantage in this type of spillway .over the open chute. In Scheme DC 4, the operating char- acteristics of a high head stilling basin are little known, and there are few examples of successful operation. All spillways operating at the required heads and discharges wi 11 event- ually cause some erosion. For all schemes, use of auxiliary release facil- ities in the dam to handle floods up to 1:50-year frequency is considered a reasonable approach to limit erosion and nitrogen supersaturation problems .. The cost of the flip bucket type spillway in the scheme is considerably less than that of the stilling basin in Scheme UC 4" The latter offers no relative operati.onal advantage; therefore, Scheme DC 1 has been selected for further study. · - 10-25 '"""""'''" ,_,~.,~.-·-··-_ ... •• ~. _· < ~ -. . 10.14 -Final Review The layout selected in Section 10.13 was further developed in.accordrince with updated engineering studies and criteria. The major change compared to Scheme DC 1 is in the central spillway configuration, but other modifications that were introduced are described below. The revised 1 ayout is shown on Plate 27. A description of the structures is as follows. (a) Main Dam The maximum operating level of the reservoir was raised to E·levation 1455 in accordance with updated information relative to the Watanc\ tail water level. This requir.es. raising the dam crest Elevation to 1463: with the con- crete parapet wall cr~st at Elevation 1466. The saddle dam '!'las raised to Elevation 1472. (b) Spillways ,and Outlet Facilities To alleviate the potential for nitrogen supersaturation problems, it was necessary to restrict supersaturatea flow to an average recurrence interval of not less than 50 years. In order to pass floods of greater freqency, an alternative type of discharge facility was required. In addition, it was considered probable that frequent and nance would be required in the concrete-1 ined plunge pool tral or•ifice spillyJays and just downstream from the dam. cal area because of· the proximity to the dam. costly mainte- beneath the ceo- This is a criti- These two considerations led to the replacement of orifice spillways by outlet facilities incorporating five fixed-cone valves, with a diameter of 108 inches~ capable of passing aAesign flow of 45,000 cfs. The chute spillway and flipbucket are located on the right bank, as in Scheme DC 1; however, the chute length was decreased and the elevation of .the flip bucket raised compared to Scheme OC 1. !"lore recent site surveys indicated that the ground surface in the vicinity of the saddle dam was lower than originally assumed. The emergency spill- way channel was relocated slightly to the south to accommodate the larger dam. (c) Diversion The previous twin diversion tunnels were replaced by asingle-tunne·1 scheme. This was determined to provide all necessary security but will be slightly less expensive than the two-tunnel alternative (see Section 10.9). 10-26 t ... ' I I I I t I I I I ' I I I I I I I I I ·I I ' I I I I I I I I I I I I I (d) Power Faci1 ities The drawdown range of the reservoir was reduced, allowing a reduction in height of the power intakew In order to locate the intake within solid rock, it has been mm,_erl_into the side of the valley, requiring a slight rotation of the water passages, powerhouse, and caverns comprisitlg the power facilities. 10-27 LIST OF REFERENCES , (l) Acres American lncorpqrated, Report on 1980-81 Geotechnical Investigations, February 1982. I I (2) Woodward-Clyde Consultants, Interim Report of Seismic Studies for th~ 1. Susitna Hydroelectric Project, December 1980. (3) Woodward-Clyde Consultants, Final Report of Seismic Studies for the Susitna H~roelectric Project, February 1982. ~ I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I TABLE 10,2: OE:SIGN DATA AND DESIGN CRITERIA FOR REVIEW OF' ALTERNATIVE LAYOUTS River flows Average flow (ave~.-30 years of record): Probable maximum flood: Max. flood with return period of 1:107 000 years: Maximum flood with return pe~iCid of 1 :500 years: Maximum flood with return period of 1:50 years: Reservoir Normal maximum operating level: Reservoir minimum operating level: Area of reservoir at maximum operating level: Reservoir live storage: Reservoir full storage: Dam Type: Crest elevation: Crest length: Maximum height above foundation: Crest width: Diversion Cofferdam types: Upstream cofferdam crest elevation: Downstream cofferdam crest elevation: Maximum pool level during construction: Tunnels: Outlet structures: final closure: Releases during impounding: Spillway Design floods: Service spillway ..... capacity: -control structure: -energy dissipation: Secondary spi11wa}' -capacity: -control. structure: -energy dissipation: Emergency spillway -capacity: -type: 8 1 960 cfs 270,000 cfs 135,000 cfs (after ro~ting through Watana 42,000 cfs {after routing through Watana 1,455 feet 1,430 feet 21,000 acres 180,000 acre feet 1,100,000 acre feet Concrete arch 1 ,455 feet 6)5 feet 20 feet Rock fill 960 feet 900 feet 955 feet Concrete lined Low-level structure with slide closure gate Mass concrete plugs in line with dam grout curtain 2,000 cfs min. via fixed-cone valves Passes PMf, preserving integrity of dam with no loss of life Passes routed 1 :10,000-year flood with no damage to structures 45,000 cfs fixed-cone valves five. 108-inch diameter fixed-cone valves 90,000 cfs- Gated, agee crests Stilling basin pmf minus routed 1:10t000-year flood F'use plug TABLE 10.2: (Continued) Power Intake Type: Transformer area: Access Type of turbin.es; Number and rating: Rated net head: Maximum gross head: Type of generator: Rated output: Power factor: Underground Separate gallery Rock Tunnel Francis 4 X 140 MW 550 feet 565 feet approx. Vertical synchronous HVA tr.9 I I I I I I I I I I I f I I I I I I I I I I I I I . I I I I TABLE 10 •. 3: . SUMMARY OF COMPARATIVE COST ESTIMATES PRELIMINARY REVIEW OF ALTERNATIVE ARRANGEMENTS (January 1982 $ Million$) Item DC 1 DC 2 DC 3 DC 4 Land Acquisition 22.1 22.1 22.1 22.1 Reservoir 10.5 10.5 10.5 10.5 Main Dam 468.7 468~7 468.7 468.7 Emergency Spillwa)' 25.2 25.2 25.2 25.2 Power Facilities 211.7 211.7 211.7 211.7 Switch yard 7.1 7.1 7.1 7 .. 1 Miscellaneous Structures 9.5 9.5 9 • .5 9.5 Access Roads &: Site Facilities 28.4 28.4 28.4 28.4 Common Items -Subtotal 783.2 783.2 783.2· 783.2 Diversion 32.1 32.1 32.1 34.9 Service Spillway 46.8 53.3 50.1 85.2 Saddle Dam 19.9 18.6 18.6 19.9 Subtotal 98.8 104.0 100.8 140.0 Total 882.0 887.2 884.0 923 •. 2 Camp &: Support Costs (16~) 141.1 141.9 141.£1, 147.7 Subtotal 1023.1 . 1029.1 1025,4 1070.9 Contingency (20~) 204.6 205.8 2Pj.1 214.2 Subtotal 1227.7 1234.9 1 ~30.5 1285.1 En(ineering & Administration 12.5%) 153.5 154.3 153.8 160.6 Total , 381.2 1389.2 1384.3 1445.7 . . l. . I L -·- I ... ' § i ... I DEVIL CANYON ROCK OUTCROP MAP ~: ·c.-~ ---- • { f LEGEND. LITHOlOGY: ~ ARGILUTE I GRAYWACKE .CONT~t . . . ---UMIT OF OUTCROP CONTOUR l,.&He:S: TOPOGRAPHIC CONTOUR INTER\JAL · -SOFEET '~~ -------~_,____ o 200 400 FEET SCALE t FIGURE 10 .. 3 . •• I I I ·I ·I I I I 1:1 I I I I I I I I tl. - LEGEND - -LrrHOl00'1~ D. M6IUrtE ... ~ - INFE'M£0 .CJIIIEN1JIT)QN Of' iEDOIN;/ . ·I"'l.JmmN, »P!UEEfT OP WHEJE.l«JJED ' lr.'~·~d'-~.(.:,.11"1 F'EUIIt ~~-~ ----WIIEJIE •,,,\, .. r ~Ttlt JltMI 10 FEET -=mu~~lllf'"'wtPE CONTAC'fS:- ---~'fOP OF WOQC --liTHillliMIIC,OUHED w.tffE·-WEMED S'I'RUCTURE: - • StEMt...WIJTH -~ WHERE GREATER THAN IU FEET ~· ~~tO=·!lHOWN W&DE GEOPttYSICAL SURVEYs: 1'sw·l5 ~EIISl:COON ~SEISMIC ltEfRAC'nOM SW·JS Hf11.SifAMfON a W11.S0H $..110"13 IMO,~~YDEalNSUL"WfTS SLii•U tiii,WOGOWMU-CLYDE CONSULTANTS .SEISMIC V£LOCrTY CHANGE . ·~ogo SEJS&Ic· VELOCITY IN FEE:T ·pp SECOND BOREHOLES: .IH•I --~ ·-ZCINE • SH£AR CH·I OSilft -~COM: Klfb!lfG aM·& .UI DIAMOND COR£ IC*INif AM ·Gt All .._.MOLE OTH£R: DC•I_. I ~IDISIECI:ION WIT~ ROLOGfe "of -~ DC•I ~ CIEDLOCUC FEATURE ODCIIIII!JED IN. ~ !IEC110H ._, Fl MAFIC DICE J)E!CRII!f::D IN !EcntlN Fn.SIC. DICE DE3CAIItD IN 'sa:TJOH MA1N SPILLWAY.- APPAOACH ~NEL I I l I i i l I I •.r F ~~ICDIKE /~JI!:tro> j t90W tY'-4-AZIMUTH ._.......,)8()• OF SECTION . ·_. lOOKING UPSTREAl!t CREST OF MAIN OW t:L.I4M DEVIL CANYON GEOLOGIC -SECTION DC..;3 ~OIVERS_ I_ON LJ TUNNEL ··-·--·· -~- FIGURE 10 .. 4 I ·~ I. I I I I I I .I I I I I. I I I I I Ms.D4.000- BORROW SI~E 0 I \ \ . . . - ~ ; -9,,~ ·"---. .. __/' ~ .. DEVIL CANYON. r GEOLOGIC MAP · - LEGEND tmfoc...oGY: o~.UNQI~. a. AAaUTE AND ~- ·.;~ ;.•. fELSIC DIU. WIDTH . ...,.__ -~ ~ ·~'l'••!' _ ~-10 ~~n. - - IWte OICE, .WIDTH SHOWN WHDI£ -~ . T*H--10 FEET. CC)NTAC'J'S:. -lJMIT OF OUT~ ~! - - SHEAR,·Wimt SOOWN. WHDE -GIIIIEATD "MMil; · lO RET, varTICAL ~ DIP SHOWN. J_ ..J SHEAR__ _ _ WIDTH LE_ SS THAN_ 10 FEET, !..a MB':. 1 r V£trrJCAL, EXT£NT WHEilE KNO'Iti · --fRACtURE ZONE. WI01'H SHOWN WHERE CftUTER _THAN 10 .F.£ET • 'VER'tJUL UfiL.DSi: · OIP SHOWN · . ~f JOWTS,INCUNED, VERTJCAL, ~ INC.~ -__ ( SETS I ANQ II ONLY • -EXC£PT .FOR OHJr, -,J{)IH'!S .) OTHER-~ oc-t t ! GEOLOGIC SECTION Loc.lTlON .1.\DCJ•I JOINT STATIO_N SCAl.E. I I I I I :I, I! I I I I I I I I I I :I •~ ' . ... j,_' ') .~· N JOINT STATIOl\'1 DCJ·2 N•fOO tEl . \1 .. · ~.. r---. z + c:::.-a.../ Q-::j- COMPOSITE .JQSNT PLOT SOUTH BANK . N•479 w s .~~~ ...,t ~-.~··~~~==~~~N~ ./.. · •• ·~-4 ·~-----···-~-~-NT-_ .... S1i-~TION---...._ · ·.·· OCJ-4 ~------DC.I-1 N•IOO .. ~ ... ~ ' . . .. .. ( • . . .• ~-----.U. SITI(A ··~. RIV;;___.-· .. ---~~~~---· ••• ~ ' DC.J·2 --..____,..,r-----.. r---....__... _ ___,_ __ ~··· J ,.. ~01NT STATION !)CJ-3 ti•IOQ DEVIL CANYON JOINT PLOTS H COMPOSITE JOINT PLOT NORTH 8ANN ,. .. 714 w NOTES I. alNTIUIS ARE PEftef;NT OF oiOIJiil'S Pfllt 1~ OF AAEA. CONTCUR tNT'EJM'L.-l,~,T,IO,IIS.a~. 2. N EQUAL! NUMBER OF DATA POIITS. 3. COIIIIPOSITE PLOTS INCQIIHIORATE .AU.. ..KliNT O.TA. .!OINT STAnCH !ILOTS · ®NTAIN DlTA •f'RQii SPECIFIC .JOiNT STAnoNS. . 4. FOR <!OINT PLOTTING METHOO SEE FIGURE 10.6 " -·~ I I I I I I I I I I I I I I I I I I I 10!50 1000 0 TYP. TUNNEL SECTION I-PRESSURE TUNNEL 900 ~------~--------~--------~------ 880 30 35 rOR-:36,000 CFS AT DEVIL CANYON DEVIL CANYON DIVERSION HEADWATER EL./ TUNNEL DIAMETER 40 FIGURE to.aiAilliJ . ---------------------------------:'( -•' ~ I I I I I I ~. . I . I I I I I I I I I I ~~----~------~------~--------------~~~~· . l075 - 1050 I -...,: 1000 u. -z 0 -... ~ IA.I ..I &&I 2 <( Q 950 ~ ~ ,'' 900 v v / ./ / v v / I I . . I ' I \,1 J . I I t ' -I l I l ' I 2 3 4 5 6 1 8 9 10 CAPITAL COST # x 10 6 DEVIL CANYON DIVERSION UPSTREAM COFFERDAM COSTS FIGURE 10.9~~~~mt . l' I I I I I I I I I I I I I I I I I -., I s- . ~ .. ;I' ~ 8 .) ...... ~ . -U) 0 - .!!( t() lri 7 -z U) - ~ 0 TUNNEL (.) /-ci -0 6 .. ~--~ 1-~ ..;~ -5 ~--------+---~~~~--------~----~--r---~--~ u:>o I ~ t X M -~ U) 0 --~ -+~COFFERDAM . . . ~ 4 ~--------~--~--~~--------~------~r--~--~~ 3 ~---------~------~~--------r-------~r-------~ 2 ~-------~r---------~'~-------r----------r--------- 0 TYPICAL TUNNEL SECTION COFFERDAM CC STS INCLUDE DAM l EIGHT r::' AE:IOVE HEAC WATER ELEVATION FO£; FREEBOARD 0 ----------~-----------------~--------._~----~ 0 20 25 30 35 40 TUNNEL DIAMETER (FT.) DEVIL CANYON DIVERSION TUNNELS COFFERDAM COST I TUNNEL DIAMETER FIGURE 10.10 [il I I I .I I I I I I I' I I I I I I I I •• 2Q 18 16 14 -U)~ - ~ 12 -t- CJ) 0 '-? 10 6 4 0 TYPICAL I TUNNEL SECTION r-- 0 .. . - ' I ' l I ~ l I ------~ i I I ' f I . ~· 17' ~- i .)/ 20 25 30 35 40 TUNNEL DIAMETER (FT.) DEVIL CANYON DIVEHSION TOTAL: COST l TUNNEL DIAMETER . FIGURE: 10.1( lii}. '' I I I I I I I I I I I I I I I I I I . .. 11 ... SELECTION OF ACCESS PLAN: This section describes the process of formulation :; . selection of an access platl for the Susitna Hydroe1ectric Project. The methodology for comparison of alt~rnative plans is outlined, and an evaluation of each basic plan is pre- sented~ considering schedule, costs, and biological and social impacts. The selected plan is described in detail, and recommendations for measures to reduce impacts are presented. · Engineering studies conducted on the alternative routes consisted of development of design criter~a, layouts of the alternative routes, preliminary field inves- tigations, cost r.:~5timates of constructing the alternative routes and logistics costs in transpor1:ing supplies and materials to the dam sites. Environmental studies included Identification and evaluation of biological impacts for each of the alternative .·cutes. The environmental studies included field investigations and assessments for all the alternative routes. Social or socioeconomic studies included a public participation program among the various studies. Public con- cerns and pre-,-~rences, particularly those of the sector that wou1d be impacted the most directly, were solicited and fully considered in the evaluation~ The evaluation of the alternative plans included development of selection cri- teria, comparisons of the alternative plans, identification of conflicts among the alternatives in the evaluation criteria, comparison of the conflicts in the criteria, and the tradeoffs made in the evaluation. 11.1 -Background (a) Existing Access Facilities As discussed elsewhere, the proposed Devil Canyon and v!atana sites are lo.- cated approximately 115 miles northeast of Anchorage and 140 mi 1es south of Fairbanks.~ The Alaska Railroad, which links Anthorage and Fairbanksl passes wi'thin 12 miles of the Devil Canyon site at Gold Creek. The George parks Highway (Route. 3) parallels the Alaska Railroad for much of its route, although between the communities cf Sunshine and Hurricane, the Highway isorouted to the west of the Railroad, so that Gold Creek is situ- ated approximately 16 miles south of the intersection of the railroad and highway. A portion of the highway between Lane and Hurricane, known as the Parks Highway, passPs through Denali State Parkfl At Cant\'/ell 51 miles north of Gold Creek, the Denali Highway (Route 8) leads easterly approxi- mately 116 miles to Paxson~ intersecting the Richardson Highway at this point:: Tb the south, the Glenn Highway (Route 1) provides the main access to Glennallen and intersects the Richardson Highway which leads south to Valdez. (b) Modifications to Plan of Study The ori gina 1 P,OS proposed that a single route \'IOUl d be selected by May 1981 to be followed by detailed environmental investigations of this route. 0 Early in the study, three main access corridors were developed. Cons idera ... tion of these plans on the basis of information received, c<1mments and con- cerns from various state agencies and a recommendation from the Sus itna Steering Committee, led to a decision to assess three ,atternative routes in more detail throughout 1981 and recommend a selected route later in the yea\r. Accordingly, this assessment included environmental studies, engi- neering studies, aerial photography, drilling and geologic mapping of all three alternatives rather than the single route initially envisaged. 11 ~2 -CJbject ives The finally selected access plan must allow for the efficient and timely under ... taking of construction and maintenance activities in order that the Susitna Hydroelectric Development can be completed and electric power be reliably and continuously provided to the Railbelt area of Alaska. In meeting this basic objective, several specific objectives were developed as a basis for· evaluation of the alterna .. ive access rt;lr~~s. These objectives are: -Tc: allow tne construction of the Susitna project to pro('eed on a schedule that would supply the necessary power to the Rail belt Area .~f A1aska when needed; -To minimize cost. including construction costs of the access facilities themselves, logistics costs for support of construction activities as well as the logistics of subsequent operation of the completed project; -T~ allow for ease of operation and maintenance to ensure reliability in-t~e power supply; -To minimize adverse biological impacts; -To accommodate the preferences of local communities; and -To accommodate the preferences of Native landowners; 11.3 -Approach The approach uti ·1 i zed to arrive at an access recommend at ion vtas an adapt at ion of the generic plan formulation and selection methodology described elsewhere in the report. ·The methodology as specifically applied to selection of the access plan is presented graphically in Figure ll.l. To aid in understanding the selection process and the various studies conducted, the following definitions are provided: (a) Corridor A strip ~f land generally 2 mile~ or greater in width leading between two points -or areas. 11-2 ,·} -· 'I' i1 ~-. -_: . :/"': .: cl I I I I - I I I I I . I I :. I I I ;I -· '.1 I; I I I 1: I I I I I I I I I I. I I I I (b) (c) Route A strip of land generally l/2 mile or less in width~ leading between t\'/0 points. Segment Portions of a route which when combined constitute one alternative route between two points. (d) Alternative Route One of several routes which will be evaluated between two points. (e) Plan An access plan which will involve one or more or a combination of existing and new al:ernative routes. The plan will also define the logistics in- volved in the transportation of supplies and materials. 11.4 -Corridor Selection and Evaluation The first step in the selection process involved identification o'f the three general corridors shown on Plate 28 and described below: Corridor ,, 1 2 3 Descri gti on From the Parks Highway to the Hat ana site via the north side of the Susitna River From the Parks Highway to the Watana site vi a the south side of the Susit:1a River From the Dena 1 i Highway to the Hat ana site. These corridors were selected based upon the use of existi1'1g transportation faci 1 ities within reasonable proximity to the Watana and Devil Canyon sites~ A general environmental analysis was undertaken for each corridor.. The majcr environmental constraints identified within each corridor are potential impacts on the following: (a) Corridor 1 -Fishery resources in the Susitna and Indian Rivers; -C1 iff-nesting ·raptors · near Portage Creek and Devi 1 Canyon; -Fur bearer habitat near Portage Creek and High Lake; -Moose habitat on the Susitna River; and -Caribou habitat between De vi 1 Creek and Deadman Creek. Tl-3 c. ·. '1 1 I I i .. , (b) Carr idor 2 I ) t c . -Fishery resources in the Susitna and Indian Rivers; -Cliff-nesting raptors near south side of the Susitna River; -Waterfowl habitat in the Stephan Lake-Fog lake areas; and -' Furbearer habitat in the Stephan Lake ... fog Lake areas. Corridor 3 ~ Caribou calving ar~a near Butte Lake; ~ Fur bearer h ;;bit at; and -Some waterfowl habitat. In addition, increased access will cause various impacts which are common to all corridor·s. ·Archaeological resources could pose a constraint; a_t this time, the locations of such resources that may exist are unknown. · ,. Finally, socioeconomic impacts will vary both in magnitude and areas of concen- tration, depending upon which access route or combination of access routes is selected, and whether a road or railroad is used. With the socioeconomic assessment of access schemes 5 there is more concern with the origin and type of access than with the actual route, because these will affect communities throughout the south-central part of the State. With a road from the Parks Highway to the damsites (Corridors 1 and 2), effects generally would be concentrated on the western side of the prcject area. An easily accessible road corridor would provide for transportation of construction materials, equipment, and labor as well as post-construction uses of the upper Susitna basin' (such as recreation). The impact of a railroad from the S<;lme side .would likewise be conceritrated on the western side. However, in every socio- economic category, impacts would be the same or less than with the road. The single exception would be in rail industry ~ctivities, which would experience major changes. With a road constructed from the Denali Highway to the damsitt~S (Corridor 3), impacts along the Parks Highway-Alaska Ra~lroad corridor would depend upon whether materials were to 1Je shipped by road or rail to Cantwell before being transported along the. Denali Highway to the access road. Impacts would occur in · the Cant we 11 area, however, regardless of transport at ion mode. 11.5 -Route Selection and Evaluation Following identification of three major corridors, a number of access routes were selected and evaluated based on engineering and economic criteria .. The selected routes were then modified on the basis of an environmental· analysis. ' (a) Design Criteria Construction of the Susitna project will require a dependable safe and efficient access route suitable for transporting personnel, consumable sup- plies and large pieces of equipment for an extendea period, in adverse weather conditions. 11-4 'I I I I I I I I I I I I I .I ~· I ·- :I I ~ ·-1 I I I ··I I I I I I I I I I I I I I I (b) (c) The preliminary design criteria adopted for access road and rail alterna- tives were selected on the basis of similar faci1tties in remote projects of this nature. Basic par·ameters were as follows; Maximum Grade Maximum Curvature Design Loadin~ -During Construction -After Construction Access Road 6 percent ·s degrees Railroad 2.5 percent 10 degrees 80k per axle & 200k total not appl icab1e HS-20 E-50 Following corridor definition., various segments that met engineering cri- teria were mapped. These segments were then jointed to form alternative routes which were compared on the basis of: -overall length; -average grade per mile; -average deflection per mile; Economic Criteria -For the early stages of corridor and route selection, the alternatives were compared on the basis of total centerline length of route, with minor ad- justments for average grade and curvature. Preliminary capital costs for · construct ion ~"/ere estimated at $1,250,POO per mile. Results A total of 16 segments, combined into 30 routes were ldentified within the three corridors. The alternatives identified as being most favorable in terms of overall length, grade and alignment are as follows: Type over a 11 1 ength Average Corridor 1 Parks Highway to Watana north side Road 64.9 miles Grade · 2.4 percent Deflection per Mile 7° 06'+ Corridor 2 Parks Highway to Watana south side Road 66.5 miles 2.2 percent 11-5 . Corridor 3 Corridor 2 Denali Highway Parks Highway to Watana to Watana south side Road Rail 39.1 miles 58.0 miles 1.3 percent 0.5 percent 1° 30'+ , .,...,..,...,.....,,.._,..~~c ... s·c . .-•. 1 .. ·. ,· .,. . ' ' . . ·.··:·.··.·· . ' ' . ";:; ' ' •••• (d) ·. Environmenta 1 Influences on A 1t~~rnati ve Routes After the engineer·ing and etonomjc assessment identified the 3 roads and 1 rail route descri'Jed above» an initial screening was made which resulted in several refinements to the alteY'native routes under consideratjon. A major refinement i nvo 1 ved the deletion of a 1 arge portion of the road access cor- ridor to the Parks Highway on the north side of the river (Corridor 1)" The segment connecting the hi~1hv1ay and the Devil Canyon site routed around Portage Creek was deleted mainly on the basis of potentially severe envi- ronmental impacts on anadromous fish~ fur bearers, and raptors. The topo- graphy in the Portage Creek area is furthermore such that the overall length of road necessary to meet the established criteria was excessive. In addition, the construction of the segment would be extremely difficult due to the predominance of steep sidehill cuts of considerable height. Another major refinement to the corridors was the routing to the west of the northern portion of the Denali route (Corridor 3). This routing was advocated on environmental grounds in an attempt to reduce potential im- pacts on the caribou subherd calving area near Butte Lake.. A final refine- ment consisted of realignment of the portion of the Corridor on the south side of the river (Corridor 2) in the Stephan Lake-Fog Lakes area to reduce potential environmental impacts to fur bearers and \~aterfow1 • · The main routes within the corridors remaining after the initial screening are shown on Plate 29 and are briefly destribed below: ( i ) ( i i} (iii) Parks Highway. to De vi 1 Canyon This route follows the existing portion of the Alaska Railroad be ... , tween Gold Creek and the intersection of the railroad with the Parks Highway just south of Hurricane. This route passes through Chulitna Pass from the Parks Highway and then parallels the Indian River to Gold Creek. The existing river valley is sufficiently wide to ac- commodate a road. From Gold Creek to Devil Canyon, the route lies south of the Susitna River,_ paralleling the river on a high ridge. Devil Canyon to Watana -South Side of Susitna River This route generally parallels the Susitna River and traverses east- west from Devil Canyon to Watana. The topography is mountainous and the route i nvo 1 ves the most difficult construction of ·the three routes, requiring a number of sidehill cuts in rock and soil. This route also includes the environmentally sensitive Stephan Lake and· Fog Lake areas. Devil Canyon to Watana -North Side of Susitna River This route generally parallels the Susitna River and traverses east- west from De vi 1 Canyon to Hat ana. This route is mountainous and in-- eludes terrain at the highest elevations of a11 routes, however con- struction of the road would not be as difficult as the route between the damsites south of the Susitna River. ll-6 -" I I I I I I I ·I I I .I I :1 :1 I '1.' . ' I I I I I I I I. I •• I I 'I I I I I I I I I I (i v) .Denali Highwa,x to Watana This route connects the Denali Highway with the t~atana site and runs. in a north-south direction. This route is the easiest to construct of the alternative routes. The terrain is relatively flat with a few wetlands involved. This route would not require .any major bridges~ 11.6 -Description of Basic Plans From the three routes remaining after the initial screening, eight pl.ans were developed. These plans were evaluated in more detail than originally planned in the original POS, as a result of information and assessments conducted during the study program, the concerns of state agencies, and recommendations of the Susitna Steering Committee (refer to Appendix D). The additional investigation and evaluat·ions consisted mainly of environmental field work, geologic mapping and subsurface borings • ' < The plans are presented below and are also shown schematically in Figur·es 11.2 through 11 .. 5. (a) Plan 1 This plans utilizes a roadway from the Parks Highway to ~latana Dam C.\long the South side of the River. This access plan is based on materials such as cement and steel being brought into the State through the port of ~~hittier. Food and other camp supplies would be imported through Anchorage via container, and fuel directly from Kenai to Al)chorage via existing pipe- line. These materials and supplies would then be carried by rail to a railhead and storage area at Gold Creek. At Gold Creek, materials \~auld be transferred to trucks for transport by road to the site. Other materials and supplies \1/oul d be transported by truck from the Parks Highway. An a 1- ternative for fuel supply would be rail haul from the refinery at North Pole:~ Alaska. (b) Plan 2 -All Rail This plan \vou1d serve both damsites by a rail line. This alternative would essentially preclude public access. Construction planning for this mode of access would be based on trains being broken down and cars dropped on the siding at Gold Creek. An engine and train crew would be stationed at Gold Creek which would allovl shuttle cars from Gold Creek to the project s1te on a daily basis. Passenger rail service would be required daily. If public access is desired after construction, the rails could be removed and the road bed graded into a single lane road with turnouts. (c) Plan 3 This plan envisages the use of a combination of rail and road transporta- tion. Construction activities at Watana would be served from a railhead and storage area at Cant\'lell by truck across the. Denali Highway and a1 ong. a 11-7 I ,j ~ . . i I 1 I newly constructed road from the Dena.l 1 Highway. Construct ion at De vi 1 Can- yon would be served by toad from a railhead at Gold Creek and road access from Gold Creek to the Parks Highway. This plan does not include a connec- tion between the two darns. (d) Plan 4 :1 j This plan serves Watana by truck from a railhead at Cantwell ana Devil Can- yon by rail from Gold Creek. In the plan, there is no connection between dams. (e) Plan 5 This plan ser:ves both dams by road from a railhead at Gold Creek. The route is located on the south side of the river to Devi 1 Canyon with i major bridge downstream from the dam site, then follows the north side of the river to Watana. There is a road connection to the Parks Highway from Gold Creek. (f) Plan 6 This plan is identical to Plan 4 except that a service road for maintenance purposes is inc ·1 uded on the north s 1 de of the river between the two dams. (g) Plan 7 This plan is thf; same as Plan 3 except that a service road would be pro- vided along thB north side of the river as in Plan 6~ - _(h) Plan 8 This plan is the same as Plan 5 except there _is no road connection to the Parks Highway~ A. newly constructed road would service Devil Canyon from Gold~ Creek on the south side of the river. A major bridge would be re- quired downstream of Oevi 1 Canyon and a new road on the north side of thE: river would connect ·the two dams~ This alternative plan precludes public access. I I •• I I I I I I 11.7-Additional Plans I Following selection and evaluation of the eight plans described above, presenta- tions were made to the Alaska Power Authority and the Susitna Hydroelectric Pro-I ject Steering Committee. These present at ions and subsequent discuss ions re-.· . sulted in the addition of the three plans described below. (ill Plan 9 I This plan is the same" u-s Plan 8 excet:t access between Gold Creek and Devil 1 ___ ... Canyon is by rail along a similar route, and the railhead is located at • - Devil Canyon instead of Gold Creek. " I I 11-8 I I I I I I I I I I I I I I I I I I- I (b) Plan 10 - This plan is identical to Plan 9 except that ~he road connecting Devil Can- yon and Watana is on the south side of the Susitna River. (c) P1 an 11 This plan utilizes a railhead at Cantwell~ road access via the existing Dena 1 i Highway, a road from the Dena 1 i Highway to Watana and a road from Watana to Devil Canyon on the north side of the River. These plans are shown schematically in Figures 11.6 and 11.7. Plans 9 and 10 suggested by the Steering Committee as a means to reduce accessi ... bility to the area thus avoiding the introduction of adverse environmental im .... pacts into the Susitna Basin. Plan 11 \>Jas added as a possible way to provide access from only one area while also alleviating the socioeconomic impacts the west side communities would feel from any access road from the west. 11.8 -Evaluation Criteria The specific objectives of the selected access plan are described in Section 11.2. The criteria used to assess the degree that any given plan satisfies these objectives are described in the following paragraphs. (a) Construction Schedule It is essential that the selected access plan be adequate to meet the over- all project scheduling requirement's. The load forecasts de~cribed in Section 5 together with the examination of the existing system· and future generating options indicated a requirement for first power from Watana in 1993. A delay in the on-line date by one year would mean that another source of fossil fuel generation would have to be constructed, combined \'lith retirement of some fossil fuel generation a year later, into there .. serve category. In terms of present 1110rth, a delay of one year would in- crease the cost of the project by approximately $50 mi 11 ion. Analysis of the construction schedule requirements for rlatana demonstrates that all-weather access route to the site is required by mid-1986 if the on-line date of 1993 is to be maintained. For the purposes of these studies, it has been assumed that an FERC licence to construct the project will be received at the beginning of 1985, and the start of permanent work on the project Will coincide with this date. In order to meet all the mid- 198"6 requirements, it is obvious that an access route to the site would have to be constructed v1ithi n approximately 18 months. A preliminary evaluation of the construction period for completion of the access plans is presented below. ll-9 Plan 1 2 3 4 5 6 7 8 9 10 11 Origin for.Watana Access Parks Highway Gold Creek Dena 1 i Highway Dena l i Highway Parks Hi ghvray Denali Highway, Go1 d Creek Dena 1 i Highway, Parks Hi gh\'tay Gdl d Creek Gold Creek Gold Creek Denal i Hi gh~-Jay -Approximate Construction Period (,years)_ 3-4 3-4 1 1 3-4 1 1 3 3 3 1 It is apparent from the above that only Plans 3, 4, 6, 7 and 11 could be constructed within the 18 month peri o.d required to maintain the over a 11 project schedulee Since this would severely limit the selection process, a scheme was developed to provide initial access to the Watana Site within the framework of regulatory and scheduling restraints. This scheme, des- cribed in more detail in'Appendix D involved construction of a pioneer road to Watana from either Gold Creek or the Parks Hi gh\'lay. The pioneer road would consist of a gravel based road with period passing turnouts and \WUl d be constructed on existing ground insofar as possible to avoid significant cuts or fills. Temporary floating Barley bridges would be used at river crossings, replaced by ice crossings in the winter. The analysis indicates that the pioneer road scheme will be sufficient to provide continuous access to the site within J8 months, and will be sufficient to support con!:t,ruction activities. until the permanent access route is completed .. Certain additional licensing and permiting requirements are associated with this scheme; these are discussed in Section 11.12. The pioneer road scheme can be implemented with Plans 1, 2, ~' 8, 9 and 10, therefore all 11 plans can be considered equivalent in terms of their abil- ity to meet initial project requirements. (b) Construction and Logistics Costs For the purposes of this evaluation, construction costs include the cost of constructing the access facilities, adjusted for any differences in ·c~st of constructing the Susitna project itself which relate to the particular access plan under construction. Logistic costs are +.he costs associated with transporting, labor, fuel, equipment, materials and supplies to con- struct the two power developments. (c) East of Operation and Maintenance This criteria relates to the relative ease of operation and maintenance of the two developments after construction is complete. Initial planning en- visa.ges operation of both developments from Watana for several years aft.er Devil Canyon is brought on-line,. after which time, both projects will be 11-10 til ·l ~i . -_: ' I I I I I I I I I I I I I I I I I I •• I I I I ~ .• I I I I I' I I I I I I I ..• ; operated remotely from a ~entra1 location. Maintenance of. two projects uf this size and complexity wi 11 obviously be an important consideration· •. Duplication of maintenanc~ facilities and staff at both. sites \~auld tesult in a substantial incr~ase in the annual costs of the overall development. Th~'2 most economic scheme) given the sequence of development$ would be to· est~ablish an operation and maint~;nanc~ facility .at Watana, \.Yith a reliable means -of access to De vi 1 Canyon 32 mfles dovmstrearn,~ In this regard, acci'eSS plans With a road connection between the tv1o sites have been eva·luated as being superior ir1 terms of ease of operation and maintenance than plhns without a road connection. (d) Flexibility in Construction Logistics and 'Transportation This criterion is used to evaluate the extenf to which an access plan con- tributes to the maintenance. of a reliable and flexible logistic support systt:m duri.ng construction of Hatana and De vi 1 Canyon. For the Susitna project, a fundamental consideration is whether or not to provide a road connection to a major highway. For this evaluation, the following alternatives have~een considered: ' - a road connection either to the Parks Highway or Dena 1 i Highway; -rail access only from Gold Creek; Plans 1 through 10 described in Section 11.3 all include a railhead and storage area at Gold Creek. Accorindgl_y, plans incorporating a road con- nection with the Parks or Denali Highways obviously provide· greater flexi- bility and reliability in case .of a transportation description invo1'ling the Alaska Railroad, compared to plans \~ith "rail only 11 access. Specific considerations are as follows: -Any breakdm·m in the rail system would result in a loss of all ground transportation, in the absence of an alternative road system •. The in- creased risk of delays has an associated cost penalty. An analysis has been undertaken· to quantify the risks associated with rai 1 access onl..v. The methodology for this risk analysis is presented in Appendix D.,· ·~ -The availability of two possible manes of transportation will undoubtedly be reflected in lower and more competitive bids for constructi'onj supply and service contracts, si nee contractors waul d othervri se include some contingency to cover trnsportation disruptions. Although significant, this aspect is difficult to quantify. (e) Environmental Considerations Exclusive of socioeconomic considerations, the objective is to develop an access plan which minimizes adverse changes to the natural environment. The cri ter·i a used to assess the degree to which any plan meets this objec- tives are described below. 11-11 0 { i) Effects of Bi.9. Gam~ A primary concern associated with' the selection of an access plan is the potential effect on the Nelchina caribou he~d, specifically the subpopul ation of approximate_ly 1,000 animals that inhabit the north ... western sect i-on of the .. Up.p£r __ Susitna. Basin.. The impacts of hunters · on moose. and bear at~e also cons ide red but as secondar-y concern_s. These impacts can be greatly 1 essened by se 1 ect ing a route other than the access from the Denali Highway. • ( i i) Effects of Fisheries In the case of resident fisheries, there are relatively isolated lakes (Butte Lake, Big Lake) and streams in the-northwestern section of the Upper Susitna Basin, and the Fog Lakes area that would re- ceive additional angling pressure if road. access was provided. These impacts can be lessened by avoiding access from the Denali Highway and the route on the south side of the Sus itn a River between the dams ites. · Since Devil Canyon acts as a natural barri.er to anadromous fish migration, there is no concern regar ... d ing the effect of improved access on this resource upstream of Devil Canyon. However, Indian River and the Susitna River below Portage Creek,-are important for salmon. Any access plans that follow or -cross these rivers could affect salmon directly through habitat disruption (i.e., sedimenta- tion) or indirectly through increased fishing pressure. These im- pacts could be les.sened by avoiding road access paralleling the Indian River. · (iii) Effects on Furbearers Wet 1 ands, important to furbearers, have been identified between the Parks Highway and Gold Creek, near Deadman Mountain, near Deadman and Big Lakes and the Upper Deadman Creek. In add it ion, the Fog Lake -Ste-phan Lakes wetlands complex is a valuable furbearer habi- tat. A red fox denning complex has also been identified south of Deadman jvfountain. Any access road crossing through these areas has the potential for negative impacts on fur bearers. Impacts on fur- bearers would be least by selecting access from Gold Creek to Devil Canyon on the south side of the Susitna River and on the north side of the River between the dam sites. · (iv) Effects on Birds Heavily forested areas between the Parks High\·lay and Devil Canyon along riverbanks are productive avian habitat. Construction through these areas ~tlould disturb this habitat. (v) · Effects on Wilderness Setting The Upper Susitna Basin is presently in a state of wilderness to 11-12 I I I I I I I I I I I I I I ,, 'I I I I ·.- ~ ••. --,_. c-~ •h ,.: ' i'~ I . I I I I I :1 I I. I I I I I I I {f) - semi-wilderness. Although continued intrusion with ATV~s from Dena 1 i Highway, potential deve1 opmeot of native 1 a nds and the estab- lishment of the Indian River remote and disposal site have the po- tential of changing the character of sections of the basin, improved public access and construction of the Susitna Hydroelectric Project will produce.~_major alteration in the remoteness of the area. Nat_- ural resource agencies an~ the local public have expressed a desire to maintain the status quo to the maximum extent possible. People from the urban centers of Anchorage and Fairbanks have expressed a stronger desire to pro vi de road etc cess and open the area for recrea- tion development. The status quo of the area would be retained to the greatest extent by pro vi ding only rail access to the damsites .. (vi) Effects 6"n Archaeol ogi ca 1 Resources ,· Archaeological resources are likely·present along all-access routes. The route from Denali crosses a substantial area of high archaeolog- ical potential. The thin soil and lack of vegetation result in a high potentia 1 .for impacts to resources a 1 ong this route. The other access routes are believed to be less sensitive. Avoidance of the Denali access link lessens the probability of the disturbance of archaeological sites. Social Considerations (i) Native Landowners Native organizations have selected land surrounding the impoundment areas and south of the Susitna River between De vi 1 Canyon and \~atana damsites. To allow for increased opportunity to develop either lands on the south side of the river, the native landowners have ex- pressed a strong desire to have a Susitna access road along the south side of the river from Watana to Go 1 d Creek, ultimately con- necting to the Parks Highway. It is considered that the basic native preferences waul d be met by pro vi ding road access to both damsites. (ii) Local_ Community Preferences Since the local communities are likely to receive many of the dis- benefits with. few of the benefits of a Susitna development, the ob- jective to accommodate local community preferences has been included in the access plan selection process. The criteria used in assess- ing the degree to which this objective is met is divided into four areas due to the differences in community preferences (refer to Appendix D). -Cant\'ie 11 The community of Cant\'lell desires economic stimulus and is in favor of the economic .changes that could result from having a major construction project in the area. The desired stimulus could be achieved by providin.g road access to the Dena)i Highway \'lith a rail head at Ca ntwe 11. 11-13 -t .:. Morth, of Talkeetna The communities along the railroad north of Talkeetna are opposed to deve\ opment in the area and strongly prefer a maintenance of the status quo. These communities have expressed a desire for -rail access only, although existing conditions in these communi- ties would probably be disrupted least with a plan involving road access only from the Dena 11: Hi gh\~ay. This conclusion is based on the consideration that if rail access only is provided, the practicality of a self contained family- status community at either of the sites would be greatly diminished and a single-status on1y camp facility would like.ly be established. If this \'/ere to be the case, workers wou1 d tend to locate their families in the nearest communities, thus increasing the impacts in these communities.- -Ta 1keetna/Trapper Creek _ Although attitudes are somewhat divided, the-majority of residents of the communities of Talkeetna and Trapper Creek prefer a main- tenance of the status quo. This can be most easily accomplished by providing access via Denali Highway. -Wi llow/Wasi la Area 0 The residents in this area are more in favor -of economic develop- ment than in other areas. -Indian River Land Disposal· Sites In 1981, a total of 75 remote state land parcels \-Jere awarded by 1 ottery in the Indian River area. Of these, 35 ~vere staked in the summer of 1981. The 35 land holders were contacted by letter through APA Public Participation Office. Of the 12 responses re- ceived to date, 11 favored retention of the remote status of the area and one favored road access to the area. This area would be most affected by road access fi om the Parks Hi gh~;1ay and 1 east a f- fected by access from the Dena 1 i Highway. (g) Agency Concerns " Correspondence, meetings and interaction with the various agencies involved with the Susitna Hydroelectric Project Steer·i ng Committee occurred through- out the study. Agency comments have been considered in the evaluation. T~e concerns of the agencies have generally related to environmental issues, with the emphasis on biological and land use impacts .. Therefore, evaluation in terms of the environmental criteria discussed previously is considered to generally include agency concerns. It is considered that the resource agencies favor a rail only access plan with a major opposition towards road access from the Denali Highway. 11-14 9.' I I ·-1-·, ' . I I J· I I I ~~·c.- • . I I I I 1 I I I I • ·-. ,~...{& .... I I I ~ ... , I •• I I I .-· I I I I I I I I I I -~ (I {h) Transmission Access plan selection has been coordinated with the transmission line studies. The transmission line studies to d,1te have identifi.ed two corri- dors, one north of_the Susitna River and one south of the Susitna River from Watana to Gold Creek. Although corridors run along the-river, ther·e is flexibility to expand the corridor to include the access road when the decision on which access route will be constructed is made. Due to more stringent engineering criteria of lines and grades for road alignments~ it was decided that the selection of a transmission line route would occur subsequent to the access road selection. The results of the transmission studies has also established that if the northern Denali acc~ss route is selected, the transmission line would not follow that route due to excessive cost and adverse visual impacts. ( i) Recreation In meetings, discussions, and evaluation of recreation plans, it has become apparent that the various recreation.plans are sufficiently flexible to accommodate any access route selected. No single route was identified which had superior recreational potential associated with it. Therefore, compatability with recreational aspects vJas essentially eliminated as an evaluation criteria. 11.9 -Evaluation of Access Plans The 11 access plans evaluated on the basis of the criteria described in Section 11.8 have bee.n grouped in accordance with the following categories 1n order to clarify the presentation. Category Plans providing access from both Parks and Denali Highways Plans providing access from Parks Highway only Plans providing access from Denali Highway Access from Gold Creek only Plan Numbers 3 and 7 1 and 5 4, 6 and 11 2, 8, 9 and 10 In addition to the specific considerations outlined in the following paragraphs, a major concern for all access plans is the creation of access to areas pre-· viously inaccessible or relatively inaccessible. Such access would lead to im- pacts to furbearers through increased trapping pressure and to big game through hunting pressure. In addition, detrimental effects could occur to all \•dldlife · through disturbance and destruction of habitat by ATVs. Cultural resources would also be vulnerable to amateur collectors and ATV traffic. 11-15 (ej · Access to Both Parks and Denali Highways (Plans 3 and 7_l · (i) Cost The costs of the 11 alternative access plans are summarized in Table 11.1 .. Given the preliminary nature of the fielddata used to develop construction costs, construction cost differences of less than $10,000,000 (approximately 5 to 10 percent of the cost of the alternatives examined) should not be considered significant. Maintenance costs are a small portion of construction costs, and large variations in maintenance costs will have negligible influence on over a 11 costs. The 1 ogi st ics costs are based on current freight ra-tes and vary by less than 10 percent for all plans. The personnel shuttle costs and contingency risk costs are necessarily approx;.:. mate but are adequate for comparison purposes. When comparing the total costs,-the plans were considered equal if the total costs were within $40 million, and a definite cost advantage was considered if there was a $50 million difference. On the basis of the foregoing~ Plan 3 is comparable to the minimum cost alternative of any of the plans. Plan 7 has approximately a $60 million cost disadvantage compared to P~an 3. (ii) Ease of Operation, ~laintenance and Construction Flexibility Access Plan 3 does not meet the ease of operation and maintenance criterion because it does not have a connecting road between \~atana and Devil Canyon. Access Plan 7 does meet the ease of operation criteria by having a connection service road between the t~t1o sites. Plans 3 c.nd 7 both satisfy the criteria for flexibility for con- struction 1 ogi st i cs and transportation by having a road access con- necting to a major highway. (iii) Biological The primary biological concerns for these two plans relate to the effects the road would have on furbearers, big game, and cultural resources .. A roadway from the Parks Highway \'lould cr'oss productive furbearer wetlands habitat between the high~'lay and Gold Creek .. The Denali segment of both these plans also crosses aquatic fur bearer habitat near Deadman Mountain~ Deadman and Big Lak.es, and Upper Deadman Creek. In addttion, a red fox denning complex south of'Deadman Mountain within one mile of the proposed road is likely to be af- fected. The primary concern relative to big game for both these plans is the Denali segment, whi~h would pass through an area that has frequently been used by ~ither major portions or all of the Nelchina herd and includes the calving and summer ranges of the northwestern subgroups 11-16 I I I I I I I ,I · .. I I I I I I I I ' I "" .;. f. I I J I I I ,•• I. I I I I I I I I I I I I of the Nelcttina caribou herd. The r·oute also lies across the late summer migration route of cari. bou moving toward Butte Lake and Go 1 d Creek and parallels a traditional spring migration route southv1ard t~. the Susitna River. " The_direct effects upon this group of caribou, shou1d Plan 3 be implemented, include disturbance to cows and calves during the road construction period, a disturbance and possible impediment to caribou migration as a result of increased traffic in the ar'ea, and the possibility of direct mortality from road kills. Ho~1ever, the presence of the road should not interfere with migration, since caribou are known to cross roads.-Moreover, interference with the calving.areas could cause a major adverse impact on the females who show an affinity to traditional calving grounds. Of greater importance than these factors, however, are the indirect consequences to this group of caribou of increased access to its range. An access road across this alpine tundra ~t/ould provide the opportunity for all terrain vehicles to push a network of unplanned trails throughtiut the range of this subherd. This new access would cause disturbance and increased mortality to these caribou from their contact with vehicles, campers, and hunters. Thusi there is a chance that this route could lead to partial abandonment of impor- tant caribou habitat. The actual magnitude of impact is difficult to assess since it de- pends on .the somewhat unpredictable behavior of both caribou and man. With an increased emphasis on management of the area and stringent _hunter control, it is technically possible to lessen the potential extent of impact. It is expected, however, that resource a-gencies would be apprehensi.ve about the success of any mitigtation plans and \>Jould strongly resist any road access from the Denali Highway. (iv) Social Considerations . Without mitigating measures, access plans with a road~1ay originating from the Parks Highway could significantly impact the westside com- munities in terms of demand for increased services, changes in popu- lation, housing availability, government expenditures and revenues, ·labor demand, and unemployment. There v1ill also be significant ef- fects on construction, retail trade, and tourism. Many of these changes wi 11 occur as construction workers attempt to t,elocate to the communities near the construction site. Depending upon commuting modes to the camp, there could be a large increase in vehicular traffic in the area. These access plans also include a road from, the Denali Highway. As such, many of the impacts which would be felt in the \vest side com- munities of Talkeetna., Trapper Creek, and Sherman would a1so occur 11-17 in Cantwell. With a road from the north, it is expected many of the worKers would settle in Fairbanks., thereby reducfng some of the im- pacts which th-e west side communities would experience. These p~l ans meet the" preferen<;:e of the pub 1 i c in Cant\'Je 11 as some changes wi 11 occur but wi 11 not meet the preferences expressed by those in the west side communities who desire no cnange. However, road access connecting the Oenal i and Parks Highway wou1 d create extensive publfc access following construction thus creati~ the maximum change in the status quo of the area. · As discussed under Section 11.13, it is considered that mitigation measures can be implemented to lessen the effects on the west ~ide communities of Talkeetna and Trapper Creek. Hith road acces·s from the Parks Highway, change in the remoteness of Go 1 d Creek and the Indian River Land Disposal sites will occur regardless of . mitigation. (b) Access from Parks Highw.ay Only (Plans 1 and 5} (i) Costs Access Plans 1 and 5 are both comparable to the minimum cost altern- ative (Table 11.1). (ii) Ease of Operation and Construction Flexibility Both Access Plans 1 and 5 satisfy the ea.se of operation crit\~ria by having a road directly connecting both sites •. Both Access PlcUlS 1 and 5 satisfy the flexibility criteria by having a road connection with a major highway. Access Plans 1 and 5 involve a shorter haul distance compared to any alternative having access vi a Dena 1 i Highway. Anchorage has been identified as the most viable port of entry for the majority of the materials and supplies (_). When comparing Access Plans 1 and 5, with plans having access from the Denali High- way, 1 ogi st i cs and cost advantage over any access from the Denali Highway. With the majority of materials and supp1 i es coming from. Anchorage, the access route from the Denali Highway \'IOUld involve an additional haul of approximately 52 miles to ~Iatana when compared to an access from the Parks Highway. The additiona1 52 miles of haul to ~Jatana, for a Dena 1 i access alternative, vwul d be a disadvantage in long-term operation and maintenance. (iii) Biological Considerations The primary conc~rns with. access from only the Parks Highway v1ere discussed in (a) above. Briefly~ the concerns are the potential im- pact"'to ~furtn:a~er habitat between the highway and Go1 d Creek and 'po- tential degradation of fisheries habitat in the Indian and Susitna 11-18 I I I I .I I I I I I I I •= I . . " I I ' I I I I I I I .I I I I I I I I I I I I I Rivers. Of lesser concern is the disturbance of moose and bear pop ... ulations and removal of their habitat caused by the northside con- necting road in Plan 5. · · In addition to these, Plan 1 includes a ~onnection on the south side of the Susitna. River bet\'-teen the two damsites. This road would pass near and through extensive \'let land areas in the Stephan Lake-FQg Lake area. These wetlands provide habitat for furbearers and water- fowl and support a large, year:-round concentration of moose. Be- cause this .area is currently relatively inaccessible, potential im- pacts include removal of habitat and increased mortality through hunting and trapping. · (iv) Social Considerations Evaluation of these plans from a socioeconomic aspect reveals that Plans 1 and 5 wi 11 result in the greatest impact to the west side communities. Because access is provided from the west onlY, the majority of the impacts v10uld be felt in the west side communities. There would be a greater tendency for people to relocate in the com- munities and perhap? in Anchorage and a .lesser tendency to live in the Fairbanks area. There would be some impacts to the Cantwell area, but fe\'/er than with a road from Dena 1 i. Impacts waul d be the same as discussed in (a) above. In terms of public preference, these plans least meet the desir·es of people living in the project area. The plans would cause the great- est change in the Talkeetna-Trapper Creek area (where residents have. expressed negative attitudes toward social change) and would mini-. mize impacts to the Cantwell area (where residents have expressed a . desire for change). The Indian River land disposal site and Gold Creek would experience the greatest change with the selection af this plan. (c) Access for Denali Highway (Pl.9JIS 4., ti and 11) ( i) Costs Table 11.1 indicates that Plan 4 is.comparable to the least cost alternative {Plan 5). The cost of Plan 6 is approximately $40 mill ion greater .than that of Plan 4 and the cost of Plan 11 is approximately $35 million greater than that ofPlan 4. (ii) Ease of Operation and Construction Flexibility Plan .4 does not satisfy the ease of operation criterion due to the absence of a road directly connecting the two dam sites. Plans 6 and 11 both have a road directly connecting the dam sites, therefore both pl.ans satisfy the ease of operation criterion. Plan 4 on1y partially meets the construction flexibility criterion. -Plan 4 i;tcludes a road connection to a major highway for the \~atana ll-19 (d) deveJopment-but not for the De vi 1 Canyon development~ Access Plans 6 a·nd 11 both satisfy the flexibility criteria by having a ~:annec­ tion to a major highway. (iii) Biological Consid.etations These three plans a11 involve road access from Denali Hi:9hvJqV to Watana damsite. Th~ potential biological and cultural impa ~ asso- ciated with this route were.discussed under (a) above. Basi tlly~ impacts could occur to portions of the Nelchina caribou herd through increased hunting mortality and potential interference with migra- tion and calving. Increased access and trapping pressure could also impact furbearers. In addition, because of treeless topography and shallm'l soi 1, disturbance and remova 1 of any cultural resources could result. ,. -Plans 4 and 6 involve construction of a rail connecting from Gold Creek to Devil Canyon. No major environmental problems were identi- fied along this portion of the route. The connection road on the north side of the Susitna River _between the t~tto dams was discussed under {b) above, the only environ~ent~l concern being the crossing of moose habitat. (iv) Social Considerations Plans 4, 6 and 11 involve the major access point of origin on the Denali Highway, rather than the Railbelt Corridor. Workers' fami- lies would tend to locate more com:nunities, including Cantwell and Fairbanks. Due to the rail access from Gold Creek, there waul d still be changes in the west side communities, but fewer than with a road originating from ·the Parks Highway~ Plan 11, involving access from Den a 1 i Highway only, waul d cause the greatest number of changes in the Cantwell and Fairbanks area and fewer changes to the west side communities. These changes would be the same as described in (a) above. Access from Gold Creek Only {Plans 2, 8, ~. and 10) Table 11.1 indicates that the total cost of Plan 8 and 9 are respectively $15 and $30 million greater than the least cost alternative, Plan 5. The substantial savings in construction costs are offset IJ.t increased personnel shuttle costs and an allowance for contingency ri.sk .. The cost comparison a 1 so shows that the total costs of Plans 2 and 10 are $55 mi 11 ion and S40 million more than that of the least cost alternative. ( i) Ease of Operation and Construction Flexibility Access Plan 2 meets the criterion for ease of operation since the dams are directly connected with a rai 1 route. Access Plans 83 9 · and 10 partially satisfy the -ease of operation and maintenance cri- teria. These plans have a road directly connecting the two dam sites, however, they do not have a connection to a major high\'lay. 11-2.0 I I I I' I I I I I I I "_,7~--•• ~. ' I I I I I t I ' I I I I I I I I I -.- 1 I I I I I I I ·.-·1 ( i i ) This reduces the flexibi1 ity in operation and maintenance of the. site~ as discussed in Section 11.8. · Access Plans 2, 8, 9 and 10 do not satisfy the flexibility criteria for construction as they do not h.ave a road connection to a major highway. · Biological Considerations These plans all preclude access from the Parks Highway or Denali Highway; t.herefore, the impacts associated vtith increased access are substantially reduced. Plans 2_and10, which involve connections between vJatana and Devil Canyon on the south side of the Susitna River, have as the major po- tential environmental impacts, the disturbance of wetland areas near Stephan and Fog Lakes,-as discussed under (b) above. The overall reduction in access and the fact there is no access con- necting with the Denali Highway to the north indicates these plans would result in the _least number of impacts to biological and cul- tura 1 ·resources. (iv) Social Considerations Thes·e plans all involve access from the west only, the only differ- ence being road or rail, and if rail, the distance into the basin the railroad extends. As such, impacts ¥muld again be concentrated on the west side communities. These impacts ~JOuld likely be concen- trated in the Gold Creek area as well as Talkeetna and Hurricane be- cause of their location at rail-highway intersections. The Cantwell and Fairbanks areas would be less affected as there would be no northerly access. The public has expressed a preference for a rail access and a main- tenance of the status quo. · Although raii access would best maintain the status quo of the Upper Susitna Basin in general with the rail access, si gni.fi cant changes could occur in the Talkeetna/Trapper Creek area as discussed in Section 5.1 (e). These plans v;ould not meet the public preferences expressed by Cant- well residents. 11.10 -Identification of Conflicts From the evaluation pres.ented in Section 11.8, it is apparent no single plan meets all the objectives or satisfies all the criteria established as part of the ,study. The basic conflicts identified were: . (a) Social and Biological tonsiderations vs Construction and Operation Logistics Rail or road access from a railhead at Gold Creek Nithout· road access from ll-21 a major highway would limit social and biological changes in the immediate project area and retain the status quo to the greatest extent possible. This option is in dire~t conflict with the requirement to provide flexibil- ity in construction logistics and transportation and to provide ease of op- eration· and maintenance •. The selection of such an option would increase the risk of high costs, schedule delays, and safety problems and decrease project reliability .. (b) Social vs Biological Considerations Social and biological objectives are not in basic conflict since limited acce5:; to tbe project area is most desirable in both cases.. If, however, the assumpt ·ion is made that road access to a major hi gh\1ay \~Ji 11 be provided, then a conflict arises. From the social/local public preference perspecttve, access from the Denali Highway is preferred. This plan v10uld create the ~conomic stimulus desired in Cantwell, reduce the potential for change in the Trapper Cteek/Talkeetna area, while retaining the remoteness·' of the Indian River land disposal site and the railroad communities north · of Talkeetna. The Denali access, however, is in conflict with biological objectives since it would allow access by hunters and ATVs to a large portion of the ll!"'per Susitna Basin and create potential impacts on the Nelchina CaribrJ,J:: other big game species including moose and bear, the fisheries in isolated lakes and streams and forbearer habitat. I-11 addition, the potential for disturbance of archaeological sites in this area is greatest. Although mitigaion measures can be employed to reduce these potential biological impacts; it is considered likely that gover-nment resource agencies would be appr·ehensive about the success of any control programs and would thus be opposed to any access fr9m the Dena]i High\vay. The selection of a Denali access plan could result in unacceptable delays in license approval or a subsequent rejection of this plan requiring a reassessment of access plans from the west. Table 11.2 broadly sunmarizes the conflicts in the evaluation,; 11.11 -Comparison of Access PI ans (a) Access from Railhead at Gold Creek (Plans 2, 8, 9 and 10) vs Access from ~1ajor Highway \r1al}s 1, 3, 4, 5, 6, 7, 11) Considerable cost, schedule, safety and reliability risks are associated with construction of a major project without road access to a major high- way. On the other hand, road access to a major highway will create addi- tional change in the status quo of the Upper Susitna Basin. If the deci- sion is made to t:levelop a large scale hydroelectric facility in the Upper Susitna Basin, it is considered essential that the orderly development and maintenance of the facility should be afforded a higher priority than main- tenance of the status quo. Thus, access plans originating at a railhead at Gold Creek only are not recommended. These considerations led to the rejection of plans not providing road access to a major highway. 11-22 c; ,, I •' I t I I I I I .I I .· I I I I I I I I I I I I I I I I I I I I I I I I I I ~. (b) (c) (d) Plans eliminated in this comparison;· 2, 8, 9, 10 Plans remaining! ~-~ 33 4, 5, 6, 7, 11 Access From Both Parks Highway and Denali Highway (P1 ans 3 ·!I 7) Vs Access from Only One Highway (Plans 1 , ·. 2 , 4, 5 , 6 , · 7 , -8 , l 0 , 11 ) The plans which optimize transportation flexibility and ease of operation involve the initial construction of a road from Denali Highvtay to \~atana damsite., .. To~allow for improved logistics during the peak construction at Watana and throughout the construction of De vi 1 Ca nyoii, road access. \-toul d also b~ created to the Parks Highway. The disadvantages of these plans are t_hat they wou1 d create the maximum change in the status quo producing both the biological impacts associated with the Denali link and the social impacts associated v1ith the Parks Hi gh~ttay 1 ink.. These impacts are further intensified with both roads since the connection of the Parks and the Denali Highway would encourage hunters and tourists to drive the complete loop. These plans are also more costly than the minimum cost alternatives. It is considered that the social and biological impacts that would result from these plans cannot be justified by the added transportation flexibil- ity and ease of operation benefits associated with road access to both the Parks and Denali Highways. These conclusions resulted in the rejection of the plans providing road access to both the Parks and Dena 1 i Hi ghv1ay. Plans eliminated in this comparison:· 3, 7 Plans remaining: 1, 4, 5, 6, 11 Road\'lay ·Connecting the Dam Sites Directly (Plans 1, 2, 5, 6, 7, 8, 9, 10 11) vs No Roadway Connecting the. Dam Sites Directly (3, 4) Plans incorporating a road connecting the dam sites directly are clearly superior in terms of ease of operation and maintenance to plans which do not directly connect the dam sites. The access p 1 ans which do not connect the damsites directly do not have advantagesi n any of the other, or oDm- bined criteria to \varrant not eliminating these alternatives from further· consideration. These conclusions resulted in the rejection of plans not connecting the damsites directly. Plans eliminated in this comparison: 3, 4 Plans remaining: (. 1, 5, 6, 11 Access to Denali Highway (Plans 3, 4, 6, 7, 11) vs Access to Parks Highway (Plans 1, 5) The main concerns associated with the Denali access are the potential effects on the Nelchina caribou herd, increased access to a large area of 11-23 (e) a 1 pine tundra vrith the associ a ted effects of disturbance by ATVs and dis- turbance of potential cultural resources. Although there are some fisheries and fur bearer concerns in the lndi an River area associated with a Parks High\'lay access, from the biologicai per- spective, Parks High~vay access is preferred to a Denali Hi ghvJay access. From a social perspective, the Denali route is clearly superior to the Parks Hi gh~tlaY route. The Denali route \'ioul d promote the economic stimulus desired in Cantwell while reducing the influence on the community of Trapper Creek and Talkeetna which has expressed· a desire to rna i ntai n the status quo. It is considered,_ however, that even with a Parks Hi gh\lfay ac- cess, mitigation in the form of se1f-contained construction camp facili- ties, regulation of commuter schedules and control of transportation modes can reduce or avoid many of the potential changes in Talkeetna and Trapper Creek. With any access plan from the \'lest, a major railhead vtould be located at· Go 1 d Creek creating suffi ci ·ent 1 ocal changes. With road access from the Parks Hfgh\'lay-to Gold Creek, changes \'/ill also ocCllr at Indian River-and disposal sites .. Based on the above discussion, it is concluded that the Parks Highway ac- cess is preferab 1 e to the Dena 1 i access plan. This conclusion is based on the assumption that: -if a Denali route Here selected, it would be Plan 6 which \vould still re- sult in significant social changes in the Gold Creek area; -government resource agencies will be opposed to the Denali route with a likely 1-to-2 year delay in schedule or denial of permit resulting; -changes in local communities can, to a large degree, be mitigated through controls imposed on contractor and construction workers; and -controls \'/Oul d be very difficult to impose upon hunters and ATV operators who would utilize the Denali's route after construction. The foregoing consideration5 resulted in the elimination of plans involving access from the Denali Highway. Plans eliminated in this comparison: Plans remaining: Comparison of Plan 1 vs Plan 5 3' 4, 6' 7' 11 1' 5 P1 ans 1 and 5 both commence on the Parks Hi gh\'tay near Hurricane and p\"oceed through Chulitna Pass and along the Indian River to Gold Creek. From Gold Creek, both Plans proceed east on the south side of the Susitna River to the Devil Canyon site. At De vi 1 Canyon, Plan 1 proceeds east on the south 11-24 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I side of the Susitna River to the Watana site. Plan 5 crosses the Susitna River at Devil Canyon and proceeds east on the north side of the Susitna River to the Watana site. Access Plan 1 has potential for greater environ- mental impacts than Plan 5. Access Plan 5 has a slight cost advantage over Plan 1, ~lso Plan 5 fs slightly easier to construct due to the difficult terrain in the segment between Devil Canyon and Watana south of the Susitna River. The only advantage Plan 1 has over Plan 5 is in Native landovmer preference\.• It is, therefore, concluded that the environmental cost and construction considerations out\~·eigh the Native lando\lmer preference, and therefore, Plan 1 is eliminated from further consideration. 11.12 -Recommended Access Plan Based on the above discussion, it is recommended that: -The selected access plan for the construction and operation of the Susitna Hydroelectric Project should comprise a road commencing near t~P 156 on the Parks Highway, proceeding southeast crossing the Sus itna River at Go 1 d Creek, turning northeast to Devil Canmyon Dam site along the southern side of the Susitna River, crossing the Susitna River at Devil Canyon, an proceeding along the north side of the Susitna River to Watana Dam site (see Plate 30). -To allm'l for continued access for project construction by mid·-1987, a pioneer road (limited access) between Gold Creek and Hatana Dam site be constructed commencing· in mid-1983. The application for permits to construct this pio- neer road be submitted to the State of Alaska and the Bureau of Land Manage- ment by August 1982, independent of the FERC license application,. -To mitigate against agency concerns in regard to the pioneer road concept and to avoid the possibility of public access to the project area in the event that the project is not built, road access between the Parks High\'lay and Gold Creek not commence until after FERC license approval. If the project does. not proceed after the pioneer road is constructed, the road as such should be rendered impassable to future vehicular traffic. -To minimize potential impacts to furbearers and fisheries resources in the Indian ~·iver and Susitna River areas, special construction techniques be . utilized (including adequate bank stabilization, revegetation and restora- " tion) when crossing wetland areas or \·/hen constructing in proximity to the Indian or Susitna Rivers. .... To minimize the effects of public access during the op-eration phase of the project consideration be given to controlling public access across Devil Can- yon Dam.. If access is provided east of Devil Canyon Dam site, restrictions should be placed on the use of ATVs and hunting. -To assist in minimizing changes in the local communities of Talkeetna~ Trapper Creek, Sherman and Curry, it is strongly recommended that subsequent decision on construction camp facilities, commuter modes, v10rk incentives, and general policies incorporate a special effort to minimize the effects of construction on these local communities. Specific mitigation recommendations are included in Section 11.13. 11-25 The foregoing is based on the fo11m·ling assumptions~ The pioneer road concept wi 11 be approved by g.over nment regula tory agencies since the pioneer road would not connect to any existing road before the issuing of a FERC license, thus not making the prior commitment to allowing public access to the Upper Susitna Basin~ -Although the native landowners have expressed a strong preference for road access from Parks Highway to both darnsites along the south side of the Susit- na Riv~r, their basic desires would be met by providing road access, from any direction~ to their existing land. holdings~ -Public access will be prohibited during the construction phase of the pro- ject. Also, the selection of Plan 5 offers some flexibility in regard to the degree and type of pub 1 i c access subsequent to 1993. -Biological and social impacts will be mitigated through adoption of the recommendations presented in Section 11.13. If permits to commence construction of the pioneer road are not obtained by mi d-1983, it may be necessary to accept a 12-to 18-month de 1 ay in the on-1 i ne schedule or possibly revert to one-of the less acceptable access plans which do not require a pioneer road. 11.13 -Mitigation Recommendations The plan recommended by Acres does not satisfy a11 the evaluation criteria out- lined in Section 11.2. In or~er to reduce potential impacts to biological and cultural resources and to alleviate socioeconomic impacts to the communities of Talkeetna, Trapper Creek, Sherman and Curry'" the following mitigation measures are recommended: Permit only on-duty construction workers to have access to both the pioneer road and access road. -After construction of the pov1er development is complete, maintain a control- 1 ed access route beyond the Devi 1 Canyon Dam. It is anticipated a coof1eY·a- ti ve agreement could be r~eached with BLM and ADF&G concerning. the number of people permitted access to the areas and the cost of any control measures. The construction camp should be as self-contained as possible, thus limiting the number of workers who could otherwise bring their families to a nearby community and commute daily. Provide incentives to encouraged workers to work the longest time possible between leaves to minjmize commuter traffic. Although the final schedule will not be known until labor agreements are established and construction commences, 1 anger work periods between breaks can be advocated. -Provide planning assistance if requestd to the communities of Talkeetna~ Trapper Creek, Sherman and Curry to aid them in preparing for the effects of increased populations. ll-26 I I' I 'I \ I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Evaluate various commuter management policies and select the one which reduces impacts .to the local communities. Socioeconomic impact assessment studies currently under way for the Susitna Project will provide important input data for evaluating possible commuter management policies.. · -Utilize excavated cuts and other construction techniques to. prohibit utiliza- tion of the pioneer road after construction of the access road. Areas used for the pioneer road which do not follow final road alignment should be reclaimed. 11 .. 14 -Tradeoffs Made "'n the Se 1 ecti on Process · (a) Basis of Selection.Process From the natural resource and local public preference perspective, mainten- ance of the status quo is probably most favored. However, it is unrealis- tic to consider that a project the size of Susi~na can be implemented with- out changing the existing character of sections of the Upper Susitna . Basin .. Access to the dam sites is a complex and controversial issue. As such, it has received considerable attention from the study team, APA, resource agencies and the public .. Although the studies have determined that there is no single access plan that satisfies all the project objectives and evaluation criteria, it has·been possible to develop an access plan which provides a reasonable tradeoff of preference. These tradeoffs are essen- tially based on the following compromises: -All entities must present a degree of flexibility, otherwise a satisfac- tory compromise is impossible. -Whenever a specific objective is partially crnnpromised, considerable ef- fort is made during subsequent decisions to compensate. -Any compromises made are clearly outlined such that decision makers t~e­ _viewing the final recommendation are aware of negotiations to date. · (b) Tradeoffs Made in the Selection Process ( i) Engineering Concessions made include: -no road access from Denali Highway which \vould incluoe a complete 1 oop connect i ng Parks Highway with Dena 1 i Highway; -no pioneer road to Parks Highway .Prior to the issuance of a FERC 1 i cense; 11-27 0 commitment to be prepared to make the pioneer road impassible if FERC license not granted; -restrictions to be placed on vmrker commuting schedules and mode; worker incentives to be provided to minimize effects on local communities; Objectives retained-include: -road access to both dam sites to allow for ease of construction, operatiorr and maintenance of the project; -maintenance of schedule through retention of the basics of the pioneer road concept. (ii) Biological Concessions made include: -road access from Parks Hi ghNay affecting Indian River area and providing partial public access to the upper basin: Objectives retained include: -no access from Denali Highway which was considered to have the greatest potential for environmental impact; -no route on the south side of the Susitna River between the dam sites, thus avoiding the sensitive Stephan Lake and Fog Lakes area; -emphasis on construction mitigation when developing road link be- tween Parks Highway and Go 1 d Creek; -retention of a degree of control on future pub 1 i c access by ac- cepting the Parks Highv1ay plan \-Jhere, due to the terrain, private vehicles are basically restricted to the access corridor bet\'leen Parks Hi gh\1/ay and the De vi 1 Canyon dam sit e. The degree and type of access east of Devil Canyon can be somewhat controlled by regu- 1 ati on of access across the Devil Canyon dam. The alternative of not connecting to a major highvJay \~as considered to have the least net adverse biological impact. The ease of opera- tion and maintenance and the construction flexibility criteria, as explained previously, was considered to outweigh this advantage. The mitigation measures and road management will reduce the adverse biological impacts associated with an access connection to a major highway, to a minimum. 11-28 ll I I I I- I I I I I ·t I I I I I I -I I I I I I •• I I I I I I I I I I I- I ( i i i ) Soc i a 1 Concessions made include: -road access to the Upper Sus itna Basin; -road access from Parks Highway which creates greatest potential for change in the Indian River land disposal si-te .. Objectives retain0d include: -Through the implementation of a relatively self-contained con- struction camp~ restrict ion of private vehicles fr·om the construc- tion site~ implementation of mass transit modes for commuting wor- kers!) incentives to encourage workers to remain on site and con- trolled public access east of Devil Canyon following construction~ it ·is considered that changes in the local communities of Trapper Creek/Talkeetna area will be minimized; -Although the western communities favored a rail access, they also favored maintaining the status quo. It is our opinion that the recommended p 1 an with its associated rni t i gat ion should produce less change in the Talkeetna/Trapper Creek area than an all-rail access plan. · Overall consensus of the local community preference favored access fy··om the Denali Highway. _The advantages of the Parks Highway ac-· cess over the Denali access in reducing the biological impacts is considered to outweigh the local community preference. In addi- tion to the lessened biological impacts, the recommended plan better meets the preferences of Native 1 and owners~ The recommended plan does not fully meet the preferences of the Native 1 and owners. They would prefer the access road between uevil Canyon and Watana be located on the south side of the Susitna River. The advantages of the road being located on the north side of the Susitna River, include, reduced biological im- pacts~ the. actual construction of the road is easier than if lo- cated on the south side, and the construction cost of the road is less. These advantages are considered to outweigh the Native 1 and owner pre: erence of having the road located on the south side of the Susitna River. 11-29 -f'!"··.·.· -· .•. ...,....., •. ,""'C""··. ~··-~·~' ••• I J". I I t'LAN DESCRIPTION: I I r I FAr.F OF NEW ROAD ,J . . TION CQ.S_l ($ x l,OOO,OOO) INlENANCE ($ X 1,000,000) lOGISTICS COST X 1,000,_0JJO) TOTAL ($ X 1,000,000) rl-K;:,uJ\!NEl SHUTTLE ($ X 1,000,000) CONTINGENCY RISK ~X 1,000,000) AL COSTS X 1,000,000) . ifRUCTION SCHEOlU F ~_J,O_R. B_RI DGES I I I I I I I I 1 z ROAI1\IIAV: · PARKS RAIL: GOLD HIGHWAY TO OCVIL CREEK TO OCVIL CANYON & WATANA CANVOI4 & WAt ANA ON SOUTH SIDE ON SOUTH SlOE Or SUSITNA OF SUSITNA 62 58 . 158 140 5 4 215 210 378 354 0 25 0 40 378 419 3-4 3-4 2 2 TABU: 11' .. 1: SUSUNA ACCESS PLANS ..$ ~4 ' 6 ROA[)IAV: DENALI ROAOOAV: DENALI ROAI)tlAY: PARKS ROA{)I{AY: D::NALI HIGHwAY TO HIGHWAY TO HIGHWAY TO OEVJ.L . HIGHWAY 10 WATANA, PARKS WATANA,. RAIL, l~OLO CANYON ON SOUTH WATANA, RAil, GOLD HIGHWAY TO CREEl< TO OEV IL SIDE OF SUSITNA, CREEl< TO OCVlL DEVIL CANYON CANYON ON SOUTH OEVIL CANYON TO CANYON ON SOUTH ON SOUTH SIDE sroc or SUSlTNA. WATANA ON NORTH SIOC OF SUSITNA. OF SUSITNA. NO NO CONNECTING SID£ OF SUSITNA. CONNECTING ROAD CONNECHNG ROAD ROAD ON NORTH SlOE OF SUSHNA~ 70 60 68 102 151 119 143 179 6 5 8 8 231 230 214 230 388 354 365 417 0 10 0 0 0 15 0 0 388 379 365 4'!7 1 1 3-4 1 0/1 0 2 0 ~ I H '':i. lU 11 • . ROAIJrlAY; DENALI ROAD4AV: GOLD RAILi GOLD RAIL: GOLD ROAD'!AV: OCNALI HIGHWAY TO CREEK TO DEVIL CREEX TO OCVIL CREEK TO OCVIL HIGHWAY TO WATANA WATANA, PARKS CANYON ON SOUTH CANYON ON SOUTH CANYON ON SOUTH CONNECTING ROAD HIGHWAY TO DEVIL SIDE OF SUSlTNA, SIDE OF' SUSlTNA, SIOC OF' SUSHNA .. BETWEEN WATANA CANYON ON SOUTH DEVIL CANYON TO .ROA~AY OCVlL ROAO/IAY OCVIL AND DEVIL CANYON SIDE Of SUSITNA. WATANA ON NORTH CANYON TO WATANA CANYON TO WATANA ON NORTH SIDE CONNtCTING ROAD SIDE or SUSITNA. ON NORTH SIDE ON SOUTH SlOE Ot SUSlTNA. ON NORTH SIDE OF SUSITNA. OF SUSITNA .. or SUSITNA. 111 54 . 58 53 86 209 93 108 123 145 9 7 .~· 5 11 . 231 214 216 214 258 449 314 '329 342 414 0 25 l? 25 n 0 40 40 40 0 449 379 394 407 414' 1 3 3 3 1 0/1 1 1 1 0 {' ,-_, .. ~ ::... I I I I I I , I I I I I I I I I I I I I TABLE 11.2: IDENTifiCATION OF CONFLICTS Criteria Costs Minimize Costs Ease of Operation and Construction Flexibilit}' Ease, of Operation and Maintenance Construction Flexibility Biological ~ Minimize Biological Impacts Social, Accommodate Preference of Native landowners Accommodate Local Community Preference 1 -Does not satisfy criteria 2 -Intermediate 3 -Satisfies criteria 1 2 3 3 3 2 3 1 2 3 3 1 1 2 3 4 5 6 3 J 3 1 1 1 3 3 3 2 3 2 1 1 2 1 2 2 2 2 2 3 1 3 7 a 9 10 11 1 J 3 3 1 3 2 2 2 3 3 1 1 1 3 1 3 3 3 1 2 1 1 1 1 2 2 2 2 3 0 J. DEFINE OBJECTIVES · SELECT ACCESS ROUTE TO HYDROPOWER SITES THAT ALLOWS CONSTRUCTION AND 1 OPERATION WHILE BEST MEETING OVERALL CRITERIA STATED IN .([J 2. ...,__,.... DESIGN PARAMETERS ROADWAY AND. RAIL ENGINEERING CRITERIA 2. ESTABLISH CANDiDATES A TOTAL OF 33 r- ROUTES ARE ESTABLISHED IN THE 3 CORRIDORS 1 2A . PUBL1C PARTICIPATION PRESENT THE OPTIONS - TO THE PUBL;C AND INVITE COMMENT 3. SCREENING PROCESS ~ TECHNICAL ECONOMIC ENVIRONMENTAL PUBL!C PREFERENCES TRANSMISSION IMPACT 4 .. PLAN FORMULATION 3 ROUTES ONE lN . ........-_ .... EACH CORRIDOR AS A RESULT OF THE SCREENING PROCESS iN IID ARE ESTABLISHED POHT FACILITIES ROADWAY OPTIONS RAIL OPTIONS LOGISTIC REQUIREMENTS 8 PLANS, WHICH UTILIZED THE 3 ROUTES ARE ESTABLISHED 5. EVALUATION ADDITIONAL STUDIES SOILS DATA ENGINEERING r+ CONSTRUCTION COSTS ........ _. II ALTERNATIVE PLANS ARE EVALUATED TO THE FOLLOWING CRITERIA LOGISTICS COSTS TRANSMISSION IMPACT -.f ENVIRONMENTAL. LABOR ORGANIZAT~ON ..-.CONCERNS AGENCY CONCERNS AS A RESULT OF r+-AGENCY CONCE'RN.S, 3 ADDITIONAL PLANS ·ARE ESTABLISHED ~ NATIVE LANDOWNERS PREFERENCES ~ LOCAL CQMMUNITY PREFERENCES - ·--- ENGINEERING ECONOMIC ENVIRONMENTAL SCHEDULING DESIRED LEVEL OF ACCESS AGENCY CONCERNS SOCIAL PREFERENCES TRANSMISSION ACCESS' PLAN SELECTION METHODOLOGY ~IGURE JU IJR· I I I I I I I I I I I I I I I •• I I I PARKS HWY. PARKS HWY. ' ' ' CANTWELL · DENALI HWY. PROPOSED ,ROAD GOLD~ --__... D.C~ --- CREEK SITE ANTWELL GOLD CREEK PLAN I DENALI HWY.· PLAN 2 --.-\_WATANA SITE WATANA SITE FIGURE ,,.21A~~m I L----------------------------~~~---------------- I : ••. I I I I I I I I I I I I I I I' ~' -- 1 I PARKS HWY. CANTWELL. DENALI HWY. HURRICANE 1!>.._ GOLDJ -- CREEK PLAN 3 PROPOSED -j D.C. SITE ANTWELL /DENALI HWY. HURRICANE G. OLD D.C., CREEK SITE ,PLAN 4 I I I I ~~ ROADS~ . I I I tWATANA SITE G I I I PROPOSED~ ROAD J I 1 ~ ,, lwATANA SITE FIGURE 11.5 !A~Im!· I I ® I I I I ' I I I I I I PARKS HWY. I I I I I I I CANTWELL DENALi HWY. HURRICANE ' ' PROPOSED ROADl_ - . _.... --....... GOLD.J- CREEK ---:cf . LWATANA SITE SITE CANTWELL HURRICANE GOLD CREEK PLAN 5 DENALI HWY. PROPOSED ROAD l I I I I I 1 I -..tWATANA SITE --------,-~-.-, PLAN 6 FIGURE 11.41 ~~~l~ l I I @ I I I I I I I I I I I I I 0 I I I I >-"-'""'"""'CANTWELL L D~NALI HWY. HURRICANE PROPOSED ROADS ---· • I I I I I I I GOLy-- CREEK ---r D.C. -lWATANA SlTE SITE PLAN 7 DENALI HWY. HURRICANE PROPOSED\ ROAD · .,.,..._ -.._._._ GOLD_j--- CREEK • __. -- D.C) l WATANA SITE SITE PLAN 8 FIGURE IL5. I I ® I I I I I .I I I I I I I I I I I •• CANTWELL HURRICANE DENALI HWY. ] PROPOSED ROAD --..._. GOLD CREEK ,-'~ ~~~~~H+~r '\_WATANA CANTWELL HURRICANE GOLD CREEK PLAN 9 DENALI HWY. PROPOSED ROAD SITE - -_. -tWATANA SITE PLAN iO ··. FIGURE 11.6! ~~~m1 I I I I I I I I I I I I I I I I I I I PARKS}_ HWY. ., CANTWELL HURRICANE GOLD_, CREEK DENALI HWY. J"' D.C. SITE PROPOSED ROAD --- PLAN II I l I 1 I J -..'-WATANA SITE · FIGURE 11.71 A~~~~ I I I I I I I I I I I. I I I I I I I I I 12 -HATANA U.£VELOPMENT This section describes the various components of the 'tJatana Development, includ- ing diversion facilities, emergency release facilities, the main dam, primary out1et facilities, reservoir, main and emergency spi'llways, the power intake, penstocks and-powerhouse complex including turbines, generators, mechanical and electrical equipment, switchyard structures,· and equipment and project lands. A description of permanent and temporary access and support facilities is also included. 12.1 -General Arrangement The evolution of the Watana gene~~al arrangement is described in Section 9. The Watana reservoir and surrounding area is shown in Plate 31. The site layout in relation to main acce,s5 facilities, borr·m'l areas and camp facilities is shown in Plate 32. A more detailed att"angement of the various site structures is pre- sented in Plate 32A. The Watana dam will form a reservoir approximately 48 miles long, with a surface area of 38,000 acres, and a total volume of 9,515,000 acre.:feet at a norma1 maximum operating elevation of 2185. During operation, the reservoir will be capable of being drawndown to a minimum elevation of 2045. The dam will be an earthfill structure with a vertical central impervious core. The crest elevation of tht= dam will be 2210, with a maximum height of B85 feet and a crest length of 4,100 feet. The total volume of the structure will be approximately 62,000,000 cubic yards. During construction, the river will be diverted around the main construction area by means of two concrete-lined diver- sion tunnels, each 40 feet in diameter, on the right bank of the river. A power intake located on the right bank will comprise an approach channel in rock leading to a multi-level reinforced concrete gated intake structure capable of operation over the full drawdown range. From the intake structure, six pen- stocks, consisting of concrete-lined tunnels, each 17 feet in diameter, \·rU1 lead to an underground powerhouse complex housing six Francis turbines with a rated capacity of 170 MW and six semi-umbrella type generators each rated at 180 ~WA. Access to the powerhouse camp 1 ex \•Ji 11 be by means of an un 1 ined access tunnel. Turbine discharge will be conducted through six draft tube tunnels to two surge chambers dO\•mstream of the povJerhouse, then by means of two 30-foot diameter concrete-1 ined tailrace tunnels. A separate transformer gallery or chamber just upstream from the po1t1erhouse cavern will hGuse nine single-phase 15/345 kv transformers. The transformers will be connected by 345 kV single- phase, oil-filled cable through two cable shafts to the switchyard at the sur- face. A tunnel spillway located on the right bank will be designed to discharge all· flm·1s resulting from floods having a return frequency of 1:50 years or less. This structure will be equipped with six fixed-cone valves at the dO\'/nstream end to minimize undesiranl€ nitrogen supersaturation in the river downstream from the dam during spillway operations. Flows resulting from floods with a fre- quency greater than 1:50 years but 1 ess than 1:10 ,OUO years wi 11 be discharged by a chute spillway a1 so on the right bank. The spillway control structute at 12-1 the upstream end wi 11 be cantrall ed by three fixed 1t1hee 1 gates 1 ead ing to a · reinforced-concrete-line chute section and then to a flip bucket at the down- stream end. An emergency spillway on the right bank will provide sufficient· additional capacity to permit discharge of the PMF without overtopping the dam. An emergency release facility will allow lowering of the reservoir over a period of time to permit emergency inspection or repair. 12.2 -Site Access (a) Roads At \~atana the main access ro~rJ wi 11 enter the site from the north. In addition to the main acces~, s.~veral additional roads will be required to the camp, village, airstrips tc.nk farm, haul roads to the borrow areas, and construction roads to the dam al'd all major structures* These roads with the exception of the haul roads c:.re snown on Plate 35. The construction roads will be 40-foot \~!ide gravel surfaced roads with small radius curves and grades limited to 10 percent. Major cut and fill work will be avoided. A gravel pad approximately 5 feet thick will be re- quired for the roads. This gravel pad will provide a drivable surface and also will protect against the sporadic permafrost areas. (b) Bridges No major ~emporary bridges will be required for the construction of the t~atana development. The crests widths of the upstream and downstream cof- ferdams \•Jill be planned to provide suitable access to the south bank of the Susitna River during construction. I , •. I ·I I I I I I I The camp l eted main dam crest wi 11 provide permanent access across the I Susitna River. One area which may require a small temporary bridge is Tsusena Creek near· 1 its confluence with the Susitna River. Currently it is envisioned that this crossing can be accomplished with one large or multiple culverts.. · (c) Airstrip II A permanent airstrip will be constructed approximately 2.5 miles north of the main construction camp (see Plate 35). The runway \vill be 6,000 feet I in length and will be capable of accommodating the C-130 Hercules aircraft, as well as small jet passenger aircraft. A road will serve the airstr'p connecting to the camp, village, and damsite, A sma11 building will t 1 constructed to serve as a terminal and tower and a fuel truck/maintenar .. -: facility will be constructed. A temporary airstrip ~vi 11 also be constructed to support the early phases of mobilization and construction. This temporary runway \-Jill be 2,500 feet in length and will be located in the vicinity of the main construction camp. The airstrip will be capable of supporting other type aircraft .. 12-2 I I I I ··I· I I I I I I I I I I I I I I I I I The temporary airstrip will eventually be incorporated into one of the main haul roads for Borro~tl Area D. This will occur after the permanent airstrip is in service~ (d) Access Tunnel An access tunne 1 wi 11 be provided to the underground po\'Jerhouse and associ- ated works. The main access tunnel will be approximately 35 feet wide and 28 feet high. The tunnel will allow permanent access to the operating development and will also be uti1ized during construction as the main con- struction ·tunnel. Construction adits will branch off to the various com- ponents of the development during construction. 12.3 -Site Facilities (a) General The construction of the Watana development will require various facilities to support the construction ·activities throughout the entire construction period. Following construction, the operation of Watana will require cer- tain facilities to support the permanent operation and maintenance of the PO'--Jer fac i 1 ity. The most significant item among the site facilities will be a combination camp and village that will be constructed and maintained at the project site. The camp/village will be largely a self-sufficient community housing and maintaining living facilities for 6,000 people during constructien of the project. After construction is complete, it is planned to dismantle and demobilize the facility and to reclaim the area. It is additionally planned to utilize dismantled buildings and other items in the camp/village as much as possible for use during construction of the Devil Canyon devel- opment. Other site facilities include contractors• work areas, site power, services, and communications. Items such as power and ccmmLnications will be tequired for construction operations independent of camp operaticns. The same wi 11 be true regarding a hospital or first aid room .. Permanent facilities required will include a permanent town or small ~om­ munity for approximately 130 staff members and their families. Othelr' per- manent facility items will include a maintenance building for use during subsequent operation of the power plant. A conceptual plan for the permanent town has been developed, however~ it is ~ecommended that preliminary design and fin?, ~esign be defered until near the end of construct ion \-Jhen more in format i 1·: .1s to the physical parcrmeters of design is available and, more importantly, the human requirements and preferences are better defined. Fuel oil has been selected as the means of heating the camp/village facil- ities. (b) Temporary Camp and Vi 11 age The proposed location of the camp and vil"tage will be on the north bank of the Susitna River bet~tJeen Deadman and Tsusena Creek, approximately 2.5 miles northeast of the Watana Dam. The north side of the Susitna was chosen because the main access will be from the north and south-facing slopes can be used for siting the structures and the location. The loca- tion is shown in Plate 35. The camp will consist of portable woodframe dormitories for bachelors with modular mess halls, recreational buildings, bank, post office, fire sta- tion, warehouses, hospital, offices, etc. The camp will be a single status camp for approximately 5,000 workers. The village, accommodating approximately 550 families, will be grouped around a service core cant ai ni ng a school, gjtflln as i urn, stores, and recre a- tion area. The village and camp areas vJill be separated by approximately 1.5 miles to pro vi de a buffer zone between areas. The hospita 1 wi 11 serve both the main camp and village. The camp location wi 11 separate 1 iving areas from the work areas by a rni le or more and keep travel time to work to less than 15 minutes for most per- sonnel. The camp/vi 11 age wi 11 be constructed in stages to accommodate the peak work force as presented in Table 12.1. The facilities have been designed for the peak work force plus 10 percent for turnover. The turnover will in- c 1 ude a 11 owances for over 1 ap of work er·s, v ac at ions, and vis i tor-·s. Th~ con- ceptual layouts for the camp and village are presented on Plate 36 and 37. (i) Site Preparation Both the camp and the vi 11 age areas wi 11 be c 1 eared in select ar·eas for topsoil, and the topsoil will be stockpiled for future use in reclamat1on operations. At the vil1age site, selected areas w1ll be left with trees and natural vegetation intact. Both the main camp and the vi 11 age site have been se 1 ected to ~:n~o­ vide well-drained land with natural slopes of 2 to 3 percent. A granul.ar pad varying in thickness from 1 to 8 feet wi 11 be p 1 aced at the main camp, covering most of the areas inside the perimeter fence. This will provide a uniform working surface for erection of the high density housing and service buildings and will serve in certa1 n areas to protect the permafrost \'/here it underlies the camp. In the village area, a granular pad will be installed only as neces- sary to support the housing units and to provide a ~uitable base for construction of the temporary village center buildings . .. I I I I I I I I I I I I I I I I I I I .I I I I I a· I I I I I. I I I All roadways within the camp/vi11 age areas will be flanked by road- side ditches; with culverts carrying water across the intersections. In general, drainage wi 11 tie through construction of a surface net- work of ditches. Peripheral ditches will intercept overland flows from adjacent non-cleared 1 and and carry them around the camps" Runoff will ~ltimate1y be directed to existing drainage channels 1 ead ing to Tsusena Creek for the vi 11 age, and the Sus itna River for the main camp. ( ii) Facilities Construction camp buildings \'lill consist largely of trailer-type factory-built modules assembled at site to provide the various facilities required. The modules will be fabricated complete with heating, lighting and plumbing services~ interior finishes, furnish- ings, ·and equipment. Trailer modules will be supported on timber cribbing or blocking approximately 2 feet above grade. Larger structures such as the central utilities building, warehouses and hospital will be pre-engineered, steel-framed structures with metal cladding. The larger structures will be erected on concrete-slab foundations. The slab will be cast on a non-frost susceptible layer at least the thickness of the annual freeze/thaw 1 ayer. Permawalks will connect the majority of the buildings and dorr:ai- tor i es ~ Th.e permawa 1 ks will be heated. The various buildings in the camp are identified on Plate 37 .. A detailed description of the ·nature and function of the buildings is presented in Appendix 08. (c) Permanent Town The permanent town \'li 11 be 1 oc a ted at the north end of the temporarJJ v il- l age (see P1 ate 35) and be arranged around a sma 11 lake for aesthet i\:c pur- poses. The permanent town vtill consist of permanently constructed building:s :and not factory built prefabricated type modules. The various buildings in the permanent town are listed below: Single family dwellings; -~lultifamily dwellings; -Hospital; School; Fire station; -A town center will be constructed and will contain the following~ . a recreation center . a gymnasium and swimming pool . a shopping center 12-5 {d) The concept of ~uilding the permanent town at the beginning of the con- struction period and using it as part of the temporary village was consid-· erect.. This concept was not adopted, since its intended occupancy and use is a minimum of 10 years away, and the requirements and preferences of the potential occupants cannot be predicted w1th any degree of accuracy. Sits:: Pov1er and Uti 1 it ies -. ( . ) 1 :t ( i i) 0 Power Electrical power will be required to maintain the camp/village and construction activities. A 345 kV transmission line will be con- structed and will service the site from 1987 onward. The 345 ·kv transmis~ion line will be operated at 138 kV while it is bringing in power to the site. After the Watana development is complete and in operation, the transmission 1 ine will supply power to the Intertie from Watana and will operate at 345 kV. Since the transmission line will be required after· construction is complete, the only cost of the line attributable to the camps will be lhe interest costs in- volved in constructing the line 6 years earlier than required for permanent operation. During the first two years of constructio~ (1985 and 1986), the pm'/er supply will come from diesel generators. ·These generators will remain on site after 1987 as standby power supply since site contractors will provide for their own construction pm·1er after this time. The peak demand -during the peak camp population year is esti- mated at 13 MW for the camp/vil 1 age and 7 MW for construction re- quirements, thus totaling 20 MW of peak demand. The distribution system in the. camp and village construction \'li1l be 34. 5 k v. . Power for the permanent town will be supplied from the station ser- vice system at the power plant. Water The \'later supply system will provide for potab1e water and fire pro- tection for the camp and village construction and selected contrac- tor's work areas. The estimated peak population to be served will be 6,800 (5,000 in the camp and 1,750 in the village). The principal . source of water will be Tsusena Creek, vlith a back up system of wells drawing on ground water. The water wi 1l be treated in accordance with the Environmental Protection Agency 1 s (EPA) pri- mary and secondary requirements. A system of pun1ps and constructed storage reservoirs wi 11 provide the necessary system demand capacity. Distribution \vi11 be by duc- tile iron pipe system contained in utilidors. The utilidors will be I ·a·". . . ;I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I .I I I plywood box sections, most of which will be integrdl with the perma-- walks. The distribution and location of major components of the water supply system are presented in Plate 35. · Details of the util- idors are =presented in Plate 38. (iii) Haste Water A waste \'later collection and treatment sys tern will serve the camp/ village. One treatment plant 'r'lill serve the camp/village, while gravity flow lines with lift stations will be used to collect the waste water from all of the camp and village facilities. The uin- camp•' and 11 in-village 11 collection systems will be run through the permawalks and utilidors so that the collection system will be pro- tected from the elements. At the village, an aerated collection basin will be installeo to collect the sewage~ The sewage will be pumped from this collection basin through a force main to the sewage treatment plant. An aerated collection basin will be needed at the village to balance out the highly cyclic waste water flows. The chemical toilets located around the site will be serviced by sewage trucks, which will discharge directly into the sewage treat- ment plant. The sewage treatment system will be a biological system with lagoons. The system will be designed to meet Alaskan state water 1 aw secondary treatment standards. The 1 agoons and system wi~ 1 be modul at to all ovJ for phased growth and contraction of the camp/ • 1 1 v1. age. The location of the treatment plant is shown in Plate 37. The loca- tion was selected to avoid unnecessary odors in the camp as tne winds are from the SE 4 percent of the time, which is cons ioe~ed minimal. The sewage plant will discharge its treated effluent through a force main to Deadman Creek. All treated sludge will be disposed iG a solid waste sanitary landfill. (e) Contractor 1 s Area The onsite contractors \vill require office, shop, and general work ar·eas. Office space for the contractors has been provided and its location is shown on Plate 35. Partial space required by the contractors for fabrication shops, mainten- ance shops, storage or warehouses~ and \•Jork areas within the camp confines has been designated and is shown on Plate 37. Additional space requir-ed for the aforementioned items will be located bet\veen the main camp and the main access road. 12-7 12~4 -Diversion (a) General Diversion of the river flo~tl during construction wil1 be accomplished \~ith two 38 foot diameter circular diversion tunnels. The tunnels will be concrete-lined and located on the right bank of the river. The tunnels are 4,050 feet and 4,140 feet in length. The diversion tumiels are shown in plan and profile on Plate 39. The tunnels are designed to pass a flood with a return frequency of 1:50 years, equivalent to peak inflow of 81,100 cfs. Routing effects are small and the peak fl 0'!/ the tunne 1 s \vi 11 d i sci1 arge a peak flow of 80,500 cfs, The maximum water surface elevation upstream of the cofferdam is El 1!53ti. -A rating curve is presented in Figure 12.1. The upper tunnel or tunnel No. 2 will be converted to a permanent low level I I I I I I outlet after construction. The local enlarging of the tunnel diameter to I 45 feet is to accommodate the low level outlet regulating devices after the tunnel is used as a diversion tunnel. (b) Cofferdams I As discussed in Section 9 t~2 nature and riverbed will require a slurry wa11 th~ough the alluvium mat2rial to bedrock. The soil/bentonite slurry \vall will be constructed through the ·c"!osure dam and alluvium material to bedrock in order to minimize the amount of seepage into the main dam exca- vation. The abutment areas above riverbed ~tJill be cleared and grubbed~ with excavation of all overburden material to sound rock prior to placement of any cofferdam fi 11 . The upstream coffer·d am will be a zoned embankment founded on the c 1 a sure dam (see Plate 40). The closure dam \'lill be constructed to Elevation 1475 . based on a low water Elevation 1470, and \vill consist of coarse material on the upstream side grading to finer material on the do¥mstream side. \'Jhen the closure dam is completed the soil/bentonite slurry wall will be con- structed to minimize seepage into the main dam foundation excavation-A dewatering system will be established in the main dam excavation to control . inflow from ~he abutments and runoff. Above Elevation 1475 the cofferdam \vill be a zoned embankment qms1st:1ng of· a central impervious core, fine and coarse upstream and downstream filters, and rock and/or gravel supporting shell zones ''lith t"ip-rap on the upstream face. The downstream cofferdam wi 11 be a closure dam constructed from approximate Elevation 1440 to 1472. It will consist of coarse material on the dO\'In- stream side grading to finer material on the upstream side, with a soil/ bentonite slurry wall can be constructed in the finer material as described above for the upstream cofferdam, 12-8 I I I I ., I I I I I I I I I I I ~. I I I I I I I I I I. I I I The upstream cofferdam crest elevation has a 9 foot freeboard allowance. This includes 3 feet for settlement and wave runup and 6 feet for upstream reserve 1 r ice protection 4 Large chunks of ice will be present in the river during the spring flood. The 6 foot freeboard prevents ice from overtopping the cofferdam, causing damage. (c) Tunnel Portals and Gates Structures A reinforced concrete gate structure \'li 11 be located at the upstream end of each tunnel, housing two closur·e gates (see Plate 41). E~ch gate will be 40 foot high by 16 foot wide separated by a center con-_ crete ~ier. The gates will be of the fixed roller vertical lift type oper- ated by a wire rope hoist. The gate hoist will be located in an enclosed, heated housing. Provision wi1l be made for heating the gates and gate guides. The gate in Tunnel No. 1 will .be designed to operate ~'lith the res-. ervoir at elevation 1,540:~ a 50 foot operating he_ad. The gate in Tunnel No. 2 will be designed to operate with the reservoir at elevation 1,540, a 120 foot operating head. The gate structures for Tunne 1 s No. 1 ail d 2 \vi 11 be designated to withstand external {static) heads of 130 and 460 feet, respectively. The downstream portals will be reinforced concrete struc- tures with slots for stoplogs (see Plate ). {d) Final Closure and Reservoir Filli!!_g_ As discussed in Section 9 one of the diversion tunnels will be converted to a low level outlet or emergency release facility during construction, It is estimated one year will be required to construct and install the permanent low level outlets in the existing diversion Tunnel No. 1. This will require the that lower Tunnel No. 2 pass all flows during this period_ The main dam will be at an elevation sufficient to allow a 100 year recur- rence period flow (90,000 cfs) to pass through Tunnel No. 2. This flow will result in a reservoir elevation of 1625. During the construction of the low level outlets, the intake operating gate in the upper Tunnel i'~o. 1 will be closed. Prior to commencing operation of the 1ow level outlets, coarse trashracks will be insta11ed in the upstream intake structure in slots provided. Upon completion of the low level outlet in the upper tunnel the inta~e gate will be opened and the low level outlets will commence operation. Upon commencing operation of the low level outlets the lower Tunnel No. 2 will be temporarily closed wi.th the intake gates ·and construction of the per""m- anent plug will also commence upon filling of the reservoir. It is esti- mated it will take 12 months to completely place and cure the plug. During this time the main dam crest elevation will have reached an eleva- tion sufficient to start reservoir impounding and have sufficient storage available to store a 250 year recurrence period flood corresponding to a reservoir elevation of 1,890 feet. . During this time "the low level o~tlets will be passing the minimum summer and winter flows of 6,000 cfs and 800 cfs. 12-9 T~.e filling of the reservoir will take 4. years to comp1Elte to full reser- voir operattr~~ elevation of 2185 .. After 3 years of fi 11 i ng the reservoir \'till be at flevation 2150 and will a11ow operation of the po~Jerplant to commence .. The filling sequence was determined from the main dam e 1 evat ion at that time during construction, the starting reservoir pool elevation at that time during construction. and the capability of the reservoir storage to absorb the tnflow volume from a-250 year recurrence period inflow without overtopping fhe main dam~ The 250 year recurrence period flood volume was selected to be consistant with the recurrence period flows and risks used for the design of the diversion and entire project. This information is presented graphically in Figure 12.2. Once closure of the upper d1version tunnel is made trashracks will be in- stall.ed at the upsj:-."eam entrance to the tunne 1. The racks wi 11 serve to prevent debris from being drawn into the intake and damaging the high pres- sure slide gates.-The trashracks will be a permanent installation with no provision for remoyal except with the assistance of divers. The trashracks will have a bar spacing of about 3 feet and will be designed for a maximum differential head of about 40 feet. To limit the maximum net velocity through the trashracks to abo.ut 12 ft/s the racks wi 11 be semi- circular. Provision will be made for the monitoring the head loss across the trashracks. 12.5 -Emergency Release Facilities The upper diversion Tunnel No. 1 will be converted to a permanent low level out- let, or emergency release facility. These facilities wi 11 be used to pass the - required minimum discharge during the 4 year reservoir filling period and will also be used for draining the reservoir in an extreme emergency case. The facility will have a capacity of 30,000 cfs at full reservoir pool and will be capable of drawing the reservoir dam in 14 months. The reservoir drawrlo\vn time incorporating the low level outlets presented graphically in Figure 12.2 for various "start 11 times during the year. Ouri ng operation, energy wi 11 be desi gr.ated by means of two gated concrete p 1 ugs separated by a 340 foot length of tunnel (see Plate 43). Bonnetted type high pressure slide gates will be installed in the expansi0n chamber tunnel plugs for the upper diversion tunnel, once closure of the tunnel is made. The gates will be arranged in groups of 3 gates in series, each group consisting of 1 upstream emergency gate and one operating gate in the upstream plug and one operating gate in the downstream plug (see Plate 43). The slide gates will be 7.5 feet wide and 11.5 feet high and will be ofwe1ded stee 1 construction. The 9 ates wi 11 be designed to withstand a tot a 1 static head of about 740 feet~ however will only be operat~d with a maximum head of about 12-10 I I I I I I I I I I I I I I I I I I I I. I I I I I I I I I I I I· I I I I I 560 feet. To reduce energy dissipation problems, all three groups of gates \-Jill be operated only v1hen the head is less 460 feet. · Uuring operation, the operating gate opening in the upstream plug will be equal to the opening of the corresponding gate in the downstrean plug. This will effectively balance the head across the gates~ The maximum operative head across a gate will not exceed about 320 feet. Each gate will have a hydraulic cylinder operator designed to raise or lower gate against a maximum head of 560 feet. Three hydraulic units will be in- stalled, one for the emergency gates, one for the upstream operating gates and one for the downstream operating gates. The nominal operating pressure will be 2,000 psi. Each gate will have an opening/closing time of aoout 30 minutes. A grease system will be installed in each gate for injection of grease between the gate leaf and gate body seats to reduce frictional forces when the gates are operated. Both local and remote operation of the gates will be possible. As no facilities will be installed for dewatering th~ area around the emergency slide gate, the design of the gate will be such that the hydraulic cylinder as \'le 11 as the cylinder packing may be inspected and rep a ired without dewatering the area around the gate. An air vent will be installed at the downstream side of the operating gate in the downstream plug. Air will be dr-awn from the access shaft. The vent \'fill be heated as necessary to ensure that freezing will not occur. To prevent concrete erosion, the conduits in each of the tunnel plugs wi11 be steel 1 ined. The upstream gate operating chamber will be separated from the downstream cham- ber by a water-tight bulkhead door. In the event of a major failure of the up- stream emergency gate, water would be contained \vithin the upstream gate-oper- ating ch ~her. 12.6 -Nain Dam (a) Comparison with Precedent Structures The main dam at Watana, as currently proposed \vith a height of 8B5 feet will be among the highest in the world. The highest embankment darns com- pleted in North America are ~lica Creek in British Columbia {794 feet} and Oroville in California (771 feet). Two dams under construction in the USSR will exceed 1,000 feet, but the only dam completed to a height in excess of 800 feet is Sulak in the USSR. A list of embankment (earth and rockfill) dams in excess of 500 feet completed, under constl~uction or proposed is given in Table 12.2. The Watana site is located in a seismically active area and the major· de-: sign features of 24 embankment dams between 350 and 795 feet in height con- structed in seismic areas are summarized in Table 12.3. The characteris- tics of the \~atana design which will be developed in this section are 12-11 included in this table for comparison~ Special site conditions, depth to bedrock, availability of materials, size of reservoir, s·ite location~ for example, all have an impact on the design and such factors account for some of the extremes quoted in the table. A further comparison is given in Table 12.4 which includes the princip~l. geometrical parameters of the core and outer slopes for high dams in seismically active areas .. Considering these various parameter: -The freeLoard ranges between 13 and 62 feet, with seven of the eleven cases quoted being 1 ess than the 25 feet proposed for Wat an a. -The crest width ranges between 33 and 111 feet. Wide crests are usually the result of non-structural requirements, i.e., a roadway across the dam. Neglecting extreme widths, seven of the ten cases quoted are between 30 and 40 feet compared with the 35-foot width proposed for Watana. -The core \'!idth ratio ranges between 0.29 and 0.56, with only Gne example higher than the 0.50 ratio proposed for Hatana. -The upstream slopes range between-2.u:l and 2.7:1. The Japanese dams tend to have flatter slopes (\vithin the range 2.5 to 2.7), while the North American dams are in the range of 2.0 to 2.6. The Watana s1opeof 2.4:1 is among the steepest, but is flatter than the next two highest dams, ivlica at 2.25:1 and Chicoasen at 2.2:1. However, special features are included in the Watana design, primarily the use of free-draining pro·:essed gravel in the upstream shell, to minimize the effects of earth- quakes on the stability of the upstream face of the dam. -The downstream slopes range between 1.8:1 and 2.7:1. Ten of the 15 cases quoted are equal to or steeper than the 2:1 slope proposed for \~atana, while only one case is flatter than 2.2:1. Complete details of the core materials used in all the dams list~d in the accompanying table~ are not available in the literature. However a number of large dams· have been constructed in Canada using similar glacial depos- its as core material. The mean grad{ng curves for these materials includ- ing that used for the core of the t11i c a Creek dam, the existing dam ge!1er- ally comparable to Watana in size~ materials, and location, are compar·ed with the mean grading curve for the core material proposed for the tiatan a dam in Figure 12.5. It is apparent f)•om this figure that the ~latana core material is \<Jell within the t'ange of materials used successfully for other large dams in North America. In summary, the proposed Watana design is generally conservative with re- spect to precedent design. Howeve.r, special features tthich are discussed in more detail later in this section are incorporated in the Hatana section to provide additional safeguards against seismic loading. 12-12 " ••• I I. I I I I I I I I I I I I I I I I. I I· I I I I I I I I I I I I I I I .. I I (b) Excavation and Foundation Preparation·-General The geology of the Watana site is described in Section 9. In summary, the existing conditions at the damsite comprise alluvial deposits in the river- bed up to HO feet deep overlying bedrock, while the lower slopes of the valley are covered \vith ta 1 us and there is overburden on the upper s 1 opes. The bedrock is jointed and \veathered at the surface with weathering along joints extending to considerable depths. Locally in shear zones and drain- age gullies the rock is weathered throughout to depths in excess of 40 feet. The frequency of joints and fractures generally decreases with depth but fractured and weather zones have been identified 1ocally at depths up to 200 feet. , Zones of permafrost occur, particularly in the south abut- ment. The dam foundation must satisfy the following basic requirements: -The foundation under the core must be stable and capable of supporting the weight of the core under all loading conditions, must not erode under the seepage gradients which wi 11 deve 1 op under the core, and must not allow excessive seepage losses under the core. The foundation under the upstream and downstream shells must be stable and capable of supporting the weight of the darn without excessive move- ment under all loJding conditions. -The core material must be pr·evented from moving down into the foundation (e.g. into cracks ~r joints) and then through the foundation under the transition zone into the downstream shell or beyond. -The contact bet\veen the core and its found at ion must remain watertight despite the distortions that will occur in the dam because of its uwn weight and the thrust of the reservoir. -Any seepage through the foundation must be contra 11 ed and discharged so that excessive seepage pressures do not develop in the downstream port ioo of the cm"e, in the materia 1 s beneath the she 11, or do~m stream of the dam. The excavation and foundation preparation necessary to meet these objec- tives are outlined in the following paragraphs. (c) Excavation (i) River Excavation The properties of the river alluvium are not well defined but it includes sands, gravels, cobbles, and boulders up to 3 feet or more. Such materials are not suitable as a foundation for the core:; pri- marily because of. their relatively high permeability. Such alluvial deposits have been left in place under both upstream and downstream 12-13 .. I shells of many of the world's largest dams (see Table 12.3). How-••. ever, at Watana .these granular materials could undergo 1 iquefaction under seismic loading with potentially catastrophic results. In- sufficient data is available to demonstrate thatthere is no pos-•... sible risk ot liquefaction of the alluvium, but further investiga- tions may provide data to support the concept of removing the allu- vium only under the central portion of the shells. However, in view 1 of the high seismicity of the area it is proposed that the river . alluvium be removed over the whole foundation area. ( i i) . Under Core and Fi 1 ters I The core and filters must be founded on sound rock to ensure that no material can wash through open joints. This will require excavation of overburden and talus on the slopes and v1!2athered rock in the valley bottom and on the abutments. The talus thickness on the abutments perpendicular to the slope varies from zero to an esti- mated 20 feet and weathered rock to 40 feet or more in some areas, Weathered rock is here defined as closely jointed or fractured rock with weathering and infilling. of the joints. The final foundation will be sound hard rock with only minor weathering, which can be grouted to ensurt. that core material cannot be washed through joints in the rock~ The maximum rock slope along the abutments is determined to some ex- tend by the valley shape. In general, lH:lV slope or flatter is ideally preferred although steeper slopes have been used. At Watana damsite, the natural slopes at lower elevations are relatively steeper but still genera 11 y less than lH !2V. It is therefore pr·o- posed that the overall core foundation slopes will be no steeper than 1H:2V below elevation 1800 and lH:lV above elevation 1800. The cross cut slopes will be lH:lOV. Local irregularities in the rock surface are undesirable because of the potential for differential settlements or strains in the core that could cause cracking and potential piping through the core~ Such irregularities also make it difficult to compact the core material to form a tight core-rock bond and they must be eliminated either by additional rock excavation or the addition of concrete to achieve an acceptable slope. Such slopes would normally be on the order of lli: 2V in the 1 ov1er sections of the dam where cant act ores- sures are higher, flattening to 1:1 at higher elevations. · The depth of excavation required to remove unsuitable rock will vary considerably over the core contact area. In scme area very little excavation may be needed, while in highly \'v'eathered zones excavation· may extend to 50 or 60 feet. On the basis of available data., it is estimated that the average excavation under the core and filter-s will be 40 feet. 12-14 I I· I I I I I I I I I I I I I I I I I I I I I I I I· ·I I II (iii) Under Upstream and Downstream Shells The shells will be founded on competent rock. Loose or detached rock or rock ribs ana highly ~eathered rock will be removed to ex- pose sound rock. Weathering along joints and local irregularities in the rock surface will be acceptable. The actual thickness of rock to be excavated vtirl vary across the site; but it is estimated that the average will not exceed 10 feet. t •'~J Dental Excavation Dental excavation over and above normal excavation is expected in zones of intense shearing or highly irregular surface£. Whereas th~ need for such excavation has been identified by investigations com- pleted to date, the magnitude has not been properly assessed because of heavy vegetation, tundra cover, and general lack of outcrops and access problems. (v) Excavation Methods It is expected that the excavation of the overburden material within the dam foundation will be performed as a multi-level operation us-- ing wheel~d loaders working with dozers. ~oulders that cannot be removed by excavation equipment will be ,blasted. On the steep work-. ing areas wi 11 be formed with material excavated from the slopes above. These working areas will be progressively lowered removing overburden and weathered rock in one operation. The excavation of the foundation will need to be complete for safety , reasons from about Elevation 1800 down to the riverbed before plac- ing of fill is commeliced. The excavation on the upper slopes will then require to be kept sufficiently in advance of grouting and f il1 placement to avoid interference of these activities by the blast~ng. Excavation of unrippable weathered rock and trimming of the r~ck surface to acceptable slopes will require blasting which will l"e- strict other activities such as surface grouting. Numerous access roads will be required throughout the dam area to reach the various working levels. Dental excavation will be none by small backhoes and final cleanup of the area under the core and filters will be carried out to a high standard by hand with high pressure water jetting prior to grouting. The rock surface under · the core and filters should be clean enough for detailed geolugical mapping and for grout leaks to be observed and caulked if necessary. The foundation must be free from snow and ice before fill material· is placed. Selected alluvial material from the riverbed may be used in the downstream shell of the dam but the remaining material, generally a mixture of weathered rock and overburden, will be wasted or used for road, cofferdam or temporary facility construction. Spoil areas will generally be below final water level in the reservoir a~ea. 12-15 (d} _§;outing and PressureR?lief A combination of consolidation gr·out.ing and cutoff curtain grouting under :~he core and a downstream pressure relief (drainage) system are proposed tor the ~atana site. Those systems will result in: Improved stabi1ity of the foundation; Reduct ion in rock mass permeability and hence seepage through the founda- tion; Reduction in the risk of movement of soil particles through joints in the rock; .and -Contra 1 and safe discharge of any seepage flows through the grout curt a in. It is proposed that the curtain grouting and drilling for the pressur·e re~ 1 ief systPm. are carried out from galleries in the rock found at ion in the abutments and beneath the dam: Det a i 1 s of the grouting, pres sure re 11 ef ·and galleries are shown on Plat~ 46. - The purpose of grouting is to improve foundation and abutm~nt rock condi- :ions with respect to load bearing and seepage considerations. Tte n~od, e:tent, and detail of grouting is dependent on site geological condit10ns, type, and character of rock, reservoir head~ and locaticn of specific structures. The diorite bedrock at Watana is competent as far as load carrying capacity is concernerL However, numerous shear zones from \1 few inches to several feet in width, have been i~entified in a general NW-SE direction. Occasionally, the width of shear zones may be ~everal tens of feet locally. Most of these zones, which are found both in the river than- nel and in abutments, contain gouge material and under appropriate condi- tions, may be susceptible to piping .. These features are discussed in ffiOl'e detail in Section 9. The permeability tests in boreholes indicate the rock ma~s permeability at the Watana site to be generally in the range from 1 x lOb em/sec to 1 x lu-4 em/sec, indicating a maximum seepage rate through the foundation of t~e order of 4 cubic feet per second. However, these permeab i1 ity values may not properly account for shear zones. For example, in Bm~ehole BH-2 on the north abutment, circulation was lost during drilling when the boring encountered a shear zone. Also, because of heavy vegetation 3 talus cover and limited access: it is possiole that there may be other shear zones not yet identified. A properly co~1ducted primary grouting program of an exploratory nature will be required i'i1der the dam and in the abrtments and, depending on the results of this rrogram, additicaal grouting includ- ing multiple 1 ine curtains rnay be required. · 12-1 f I I .- 1 I I I I I I I I ' I I I -1 I I I I I I I I I -I I I I ·t I I I I I (i) Consolid,a~ion Grouting The rbck under the core, upstrean filter, and downstream filter will be consolidation grouted to provide a zone of relative1y impermeable rock under the entire contact. Locally, the rock may be sound and. free of any discontinuities resulting in virtually no, grout take; rievertheless, the Joints and shear zones are generally steeply dip- ping and any particular vert ica1 plane is 1 ikely to intersect these zones which are estimated to be lS to 20 feet apart. Consolidation grouting is estimated to require 30 foot deep holes on a 10 foot by 10 foot grid. (ii) Curtain Grouting The design of g~out curtains under dams is largely empirica1, though based on data from boreholes. At the Wa.tana site, only borehole DH-21 extends to a significant depth below the river to elevation 876 -feet, approximately oOO feet be1oVJ dam foundation level. Shear- ed and highly fractured zones are indicated at an aver~g8 of 50 feet intervals to the bottom of the hole and the upper zones ~hou 1 d be grouted to reduce seepage losses. The average rock permeability de- creases significantly around 200 feet depth~ A grout curtain is not -expected to be 100 percent effective in eliminating seepage, but rather to increase the length of the seepage path. Flow net analy- sis indicates that a positive ct..toff of 350 feet deep increases the potential flow path by a factor of 1.7, decreasing the average hydraulic gradient from about. 0.9 to 0.5. _For the purposes of this study, a double }~ow grout curtain to a depth of 0. 7H, where H is the head of water behind the dam at ::1 par- ticular location, \'lith a maximum depth of grout curtain of 35u feet has been assumed. Grouting \·lill be carried out from a series of underground galleries which will also serve the drainage syste~n pressure relief. It is likely that in some areas the grout take at depth will be very lov1. Primary holes will be considered as exploratory holes a!na will be core dri11ed. On the basis of the core and vtater pressure tests in the exploratory holes, the depth of secondary holes can be ae- cided. The exploratory holes may also identify areas that need additional grouting. All h0les will oe water pressure tested in stages and the grouting program will be determined using these results. Grouting will be c-arried out using split spacing with the primary holes at 40 feet spacing. The secondary, tertiary and quaternary holes \·/Ould hr·ing the final hole spacing to 5 feet if required. ~ Permafrost in the area to be grouted will have to be thawed befor"e water pressure testing and grouting can be done. The greatest depth of permafrost so far recorded was in Bh-8 where the hole froze up to 12-17 0 . . 175 feet depth, Additional boreholes may be required for the thaw- Rock wi 11 be ho 1 es. ing of sufficient rock to form an effective curtain~ thawed by circulating clean river· water through -drill The effect1veness of the initial thawing and gtouting may be diffi- cult to assess but the permanent galler·ies under· the dam will enable additional grouting to be carried out at any time during the follow- ing reservoir filling and subsequent thawing of the foundation and abutments. It is desirable for uities as possible. from 80° to 60° and drill the boreholes the grout holes to intersect as many discontin- The dip of the main joint sets and shear range its is therefore considered most efficient to vertically or at an angle of 45°. A major shear zone approximately 600 feet wide trending in a NW direct ion intersects the left edge of the dam and reservoir area and the curtain should extend into the abutment to provide.a positive cutoff of this zone. The ground surface rises to the south of the dam and the surface express ion of any shears to the south of the major zone will be outs ide the reservoir area and are un 1 i kely to cause appreciable seepage. The extent of the grout curtain through the shear zone will be de- termined by t:xploration from the grout gallery. Artesian water pressure was observed at the shear zone in BH-12 in- dicating that materials with high permeabi1ities exist in the shear zone which must be effectively grouted. No major shears have been found on the right abutm 1t where the rock is of good quality. The grout curtain will extend ·rom the spillway intake structure 400 feet into the abutment with the depth of the curt a in set at a minimum of 200 feet. The spillway control structure is located on the dam centerline-and the grout curtain will extend beneath the structure with drilling and grouting from the gallery formed within the concrete· roll·Nay. Drainage will be provided behind the grout curtain with holes drilled from the gallery. (i·H) Drai!§ge and Pressure Relief -I . I I I I I I I I ..... II I I I I Drainage features are included beneath the dam foundation and the I abutments to intercept seepage through the grout curt a in and re 1 i eve pressure. Ccm~on drainage and grouting will be ~onstructed with grouting from the upstr·eam side and drainage from the downstream ·• side of the galleries. The use of galleries is recommended for the following redsons: 12-18 I I ., I I I I I I I I I I I I I ( i v) I, I I ·I· I -Curtain grouting from the gallery can be carried out independently of the construction of the darn. This can shorten construction times. -The grouting can continue longer into the winter than would have been possible with surface grouting. .... Permanent access is av a i 1 ab 1 e under the dam for inspection. Additional grouting or drainage holes may be drilled after construction of.the dam which is an important consideration where there is permafrost. The tha\'ling effect of the reservoir may require r·emed i a l grouting after impounding has commenced. - -Higher grout pressures can be used if required because of the overlying weight of embankment. -Drainage holes drill€d dmvnstream of the grout curtain \vill be discharged into the gallery enabling flow from individual holes to be monitored. This system will prevent the outlets of the drainage.holes freezing which is an essential requirement. Gallery drainage is more effective because pressures are relieved at a lower l eve 1 . -The galleries may be used for the installation of instrumentation and provide access for the repair and replacement of instrument at ion. :> -Tunneled galleries provide the great advantage as an explor-ation tunnel for the rock of the dam found at ion. The tunnel gives the best opportunity for understanding the nature of the rock along the grout and drainage curtains which will be invaluable in the f au 1 ted and sheared zones. The drainage/pressure relief holes will be drilled after all gt'OUt- -ing is complete. They will be 3 inches in diameter spaced at approximately 10 foot centers. Generally the ho 1 es wi 11 be open but any penetrating fractured or sheared rock may require perforated casing to prevent caving. Construction Methods -Grouting and Pressure Relief The schedule of work is of particular importance in this phase of the work. The excavation for the g a 11 er i es must be carried out be- fore consolidated grouting because the grouted rock mo.y be disturbed by the blasting for excavation. It will also be preferable to com- plete excavation of the dam foundation in a particular section be- fore excavation of the gallery so that the surface rock profile may be confirmed before tunnelling. 12-19 Hock temperatures will be measured any any areas of permafrost thawed prior to .grouting. Grout holes 't/ill generally be 1-1/2' inch in diameter~ Large hole sizes wil1 be drilled where explotatory cored holes are required or do\vn-the-hole hammer equipment is used. All holes will be washed and pressure tested before grouting. Grouting will be done with Type II Port1 and cement with 2 percent addition of bentonite (by weight of cement). The water/cement ratio and grouting pressures will be varied accordirtg to the conditions encountered. Grouting will be carried out in stages using packers. Some redrill ing between stages will be required. To allow greatest flexibility of the ~chedule, most curtain grouting, which will in- clude up-hole grouting will be done from the galleries. In the in- clined galleries special platforms will be required for drilling and grouting equipment. Primary grout holes will be treated as exploratory holes and core drilled with further core drilled holes as required to test the effectiveness of the grouting. The grouting progr ain will be modi- fied according to the rock conditions encountered as the work pro- ceeds. (v) Gallery Construction The layout of the galleries are shown on Plate 46. The. horizontal and inclined tunnels will be excav(lted by conventional drill and blast methods. Vertical shafts will be raise bored providing a smooth excavated profile with little support required. It is ex- pected that the majority of the gallery length \vill not require any support but from avail able geologic data it is estimated that aoout 25 percent will require rock bo 1 t and shot crete support.. Stee 1 arches vlill be required at the portals ano at tunnel junctions or in highly fractured Ot" shear·ed zones. A concrete slab will be c~st in the tunnel invert to provide an even working surface and to fo~"r.l the drainage channel. Measuring weirs will be cor structed in the drainage channels i~ order that the volume of sewage water may be monitored. The seepage water will be discharged from the gallery just above tail water l eve 1 through drainage tunnels extending to the downstrearn toe of ··he dam. The drainage outlet of these tunnels \vill be located under taih1ater level to prevent icing up of the outlet. Inspection access will be provided at the down stream toe of the dam but from a separate porta 1 above water level. Lighting for inspection of the galleries and ventilation wil1 te re- quired. The fresh air intake during the winter must be heated to prevent freezing of seepage water within the tuPne1 s. The ventil a- t ion will only be required occasionally when personnel are in the tunnels. Elevators will be installed in the vertical shafts to- gether with emergency stairs and cable hoists installed in the in- c 1 i ned tunne 1 s for· movement of equipment. 12-20 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I {e) Main Dam Embankment The main dam v1ill consist of a central compacted core protected by fine and coarse filters on the upstream and do~tmstream slopes. ·The downstream outer shell wi11 consist of roc'f< fill and alluvium gravel; and the upstream outer shell of clean alluvium gravel .. A typical cross section is shown on Plate 45~ (i) Comparison of Vertical and Inclined Cores The design of the embankment is dependent on the type of core chos- en, either a vertical core or an inclined core, and its location, upstream or central in the embitnkment. The advantages to each type of cote are as fo l1mvs: -Vertical Core Provides better contact with the foundation; Provides slightly more thickness of core for the same quantity of the core material; ana Settlement of the core will be independent of the post- construction or seism.ic displacement settlement of the do\vnstream shell. -Inclined Core Can place bulk quantity of downstream shell before p1acing core material; and Can carry out foundation treatment duri119 placement of she1~ material. The major disadvantages for each type of core are as follow-s: -Vertical Core Placement of core material controls placement of filters ana shell mat~rials; and Possible arching of a thin core oy transferrins weight to c;~jacent filters and shell materials during settlement or seismic d~sp1ace­ ments. -Inclined Core Excessive ~est-construction settlement or seismic displacment of downstream shell may cause rupture of core; and Locatiorr of core may effect upstream slope by ma~ing it flatter for stability reasons. '1? 2 .. 1~:.--I A central vertical core was chosen for the embankment based on a re-- view of precedent structures discussed above and.the nature ~f the proposed impervious material. The proposed impervious material is a combination of glacial outwash and tills with a \'lide grain size distribution~ This material is nonplastic and would tend to crack rather than deform under tensile stress and-hence may be susceptible to erosion. For a sloping core the possibility exists of cracks developing in the core for a non- plastic material because of lateral settlement or displacement dur- ing a seismic event. It also becomes difficult to avoid high ten- sile and shearing stresses in an inclined core. Settlement data indicates that the magnitude of water load settlements in rockfill dams may increase at a rate greater than direct pro port ion to the height of the dam. For these reasons a central vertical core will be used in the \~a tan a Dam cross sect ion. (ii) Earthquake Resistance Design Features Bee a use of the apparent lovl p 1 ast i city of the materia 1 to be used in the impervious core and the requirement for an earthquake resistant design, the following design features will be incorporated into the main dam cross section: -The cor--e-foundation contact will be widened near the ends of the embankment to ensure seepage control during normal operating con- ditions and any seismic event. -Thick filter zones will be placed upstream and downstream fr..,m the impervious core to prevent breaching of the core from either post- construction settlement and cracking or from any cracking result- ing from a seismic event. The filters will be designed to be self-healing in case of trans- verse cracks in the core resulting from either post-construction settlement or a seismic event. -The downstream filters will be designed to be capable of hana1 ing any abnormal flmvs which could result from transverse crack~ng at the core from post-construction settlement or a seismic eve~t. -The proposed ~tlidth of the core will prevent arching of the core caused by transfer of load to the shell or filter materials. I I ·I· I I I I •• I ··, I I -Compacted river alluvium gr·avel and rock fill vlill be usect to con- struct the downstream outershell. Compacteo processed clean rivet alluvium gravel of high permeability will be c~~d to construct the upstream outershell to minimize settlement displacem~nt ana the build up of pore pressures during a seismic event and to ensure I rapid dissipation of any pressures which may occur. , · 12-22 . . _, . . . ,, I I f -. . . . , . . . . . . I I (iii) I I I I I I I I I I I I :I I • , I I ··~ .. Freeboard and Static Settlement ~--" 't'he minimum required crest elevation of the ~·1atana Dam, not includ- ing static and seismic settlement, was determined for each of the following conditions: Norma1 maximum reservoir el ev at ion Storm surcharge Hater e 1 ev at ion Wave runup allowance Dry freeboard allowance Elevation top of cbre Roadway thickness Minimum crest elevation 1:50 Year Storm 2185 b 2191 6 3 2200 3 2203 1:10,000 Probable Year Storm Maximum Flbod 2185 8 2193 6 2199 3 2202 2185 17 2202 3 2205 These elevations refer to the maximum section of the dam and are based on a normal operating reservoir level of 2185 feet. The gov- erning minimum crest elevation excl.',..lding static and seismic settle- ment is 2205 feet at the maximum sect ion and at the abutments, This is the lowe5t elevation acceptable for the dam crest and allow- ances must be made for static settlement of the dam following its cnrnpletion, settlement on saturation of the upstream shell, and pos- sible slumping because of seismic loadin9. It has not been· pcssible to perform detailed calculations at this time to determine the like- ly settlements since no test data is available. For purposes of this feasibility study, it has been assumed that seismic slumping could be up to 0.5 percent of the height of the dam and the design crest elevation at the center of the dam is, t~nere­ fore, shm'ln at 2210 feet, 5 feet abuve·, the required minimum e:~va- t ion of 2205 feet. An allowance of 2 feet has been made at b~th abutments and hence the design crest elevation rises from 2207 at the abutments to 2210 feet at the center. Thus, under normal nper- ating conditions the minimum fr"'eeboard relative to the maxim~t.ri oper- ating poo 1 e 1 ev at ion of Llo5 wi 11 be 22 feet at the abutments and 25 feet at the center of the dam. If for any reason the crest settles below 2210 feet; more material should be added to maintain the safety margin-of 5 feet to al 1 cw for seismic slumping . 12-23 ... An additional allowance must also be made for post-construction settlement of the dan under its own \'Ieight and because of the effects of saturation on the upstream gravt:1 fill when the r·eservoir is first filled, This allowance is not shO'tll1 on the drawings since it is not a permanent requirement~ However~ for initial estimating purposed;, 1 percent of the hei~ht of the dam has b.een allowed~ Hence, at the end'of construction the dam crest at the center-of the dam would be at 2,210 feet plus 9, or £..,219 feet. The additional height constructed into the dam wou1d be achieved by steepening both slopes above approximately elevation 1850. Further margin agq.inst overtopping of the mafn dam is provided by the emergency spillway. Under normal operation this spilhvay is · sea 1 ed by a fuse p 1 ug or dam across the entrance channe 1. This p 1 ug is simply a gravel dam with special design of the core and strict contra l of the she 11 mat_~r i a 1 s to ensure that it wi 11 et·ode rapidly when ov~rtopped, allowing flood flows to be discharged freely down the emergency spillway. ' The location and typical cross section through the fuse plug are shown on Plate 53. The fuse plug has a total length of 310 feet and a height of 31.5 feet. A pilot channel 20 feet wide with an invert 1.5 feet lower than the crest, is provided at the center of the fuse plug, to start the washout at a predesignated location. The. loca- tion of this pilot channel is chosen so that the flow of water dur- ing washout vd 1l be smoothly channeled into the spillway chute •. {iv) Typical Cross Section The typical cross section of the main dam is shown in Plate 45. The central impervious core slopes are 1H:4V wlth a crest width of 35 feet. The thickness of the core at any section will be slightly more than 0. 5 times the head of water at that sect ion. Nin imwn core-foundation contact will be 50 feet requiring flaring of tae cross section at each end of the embankment. The upstream and dmvnstream fi 1 ter zones increase in thickness from 15 or 30 feet near tne crest of the ctam to a rn .dmum in excess of 60 feet. They are sized to provic .. protection against possible piping through transverse cracks in the core that could occur bt::cause of settlement or resulting from disp1ao"'"::;-2nt during a seismic event. The shells of the dam will consist of compacted alluvium gravels. To minimize pore pressure generation and ensure rapid dissipation during a seismic event, the saturated upstream shell will consist of compacted clean alluvial gravels pr'Ocessed to remove fines so that not more than 10 percent of the materials is less than 3/8-inch in size. The dmvn£tream shell will consist of compacted unprocessed alluvial gravels and rockfill from the excavations for unctarg}~ound v10rk s i nee it ltd 11 not be effected by pore pressure generation during a seismit event. 12-24 I I I, I ,: I 1: I I I I I I I IJ I I I I ·I I 'I . I I I I I •• I I I I I I I I I Slope protect ion on the upstream s1 ope wi 11 consist of a 10 foot zone of oversize material up to 24 inches in diameter, placed and compacted by suitable equipment. The typical crest detail in shown in P1ate 45 .. Because of the ttar- rowing of the crest dam, the filter zones are reduced in width ana the upstream and downstream coarse filters are el iminat>:?d, A 1 ayet of filter fabric is incorporated to protect the core material from dam age from frost penetration and dessication, and to act as a coarse filter where required . (v) Core Material Properties The core material wi1l be. obtained from Borrow Ar~t:a D, locatec on the right bank of the river, upstream from the dam. The area con- sists of a series or glacial deposits separated by alluvial an.: la.- custrine deposits. A generalized surficial stratigraphic colu~1 in Area D, based on all investigations to date including seismic lines and deep drilling, is given in Table 12.5. Typical gr·ading curves for each unit are presented in Figures 12.6 through 12.10, which also indicate the range and average moisttire content of each unit. It is proposed to blend material from the various units as required to provide core material with a max~~um particle size of 6 inches Qnd within' specified limits of moi~ture content, gradation (as shown in Figure 12.11) 3nd plasticity~ The composite gradation curve from Borrow Area 0 is shown in fig~n:: 12.12. The Atterberg limits will be within the following ranges: -Plasticity Index-0 to 20 -Liquid Limit -10 to 45 Permeability tests indicate a permeability on the order of 10° em/sec, which is within the normal range for glacial deposits ~sed in simi 1 ar dams. Modified Proctor Compaction tests on material passing 3/4-ir . .:::~i sit:~ve indicate an optimum moisture content oA' 7.5 percent with a ma;{·hnum dry density of 135.5 pcf. Standard Proctor compaction test r~sults on material passing No. 4 sieve indicate optimum moisture cont~nt of 10.4 percent with a maximum dry density of 127.6 pcf. The tes~ re- sults are plotted in Figures 12.13 and i2.14. The natural water cont~nts of samples tested range from 7 to 21 per- cent with occasional samples of finer grained material up to 4u per~ cent. Blending and processing of the core material will be ne~es­ sary vlhile pockets and layers of very ~1/et material ',vill be left in the borrow areas or otherwise wasted. 12-25 Consolidated undrained test results (see Figure12.15) at 95 percent ~1odified Proctor density and 2 percent above optimunl moisture con- tent, indicate the angle of shear strength resistance {VJ) equal to 37~, with a cohesion intercept (c) of zero~ Pinhole dispersion tests inaicate that the proposed core matet'ial is a non-dispersive mttterial. Consotidation tests indicate a compression indec (Cc) of o .. Ob and test results are shown in Figure 12.16. (vi) Excavation, Placement~ and Compaction of Gore Material -. The borrow area wi 11 be exc ~v a ted to a depth of approximate 1 y 30 feet working vertical faces. Processing and blending of the mater·i- al will be done during excavation. Oversize material (greater than b inches) \'/ill be removed by grizzlies or raked out of' the fill dur ... i ng spreading. Frozen mater i ~1 \<Ji 11 have to be 1 eft in place or . loosened by blasting and ripping for haulage to waste area. ~1ois­ ture conditioning will be done in the borrow area. Material will be placed in 8-inch compacted lifts at a maximum mois-- ture content of 3 percent above optim~m moisture content, and com· pacted to 95 percent of the maximum density obtained from the ::;~an­ ·ctard Proctor test. Type of roller, number of passes, thickness of 1 ift and moisture content can be adjusted basr=d on field tests and equipment to be usea. (vii) Fine and Coarse Filter Materials . Fine and coa)~se filter matflria1 wi11 be obtainE:d from narrow ;'-\raa E:, The material wi11 be proce:;.;se· to provide ths:; required gradat ivns. Frozen material will> wh~~0 ~ossible, oe allowed to progressively thaw insitu, with a system of surface ditches to accelerate .. :a"ainage of the thav1ed material. Where this is not practical for scheduling reasons or other considerations, the foreign material will be loos- ened by ripping or blasting and hauled to a disposal area. ~nisture conditioning will be done in the borrow area. Criterion 1: The 15 percent (015) of a filter material must :Oe not more than five times the 85 percent size (085) Df a prot2cted so i 1 . Criterion 2: The 15 percent size (015) of a filter material should be at least five times the 15 i)ercent size {015} of. a protected soil . Criterion 3: The 50 percent size (D5U) of a filter material must be not more than 25 times the 50 percent size (050) of a protected so i 1 ~ 12-26 I I I I I I' I I I I I I I I I I I I I I I I I I I I 'cl I I I I I I I The required gradations of the fine and coarse filter material to satisfy the above criteria are sho~'fi T Figure 12.11, whi1e compos- ·.· i te gradation:; for Borrm'l /l,rea E a 1M1 in Figure lc .ltL Permeab i 1 i ty of the fine f i1 ter and ... ;se filter is estimated to be greater than 1 em/sec and 10 em/sec, respectively. Permeability will be verified by large scale field or laboratory tests. The fine and coarse filter material are assumed to have an angle of shearing resistance (0) of 35:. for the purposes of these studies~ Actual properties will be determined from large scale triaxial tests ard/or modeling the gradation for standard triaxial tests for final de~ign. (viii) EY.:avation, Placement,; and Compaction of Filter Material The borrow areas will be developed utilizing scrapers and dragli·,~s which will supply the required amounts of fine and coarse fi1ter material construction. Material will be processed by screening and blending Gsing wet screening methods. Oversized material will have to be removed and either used as an aggregate source or possibly used in the outershell of the dam~ The method of placement and compaction will depend on the r.c~;.i 1 ts of full scale test fills to be done prior to construction using the proposed equipment ana materials. It is assumed that 12-inch lifts with four passes of a large vibratory roller will provide there- quired compaction. ( i x) A 11 uv i a 1 Fill l'tlat erial The alluvial fi 11 wi 11 be obtained fro'il Borr0\'1 Areas E and I, The upstream shell of the dam will be constructed using processed allu- vial gravel and the downstream shell of unprocessed alluvium fill material mixed \'lith rock from the various excavations, when a\rail- able. Any oversized material (greater than 18 inches) will be either used in the rip-rap zones or crushed for concrete aggreg·ate. The gradation of the available alluvial fi11 material will be as shown in Figures 12.16 and 12.17, while the: required grading limits for the upstream shell are shmvn in Figure 12 .11. The do~tm·st.~-.cam shell material will not require proces· ing. ~·!aximum size of r·iver gravel will be 18 inches in the greate5L dimension. Permeability of the processed alluvial fil1 is estimated to be greater than 100 em/sec. An angle of shearing resistance of ~ equal to 35Cl has been asstnned for the alluvial fill material. 12-27 • I Actual properties ~.,n1 be ·determined from 1arge scale triaxial tests ·I and/or modeling the gradation for standard triaxial tests for final design. (x) Excavation, Placement, and Compaction of Alluvial fill Naterial The alluvial f i 11 materia 1 wi 11 be obtai ned from the main dan foun- dation excavation and from dO\'!nstream from the main dam. ~~ethod of excavation will be by scraper operations above the water table and dragl ine operation below the water table to a maximum depth of ~0 feet. The mater i a1 wi 11 have to be processed to remove the undet·- s ized and oversize material for the upstream she11. All material in the si·:..ll:; must be \1/ell compacted to minimize post- construction settlement and seismic slumping. The method of p1are- ment and compaction will depend orythe .results of full scale test fills to be done prior to construction using the proposed equipment and material~ It is asst:n'ed that 24-i(,.:t, lifts for alluvium fill material with four· passes of a large v1bra:-ory roller will provioe the required compaction. (xi) Rip-Rap Material The rip-rap material (slope protection) will be obtained from tne oversize material from the various borrow areas~ Quarry A and an,y other rock excavations. The rip-rap material will be placed an the upstream slopes and in c~t'ta1n areas on the dovmstream slopes of the dam exposed to wave action. The gradat·ion of the rip-rap matefial, Figure 12~11, is based on a 6-foot vJave heigr.t us'!ng a r;omorgraph, Figure 5-6, from Eivllll0-2- 2300. The maximum size of rip-rap material will be 24 inches .. It is assumed that 36-inch lifts for the rip-rap zone with four passes of a large vibratory roller will f.:!Ovidt! the required compact1tm~ (f) Stability Analysis (i) Methodology Static and dynamic stabi 1 it_y· analyses have been performed to cnnfirm the stabi1 ity of the upstream 1d downstrearn slopes of the Watana dam. Ttie analy5HS indicates stable slopes under a11 conditions fo'." a 2.4 horizontal to 1.{; vertical upstream slope and a 2.0 horizontal to 1,0 vertical down~t""eam slope. The static analyses have been done using the STr.UL computer p~~O!}r'am developed to handle general slope st~bil h.y prob1ems by adaptation of the ~lodified Bishop method and a finite element program fJr sta- tic ana1ys·is of earth and rwckfill dams (FEADAf'l) to determir1e the initial stress•"s in the dam during rormal operating conditions. The r~esults and c~nclusions from both the static and dynamic analyses ar~ given in Appendix D. 12-28 I I I I I I I ,I I II I I I I I I I I I I I I I I I •• I I I I I I I 1: ,I I The dynamic analyst:; have been done using the QUAD 4 finite element program which incorporates strain dependent shear modulus and damp- ing parameters. The design earthquake for the dynamic analyses was developed for"' a Benioff zone event. An asses$f!ient of the static and· seismic response of the l,;latana dam for the static and postulated seismic loading involves the follow-- irg: -Finite Element Model The finite element model consists of 20 lt1yers of elements \'lith approximately 550 nodes and 520 elements. uif·Ferent soil para- meters as described in the following sections have been chosen for the core, transition material, and the shell material. The tran- sition material will comprie the fine and coarst:: ·filte_r zones. -Static Analysis The slope stabi-1 ity and analyses were done using the STABL com- puter program for the general solution of slope stability problems by a two-dimensional limiting equilibr1um method. The calculation of the factory of safety against instability of a slope is per- fo.rmed by an adaptation of the f'lodified oishop method of slices v1hich allows the analysis of trial failure surfaces other than those of a circular slope. Soil properties used in the analysis are given in Table 12.6. The following conditions \>Jere analyzed~ Condition ~ Construction Normal Operating Rapid Or awdovm Normal Operating with Maximum Pool Required f•1i nimum Factor of Safety 1.3 1.5 1.0 1.3 Calculated 'factor rof Safety U/S Slope DIS Slope 2.2 -2.2 . 2.0 1.8 -2.0 2.0 2.1 1 7 --1,7 1.. 7 The calculated factors of safety as shown in the above table indi- cate no general slope stability problems under static loadi11g. Further analysis, using the finite element program for static analyses of earth and rockfill dams (FEADA~I), determined the ini- tial stresses in the dam during normal operating conditions. The program calculates the stresses, strains, and displacements in the dam simulatin~ the actual sequence of construction operations. Approp~iate nonlinear and stress-dependent, stress-strain proper- ties for the soils were taken from information compiled in Table 5 in Duncan et al (1980). Table 12.6 presents the values used in the-an a lysis. T\vo an a 1 yses were performed to show the effects of relatively soft vs stiff core materiaL 12-29 ... Oyna!llic Ana1ys is The dynamic analysis was ctone using the QUAO 4 computer ptogr<!m. The initial values of shear modulus .and damping ra:,io to b~ used in the analyses were derived from typical values available in tianerjee et a1 (1979} and are as follows: ZGI\IE Core t-1ater i a 1 -Soft -Stiff Transition Naterial Shell filater i a 1 90 120 150 180 Uampi ng Shear Type Curve sand sand sand sand The· design earthquake time hi story wa·s deve 1 oped by Wood 1t/at~d­ Clyde Consultants and is shown ·in Figure 12.18. The significant featur-es are as fo 11 ows :. -Magnitude 8.5 Richter; -Location 40 kilometers below site (Benioff Zone); -Maximum acceleration of 0.55g; -Uuration of strong motion -45 sec; and -Significant number of cycles-25. The preliminary dynamic analysis had peak output values occur- ring about 24 seconds into the earthquake acce1erat ion time his- tory. Based on these results, the three iterations for the pro- posed dynamic analysis were performed using the following sec- tions of the e~rthquake time history: Iteration 1: Iter·ation 2: Iteration 3: From 10 to 30 seconds From 10 to 30 seconds From 10 to 70 seconds Conclusions: (Infor-mation to follow on completion of computer anaysis in mid-February 1982). (g) Instrumentation Instrumentation vlill be installed within all parts of the dam to provide monitoring during construction as 11ell as during operations. Instruments for measuring interval ver~ical and horizontal displacements, stresses and strains, and total of fluid pressures) ~s well as surface monuments and markers \vill be installed. The quantity and location will be decided during final design~ Typica.l instrumentation is as fo1lo~t/s: 12-30 I I u I I I I I I I ·I I I I I I I I I I I I I I I I -1 I -I I. I I I I I I I -Piezometers Piezometers are used to measure static pressure of fluid in the pore spaces of soi 1 rockfi 11 and in the rock found at 1 on. -Internal Vertical Movement Devices Cross-arm settlement devices as developed by the USBR • . VariO!JS versions of the taut-\ii re devices have been developed to measure internal settlement . . Hydraulic-settlement devices of various kinds. -Internal Horizontal Movement Devices . Taut-wire arrangements. . Gross-arm devices . . Inclinometers . . Strain meters. Othe~ Measuri~g Devices . Stress meters. . Surface monuments and alignment markers . . Seismographic records and seismoscopes . . Flow meters to record discharge from drainage and pressure relief system. 12.7 -Relict Channel TY·eatment (a) Site Conditions Ear 1 i er stud-ies i dent i fi ed a buried ch anne 1 r·unn 1 ng from the Su si tna ~1 ver gorge immediately upstream from the proposed damsite to Tsusena Creek~ a distance of about 1.5 miles. Boring by the Corps of Engineers penetrated 454 feet of glacial deposits overlying bedrock ltlhich v1as encountered at elevation 1,775 fe-;t, while the surface elevat1on of the lov1est saddle is approximately 2,005 feet. Additional investigations during the curre~t study further de 1 i neated the channe 1 and fu 11 detai 1 s are given 1 n the Task 5 Geotechnical Report. The channel represents a potential source of leak- age from the Watana reservoir. Along the buried channel thalweg, the high- est bedrock surface is some 450 feet below reservoir level, while along the shortest leakage path between the reservoir and Tsusena Creek the highest rock surface is some 250 feet below reservoir level. The maximum hydraulic gradient along the bured channel from the edge of pool to Tsusena Creek, is approximately 9 percent, while between existing riverbed levels it de- creases to about 6 percent. There are surface lakes within the channel area and while some drill holes encountered artesian water, others pene- trated highly permeable zones resulting in complete loss of drilling fluid. Zones of permafrost have also been identified throughout the channel area. 12-31 (b) Although the glacial history of the area is not clearly understood, a se- quence of events has been postulated in the Task 5 Geotethnical Report, based primarily on the investigation of the Borrow Area D adjacent to the buried channel. The generalized sui"ficial stratigraphic column is given in Table, 12.5 and the gradation of the soils in the various units are given 1n Figures 12~55 through 12.18. Of particular relevance to the buried channel problems are the alluvium at the base of the channel, encountet~ed in one deep borehole beb·:een 292 feet and bedrock at 454 feet below ground, and the unconsolidated outwash, allu- vial and fluvial deposits. The deep alluvium offer·s a potential leakage path, its high permeability being indicated by loss of drilling fluid, vlhile the unconsolidated, primatily sandy deposits may be subject to lique- faction following saturation. Potential Problems The major potential problems associated \'lith the buried channel are leak- age, both surface and subsurface flows; piping at downstream outlets to Tsusena Creek; the impact o: permafrost and the long~term effects as heat fr-om the reservoir thaws the ground thl~ough the channel area; and i nstab: 1- ity of soil slopes on saturation, thawing, or seismic loading leading to a breach of the rim of the reservoir. (i) Surface Flows During the study of alternative layo'Jts for Vfatana) the maximum op- erating reservoir level was higher than the cr1tical ground e1eva- t1on of 2005 in the bm~ied channel :J.reci.. Th~~e layouts, therefore, incorporated a saddle dam about 40 feet high and 2!1500 feet long across the critical section of the channel. TI1e foundation condi- tions for such a s add 1 e dam are not we 11 defi n-::d at this time hut because of the variable nature of the glacial deposits, the effects of permafrost and potential for liquifaction vlithin the foundation were addressed. It \'las concluded, however, that in any event there. was a strong possibility that settlement of such a dam could n.ot be adequately controlled and there vmuld be a real risk of transvers€ cracking occurring through the dam. With the reservoir level-above ground sJrface, any such cracking could lead to surface flows and - subsequent channeling through. the unconsolidated deposits. {ii) Subsurface leakage No field permeability tests have been conducted, but it is antici~ pated that the total subsurface leaking wi 11 be relatively small and economically insignificant. For example, if the average permeabil- ity of all material in the channel were lo-2 em/sec, the total 1 eak age flow would be' 1 ess than 100 cubic feet per second. By in- spection of the grading curves, the actual permeability is certainly less than lo-2 em/sec, except possibly in the channel bed allu- vium, and a more realistic leakage flow would be about 10 cubic feet 12-32 I- I I I I I ·'I 'I I I I I I I I I 1·-, . . . . .. I -· I I I I I I I I I I I I I ,. •• I~ I '. . per second. The capital value of this leakage is about $4 million. However, any leakage may be concentrated in the discharge zone in Tsusena Creek, and there is potential for· piping which could lead to large-scale.erosion cutting back to the high ground forming the rim of the reservoir. ( i _i i) Permafrost Thawing of permafrost wi 11 result in higher seepage rates and pos- sibly settlement of the surface as excess water drains from the thawed soils. (iv) Liquefaction Filling the reservoir will lead to the saturation of some of the glacial deposits within the buried channel area, including the upper slopes of the Susitna River valley., and produce the potential for liquefaction of these deposits under seismic loading. Under extreme circumstances, 1 iquefact ion could lead to mass movements of soils into the reserve ir and breach of the reservoir rim in the area of the freeboard dike. For this situation to occur, it would require a large, continuous deposit of loose, saturated, granular material with sufficient ground surface slope so that the soil above the liquefied zone wcula move under its own height. Although such a scenario is consid~red most unlikely, the investigations to date are not sufficiently de- tailed to preclude the possibility. In view of the potentially catastrophic failure that would result from a breach of the r~ser­ voir rim, further investigations must be carried out prior to con- struction to confirm the stratigraphy and provide adequate data to properly assess the need for and design of remedial treatment. : .(c) Remedial Measures Since the stability of the section of the buried chdrmel forming tne rim of the \4atana reservoir is essential for the feasibility of the Watana ~evel­ opment as outlined in this report, practical solutions to all possib1e scenarios, including extreme combinations of the problems outlined above, must be identified. (i) Surface Flows To eliminate the potential problems associatea 'tJith settlement and breach of a saddle dam allowing surface flows through the buried channel area, the maximum operating level of the reservoir has been lmvered to 2,185 feet leaving a width of at least 1,500 feet of 11 dry11 ground at the saddlE: above this elevation1 A freeboard dike ~ith a crest elevation of 2010 is required to provide protection against extreme reservoir levels under probable maximum flood condi- tions. The shor-test distance between the toe of the dike and the edge of the 2185-elevation reservoir pool is at least 450 feet, ana 12-33 I and under Pt•JF flood, the \'later 1 eve 1 wi 11 just reach the toe of the dike. I (ii) Subsurface Flows Progressive piping and erosion in the area of discharge into tne Tsusena Creek \vi 11 be centro 11 ed by the placement of properly graded granular materials ~TTke ·a f'i Yter blanket over the zones of emer- gence. Fie1d investigation will be car-ried out to define critical areas, and on 1y such areas wi 11 be treated. Gont inuous mon i tor·ing of the out 1 et a) .. ea will be necessary 1 s i11ce it may take many years for equilibrium vlith respect to permafrost to become estab1 ish~d in the buried channel area~ If the permeability of the base alluvium is found to be excessive~ grouting of the upstream inlet zone could be carried out to reduca the total leakage. (iii} Permafrost Thawing of permafrost wi 11 occur that may have an impact on subsur- face flows and ground settlement. No specific remedial work is necessary; but flows, ground \'later elevation, and groJnd surface elevation in the buried channel area must be monitored and· any nec- essary maintenance work carried out to maintain freebcard an<1 con- trol seepage discharge. (iv) Liquefaction To guarantee the integrity of the reser·voir rim through the channel area requires that either: There is no potential for a liquefaction slide into the l"'eser·voir \vhich cou1d cut back and br·each the rim, or If there is such potential, there is a sufficient volume of stable material at the critical section that even if the upstrearn ::i:ateri- als were to slide into the reservoir, the failure zone couln not cut back to the reservoir rim. Any remedial treatment required will depend on the location .and e>:.- tent of critical zones and cou1d range from stabilization by compac- tion (vibroflotation) or grouting techniques, either cement ~·~ chem- ical grouting, or in the limit, removal of materia14 The stratigraphic column indicates that the two lower till deposits I and K have been overconsolidated by glaciation, and it is unlikely that these deposits could 1 iquefy under any circumstances~ The overlying Unit H is a medium fine sand ',-~fit~ silt and is probanly the most susceptible to liquefaction of all the materials sampled .. This unit has been identified up to 40 feet thick in places, with the top of the layer ~stirnated to be about 100 feet below ground sur--face at the deepest point, as shovm on Plate 6.34 of the Task 5 Report. All 12-34 '1·.··.' : ll .. I' I I I I I I I· I I I I 'I I I ·I d, I I I I I I I I I I, I I 'I I I ·.,I ,I ·•.··•·· ·' . ' . ·~"-·-.·· materials above this unit, are normally consolidated water/1ain till, nutwash, alluvium, and fluvial deposits which could include zones of critical materia 1-s. . . There are insufficient data available to identify the fu11 extent of such critical materials; henc;e,_.Jt __ _i§ -~ot possible to precisely de- fine the remedial work necessary-·at-lTfis time. Available a1tel'"f1a.-. tive methods include: -DensificQ.t ion Layers within about 100 feet of the surface could be compacted by vibrof1otation techrdques to eliminate the risk of 1iquefacti(ln . · and prov id€ a stable zone. Stab i 1 i z at i on , Critical layers at any depth could be grouted, either with .c.e·:nent for fine gravels and coarse sands m" by chemical grouting ftt.r fine sands and s i 1 ts. -Removal This could range from the replacement of critical material n~~ar the valley slopes with high-quality~ pro-cessed material, whi.z:h would stabilize th~ toe of a potential slide and so prevent the -initiation of failure that might otherwise cut back and caus~ major failures, to the excavation blending and replacement of large volumes of material to provide a stable zoneA The ultimate tr-eatment \'lill be based on an engineering and cost analysis study of the appropriate alternatives during the design phase of the project when the site conditions have been more closely defined. However, to confirm the technical and over.p.ll finaoc~al feasibility of tr~e project at this time, it is necessary to C.;,lnsider a solution to tht: worst conditions deemed possible. On tht~ basis of avail able data, such conditions are: ... That the a 11 uv ium Unit H enc0untered between elevations 2lu:JJ and 2140 in drill hole DR22 is a homogeneous loose, silty~ fine: sand~ That it is of 1 arge areal extent and cant iouous frorn beyond the '· saddle out to the Susitna valley slopes .. -That is of such thickness that a fa i 1 ure plane could be contained: fully vdthin its boundaries,. With these conditions, liquefaction of the unit under seismic load- ing following saturation fr0J11 the reservoir could result in the overlyi.ng materi.aJ sliding on the 1 iquefied zone into the res~l~voil~. 12-3.5 ' . -. _, --j_""-~ • ..:, •• _ ... : •.• :,:..,_ , -~.t "-··-·~~:~'~,...a,t,1'~;,_u . ..:o.·.~~ ~~-·c··-.{'~ ..... ~_:.~:~.~..,, '""~~ _· ... ~ '. (d) Catastrophic. failure would d~velop i..f the back scarp of the failure ;. surface through the overlylng materia1s broke ground surface on the downstream side of the saddle belovJ reservoir water leve1.~ The most positive solution·to such a situation wou1d be the replace-· ment of the critical zone with material that wou1d not 1iquefy •. This would involve~ in effect; .the rearrangement of the in-place ,: .. materials to create an underground dam sect ion founded on the den~e.. · till layer beneath the critical alluvium. Such an operation would invo 1 ve the excavation of a trench up to 135 feet deep with a sur- face width up to l::OOQ.feet. Selected materials would be compacted to form a central zone, with 2 horizontal to 1 vertical slopes. Surplus and unsuitable materials would be placed on both sides of this central 11 dam 11 to complete backfilling to.ground surface. The central zone would be desioned to remain stable in the event that all material upstream did s1 ide into the reservoir~. Preliminary estimates indicate that such a structure would need to be 5 ~OOO feet long, with a total cut volume of about 13 million· cubic yards, of rlhich 4-1/2 mill ion cubic yards could be used in the compacted center zone~ The cost of such \vork is estimated to be about SlUO million. The need of such e~penditure is considered to be most likely and is deemed to be covered by the.overa11 project contin- gency sum~ Further Investigations 0 Additional site investigations are necessary in the relict channel area to more closely define the follov.;ing; Confirm and/or refine the stratigraphy throughout the area. -Thickness, extent, density, continuity, and permeaoility of the alluvium identified in DR22 immediate1y above bedrock. The .investigations shou1d include pumping tests and dye injection tests to check the continuity of this unit along the buried channel, since this is deemed to have the ·greates~ potential for leakage. · · -Density of the lov1est till layers I and J which have been subjected to overconsolidation by glaciation to confirm that theyv;ould not liquefy under earthquake loading. -Density, gradation, extent, and continuity of the sandy silt alluvium, Unit H. -Extent of any other units which may be subject to -1 iquefaction. -Conditions in the outlet area of the relict channel into Tsusena Creek. -Ground \l!ater regime throughout the channel at'ea v1ith part icu1 ar refer~nce t~ the source of artesian or confined aquifers and the drainage outlets · from such aquifers. 12-36 ·1.··.,.-·.1 . 1 . .1 I >~ .•• ;:'~~ . . I I I \. I I I :1 I I ~I f J I l ( •.. -.. · I .· I ::'1 'I I t' I I :I I I' '.•1·. . , 10 I ·I ,I I I I I (e) Construction Methods For the purpos~ of thts feasibility study, it has been assumed that treat~ ment of the relis_t channel.· will involve ti;e following: ... Construction :o( the freeboard'tfike at the c:.~est ·of the saddle involvit1g stripping of orqanic material and construction of the dike with impervi- ous c(~e and gravel or rock-fill shells; and -Stripping, grading, and placement of a filter blanket over the outlet area of the channe1 in T.susena Cret;k, This b1<H1ket is estimated ta be 7 feet thick and cover an area of 460 acres •. Allowance for the cost of these items is included in the estimates .. How ... ever, it is considered likely that the actual area of downstream blanket required wilT be less than allowed, but that other~emedial work in the sadd1e area vdll probably be necessary~ · The allo~t/ance for the downstream blanket is therefore considered a ~=ason­ able estimate to cover all work required in the channel area excl udi;ng the extreme situation involving major excavation as discussed above. 1,.8-Primary Untlet Facil·ities The primary function of the outlet facilities will provide capability to ~is­ cil argE floods with_ recurrence frequencies of up to 1:50 years after they h.ave ~ ·been routed through the ~~atana reser-voir. Downstream erosion will be mi~ima1 and the dissolved nitrogen content in the discharges will be restricted as much as possible to avoid harmful effects on the downstr-eam fish population .. J. sec- ondary function of the facilities will be to enable the rapid drawdown cf the reservoir by up to 150 feet during an extreme emergency situation . .The structures will be located on the right abutment, as shown on Plate 4~, and will consist of an i~take structure~ pressure tunnel, and an energy dissi9~­ tion and control structure housing six fixed-cone valves v1hich \"fill dischrarrge into the river 150 feet below. {a) Approach Channel and Intake .. The approach channel to the outlet facilities wi1·1 be shared with the power intake, The channe 1 will be 400 feet wide and excavated to a depth r;;.f approx~nately 150 feet in the bedrock with an invert elevatio~ of 2U~7. The intake structure will be founded deep in the rock at the end of the than- nel. The single inlet passage will have an invert elevation of 2003 .. It will be divided upstream by. a central concrete pier which wi 11 suppct·t ~teel trashr•:~.cks located on the face of the structur~, spanning the ~pcn­ ings to the water passage. The racks \vi11 screen out suomerged debris which could damage the downstream valves. Th"e racks will remain in place with no provision for their withdrawal,. which would be e!(per'sive and is considered unnecessary because of the infrequent use of the outlet. If 12-37 /) (b) (c) there.ever should be .a blockage) this 'tlOU1d have to be cleared by divers. Dovmstream of the racks, located between the pier and ·eacn of the side-· · walls~ 'VIill be two fixed wheel gates operated by a m~chanica1 hoist mounted (at the surface) above the deck of the structure. r~e purpose of the fixed \·Lheel gates will not be to control flows thrcugh the out'iet, but to close . off the downstream tunnel to allow devfater·ing for maintenance of the tunnel or gates within the discharge structu1"a~ Stoplog gate:s wjll ba pr·ovided just upstream of the gates to allow insertion of stoplogs for dewate~ .. ing of the structure and access to the gate guides for maintenancB. Intake Gates and Trashracks The gates will be of the fixed wheel vertical 1 ift type with downstream skinplate and seals. The norninal gate size \~ill be 16 feet wide b_y· 32 feet high4 Each gate will be operated by a single drum wire rope hoist mounted ·;n an enclosed tower structure at the top of the intake. The height of the to,tJer structure \·:li 11 permit raising the gates clear of the intake concrete ·for inspection and maintenance. The gates wi11 be capable of being lowered either from a remote control room or locally from the hoist area. \:late: raising will be from the hnist area only. The trashracks will have a bar·spacing of about 7 inches, and tttill oe de- signed for a maximum differential head of about40 feet. The maximurn net velocity through the racks will be about 6, ft/s. Provision \vi11 be made for monitoring the head loss across the trashriicks. Shaft c.:ld Tunue 1 Discharges ~vill be conveyed from the upstream gate structure by a contrete- 1 ined shaft and tunnel terminating in a steel 1 iner find manifold. The man-- ifold wil·l branch into six 78-inch diameter steel-lined tunne1s \<Jhicil \'!ill run through the main spillway flip bucket structure to the va:ves mo~nted on the dovmstream face. The water passages v.;ill be 28 feet in diameter up to the steel manif:gld. The upstream concrete-1 ined port ion wi 11 run a short distance hori Z)ntally from the back. of the intake structure befor·e dipping at an angle of s:/' to a 1ovJer level tunnel of similar cr--oss section. This angle of 55° is con- sidered the fl~test slope at which the tunnel can be "self-muckingn during construct ion and is cost effective in the r~esu1tant ratio of shaft to tun- nel length. The lower tunnel will run at a gradien~ of l~lOtoth.e point where the overly1ng rock is insufficient to v.fithstand the large hydrostatic pressure which vlill occur within the tunnel. Downstream of this point the pressure will be transferred throughout the mass concrete and resisted jointly by the surrounding rock and the stee1 liner. The steel liner wi11 be 26 feet in d i arneter and surrounded by mass concrete filling the space between the 1 iner and the surrounding rock, The area of the outs ide face of the 1 i ner and the <'Oncrete \-Ji 1l be grouted to fin all voids and ~"educe non-uniform external ground water pressure build up from the ground water. 12-38 •:'.·.·: . ' I . ._ I I -. -:-~ I. I I I I , •. I •• I ·~ •• I I I I I ; •• ,, I I I I I I I I I I I ,I I I I ., I Upstream from the discharge structure the 1iner will terminate tn a. steel manifold vtith six parallel 8 foot diameter steel-lined branches. These will continue through the back face of the discharge structure. and termi- nate in discharge valves mounted at the downstream end of the structure. (d) Discharge Structure The concrete discharge structure is shown on Plate 52. It will form the flip bucket for the main spi 11\-vay and wi 11 house the fixed cone valves and individual upstream ring follov1er guard gates. It wlll be founded on sound rock high above the river The valves will be set with a centerline eleva- tjon of 13560 feet and will discharge. into the river approximately 85 feet below. Openings for the valv~s will be formed in the concrete and the valves will be recessed within these openings sufficiently to allow enclos- ure for ease of maintenance and heating of the moveable valve sleeves. An access gallery upstream from the valves wi 11 r·un the length of the dis ... charge structure, traversing the steel liners upstream of the valves~ and wi 11 terminate in the access tunne 1 and access road on e1 ther side of the structure. Housing ring follovJer guard gat-es will be located upstream from the gallel"'Y gate chambers. These gates wi 11 operate on the steel 1iner·s and wi 11 serve to i so 1 ate the discharge valves. A common monorai 1 hoist wi 11 be located above each valve and gate assembly to provide for their removal and transportation to the access gallery .. From the access gallery they can be maneuvered on a low trolley to the service area adjacent tc the end of the gallery. (e)-Fixed Cone Discharge Valves Eight 78-inch diameter fixed cone discharge valves will be installed :at the downstream end of the outlet ,nanifold, generally as shown on Plate 52 .. The valves were selected to be within current experience, considering the valve s 1 ze and operating head (see Figure 12 .20) ~ The fixed cone valves ar~ si m- i lar to Hm•1ell Bunger valves except that the cone support valves ext~nd further upstream and are more streamlined. The va 1 ves have a slightly higher discharge coefficient than Howell Bunger valves and are less prone to vibration. · Electric jacket heaters will be installed around the cyllndrical sleeves -of the valves which extend outside of the valve room, and since the valves will be located within a heated valve room, they will be capable of year- round operation. Normally, when the valves are closed, the upstream ring· follower gate vdll also be closed, so that freezing of leakage at the valve seat should not be a serious concern. The va 1 ves will be operated either by two hydraulic cy11 nder operators or by a sere\~ stem hoist. For preliminary design purposes, hydraulic opera- tors have been assumed. The valves may be operated either locally or re- motely. At the time of valve design 3 careful consideration must be given to prevent vibration. Considerable research will be carried out concerning experience 12-39 r ~. ~ and design of existing installattons, and model tests will be necessat'Y to help ensure satisfactory VrilVe operatton. Adequate design of the fix~d vanes wi11 be of prl,me importance. In si.z1ng the valves for the· pre1 imtnary design, it has been assumed that ·the valve gate_opening will be restrh:ted to 80 percent full stroke because of v1 brat ion conslderati mrs. Careful considerat1on must also be given to machining and surface finish of the valves, in order to prevent ~avitation and erosion of the valve seals resulting from th.e extremely high water velocities. · (f) ·{Ring Fo 1lovJer Gates A ring follower gate will be installed upstream of each valve a~d will be used: -To permit inspection and maintenance of the fixed cone valves; To re 1 i eve the hydrostatic pressu·re from the valves ~tJhen they are in the closed position; and~ , ~ -To c 1 ose against flowing water in the event of ma 1 funct 1 on or fa 11tn'e of the yalves. The ring follower gates will have a nominal diameter of 90 inches and \~i1l .. be of welded or cast steel construction. The gates will be designed to withstand a total static head of about 630 feet. Existing large diameter high head ring follower gdtes are summarized in Table 12.:6. ~The ring follower gates wi 11 be designed to be lowered under flowing Nater conditions with maximum head, but they will on1y be raised under balanced head conditions. Valved bypass piping will be used to equalize the pres" sure on both sides o.f the ring follower gate before raising. The gates will be operated by hydr·aulic cylinders with a nom·inal operating presstrre of 2000 psi . Either loca 1 or remote operation of the ring fo 11 ow2r gates will be possible. A grease system will be installed in each gate for injection of grease be· tw€en the gate 1 eat .and the gate. body seats to reduce fr1 ct 1 on a 1 forces when the gotes are operated. (g) 01 sc;h arg e P..rea Immed1 at ely downstream of the discharge struc;ture, the rock wi 11 be cut at a slope of 2H:3V to a lower elevation of 1,510 feet. This face will be heavily reinforced by rock bolts and protected by a concrete slab anchored to the face. The lower level will consist of unlined rock extending to the river. Much of the lower trajectory of the discharge jets \·li 11 impinge on this area in the form of a very heavy spray, while much of the upper part of the jets will carry as far as the river. Because of the high degree of dispersion of the discharges and the infrequency of operation of the valves, it 1s not anticipated that erosion will be a problem requiring other than occasional minor maintenance. 12-40 I I I I I I I I I I I .-. . \ I I I I I ·I I,. It [I I \'• J( ~~.{···!· -.· ~ ---\; I I I I I I I I I I I I I l I 12.9 --Main-Sei11wax {a) General The main spillway 'vill provide discharge capabi1 ity far flows exceeding the capacity of the out.1et facilities, passing the excess of floods with a corresponding return :period of less than 1~50 years. The comtdned total capacity -Of the main spilh1ay anJ outlet facilities ,,;111 be sufficient to pass routed floods with recurrence frequencies of up to 1 in 10,000 years. The spillway, shown on Plate 49, is located on the right abutment and con- sists of an approach channel, a gated agee control structure, a concrete~ lined chute, and a flip bucket. (b) Approach Channel and Control Structure (c) The approach channel is excavated to a depth of ctoproximately 100 feet ·into rock. It is adjacent to the p0\1er facilities approach channel, and in order to minimize its length, it partly offtakes from this channel, inter-- secting it at elevation 2125 feet, The control structure is a massive concrete structure set at the end of the approach ch anne 1 . Flows are contro 11 ed by three 42 feet· high by 36 feet wide vertical lift gates. As shown on Plate 50, each gate is contai:n~d ~tlithin a separate unjt consisting of an ogee overflow weir, piers~ a~d an integral roadway deck~ The units are of amonol ithic box type structul'"e with split pier construction. The box configuration will give rigidity during seismic shaking, and the spl1t piers will allow for. some rala~ive motion and stress relief during earthquake motion in order to minimi;;:B the possibility of the gates jamming in the closed position. Model tests will be 1ecessary during the f,inal design stage to determine final geometry and dimensions such as pier noses, crest shape, and p3er length. The structure wi 11· be located adjacent to the right dam abutment in ~ ine w.ith the dam crest. The main access route from the Denali Highway 'n'i1l pass across the spillway deck and along the crest. The ma.in dam grout curtain and drainage system \'/ill pass beneath the struc- ture~ Access to the grouting tunnels will be via a shaft within the struc- ture and a g a 11 er y running through the ogee \\'e i r. Spillway Gates and ·stoploas ...:.:...;_ The three spillway gates will be of the fixed wheel vertical 1 ift type op- erated by double drum wire rope hoists located in an enclosed bridge struc- ture.. The gate size has been se 1 ected as 36 feet wide by 40 feet high, \n· eluding a 3 foot freeboard allowance above maximum normal reservoir level~ The gates will have upstream skinplates and seals ~vi11 be totally encJosed to permit gate heating in the event that winter operation i.s necessat-y~ Provision \vi11 also be w;.Je for heating the gate guides~ 12-41 ,_ ,_. (d) (e) The height of the to\'ler and br·idge structure will be selected to ·permit raisi'ng of the gates above the top of the s.pillwny pier fm~ gate. inspection .and maintenance. An emergency ga?o 1 i ne engine 'lf.Ti '11 be provided to enab 1 e the gates to be r·aised in the ~vent of loss· of power-to the spi1hJay gate hoist motors. A set of stoplog guid_e;:; will be installed upstream of each of the three spillway gates. One set. of stoplogs wi11 be provided to permit raising the gates with maximum reservoir levels vlithout discharging water over the spi 11way. The stoplogs \'iil1 have dov-mstream skinpl ate and seals and will be sr'f'anged in sections suitable for handling by a mobi1e crane with a follower. Spillway Chute The control structure will discharge dovm an inclined chute that tapt:!rs slightly to a width of 80 feet, which is maintained over the remainder of its length .. The chute section will be rectangular in cross section~ ex~avated in rock.~ and lined with concrete which will be .anchored to the rock. An under- drainage system will be constructed beneath the slabs to relieve uplift pressures and will be formed by drilling holes from a centrally located gallery, in the rock, to intersect box drains located oeneath the chute floor slab. Provision will be made at two locations dovm the chute to aerate flmvs and prevent cavitation of the concrete floor. Aeration vti1l be attained by means cf anincline.d step into the flow. Air wi11 be drawn of from a transverse 1o\ver gallery via ducts \'1hich exit at the downstream vert ica1 face of the step .. Convergence of the chute r1alls at the upper end of the spillway will be gradual to minimize any shock ·aave.deve1opme!lt. Nadel tests will determine the maximum allowab1e convergence _of the 1~a11 s to assure both economy atld sat i.sfactor·y hydrau1 ic performance~ ivloue1 tests will also determine aeration requirements (number and size of aeration ducts) a~d ~he shape of the chute floor. Flip Bucket ·, The function of the flip bucket will be to direct spillway flows clear of the spillway and \''e1l downstream into the river belov1. The jet issuing from the bucket will be partly dispersed during its passage through the air with a corresponding loss of energy. The remainder of the energy \v111 be dissipated on impact with the plunge pool~ The mass concrete block, as described in Section 12.8, will house the gates and discharge valves for the out1et facilities, and a1so will form the f1 ip bucket for the main spillway .. The elevation of the structure will be con- trolled by the elevation and head limitations of the dischat·ge valves. Final geometry of the bucket will be determined by model studies, as well as dynamic pressures on the floor and v~all s of the structure. Although the structure shown on Plate 52 shows a simple) cylindrical type of bucket, it 0 12-42 I" I 'I I I I I I I I I I I I I I I ·I I I I I I I I I I I I -·, I I I I II I I I ;..-' ~' 'f; J it is foreseen that; a more effective, dispersive type bucket ~tlill be deve1- oped.during model tests. _Such. a bucket was Clev-e1or:~:d for_ the Portage f'~ountain Project in or it ish Columbia, 'tlhen.:· flows u-p to cfs under a .str1tic head of feet af'e ct'~scharged at a 90~ angle into the river below., In .order to prevent ero- s ii:.tn of the oppos 1te river bank, the jet is spread by means of a disk-· shaped bucket so that the area of impingement in the river is spread in a narrow 1ine parallel to the bank and normal to the chute centerl ire. 12.10 -Emergency Spill way The emergency spillway wi11 be located on the right side of the river beyond tr1~ main spillway and the pm'ler intake structure (see Platg 53}. The spilh-1ay \·d11 consist of a long straight chute c·ut in the rock and leading in the direction of Tsusena Creek. An erodible fuse plug, consisting of irnper·tious and fintl gravel materials, wi11 be constructed at the upstream end; it will be designed to wash away \•then overtopped by the reservoir" thus releasing floods of up to 160,0UO · cfs in excess of the combined main spillway and outlet capacities and preventing overtopping of the main dam. - (a) fuse P1uq and t\pproach Channel The appr·oach channel to the fuse plug will be excavated in rock and wil1 have a width of 310 feet and invert elevation of 2170.; The main accas-s·-- road to the dam and power-house wi11 cross the channe1 by means of a b}~iage~ The fuse plug will c 1 ose the approach channel) and \'li 11 have a maximum height of 31.5 feet with a crest elevation of 2201.5 feet. The plug \xi11 have an impervious core up to l.O feet \vide, steeply inclined in an upstream direction" with fine filter zones upstream and downstream. It \vi1l be sup- ported on a downstream erodib1e shell of crushed stone or gravel up to 1,5 inches thick~ The crest of the plug will be 10 feet JtJide and ~vill be tt"a- versed by a 1.5 foot deep pilot channel. The zoning of the plug will te similar to that designed for the New Melones Lake project in California~ For this project a half scale version of the plug 'tlas constructed and tested to destruction. The.p1ug failed completely in 1essthan an h.:our and the water level rose no more than 6 inches above the invert of the p11ot channel~ The principle of the plug is based on erosion progressing r-apidly dmvnward and 1 aterally from the pi1 ot channel as socn as'' it is overtopred. The channel sect 1on at the fuse plug is cons 1dered as a broad crested \•;eir with a coefficient of dischar'ges of 2.7. A gated control structure \'ias considered as an a1tetnative to the fuse plug, but this would give higher construction and maintenance costs and 't~ou1d not provide the automatic breakirl;g and discharge of the plug. (b) Discharae Channel The rock chanoel dm·mstr--earn of the fuse plug will narrow to 200 feet and continue in straight line over a distance of 5,000 feet at gradients of 1.5 percent and 5 percent in· the direction of Tsusena Creek. The channe1 !.~ill 12-43 ' .· --~·. ·~ ,. ': •• "~' ,, •• -J,• -,' '.: • --·,_ 7 ,:<>' ''-:::~, ·,:__. "\ :..·, d isc.harge in a buried Valley 110tl the ,pownstream Side of the main re1 ict channel from \'lhich flmvs will r·un down into the 'creek. It is estimated that flows down the channel wi 11 cent inue. for a period of days unaer probable m~imum flood conditiohs" Some erosion in the channel wfll be 'acceptab1_e,fbut the integrity of the main dam will be maintained. The reservoir-will be dra\··m down to Elevation 2170) and reconstruction of the fuse plug wi1.1 be required prior to refilling of the reservoir. 12~11 -Intake (a) Genera 1 ,I) The intake structure at Watana will be required to fulfill the following functions: To pt~ovide independent power fl0\•1 to each of six Francis turbines, up to a maximum flow of 3,800 cfs per unit, for any reservoir level from EL 2000 (maximum extreme flood level J to the minimum dr·awdown 1eve1 {Selec- tion of drawdown level is reviewed below); ... To provide an upstream control gate on each penstock to perrnit de'natet'ing of the pen&tock and turbine water passages for routine inspection and maintenance; and To contro 1 the temperature of water d i sch ar·ged ftom the reserve ir tt•d thin acceptable 1 imits to mitigate the environmental impacts of the Susitna development on dmvnstrearn fisheries and vegetation. (b)· Environmental Constraints The seasonal operation of the Watana reservoir will have two major tapacts on downstream flows: -In summer, the temperature of dm•mstream releases wi 11 be cooler than the normal river regime; and In winter, the temperatures \-J11 1 be warmer than the normal r·iver r·e:gime. Cooler water in the summer months could have a significant impact on down- stream fisheries, particularly in July and August when sa1mon are mnvit'lg into the sloughs downstrea11 from Devil Canyon to spa't'm~ ~~armer water in winter will-affect the formation of ice, resulting in extensive opcln 1.111ter downstream from the._reservoirs. Temperature simulation using a Corps of Engineers Hydraulic Engineering Center (HEC) progra.'TI was used to model the dO!tmstream effects of reser·voir operation using a variety of differ·ent pov1er intake designs at watana ano Devil Canyon .. These studies have indicated that dovmstr·ea1l temper·atures wi 11 remain constant in winter at 39°-F reg~rd1ess of the type of po'.';er in-, take design used. Hm·1ever,. the do'lmstream river temperatures in the summer months can be significantly improv€d by power intake design at ',"/atana which 12-44 l ···::t l ~ ' .. I I I' I I I I ,~ I I I •• I I I I I ;I ~~· r::, e I I I I I I I I I •• I I 'I I I I I I (c) {d) months can be significantly improved by power intake design at Watana which· wou 1d ·permit water to be dr a\·tn ·off to the reservoir surface at an times . . The power intake design at Oevi1 Canyon ·will be 1ess·significant because ·the maximum drawdm·m is only 50 feet. The se 1 ected power intake design at Wat an-a \·li 11 perm1 t water to be dt"awn from the reservoir at four distinct levels thr·ough the antfcipa:ted range of dra\'Jdown to mitigate the environmenta1 impacts on do;.mstream river t~mpera­ tures {see Volume 2}. Details of the reservoir temperature modeling are presentad in Appendix 84. Orawdown The maximum drawdown .at the power intake contra 1 s the 1 ive storage of the reservoir (the volume available for seasonal and over year regulation of the r'iver flows). With no drawdown capability, trH~ dependable (firm} energy from Hatana would be controlled by the oependable flow of recor·d to the required reliability criterion; this has been defined as the lowest fl mv in the second driest year' of record, \vi th a recurrence of about. 1 in 70 yeat·s. As the drawdmvn is increased> the firm flow from the reservoir \'li 11 in- crease. The firm energy \vi11 also increase fOt"' dra~t!do~·m up to about 140 feet. If drav<~down is increased beyond 140 feet, hO':tever, the firm f]cw will increase but the firm ener~gy will start to reduce, since depend.able energy is governed by a combination of flow and head. Average energy avail able from the reservoir shows a very slight decrease \·lith increasing drawdown because of the imposed constraint in the computer simulation that the reservoir should be fu1T at the end of the 32 years of recorded flows. Hm~Jever, the usab ~e energy \'li 11 increase with dratid~wn be- cause the storage available l'lil1 permit excess energy in summer to be stored for use in the winter. Costs of the intake structure and approach channel will increase sig~ifi­ cant1y with the depth of-dra\~down, while the cost of penstocks wi11 te re- duced. A detailed study has shown that the extra benefit of increasad firm energy from ~~latana will be in excess of the required incremental cost of the intake, approach channel, and penstocks for any drav;~owo up to l..:i,J feet .. Therefore, the maximGm dravldo\·m selected fof' preliminary design was 140 feet. Desi~n The power intake wi 11 be a free-standing concrete structure located in rock excavation at the upstream end of the approach channel. Access to the · structure ~<~ill be the same as access to the intake for the outlet \IJQ~"'ks, since the two structures have a common approach channel. · In order to draw ~'/ate:r close to the reservoir surface over a ctrawdmvn range of 140 feet, four openings wj11 be provided in the upstrearn concrete wall 12-45 (e) i. I of the structure for each of the six independent power intakes.. The uppet· opening wi11 always be open~ but the lower three openings can be closed off by s1iding stee1 shutters operated in a ccommon guide. ~All openings wi-11 be protBcted by upstream trashracks~ ·A heated ice bulkhead will be provided . for winter operation .. An intake contra 1 gate will be pr·ovided on each penstock.' A s iog1e up..- stream bulkhead gate will be provided for· routine maintenance on the six intake cont~--ol gates~ I~ an emergency., stoplogs can be insta11ed in the upstY'eam wall of the power .intake for work on the trashracks or shuttet guides. The vJidtb of the intake wilLoe controlled by the minimum spacing of pen- stock tunne1 excavations, taken as 2.5 times the excavation diameter. The upper level of the concrete structure will be set at EL 2200, corres- ponding to the maximum anticipated flood level. The mi11fimum structure .. level will be governed by the physical dimension of the penstock tunnel ex- cavation and the vortex criterion for flo~l 1-nto the p-enstock from the mini- mum reservoir 1 eve 1 EL 2045. The found at ion of the str-ucture will be an out 150 feet below existing ground level and will be expected to be generally in sound rock. Mehanica1 equipment will be housed·in a steel-frame building on the upper 1 eve1 of the concrete structure~ The general arrangement of the pC\w~:r in- take is sho\vn on Plate --- Approach Channel The \'lid th of the approach ch anne 1 wi11 be governed by th-e combined \'tiuth of . the power intake and the intdke to the outlet fac·n ities, which in t~trn will be governed by the minimum spacing of the penstock and outlet turme1s, The overall width of the channel wi11 be about 350 feet. Excavated s1.:lpes in sound rock will be generally 1H:4V. The maximum flow in the intake approach channel \'li11 occur-when six machines are operating and the uutlet facilities are discharging at reaximum design capacity, with ~he reservoir· drawdo\·ffl to EL 2045. Under these con:- ditions the maximum ve1ocit_l/ of flow in the approach channel wil1 be 3.5 ft/s, which wil1 not cause any erosion problems~ Higher flm•Js will be possible where the intake approach channel intersects the ·approach channel to the main spi 1lway and the approach channel. to the emergency spilhvay. The maxi_mum velocity of f1ow 't;ill be about 25 ft/s'! and excayated slopes in this vicinity may need increased support at local- ized are-as of sheared or fractured rock. Excavations in overburden wi11 generally be trimmed at 2H;1V; rip-rap pro- tection v'li11 be required in the areas \~there high-flO\'/ velocities ar·e anti- cipated_ 12--46 I I ;I'_· __ ·_ ~. . --· I j " .' ~ ' 1··_. I I I I I I I I I I I I I I I I I I I I I I I. ·I I I I . ·I I I I I I I (f) (g) Geotechnical ConsidEwations The excavationw.f11 be over 200 feet deep in ruck in the northwest corner-, with a total~ excavation d~pth of 240 feet. The southern end of the struc- . ture will be located in tne ao feet vJi th shear and fracture zone. The ex~ cavat ion depth at the north end of the str"uctura wil1 be 120 feet. \Hth sufficient rock support, main1y fr·om rock bolting, tne rock -:lopes can be cut nearly vertical, \1/ith the pussible except.ion of the southern end., v.lhere the excavation wil1 "intersect the fractur~ and snear zor1e. If it proves impracticable to support this face nearly vertically~ it will have to be trimmed back to a stable slope. The int-ake structure wuulcl therl bd pa\"tially free-standing .. The spillway tunnel portal will also be 1ocat-ed in this zone of fractured rock and will r·equire substantial rock support installed in the portal face. Since the intake structure will~ when ccm- plete, support this rock face, the required support will be temporary. The found at ion wi 11 be in sound rock~ but the shear and fracture zones. at the southern end may require consolidation grouting. t11inor shears and . fractures exposed in the remainder of the found at ion area may require "!oca 1 grouting and/or dental concrete. Mechanical Arrangement ( i) Ice Bulkhead A heated ice bulkhead will be installed in guides immediately up- stream of the trashracks for each of the six pO\•ter intakes. Tb: ic~ bulkhead will be operated by a movable hoist with a double point 1 ift and will be automatically raised and lm1ered so that it w111 always be at r·eservoir level. The ice bulkhead wi11 serve to~ -Ninimize ice accumulation in the trashrack and intake shutte·~ area; and Prevent thermal ice-loading on the trashracks ~ The bulkhead will incotporate fixed \vheels which will serve t,:: transmit thermal ice loads to the embedded guides. The bu1knead \'lill be totally enclosed and heated~ The power supply cable t:: the bulkhead will be located on an electric, motor-operated, tak~-~~P reel \vhich will operate in conjunction with the bulkhead hoist, The bulkhead ~vill be handled by a movable hoist to facilitate re~r:\.~',ial of the bulkhead for trashrack cleaning. In the unlikely event that it is desired to dewater the compl~t2 intake for inspection of the shutter guides, stoplogs can be pJ~~­ chase;d and installed in the ice bulkhead guides~ The height uf the stop 1 ogs will depend on the reserve ir 1 eve1 at the time of inspec- tion. 12-47 ·I {if) Trashracks Each of the six pmver intakes 'Nil1 hav:e four sets of trashracks, one set ·;n front of four intake openings.. Each set of trashracks wi11 be in two sections to f ac n i tate hand 1 i ng by the intake service crane with a follower. Each set of trashtacks wil1 cover an opening 30 feet wide by 24 feet high.. The trashr acks wi 11 have a oar spac- ,..ing of about ti inches and \'fill be designed for a maximum differe-n .... tial head of about 25 feet. (iii) Intake Shutters Each of the six pm,ter intakes will have three intake shutters h'hich will serve to prevent flow through the intake openings behind which the shutters will be installed. As the reservoir level drops) the sliding shutters will be removed as necessary using the intake ser-- vice crane, Each of the shutters will be designed for a differential head of about 15 feet~ Suitable alarms and generating unit shutdown devices will be installed to activate when the differential head is atl01Jt 1/2 th·<: value. In addition, at least one of the shutters at each power intake will incorporat~ a flap gate which, with 15 feet dif- ferential head across the shutter, Nill a11ow maximum turbine flmv through the flap. This will prevent fdilure of the shutters in the event of accidental blocking of a11 intake openings be1ow wat~Y" leveL The shutter guides will be heated to facilitate removal in s~b­ freezing weather. In addition, a nubb1er system will be pr·ovided in the intake behind the shutters to keep its water surface free of ice in this area. The bubb1er nozzles will be located at several 1eve1s to per1nit bubbler system operation as the reservoir is drawnd~wn. tiv) Intake Service Crane A single, overhead, travel ing-br·idge type intake service crane_ t~i11 be pr·ovided in the intake ser-vice buildings. The crane will be used: For serv1c1ng the ice bulkhead and ice bulknead hoist; -For handling and cleaning the trashracks; -For handling the 'Hater intake shutter-s; -For handling the intake bulkhead gates; and -For servicing the intake gate and hoist. The overhead crane will have a double point lift and will have fol- lowers for handling the trashrack shutters ana bulkhead gates. The crane will be radio-contr-olled with a pendant or cab control for backup. A grappling hoist rlill be provided for cleaning debris from the rack area* 12-48 I I I .; I I I ' .•. . . I I I I I I I I :1 I I I I •• I I I I I I I .,. I I I I I •• I I I . {v) Intake Hulkhead ti~tes One set of intake bulkheads, consisting of two gate SciCtions, will be provided for closing any one of th{~ six inta~<e openin9s upstream from the intake gates~ Tt11~ gat~s ,,,; 11 be used to permit inspect ion ·and maintenance of the intake gate and intake gate guides. The gates will be raised and lowered under balanced ¥·Jatet conditions only. To balance water pressute ·in order to raise the intake bulk- head gate, the space between the gate and the dovmstream contro 1 gate will be flooded by a follower-operated bypass valve oti the top gate section; an air valve will be provided in the top of the ~ate. The gates wi 11 have a downstream sk inp 1 ate and wi11 seal on the dovmstream sid-e. The gate \vil1 be oes igned to withstand full dif- ferential pressure. (vi) Intake Gates The six intake gates!! one for each intake, will e3ch be provided to close a clear opening 17.3 feet ~tlide by 16.7 feet high. It is pro- posed that the gates \vi ll be of the vet·t ical fixed wheel 1 ift type with an upstream skinp1ate, and they will seal on the upstrea~:1 side. Each gate \~ill be operated by a hydraulic cylinder type hoist~ mounted below the 2,200-foot deck level. The length of a cyli~d~r will allow withdrawal of the gate from the water f1m·t. The c): illder and gate will be connected by a series of hooked links. The i::take ser'vice vli11 be used to raise the gate, cmnplete ;·lith links aP~ cyl- inder, by dogging and removing each link in turn until the gate is raised above deck lev~l for maintenance. The gates wil 1 normally be closed under balanced flow conditions to permit dewatering of the penstock and turbine water passages for inspection and maintenance of the turbines~ The gates will also be designed to close in an emergerfcy with full turbine flov1 conditions in the event of 1oss of control of the turbine. The hoist pumping unit will have an operating pressure of l,OCD to 2,000 psi.. The hoist will be designed to allow gate closure ",;·;ithout any ac power supply. A heated air vent will be provided at the intake deck to satisfy· air demand requirements when the gate is closed with f1owing water-.. 12.12 -Penstocks The general ar·rangement of the penstocks is shown on Plates 54 and 55. The maximum design static head on each penstock is from normal full reservoir" elevation (EL 2185) ·to centerline distributor level {El 1422). An allowance of 35 percent has been tnaae for pressure rise in the penstock besause of hydrau) ic transients. The maximum design head is therefore 1,030 feet. Maximum ext~sme head (corresponding to maximum reservoir flood level) is 1,050 feet. 12-49 :;.::.· .. ·-. (a) Steel liner It has been assumed that the r·ock. adjacent to the powerhouse cav.ern wi 11 b~! incapab\~J:~J. long""'term restr·aint against the forces transmitted from pen- stock t;:Ydt"'ati'tic pressures, Consequently, a steei liner will be requ·ired which ·wi11 wholly r'esist the maxi-mum design head, without suppor"t frc~:n the surround rock.. This section of steel liner wf11 extend 50 f~et"frcm t~1e powerhouse. Beyond this section the steel liner will be extended a further lbG feet, and allowance in the design wHl be madf.! for partia1 rock StiPi)Or't to 'mitigate the maximum design stress. For pre11minary design purposes it is assumed that not more than 50 percent of the maximum. design _head w111 be taken by the rock support over this transition length,. The steel 1 iner \<till be surrounded by a concrete infi11, \viti1 a minimum thickness of 24 inches. The optimum internal diameter of the steel lining will be 15 feet based on the minimum total cost of construction and the capitalized value of anticipated annual energy losses~ A tapering transi ... tion will be provided to increase the internal diameter of the steel liner to 17 feet at the.junction bet\•Ieen the stee1 linsr and the concrete 1inet". (b) Concrete Lining The penstocks wn 1 be fully 1 ined with concrete from the intake to the steel lined section:1 the thickness of lining varying with the design head. The optimum internal diameter of the concrete l_ined penstock will be 17 feet, based on the minimum total cost of constr-uction and the capitalized value of anticipated annual energy losses, and assuming an av0rage coocrt.:te lining thickness of 18 inches. The minimum linin_g thickness \1i11 be 12 ,. f . -1ncnes. (c) Grouting and Pressure Relief A comprehensive pressure relief system wi11 be required to protect t~~ undergt~ound caverns against seepage from the high pressure pens tcck ~ The system will comprise sma11 diameter bm~eholes set out in patterns anc t:ur- tains to intercept the jointing in the rock. Grouting around the penstocks will be provided to: -Seal and fi11 any voids bet\11een tbe concrete lining and the steel ~-~h-ich may be left after the concre:-J:e placing and curing; and- -Fi 11 joints or f~~actures in the rock surrounding the penstocks to reauce flow into the pressur·e relief system and to consolidate the rock .. 12.13 -Powerhouse ( a), Gen·er al The underground_ pm·Jerhouse complex ~tli 11 be .constructed underground in the -right 7 abutment.. This will require the excavation in rock of three :najor 12-50 . . ·::- I ---. -- l I I I I I I I I I I I I I I I I I I I I I I I I I ·~ I I I I caverns, the powerhouse, transformer gallery~ and surge chambers 1.v~th Hl- terconnecting rock tunnels for the draft tubes and iiolated phase bus ducts. Unlined rock tunnels will be reauired for vehicular access to the three main rock caverns and the penst~ck constructi~n adit. Vertical shafts wil1 be required for personne 1 access to the undergrounti po·:;erhou';~) fer cable ducts from the transformer gallery, for surge chamber 1enting and f0~ the heating and vent i 1 at ion sys tern. The gener-al 1 ayuut of the povverhouse comp 1 ex is shr;wn in p 1 an and se•:t ~on in Plates 54 and 55~ and in isometric projection in Plate 5b. The t?s~s­ former gallery will oe located on the upstream s~d~ of the po·dm·huus-J cavern; the surge chamber 1Nill be located on the do'tmstream side. S~~a~~ dimensions bt=twaen major roc~ excavations have been :set at 1.5 t hm:.:;;:. :~~:.~ main span of the larger excavation. This criterion contrc,ls not O~i:} tnt: minimum distance bet\veen caverns, but also the spacing between tr·an~f"~"\Ht:J' gallery and penstoc~, beb·teen Lus shaft and penstoc<) and the min~mt,::~ s~;dc­ ing of penstock and tailrace tunnels. The draft tube gate gallery and crane will oe located in the surga .. ~ .. ~.;;:~;~e~~ cavern, above the maximum anticipated surge level. Provision wil1 a~s: o~ made in the surge chamber for ta i 1 r cce: tunne 1 intake stop 1 ogs, \vh h::h ,..;~ 11 be handled by the draft tube crane. (b) Layout Consideration The location of the powerhouse \'/as selected from consideration of th~~ following data: Plots of the known major faults and shear zones on the·right abut~~~:; -Estimated cost of approach ch.:mne1 excavation) intaKe str·ucture, ~'e'i­ stocks, .and tailrace; and An ass!lmed argle of 55Q to the horizontal for the inc1 ined section ,)f' penstock~ Prel iminal~y cost estimates indica~e U1at the intake ::tructun? and .a~-~~ .... -ac~: channel excavation are the most sionificant items in the overall ay'T~nce- ~ ~ ~ent of the power facilities; the underground powerhause costs are a~;~n- dent only on installed capacity. The optimum atTa'hgement has the~'t.:f,.,,~~t: been determineo by adjusting the position of the i~take to giv~ the :~ast cost for intake, penstocks, and tailrace~ Since the costs of tunne: ~-,~ ar·e- small compared to the intake costs, the intake will ne sited as far ~J­ stream as possible) consistent \'Jith the required minhnum draN:.im·m 1tE;•,21 .. and a reasonable 1 ength of access tunne 1 s ~ The underground transformer gallery will be located on the upstremn s~~e of the powerhouse.. This arrar1gement gives the minimum possible d1stanc-:: bet\<Jeen the turn ines and-the surge ch ar.1ber, for mt1X imum protect icn cf the 12-51 (c) (d) draft tubes under transient load conditions .. The trans· .. ormer gallery and- the powerhouse cavern vii 11 be pr·otected against high pr ;ssure seBpage from the penstocks by a .200 foot long stee1-1ined section arid an extensive pres- sure relief system (see Sect ion 12~12) 4 Access Tunnel~ and Shafts Vehicular access to the underground facif)~·+:ies at Hatana will be provided by a single unlined rock tunnel f}*Om tilt: right bank area adjacent to tht? diversion tunnel portal. The access tunnel will cross over the diversion tunnels and then aescend at a uniform gr·adient to the south end of the powerhouse cavern at generator floor level~ at El 1463. Separate branch tunnels from the main tunnel will give access to the trilnsfot?mer gallery at EL 1507, the penstock construct 1 on ad it at EL 1420, and the draft tube gate gallery at EL 1500.. The maximum gradients will be o.l percent on the con .. struction access tunnel, and 6.9 percent on the permam.mt access tunne1s, The common access tunnel will be size.d to provide passing c1earanc2 fer the construction-plant used during penstock construction. Tne size of articu~ 1ated trailer required to de1 iver heavy items of machinery such as the tur'- bine runner, turbine spiral case, and generatot rotor, will be less crit1- cal with respect to tunnel size, but will dictate the minimum radius of vertical and horizonta1 curves. Fat~ preliminary design, tr1·e cross section of the access tunnel has a modified horseshoe shape, 35 feet wide by 28 feet high. The access tunn.e 1 branch to the surge chamber and draft tttbB gallery will have a reduced section, consistent with the anticipated size of vehicle and lo-ading tequir·ed. The main access shaft ~vi 11 be at the north end of the poworhouse c ~ivern ~ providing personnel access from the surface control builtJing by elevator. Access tunnels will be provided from this shaft for pedestrian access to the transformer gallery and the draft tube gate gallery. Elevator access wi11 also be provided to the fir8 protection head tank, located about 250 feet above powerhouse 1 eve l. Powerhouse Cavern The main pO't'.fer·house cavetn 15 designed to accormn0date si:< vertical shaft Francis turbines~ in line, with direct coupling to coverhung generators. Each unit is designed to generate 170NW at a rated head of 680 feet~ The vertical dimension of the pow·erhouse cavern is determined by the physi- cal size of turbine and generator, the crane height reqairea for routine maintenance, ano the design dimensions of the turbine draft tube. The length of the cavern \vill allow for a unit spacing of tiD feet, tttith a 110- foot long service oay at the south end for routine ~aaint.:::nance and for con- struction erection. The width of thecavern allovts for the physical size of the generator plus galleries for piping and air-conditioning~ electrical cables, isolated phase bus ducts) and generator circuit oreakers. Contin- uous d~ainage galleries will be provided to a low level SJcp. 12-52 I I 1 .. '--.::: I I I I I I I I I I I I I I I I I I I ·~ I ·- I I I •• I I, I I I ., I I 1- Vehicular access will be by tunnel to the generator floor at the south end of the cavern; pedestrian access win be by e1 eva tor frone the s_urf ace con- trol building to the north end of the cavern. Multip1~ ~tairway access pqints '.vi11 be availaole from the mairr·f1oor to e:tch gai18ry level. Access to the transformer gallery from the pmverhouse \·+ill be ·by tunnel frcm -the main access shaft; or by stairw.ay through each of the h:~1ated phas~ ~)'JS shafts. A service el~vator will be provided from the maintenmtce area on the main floor level to the ma~.:hine shop and stores i'lrr2'l t;n th~ turb~ne floor level~ Hatches \'lill be provided ttrcough all main floors fnr in5tii1latic..m anJ main ... tenance of heavy equipment using tha over~head traveling cranes. (e) Transformer Galler_x The transformers will be locat~d underground in a separate gallery, 120. feet upstream from the main pow::=rnouse cavern, with three connect in;1 t"tjn- nels for the isolated phase bus: Tl1ere will be nine sing1e-phase tt"2!!1Sfm~­ mers rated at 15/345 kV, 122 MVA, installed in groups of three one gr0up for each pair of turbines. Generator circuit breakers wi11 be required, and wi11 be installed in the oo'.-Jerhouse on the lm..;er Clenerator floo~~ 1eve1. . ' - High voltage cables wi11 be ta~~en to the surface by tltiO cable shafts, each with an integr·al diameter· of 7.5 feet. r'rovision has been mad~ fo;--~"1sta1- 1ation of an inspection hoist in each shaft. A spare transformer w~~~ be located in the transformer gal18ry, and a spare HV circuit will also~~ provided for improved re1 iab11 ity~ The stat hm service 1 11Xil iary b"J.t~sfor­ mers {2 MVA) and the camp services aGxiliary transformer (7.5/10 MV~~ will be located in the bus tunnels. Generator· exc-it.:iticn tran:;,formers .... d]1 ne located in the powerhouse on the main floor. Vehicle access to the transformer gallery \'li '11 be the main power'hous::: 3C- cess tunnel at the south-end. Pedestrian access wi11 be from the ';1::t~f'J ac- cess shaft or through each of th2 three isolatud phase bus tunnels. (f) Surge Chamber A surge chamber will be provided 120 feet downstr~~l frcrn the power~:~se cavern to control pressure rise in the turbine oraft tubes and tai~r3:B tunnels under transient load conditions, and tJ prlviae s:~rage of water for the machine start-up sequence. The chamber vii ll oe ~Jzmon _to aJ: six draft tubes~ and under normal operation will discharge equally to th~ two tailrace tunnels. The draft t~be gates gallery and crane will be located in the same Cai.'8rn.) above the maximum anticipated surge 1evel. The draft tti:Jc gate crar.t? ha;, also been designed to allow installation of tailrace tunnel intake st~plogs for emergency closure of either tailrace tunn~l. The chamber will generally be an unlined r·ock excavation, ,~· .. il'!:h loca1i.:~ed rock support as neces s ar~y for st abl1 ity of the roof arch and ~tla 11 s. The ., 12-53 gate guioes for the dr~aft tube gates and tailrace sh;p1og;; ;,till be of :~t'in ... for•ced concrete) anchored to the rock by rockbo 1 ts. Access to the draft tube gate ga 11 ery wi 11 bt; by an aai t ft'Ohl the 111a in iV.> cess tunnel; the tunnel will be widened locally for storage of draft tube bulkhead gates and stoplogs. 12.14 ... Reservoir -- The Watana reservoir, at rormal operating level of 2,18~ feet~ w~ll be appr~xi­ mat~ly 48 miles l0ng with a maximum wioth in tha ord~r of 5 miles. The total watel~ surf ace area at normal operating leve 1 is 38,000 ::1cr·8s. .Just up-s t~"e,}.;li from the dam, the maximum water depth ;vill be appro;dmatt~1y 6:W feet. The i~~~ni-­ mum reservoir level \vi11 be 2,045 feet during normal fJperat1on, y·esulting ;11 a maximum dra.wdm~n of 140 feet. The reser·voir will have a total capacity uf 9,515,000 acre-feet of which 4,210,000 acre-feet will be live storage. Prior to reservoir filling, the area below Elevation 2190, five feet above ma~i­ mum operating level, ·wnl be cleared of all trees and brush. A field r€C\1nnais-- sance of the proposed reservoir area was undertaken as part of these studies. This work included examination of aerial pnotographs and maps, an aer'ia1 uV-2¥"'- flight of the reservoir and collection or recent (1980 field season) for2st in- ventory data fr\un the U.S. Forest Service. Most of the vegetatal material :.dth- in the reservoir consists of trees, with very little in brush. The trees are quite small, and the stands are not very dense. In the Watana reservoi~ ar~a, an estimated U;;;UUU,OUO cut>ic fe~~t of \~·ood exists averagi11g appr:>ximatt;ly 5JO cubic feet of lo•tJ comlller·c.i3.1 quality, and some ver-y significant 1ogginq prJn1ems would be poseo oy the steep s1opes arld incised tt:rrain encot.H1tm·ed in trh: aJ-ea. Approximately 87 perc2r1t of the avail ab1t2 timt;~.;r-an~ soft ltiJO~"l~~. TtH} r·~,is:.;~ts of the timoer reconnaissanct: studies ar-t: aescriDi;;d in more detail in Auot?.nuh. C3. . . . , . The comDination of steep terrain, moderate-1 ight tr~e stocking levels} Siila11 trees, :2rosive potential 'Jf the rese~~voh· slopes, rt:!mutt:.:ness, and very re- stricted access to the reservoirs ~·~ major factors ~ff~cting the choice of har- vesting systems to be utilized for this project. Sucr1 syste:ns include h1;;n lead, skyline) tractor, whole tree logging with or without chippers, ba11oD~ and helicopter. fach system has its O\'m advantages and disa.:wantages and set :rf conditions under which its used is optimized. Present mar·ket demand for the timber at Sus i tna is l 0'11, ho.·:ever the wor 1 o~·~ ~de demand v;ood fluctuates consicerably. It is anticipated that use of the hs.r- vested material would be limited to either sale as wood-waste products ~~d as fue1. Slash material including brush and small trees, v1hich will be unsuitab1e ::-0-r~ either of the above uses, will be either burned in a carefully control~ed ~anner consistent with applicable laws and rt:gul at ions, or-rl3u1ea t~ a disposal sit.e in and adjacent to the reservoir. Nate:ia1 placed in dispos:11 areas will be cov·· ered ~'lith a earthfill cover suffic ier"!t t•J pr·event erosion and subsequent expcs- ure. 12-54 I I I I 'I I I I I I I I I I I I I I 1. f l I t· I I I I I 'I I .. , . • •• I I I I I I I 12 .15 -Ta i 1 race Two tailrace pressure tunnels will be provided at Watana to c~rry water from the surge chamber to the river. The tunnels wi 11 have a modified dL:HilE!L;r cf horse- shoe cross-sect 1 on \vi th a major i ntt:~rna 1 d:imens1 on of 35 fe.et. For prel imi nar·y design tiley are assumed to be fully concrete-lined throughot:r., with a miflirm,;_m c-oncrete th1ckness of 12 inches and a lennth of 1,800 feet .. The tailrace tunnels \•Jill be arrang~:~d to discharge into the~ r~'JE!r between the ma1n dam and the main spillway. In viev-t of the SGver-e limitations on space i·1 thjs area, one.tailrace tunnel will be designed to discharge t~rough one of t~e diversion tunnel portals. The cross section of the tailrace !:unnei will be ~~t--:d- 1f1ed over the common length of 300 feet to the shape of tre·::nversion tunne1 in order not to impair the hydr"aul ic per~fm'r.Mnce of the tailrace tunnel. P\fte!~ commissioning, the diversion tunnel upstream section will be plugged w'th t~n­ cn:te. The size of the t\10 tailrace tunnels ;·1as selected after an e::onomic study of the cost of construction and the capitalized value of average annual energy losses cause-d by friction, bends, and changes of section. In an emergency, however:. the station can be operated using one tailrace tunnel, with increased head losses. For such an emergency condition, tailrace intake stoplog guides will be provided in the surge chamber. The surge chamber ~vi 11 be designed for ~u 1 ~ 1 oad rejection with either one or b'iO tailrace tunnels in operation. The tailrace portals will be reinforced concrete structures d~~igned to r~~~ce the out 1 et f 1 ovt ve 1 oc i ty ~ and hence the vt: loci ty h~:ad 1 oss a: the ;;;x 1 t t a t ·;~ river. The.mtnimum rock cover required above the tunnels ~il1 be 1.5 times the major excavated dimension (about 5,~ feet}, and the pcrtais ·.dil also provi-::~2 the necessary transition length to the river where the rock ccv~r wo~ld be 1ess than 54 feet . 12.16 -Turbines and Generators (a) Unit Capacity The Liatana powerhouse will have six generating unlts ,n:n a nominal c5na- city of 170 NW. This is the available capacity with r~nhnum Decembe~ .. ~~es­ ervoir level (El. 2112) and J cor~esponding gross hea~ of 562 feet on ~he station. The head on the plant wil1 vary f~~om 735 feet !TJaxiillu;n "724 feet net h~s.d) to 595 feet min 1 mum (584 feet net 1 eve l) . Because na;dnum turbine o:r:put varies approximately with the 3/2 power of·head, the ~aii~um unit out;ut will change with head, as shown on Figure 12.21. The rated head for the turbine hds been established a~ 580 feet, which is 'the weighted average operating head on the station. · ~~lowing for generator losses, the rated turbine output is 250,000 hp (186.5 ~,:~;). 12-55 .J I The generator rating has been selected ·as 190 MVA with a 90 percent power factor, whi-ch corresponds to a power output of 170 MW. fhe 9€nerat;:,rs ·.·ri11 be capable of a continuous 15 percer.t overload; ·this w1 1! anoN a unit OIJ~­ put of 196 t1H. At max 1 mum resetvo 1 r water" 1 eve 1, the turb i n~s !fli 11 be ope~~­ ated below maximum (full gate) output to avoid ov~rloading of th~ g~nera~ tors. (b) Turbines The turbines will be of the vertical shaft Franci~ tvoe with steel spiral casing and a concrete elbow-type drnft tube. The dr·'dft t.Jbt; will c~Hr:pr'U s:? a single water passage without a no renter pier. The rated output of the turbines wi 11 be 250,000 hp at 6;;D feet rat..;;d :iet head. Maximum and minimum heads on the units will be 724 fe~!t and 584 f2~~·~ respectively. The full gate output of the turbines w111 he about 275 .,GOO hp at 724 feet net head and 200,000 hp at 584 feet net he~d. Overgatin9 df the turbines may be possible, providing approximately 5 percent additional pm'!er; hm1ever, at high l}eads the turb1ne output ~vill be rr::stl"icted to avoid overloading the generators. The best eff~ciency poi~t of the tur- bines will be established at the tjme Gr preparation of bid documents for the generating equipment and will be based on a detailed ana1ysis of the anticipated operating range of the turbines. For preli~inary design cur- poses, the best efficiency (best gate) output of the unlts has been a~s:.n~r:d as 85 percent of the full gate tu~b1ne output. This percentage may v~r from about SO percent to 90 percent; in general, a lower percentage rec~(~S turbine cost. The full gate and best gate efficiencies of the turbines will be about 91 percent and 94 percent r2spectively at rJted he~d. Tho effi~iency w~l1 be about 0.5 percent lm.;er at maximum head and 1 p1::1"cent 1t.Mr;r· at minitt'!~;::~ head. The preliminary performance curve for tho turbine is shown on ~i~ure 12.22. A speed of 225 t~pm has been s~1ected for the urdt for preliminar·y des.~;n purposes. The resulting tarbine specific speed (Ns) is 32.4. As sh0~~ on Figure 12.23, this is within present day practice fDr turtdnes op~~·?:tinq under a head of 670 feet. In general, a lower speed ~ac~ine will inc~eas~­ the cost of the turbines and generators as well as the ~ow~~house civ~: cost because of the increased physical size of the ge~er~ti~g equi?ne~~. A hjgher speed unit, on the other hand, requires a deeper ~nit setting ~rd is generally considered to be a less conservative design with increased ~is~ of vibration and rouah oueration. The differencE~ in efficiencv het\-.ee·: t,he .. , ' ..... higher and lower speed machines at this head ra~ge is v~ry s~all, w~tn the increase in efficiency which is associated with a phys~:31ly larger (~~w~r speed) runner offset by higher disc friction and sea1 !ea%ag~ losses. For an undergn:)und powerhouse, the incremental cost of I r:cre~s i ng the ~mit setting is usually relatively inexpensive; therefor~, as;~rning no ch3~9~ in efficiency, the trend in unit selection is to choose as hi9h a speed as possible consistent with .satisfactory precedent and g0od operating e:x;,?er·i- ence with similar specific speed turbines. Draft tube vo:tex and surge 12-56 . , I, t I I .• , I I I I I I I I I I I I I I . ' .. ,. ··,. I 1·~. . I '1: I I I I I I I I I I I I phenomena may also have an influenc~ on the selection .of u~lt speed as discussed below. The ttwbine data is summarized in Tat71e 12.7~ On the basis of lnformat1cn from turbine manufacturers and the studi~s on the po\'ter plant layout> the centerline of the turbine distributor· has bt1en set at 30 feet below min1mum taih;atet level. The final s~:tt~ing of the unit 'v'lt11 be established in conjunction w1th the turbin:; !i1anufacture~· ·~·;hDn the contract for the supply of the t:.n~bine e.quipment h,j-; b~?en a~·i~trl\2·~~. Tht turbines will be of conventional design, f1enera11y >if welded Or' ·c15t steel construction with forged steel shafts and pins. 9ecause o~ t~e remote 1ocat1on of the project and the desir·ed high rG·~~ability/av~i,.,atli 1- ; ty of the eqtd pment, spec i a 1 .cans i deration shou 1 d b~ q Lten t n red:.;c h~~ cavitation pitting on the turbines. This will include: -Provision of weldable stainless steel runners; Careful profiling and finishing of the water passage runner ~·ticket gates and stay r·ing; - A conservative unit setting; and -Extensive cavitation tests on the turbine model. surf .1ces of Bulkhead dm?tes wi 11 be: provided with two of the turbines (Un1ts 3 ar:.d .:!) to be installed at the bottom of the draft tube liner at the time of t~~~~ne installation. The domes perm1t work to continue on turbine insta11a::Jn after thd tailrace, surge chambe:~, and draft tubes are f1ood•:d (pi"L:r !~.:> startup of Unit 1), without installing <.h~aft tube gates. Because of the relative1y short lengtn of tho intake penstock and a s~rge tank location immediately downstream of the powerhous~?, the hydr·au1 L: transient characteristics of the turbines are favorable. Assuming ~c~~al generator iner·tia (H == 3.5 f/1~4-Sec/MVA), a preliminary ~:1alysis has i~~:;i~ cated the following: -t·Jater star·t ing t tme (T'.'I) ............................ 1.6 seconds -r.·lechanical starting tim~ (Tm) ....................... 6.6 seconds -Regulating ratio (fm/Tw) ............................ ll.l -Governor time .......................................... 6.0 seconds -Speed rise on full load rejection ................... 42 percent -Penstock pressure use on full lead rejection ........ 30 percent The regulating r·ati0 is above the minimum recor:1~ended b~: the USBK fc•"' ~-~od regulating capacity. Also) unit speed rise and penstock pressJre r~s~ are all well within ·normal accepted values. Because of the deep unit set~1~y and the relatively short distance bet'.veen the tw~bin::: and the ta11r.~,..'=~ surge tank, there will be no problems with draft tube water coJunn s~para- -~-. ... lon. The Hata11a project will form a 1 a.rge port ion of· the over a 11 system g2~'erat­ ing capacity in Alaska; therefore) satisfactory operatic~ of the units over a very wide range of loads wi 11 be 1mportant. Althou~~h there are Frc.r~.~i s turbine installations which operate for considerable periods at very s~all gate openings, operation belo~1 about 50 per·cent load gEr.era11y beco1: .. Js increasingly rough because of reduced efficiency of the t~rbines. 12-57 0 The ab i 1 i ty to op~rate .at par't 1oad vii 11 a 1 so dt=pend 0n the draft tube surge ph2nomena JJid ·associated generator· PO\'ler s·.-Jings ... The su~ging occurs on many turbines, partit:u1.ar1y thr. Francis type, and narma1ly has a fre- quency of about 1/3 to 1/5 of the turbine rotational s;1~ed. Thesa S:.H"'ges, ,,,h ich may occur from abtHlt 30 percent to som~t imes as h i1h as oO pet .. Ce!1t wicket gate open;ng., result in puh~ating tor·q~Je on the turbine runfl~r· and corresponding g~ner a tor po.wc:r f1. uctuat ion$. Thr; cond 1t ~on tH;.comes rnore severe when the sur·ge fn1quency c1os~ly cncr"'!sponds t'1 the rnt.;ral f~-e-­ quency of the gener JtiJr. To rt:thh>~ the pos~:; in 1 'I i t,y of •.~n rlCt>:-pt 1ti 1 ~:: pO'•'l.::O~" swings and unit vibration, it is lif~sir·able to hj.Vt~ V·~ ~wr~;.:; ftd~~WL!ncy d,'f,t:;~rent r,-.ro1~ t'h·e o· (.'JJ"'.-+~rJK pr·li'C'Jy.--• •li;,\I.M. r"~,.._.l.l r···~~~v "t•..-4 trj,•· /''"r't'<J""'"'">.>' I ..,. . h , ..:..11 .:> t.IJ ... 1\ t! ..') '::' '-.0 t: •n:<, ':::: • l";..l.'l})~.!lt\..., ~.,>I.< ,., F,i ,p.;; ,r;. C ""'·'' natural frequency. Ti1e estimated no.tura1 fr·equency or tne gener~otors Ni ~ 1 be at;:Jut 1. 3 cycles/s~ \'lhich is unoesirdb1e •·,.hen cons ider"ir1g a poss io1a s.ycge frequency of about·o.7S to 1.25 cycles/s. The selection of a lO'tter' unit S·peed 'tiould separate the draft tube sw~ge frequency from the gener,atcr natural frequency; hm·Je':er > this ~·li 11 1ncrease the generating equipment as we 11 as pm"·erholise cost. lt may be poss it 10 to increase·th? natural frequency of the generator by 'feducing the generator' inertia (wH2) as much as possible; however, the lowe~ ~R2 has an adverse effect on the transient ch-aractel-istics of t-::e Jnit and may aff2ct electrical syst~m stability. Careful model studies of the turbine to accurately ~r~dic~ draft t~~~ Vo rtex/s·u·rgi=l Pt1r-lt'•o·",",·.!lln?<-01 t'"""~ng ~.<L1-s''1"'··-.,,; .:.r:c; .. ,., .. ; .. t.•")~'l t·a·l ~"':::s·'·!·r-"' , -~ -.. I"'-"~ o c ~~"...-, .. ,, .t!:~ t. t~t l::! , ~.~_r.:t: u-. \.,..-! .... · ~· :.~~o~ ,,.... ~ ·).~" ·'~ ~-..., • ~ -U~--t'::ll\:'i surge problems; howeVef, a r·~d~JC t ion of pt.a~t-1 0 ad ;;;~f ic h~~Ky mawv f~St~} t. -Pr~ovision for air admission to the draft t.~lbe. Th~5 is done on rncP~~ or' less a trial and error basis ana mav includt! inie~:":: . cf 5ir frc1:! ':h2 station 100 psi COt:tpressea o.ir-syst~m, spc!C iGJ.1 1t.•·i iJress~.r,.·e compn:?:SS:Jrs •d • C • 1 1 f ~ b • l ' • I • • ~ prov1 ed specn 1ca. 1Y or ufart tu e a11~ ao!ll1$S1Gl1~ ac;aicw prov1s1 ot an 11 air-head 11 ~·mich allo~·Is atr:~Dsph2ric a.ir to b·.~ c:~·::M: dc.m the ge:-;-,;;r·att)f and turbine shaft aod through the runner o.::nt:.:. -Additions of fins t8 the i'lt'aft tube cone h'm~~dia.te~J bela~·; the rLm::~t·. \4hi1e this has ne~~n us,~d on 1:~any insta11aticn:;~ ::~ere have teen i~s:ances of structural failure of the fins. "not· n' aY. ::~nnro;lrl., r'tflr"C.l1+-1y ••nn~., .. s·blr1Y, ~)v rn·· nq~:.,: ~r. '-n . .-1;::~ ... ~·r'1o t')J" .• ·.,;_;,.. .n ~t ~tJfl \.A'l.#tJ '-"'t._~ · "'-'~ t_. ~~ u._..; ._ y\.. "\J -J ;..,;;\Jt': \o-.J..J-~i\ "woP w'u ...t·~ .. ;.p 'r..-._ ,t'""V4i-'t-..,..,;; oscillations resulting frcm the draft tube sJrges by va~yi~g the generst0r excitation~ Computer simulation irwicatas that is f~asioie tJ damp~n th~ large oscillations; however> this has yet to be t~sten in a prototype unit. Employing one or' fiiure of the above approaches, a d>2s;gn may be achievt:d that operates satisfactorily \vithOllt serious gt:nerat:=..:r power· svJings. The potential problem, however, must ae given serious consid~ration in th~ design stages~ 12-58 t I 1 ... . I I I I I I I I I I I I I I I I l I ... . . ~· I I I' I I I I I I" I I I I I I :..,_,. _ _ , .. · . I . {c) .::, Gener a.tors (i) Tvpe and Ratino The six generators ;,'l the Watana powerhouse; w:n l be of the vert :cal shaft, overhung type .l·\rectly connected to the ver-tical Francis tur- ·Lines~ The arrangement of the unit-s is shown in p],ates ana the ""''t~J!~- s i ng le 1 i ne diagram i ~. sho\·JO in P 1 ate . - The optimum arrangement at \~,~tana will consist 'If tv,;o gsnerator~s p~r transfnrmer bank:~ with each transformer bank ctJmpris ing three singl-e-phase transformets. (Development of this scheme is desct·ioeu in Section 12.18)4 The generators will be connected to the trans- formers by iso1 ated ph as~ bus through generator circuit or~ake~s directly connected to the isolated pnase bus ducts. Each generator wi11 be proviclt-~d \vith a high in'itia1 respo11se st:=tic excitation system, The units wi11 be contro11ed from the viatJn~ surface control room~ with local control facilfty also provide-a at the powerhouse floor. The units will be designed for black start operation. The generators ar·e ~"'.:\t~~d as fo11m•ls: Rated Capacity! Rated Pov1er ~ Rated Vo 1 tage: Synchronous Speed: Inertia Constant: Transient Reactance: Short Circuit Ratio: Efficiency at Full Load: 190 MVA, 0.9 power factor 170 Wtl 15 kV, 3 phase, 60 Hertz 225 rpm 3.5 kW-sec/kVA 2B percent (max imur.t) 1.1 (minimum) 9B percent (minimum) The generators \>Ji11 be of the air-~coo1ed type, ~lith water-ca;..,:::;:J heat exchangers located on the stator periphery. h'1e ratings ;:iven above are for a temperature rise of the stator and rotor 'tlinc~·1;s not exceeding 60°C \vith coo1 ing air at 40°C- The generators wi11 be capable of delivery 115 percent of rate·: :..VA continuously (195.5 N~l} at a voltage of +5 percent ~·rithout ex=-;;~ding BOOC tempei~ature rise in accordance ;·lith t-\NSI Standard CSO.lC. The generators will be capable of continuous operation ftS synchron- ous condensel"S Whtln the turbine is Um'latered, ltdth an underex~ited reactive power rating of 140 /ltiVAR and an overexcited rating c-t lliJ t~1VAfL Each geoerntor ,,,lill be capable of ener•gizir.g the trans;-:13ssion system without risk of self-excitation. ·The design data of the generators stated above should be revie.-~ed during the detailed design stage for overall economic ana tec~nical design and performance r·equ 1rements of the pm·;er p1 ant and the powe; system. 0 ( i i) The generator will be of a mad ified umbrel1 a type ov'~~rhung const""tlc- tioni) Nith a combined thrust and guide bearing be1o•;t th~ roto~--and a guide bearing abovt;~t_he rotor._ The lower oearit1g brack>llt wi 11 s~p­ port the rotor and turbin1-1 tum1er weights and the unba1t1nr:ed hy,jrau· lie thrust of the runner. All removable parts, inchtding tur·bin~: p_arts~ wi11 be designe·j for removal througi1 the ':]erwr1t>:w st~tur. Approximate dimensions and weights uf the pr·incipal P·:trts ~f generator are given below: Stator pit diameter Rotor' diameter Rotor length (without shaft) Rotor weight Total 't'le .;ght 36 feet 22 feet 7 feet 3B5 tons 660 tons It should be noted that these are approximate figures and th~y \dl1 vary between manufacturers.} sometimes considt:rab1y~ However·) at this stage of design feasibility and planniYlg) the dimensio~is and weights are considered appropriate and repr2sentative. The generator stator windings will be insulated with G1ass b ins~la­ tion as clefint:~d by ANSI Standard CSO.lO) of eJ;oxy resin tron;::~·J :.:.J~e. The stator windings will be wye-connact~d fur gr0unded op~~~~=oo through a neutra 1 ground 1 ng trans formet' 1 uc i.ite1 in t.h~.: ~t.~:h:~ .... ~t~r neutral cubicle. The sta!:nr windings and lari!ir:ated-c~we v:n~ be shop-assemb l i?.d in thre~ or four sections f;Jr f ac i 1 i t:y of tr· -~~·1s;:cw~ and erection in the powarhous~. The rotor wi 11 be designed to s :1fe 1 y '.4i th:Jt : .. H1:1 tne maximum ~· :~·~ .:t'day speed of the turbine. The rotor hub, yok~J, ~·-;:i laminated ~"'~n :'l1i11 be designeo for assemb:y at the powerhou::.t~. 7•H:: assen:b1~d i~~~t.;r· will be erected in the genc:rator pit without '!:he shaft, :ne;"~:>y f'e- quiring min iri1um crane 1 ift ano a cons idl;!t~at 1~ t>2duct ion of t.:~~ powerhouse cavern height. The rotating parts of the generator and turtine will be desi~nej sJ that the critical speed exceeds the run a·t~a.t Spee:d of the un i ~ Jy at least 20 percent. The design of the plant ard po'Jier system ~..v~! 1· not require ad~itional inertia in the rotating psrts; the inertia con- stant specified thus will correspond to the unatura1 11 inert·f3 of the machine. Damper windings of low resistance anj rugged constr~:tion wi11 be provided on the po 1 e shoes and des ig;;~.:; to pr~ov ide a~~~quate damping currents for stabilized operation. The thrust bearing '."''ill be of the adjustable srH::e (Kingsbur}) or precompressed spring (General Electric) type, oil-cooled, with high pressure oil inject ion our ing :>tart ing and st-:>pping. 12-60 :· -~- I ' I . ;: .. I I I I I I I I I I I I I I I " I "': I I ···"·. . . I ••• • ' I I I I I I I I I I I I I (i li) tienaratOt~ Excitation S.1:2.~ei!~ The generator will De provided with a high initial r~spons~ type 't .t. ::. . it:)""·~. ,...,_,....;.. ·r. .• t ··1 i ::~4 ., • .J·H 't' .... C:f'~ ~"" -.. v-·~ ::..1..: ·n n,...h!e"" sa 1c exc .... ~.-H.m ::.>J:.,:.t: •• l Sd~"'IP .t-~~ H•·~·1 r~c~..~,,~H t.ht_l~.-'.A;.,o. ~-'""'l1. fr-om tJ~ansformers connected ah·ectly to the genetiJtor· t~:H·minals, The exc:itat ion syst~m wi11 be capabh;! of suppJyir19 200 petcent of rated excitation field (ceiling voltage) with a generator terminal voltage or 70 pet·cent. The power rectifi~rs t-Iili hava a one ... third • .. • ~· t ... ..t..o"" .1 ... spare capacity to ;~t:nnt211n :F:rwr·a~.1on t3'1en (Jl.H'H1g nu ure tJr a com- plete rectifier modul~. The excitation system wi11 be equipped \·d!.£1 a f:;l ty static vo1taye regulating system maintaining output fro~ 30 percent to 115 p~rcent, ~·lith in +0.5 percent accuracy of tne· voltJge s,2t~ ing. ~ianua1 :ootr•o: will.be possible at the excitation board locat~d on the powerhJ~s~ floor, a1 though the unit wn 1 ntH'ma·l_ly b~! undt:!r remote control 'I as described in Section 1~.18 coverin9 tht~ control syst;:-}mS of the plant. The static excitation system ;·1n l also includ~ :n~:dmurn and mHn:nun; 1 imi ters and react ivt! current compensator, and ~·~i! l be su itao 1e fO!'' paral1e1 joint cont~~ol of the units. Fi~ld fl::shing during st:lrtup will oe from the 125 volt de station battery. (iv) Erection and Tests A . , -1 h ~ t . t ,, l. • .. , ' s 1 s norma. ror i :ir·g~ .r:!n)e f ec r 1 c generd .ors, -cne macn mes :,-;1 1 ~ not oe assembled Ci.1[;~p1ete1y and tested in the factory~ The e"""eCtion and tc;sts of the generatot~s at the powerhouse, therefore, wil ~ as s.ume greatBr impottance in the successful ccm;r: is~ ion i ng of tht: station and should ~e carefully coordinateo witn that of the tur- bines and civ1l works. The assembly of the stator sections will be aone in the pit. The rotor 11'1i 11 be assemo 1 ed in the er·ect ion bay. -:-he powerhouse c~~a:ll~ will be capable of 1 ifting the comp!eted rot1~r ass~mb1y and l:t.lier·ing it into the stator~ m1d onto the thrust be?.ring and shaft ass~r;b1y on the bracket supports. Alignment and tests of the rotor) t~r~i~2 runner, and shaft will be done to tol·~ranc::;s specif~ed in PtB\;h:ANSi -c-~ ' . ' d .)J.. an-l1 ar s. The generators will ne fu11y tested aftar assembly and mecha~L:a1 run tests, including die~ectric tests, satura~ion tests, heat ·un, efficiency, and fu11-1oad rejBction tests. (ei1ing voltage ,:.,;" . .1 ~,;e­ sponse of the excitation sys:em will be tested. ~perati~n cf the unit within specified vibration limits will be chec%ed. (d) Governor System The governor system which control the generating unit will include a gover- nor actuator~ and a govarnor p..tmping unit. A single system will be p~"':Jvioed 12-61 fur each unit. The governor syste~ operdting pressure w~11 psi, as t·ecorr.m-anded by tha gov~rnor· system r.:anuf ao:t ~u-r~r. The gove:"nor actuJtor wi 11 be the electr'ic h,yd~·1n~ i:; t.y!Jt: .1~Ht ·\-;; ., 1 ne con- ... t•r:.t~n +.:J ttJ~:~ comput£_,rl·z~~d r.::t;1t:1.rJn. coqtr·ol -:J-..· .. ":'hp ;;H,~v.::.,...n.,r· n.f't'i'"'~.;,"' It---..... ~..J. _.._ -'-~-w-¥ -....-.... u ..;J' ...,; 1t ~at,.._ .,:j·:...l "J' "lV :t~~'" .Jl:-l(:o::¥¥-c;~ _ unit wiil include governor pumps, an accumulatol t·)n~., .an(i a s~,~Np t::~n~. Each unit wn 1 have. three gavern,;r pumps: two ir~ ~.H.ll;'p;, ;·;h ich \}pr,;t• ~t;;:: intermit t;,:'!nt 1 y, and one jock~;y lJWP.p which ope~· ~Jt>£s c;)n t i w-t>Jus 1 y ~~hi h:~ tb;;; turbine 'tjicket gates are op• . .::n tmd intermittently litl~J~~ t'E= ~ptes Jr'.:! f";11y c1o~ed. 1- 12.1?·-Niscen~r:~~ous ;-.1-::char~ic:al Ecg,dpment I I I I (a) ~ ,_ .. ____ -- Powerhouse Cranes Two overhead traveling bridge type powerhousE~ c:~an';r': ·wi l! .o~ inst.J.1 h: . .:i 1n the powernouse. The cranes will be used for: Installation of turbines, generators, '"'r·d o"-"'· .,, I..J. l L:lt!t Subsaq~ent dismantling and reassembly of overhat~i s. cranes. I , .. I Each c ran;, will h;}, e a ;na in an,J aux i1 i ilt' y h•J L;:: . Th ~ c·;·.:.!:d :·rc:d c ·1:1 • • :} of I tne main hoist for both CtJJL:~s will be suff1ch:nt f•.~r ~:h17• !i·.~~Vh!-;t ~·: ... : .. ·~-. . ~n.snt lift, which will De the y2neratnr 'fOt.:lr', pl j!t., •?•,J•~.i!tzing tH::,l:~. A tentative crane capacity of 205 tons has bet:'l c:·,t,.,;,:isl-:f!•l. Th·~ auxi:i~fy J noist capacity \•.'ill be about ;~:'? tons. · The powef·hou se o~ arH~s \1/ i 11 bt3 cab con tn) l12rl~ Cor~: L:~2r ::t ~orl may given to proviaing radio control for the crdna~. a '!~"'-""" ll _:)· ..•• ;. Jraft t~b2 gates will be provided to pennit d~~d~erirg cf the turbin~ ~ater ...,~S"'a'"'0 ~ ,c,Y' ;,..S.,e-""l'on and "~'a1·ntan:':lnre·· Of-.J-\.-12} ..... t·~•~r.:.~ Tne -lraf.~. '"'·~·.::.'> ~'\.4 ~ ~~..J !v, lll .,._.,. ~~ ... t )t ~J:>i v· .... !J. 1,.; ....... ro;:: \..1..1 LttJ~-.:). ~~f U' ~ :.:..-...i'""'~';: gate openings tone opening p~?r· uni~) will oe 1ocat.~d ir the sur·ge ct::·~.::~.~r·. Th:::~ nar"';;:;. ''111 hr.) of th;:l bul!/h~~ri t··;·f)O ,·,,..,,sta11c,,.l :·(·,··~;:..·!'" h:;:-';;~nrpr! ·n· >:)"":'"' '~n t.Jtt".. ~~.~c_. -.,f, 1 u--.. ""'"'-"",.-t;,..l ..... t......, -, 1 11-J -.A•---:..r..-.tM v-u ._.:._:..Ji \.,"• . ._ ditions ~sing the surge chwnber crane descrioeti o~~o~. Fo~r gates n~~~ been ass~ffif2d for the six units, ·.-Jith each gate a 3 iP;;,; 1e :~~f, 2U fee: ny 20 feet. t~hen Un ;: 1 is ready fm" startup, 5, and b~ with one gate avail ab1e h.~ 1·n~+~1 ~~a· ~~ 'u•1~~-~ ~nd· 4 '"'-H -'H,• ,.\.# i ~· !. , l I t,. ::::> v v.. • • 12-62 I I I I I I I I I I I I I I I I ' •••••• .~.:. I .. I I I I I I I I ',' "- (c) Surge Chamber Gate ~rane A crane will be installed in the surge chamber for installation an<!, removal of the draft tube gates as well as the tailrace tunnel intake stoplogs .. The crane will either be a monorail. (or twin "monorail) crane, a top running crane~ or 'a gantry crane. For the· .pre1 iminary design, a twin mOnorail crane has been asstJned. The crane will be about 45 .tons in capacity, pendant operated, and will have a t\'MJ point lift.. A follower wi11 be used with the crane for handling the gates and stoplogs. The crane will nor- mally travel along the upstreMl side of the surge chamber; however, the crane runway will have a transfer mechanism for mov;.ng the crane to the downstre~ side of the surge chamber for install at ion or removal of the tailrace tunnel intake stop logs.. The crane runway will extend over the tailrace tunnel stoplog storage area at one end of the surge chamber. (d) Miscellaneous Cranes and Hoists In addition to the powerhouse cranes and surge chamber gate crane, the following cranes and hoists will be provided in the power plant: -A 5-ton monorail hoist in the transformer gallery for transformer main- tenance; -A 4-ton monorail hoist in the circuit breaker gallery for handling the main circuit breakers; -Small overhead jib or A-frame type hoists in the machine shop for hand- 1 ing material; and -A-frame or monorail hoists for handling miscellaneous small equipnent in the powerhouse. (e) Elevators Access and service elevators will be provided for the power plant as follows: -An access e 1 ev ator from contra 1 bui 1 dings to powerhouse; - A service elevator in the powerhouse service bay; and -Inspection hoists in the cable shafts. For preliminary design purposes, a 12,000-lb, double-deck elevator has been assumed for access to the powerhouse from the control building.. The ele- vator will be locat-ed in the access shaft and will trav.el at a speed of about 500 ft/min; it will be operated by a friction type hoist located above the elevator shaft.. The elevator will have a single landing at the control building plus four underground landings. ·The service elevator in the powerhouse service bay w·ill have a capacity of 2,000 to 4,000 pounds and will provide access to the various powerhouse floors. The elevate,-will travel at about 100 to 150 ft/min and will be operated either by a friction hoist or a hydraulic cylinder. I "Alimaku type rack and p1n1on man hoists have been assumed for the cable 1· sh~fts, to use in inspection and/or maintenance of the oil-filled cables ; . '· and control ca.bJes. The hoist would also provide emergency access from the power plant.. Each hoist will have a capacity of about 900 pounds and 1,,·~· t rave 1 at a speed of approx in!itte 1 y 130 ft/mi 11 • . (f) Power Plant Mechanical Service Systems. The mechanical service systems for the pOwer plant can be grouped cinto six major categories: -Station water systems; -Fi~e protection; -Ca~npressed air; -Oil storage and handling; -Drainage and dewatering; and -Heat;ing, ventilation and cooling. ( i) Stat ion Water Systems The station water systems will include the water intake, cooling water systems, turbine seal water systems, and domestic water sys- tems .. The water intakes will supply water for the various station water systems in addition to fire protection water. The water can be taken frooi the penstock; however, pressure-reducing valves will be· necessary because of the high pressure of the water {about 330 psig. maximum).. Alternatively, water can be supplied from the draft tube using pumps to provide suitable pressure. For preliminary design purposes, the 1 atter approach has been adopted with a water intake at each draft tube. The water will pass through an automatic back- wash strainer which will limit the maximum particle size in the water to about 1/16 inch. An interconnecting header will permit a strainer to be taken out of service without affecting operation of a generating unit.. Each strainer will be .sized to handle the water requirements for two units. On a unit basis, cooling water will be required for generator air coolers, turbine and generator bearing cooler~, transformers~ and powerhouse unit air coolers. The total cooling water requirements for each unit will be about 4,000 gpm. In addition, the compressed air systems in the service bay will require approximately 100 gpn of cooling water .. One cooling water pump will be provided per unit which will take water fram downstre.am from the water intake strainer. To ensure suit.able reliability, the cooling water pumps for two units will be interconnected, with each pump capable of handling the flow for both unitso Two cooling water pumps in the service bay will handle compressor cooling water requirements. The cooling water for each unit will discharge into the turbine draft tube, while the compressor cooling water will flow into the station drainage system~ 12-64 I I ;I r I. ., I I I I I I I I I •• I I I I I I I' I I I I I I I '~'· I I I I I Turbine seal water wi 11 be supplied to the seal on the main shaft and to the runner seals when the-unit is spinning in air(i .. e •. , in spinning reserve mode)~ Filtered water may or may not be required, depending on the type of shaft sea.l. If no filtration is needed, the seal water will be _taken directly from the high-pressure side-of the cooling water pumps. If filtration is necessary, a single · system will be provided for the powerhouse .. -The system will have two filters and two pumps which will take water from downstreClll from the water intake strainer and distribute the water to each unit via a looped header .. Domestic water wfll be req4ired for the washrooms, lunch rooms, drinking fountains, and a service sink and emergency eyewash in the battery room. Peak domestic water requirements are expected to be about 30 gpm. The system will-have two pumps and a hydropnellflatic tank Q Water wi 11 be taken from the water intake system and will be treated by chlorination or-other means as necessary. (ii) Fire Protection System The power plant fire protection system will consist of a fire pro- tection water system with fire hose stations located throughout the powerhouse and transformer gallery; sprinkler systems for the gener- ators, transformers, and the oi1 rooms; and portable fire extin- guishers located in strategic areas of the powerhouse and transfer- . mer gallery.. Carbon dioxide could be used in the generator r-ather than a sprinkler ~Jstem; however, the water system is re,commended because of the $a~ety hazard of C02· Fire protection water will be taken fran the station water intakes. Pressurized water will be provided by a pumped system with two main fire pumps as well as a jockey pump, or alternatively by a head tank with two supply pumps which keep the head tank full. For prelimin- ary design purposes, a system with a head tank has been selected because of the increased reliability of the system. With an under- ground powerhouse, a head tank can be provided quite easily at a suitable elevation as an ad it to the access shaft. The capacity of the head tank will be about 100,000 gallons; the tanks will have two compartments to permit draining of half the water for inspect ion and maintenance. For reliability, the water supply ptJnps will have two electrical power sources. Fire hose stations will be provided on all floors of the powerhouse, in the transformer gallery, and in the bus tunnels. Service water · outlets will be in.stalled at the various fire hose stations to supply water for washing downs floors or equipment. The sprinkler systems for generators, transformers, and oil rooms will be the dry deluge type, operated by a solenoid valve which in turn will be activated by detectors in the respective area. \) ll-65 The portable fire extinquishers will generall.Y be carbon dioxide or ,a dry chemical type. · (iii) Compressed Air Systems " Compressed, air will be required in the powerhous·e for the following: -Service air·; -Instrt~nent air; -Generator brakes; . -Draft tube water leve-1 depression; -Air blast circuit bra.:.:.~a~rs; and -Governor accumulator tanks. For the preliminary design, two compressed air systems have been assuned: a 100-psig air system for service air~ brake air, and air for draft tube water level depression; and a 1,000-psig high- pressure air system for governor air and circuit breaker air. For detailed plant design, a separate governor air system and circuit- breaker air system may be provided. The service air systems will have three aiv-compressors of the ro- tary screw or reci procafing type, each with a capacity of about 200 cfm. The sY5~em will have four air receivers, two with approxi- mately 800 ft capacity used for the draft tube water level de- press ion system, and two with approximately 150 ft3 used for ser- vice and brake air. The system will be designed to give priority to the brake air system!! Service air piping with air hose stations will be located on all floors of the powerhouse and in the trans- former gallery. The high-pressure governor/circuit-breaker air system will have three reciprocating air compressors with approximately 30-cfm capa- city each, and three small air receivers •. The governor air system wi 11 supply air for initial filling of the governor system accumul a- tor tanks and for makeup air to replace air lost through leakage and air dissolved in the governor system oil. The circuit breaker air system will provide compressed air for oper- ation of the main breakers. To insure dry air for the breakers, the air will be stored at 1,000 psig and then reduced to about 350 psig for operation of the breakers. Instrtment air will also be taken from the high-pressure air system. ( iv) Oil Storage and Handlin.9 I I ·1,,;· .. ' ;:' , .. f 'I I I I I I I I I I I ... . ..~~- ' -----·-----...-;10'·- Facilities will be provided for replacing oil in the transformers and for topping-up or replacing oil in the turbine and generator bearings and the governor pumping system.. For preliminary design purposes, two oil rooms have been assumed, one in the transformer gallery and one in the powerhouse service bay~ 12-66 1 I I •• I I I I I I I ,f I I I I '"· I t ,, I "· I The transformer gallery will have two oil storage tanks, one for filtered oil and the other for unfiltered oil.. Each tank will have a capacity at least equal~""to the volume of oil in one transformer (about 8,000 gallons). · A header with valve stat ions at each trans- former will be used for tr·ansferring oil to and from the transfer- . mers. Oil ~·fill be transferred by a portable pomp and filter unit. A similar system will be provided in the powerhouse with a filtered and unfiltered oil tank and distribution header with valve stations at each unit. The oil tank capacity will be equal to the total oil vo'il.l!le for one unit (about 3,000 gallons). During the detailed design stages, consideration should be given to the use of mobile oi.l tanks located in a parking area near the po~erhouse and transformer gallery, near the access iunnel. (v) Drainage and Dewatering Systems The drainage and dewatering systems will consist of: -A unit dewatering and filling system; - A clear water discharge system; and - A sanitary drainage system. The dewatering and filling systems will consist of two sumps each with two dewatering pumps and associated piping and valves fran each of the units. To prevent st-ation flooding, the sump will be de- signed to withstand maximun taiiwater pressure. For preliminary de- sign purposes, submersible dewatering pumps have been assumed.. Ver- tical turbine type pumps can also be considered; however, since the dewatering system acts as an emergency drainage system, the pllllp columns would have to be extended so that the motors are above maxi- mLm tailwater level. Another optinn is turbine-driven pumps, but these are generally very costly. A valved draft tube drain 1 ine will connect to a dewatering header running along the dewatering gallery. The spiral case ·will be drained by a valved 1 ine connect- ing the spiral case to the draft tube. Suitable provisions will be necessary to insure that the spiral case drain valve is not open when the spiral case is pressurized to headwater level. The de- watering pllllp discharge 1 ine will discharge· water into the surge chamber. The general praced ure for dewatering the unit wi 11 be .. ' close the intake gate, drain the penstock to tail water 1 evel thrc. ~~h the unit, then open the draft' tube and spiral case drains to dewater the unit. Unless the drainage gallery is below the b.Jttan of the draft tube elbow, it will not be possible to completely unwater the draft tube through the unwatering header. If necessary, the remain- der of the draft tube can be· unwatered using a submersible pump low- ered through the draft tube access door. Unit f i 11 i ng to t a i lwater 1 eve 1 wi 11 be accomp 1 ished from the surge chamber through the dewatering pump discharge 1 ine (with a bypass around the pumps) and ~then through the draft tube and spiral case drain 1 ines. Alterna- tively, the unit can be filled to tailwater level through the draft tube drain line from an adjacent unit. Filling the unit to head- water pressure will be-accomplished by 11 Cracking" the intake gate and raising it about 2 to 4 inches. The clearwater drainage _system wi)l handle normal drainagec-into~-the po~er plant .. · Drainage wi11 be collected by a network of floor drains, trench drains, pressure relief drains, and equipment drains which discharge into gravity drainage sumps where items are pumped to the surge chamber. The station will have three main sumps, two in the powerhouse adjacent to the dewatering sumps and one in the transfonner gallery. Smaller sumps will be located in appropriate areas such as the elevator pits and the upstream drainage gallery .. The sumps in the powerhouse will have submersible pumps for the same reasons as discussed above for the dewatering system. The transfor• mer gallery will have vertical turbine type p1.111ps.. The drainage sumps in the powerhouse will have an overflow line which will dis- charge water· into the adjacent dewatering sump should inflow into the drainage sumps exceed the capacity of the drainage pumps. The overflow 1 ine wi 11 have a flap valve to prevent reverse flow from the de~atering sump. Particular care will be taken to prevent accidental oil spills from being discharged into the powerhouse. The following provisions will be made: -All three mair~ sumps will have oil contamination detectors to ob- tain the pressure of oil in the sumps; -Drainage into the sumps will_ first pass through an oil separator; -Controls for the drainage pumps into the transformer gallery will be interlocked with the tr.ansformer fire protection sprinkler sys• tern. Activation of the sprinklers, which signifies a transformer fire. and the possibility of a major ail spill, will prevent the drainage pumps from starting until the drainage sump is almost full. It wi 11 be pass ib le to retain about 40,000 gallons of oil I water in the sump before the pump start (each transformer holds about 8,000 gallons of oil). In this manner~ it will be possible to retain a large anount of oil ~~:~ the sllflp where it may be skimmed off; and -Suitable oil retention curbs will be provided in the oil rooms .. Sanitary drainage fran the washroomslt lunch room, and drinking foun- tains wi 11 drain to a packaged sewage treatmen-t; plant and then will be. discharged into the surge chamber via sewage lift pumps. 12-68 I I 1~7:" ,, .-I I I I ....... I I I "'··· I I I I -~ I I I . "" I '' ' •• I I I' '• I '>t<"""'' I I ~...., I ·~ ....•. : ' I I l I l I I. ' I I I (vi) Heating, Ventilation and Cooling The heating and ventilation_ system far the underground· power plant will be designed primarily to maintain suitable temperatures for equipment operation and to provide a safe and -comfortable atmosphere for operating and .maintenance personne.l.. Air wf11 be drawn into the power facilities through one or mote shafts or tunnels, circulated throughout the power plant, and discharged from the power plant through other shafts and tunnels. For preliminary design purposes -it has been asstJned that air wi11 be drawn down the access and the cable shafts,. and discharged out through the access tunnel; however, -the actual arrangement will depend upon the final design. The power plant will be located in mass rock which has a constant year around temperature of about 40•F.. Considering heat given off from the generators and other equipment, the primary requirement will be for air cooling.. Initially, some heating will be required to offset the heat loss to the rock, but after the first few years of operation an e{luil ibrium will be reached with a powerhouse rock surface temperature of about 60 to 7o·F. Air cooling will be accomplished by providing suitable air changes incorporating cooling coi 1 s in the air circulation system-Cooling water from the station service water supply will be circulated through the cooling. coils.. In winter, some heating may be required to moderate the temperature of the incoming air into the power plant. Allowance must be made in the design for the possibility · that large quantities of air (up to about 6,000 cfm per unit) may be required for turbine aeration. Other factors which· must be considered or incorporated in the design are: -To prevent or minimize the circulation of combustion products in the event of a fire, powerhouse ventilation should be sepm-ate from transformer gallery ventilation and provision should be made for i so 1 at i ng the two areas; and -Suitable air locks will be necessary to preclude adverse chimney effects in the shafts. (g) Surface Facilities Mechanical Servi~e Systems The mechanical. services at the control building on the surface will include: - A heating, ventilation, and air conditioning syst1.!m for the control room; -Domestic water and washroom facilities; and - A halon type fire protection system for the control rc,!Jm. 12-69 Domestic water will be supp 1 i ed from the powerhouse domestic water system, with pumps located in the .powerhouse and piping up through the access · shaft. St1nitary drainage from the cont.rol building will drain to the sewage treatment plaat in the powerhouse through piping in the access tunne]. The standby generator building will have the following services: - A heating and vent i 1 at ion system; -A fuel oil system with buried fuel oil storage tanks outside the bu.ilding, and transfer pumps and a day tank within the building; and - A fire protection system of the carbon dioxide or halon type. {h) _[achine Shop Facilities A machine shop and tool room will be located in the powerhouse service bay area with sufficient equipment to take care of all normal maintenance work at the plant, as well as machine shop work for the 1 arger components at De~il Canyon. For preliminary design purposes, an area of about 1,500 ft (las been allocated for the machine shop and tool room. The actual · equipment to be installed in the machine shop will be decided during the design stages of the project.; however, it will generally include drill presses·, 1 athes, a hydraulic press, power hacksaw, shaper, and grinders .. 12.18 -Accessory Electrical Equipment The accessory electrical equipment described in this section includes the following: .. Main generator step-up ~5/345 kV transformers; . Isolated phase bus connecting the generator and transformers; .. Generator-circuit breakers; · • 345 kV oil .. fi lled cables from the transformer terminals to the .. Control systems of the entire hydro plant complex; and .. Station service auxiliary AC and DC systems. switchyard; Other equipment and systems described include grounding; 1 ight ing system, and communications. The main equipment and connections in the power plant are shown in the single 1 tne diagram, Plate 60A. The arrangement of equipment in the powerhouse, transformer gallery, and cable shafts is shown on Plates 57 through 59. (a) Selection of Transformers and H.V. Connections I I I . ¥,;;l . .,_ .. _ ~ .. -'"' I I . . --..,;,:~.- 1 ·-.. ,~ I I ~-.. ~·-"·· " •. ·-· I I t.,.- 1 .., .... ' . I ... ,.. .... ~ ... ~~-..._. ... --..,..._\ - (i) General Nine single-phase transformers and one spare transformer will be lo- cated in the transformer gallery. Each bank of three~single-phase 12-70 t I I I ' I I I I ....,_ ' I .._,, r I t I-I ·' ,, ; ,. I ~ I· ......,._. I ..... _ I I - '< I I ..., -· tran.sformers will be connected to two generators through·generator circuit breakers by 1solated ph.ase bus located in--individual bus tunnels~ The HV termtnals of the transformer will be connected to the 345 kV switchyard by 345 kV single-phase oil-filled cable installed in 700-foot.-long vertical shafts. There will be two seats of three single-phase 345 kV oil-filled cables installed in each cable shaft.. One set wi 11 be maint:ained as a spare three phase cable circuit in the second cable shaft. These cab1e shafts will -also contain the control and power cables between the powerhouse and the surface control roorn, as well as emergency power cables from the diesel generators at the surface to the underground facirities .. A nllllber of considerations led to the choice of the above optimum system of transformation and connections. Different alternative methods and equipnent designs were also considered. In summary, these are: -One transformer per generator vs one transformer for two gener- ators; -Underground transformers vs surfa~e transfomcrs; Direct transform·ation from generator voltage to 345 kV vs inter- mediate .step transformation to 230 kV or 161 kV~ and then to 345 kV; -Single-phase vs three-phase transformers for each alternative method considered; and -Oil-filled cable vs solid dielectric cable for SF6 gas-insulated bus. ( i i) Re 1 i abi 1 ity Consider at i_OJ'!! Reliability considerations will be ~ased on the general reliability requirements for generation and transmission described in Section 15 regarding the forced outage of a single generator, transformer, bus or cable in addition to planned or scheduled outages in a single contingency situation, or a subsequent outage of equipment in the double contingency si·tuation.. The system should be capable of re- adjustment after the outage for loading within normal ratings and for loading within emergency ratings. The generators will be rated with a 115 percent continuous overload capability. All main connections and equipment including the trans- formers, circuit br.eakers, isolated phase bus, and 345 kV cables will be rated for continuous operation at the 115 percent overload rat tng of the generators .. Emergency ratings are different for different items of ·equipment and emergency periods. It generally varies between 110 to 130 percent 12-71 in s~.m~~er to 120 to 140 percent in winter for a 4 to 12 hour ,period-, with somewhat higher values for very short (1 hour) emerg~!l~JI . periods. (iii) Technical and Economic Considerations The use of surface transformers connected directly to· the under- ground generators by i so 1 ated phase bus was ru1 ed out at the outset due to significantly higher costs and higher losses associated with generator isolated phase buses. The incremental cost could be de .. creased if three units were· connected to one-transformer, but such a compromise is not acceptable due to reliability considerations. In general; 3-phase transformers are preferred to single-phase transformers because of their lower overall costs, smaller overall dimensions ·and smaller underground gallery dimensions.. However, transport limitations seriously affect the use of the larger size 3-phase transformers, both in dimensions and weight. The following are the road and rail data avail able: -Parks and Denali Highways Maximum load -150,000 lb Overweight's require special permit~~. -Railway Maximum Weight -263,000 lb Dimension Limits -16 feet high, 10 feet wide A further check of these design limitations for the selected sizes of transformers is recofl111ended during the detailed design stage.. A careful route reconnaissance study is also required. Single-phase transformers are therefore recoomended for the 6-unit power plant. The grouped unit arrangement with two generators per transformer will allow a smaller gallery length, with center-to- center spacing comparable to the generator spacing. The grouped unit arrang.ement is the recomnended arrangement. The alternative with one transformer per generator will requie a gallery about 300 feet longero One distinct advantage of single-phase transformers is that a spare transformer can be provided at a fairly low incremental cost .. The double-step transformation scheme (15/161 KV generator- transformer, 161 I<V cable and 161/345. KV auto-transformer at the switchyard) is economica111y competitive with the direct transforma- tion scheme· (15/345 KV), resulting from a number of tradeoffs: cost/MVA per transformer is lower; also dimensions, weights and cav- ern dimensions are lower; but the intermediate-voltage transformer costs are additional. 12-72 ~~-.. _, ,. ·. I f, ' I I I I I I ""~r I 'I ~ I :,..../ I I ,....._ .t I 'i ·8' I . ' '' '' ~ .; ·"-·~·· .".k._ ... ,.-..~~,. ....... ·:~~~,.-.._.:v--~~.i&H'"-""""h;·~-b~~~-~a'~''$tl~ ' .'-"! , .. I '.;' I I '-r· I I; "-.··} t ' ,. I '-'" I I '\.<-' ' I ' I I I (c) Direct transformation {15/345 KV) is. better from system transient stability viewpoint since the overall Jmpedance of the generator unit to the 345 KV bus is lower. Furthermore, it has a better over~· all rel iallility since there is: no one less voltage level and, there- fore, less equfpment in the generating nchain" of equipment. This scheme costs about $2 mill ion less in over.a11 costs compared to the double .. step transformation scheme .. The comp.arison between 345 KV oil-filled cab1es and other .345 KV cable and bus system is made in Section 12.18. The SF6 bus is about 5, to 6 times the cast of the oil-filled cables. It also requires a larger diameter cable shaft. The oil-filled cable is well proven at a nliTiber of underground power installations and was therefore sel- ected for both technical and economic considerations. Main Transformers (i) Rating and Characteristics The nine single-phase transformers (three transformers per group of two generators) and one spare transformer, will be of the two wind- ing, oil-immersed, forced-oil water-cooled-{FOWT-type, with rating and electric characteristics as follows: Rated capacity: High voltage winding: Basin insulation level (BIL) of H .. V .. winding: Low voltage winding: Transformer impedance: 145 MVA 345 I 3 kV~ Grounded Y 1300 kV 15 kV, Delta 15 percent The temperature rise above air ambient ·temperature of 40~C is 55°C for the windings for continuous operation at the rated kVA .. (ii) Construction The transformers will be of the FOW type with water-cooled heat ex- changers which remove the heat from the oil circulating through the windings. A one-third spare cooler capacity will be provided.. The transformer will be of the forced oil directed type with a design aimed to achieve minimum d-imensions and weight for shipping pur- poses. The low voltage terminals will be connected to the isolated phase bus,. and the high voltage terminals to the 345 kV oil-filled cable box termination at the transformer. Lightning arresters will be connected directly to the high voltage terminals. The transfonner installation in the gallery will be de- signed to provide the necessary ground and safety clearances from the· live 345 kV terminals to all nearby equipment and structures. 12-73 (d) The tank underbase wi 11 be provided with. flanged wheels for trans- port o.n rails •. The spare single-phase transformer wfll be exactly identical to the remaining ·nine single-phase transformers .. · It will .be maintained tn a state of .maximllll readiness, for connect ion in the shortest practical time to replace any of the main transformers .. Tha ttansformers wi 11 be ful'ly tested· andi}inspected. in the factory according to ANSI/NEMA StandardS~ -They will be shipped without. oil and filled with inert gas for protection. At the site, erection would be mainly for external fittings such as bushings, 1 ightn ing ar~esters, heat. exchangers, piping, and electrical connections. {iii) Fire Protection Fire walls will separate each single-phase transformer.. Each trans- former will be provided with fog-spray water fire protection equip- ment, automatically operated from heat detector.s located on the transformer. Generator Isolated Phase Bus. ( i) Ratings and Characteristics The iso 1 ated phase bus main connections wi 11 be 1 ocated between the generator, generator circuit breaker, and the transformer. Tap-off connections will be made to the surge protection and poten- tial transformer cubi.cle, excitation transformers~. and station ser- vice transformers. Bus duct ratings are as follows: Rated current, amps Short circuit current mcn11entary, Clllps Short circuit current, S)«'dlletrical, amps Basic insulation level, kV (BIL) Generator Connection 9,000 240,000 150,000 150 Transformer Connection 18~000 240,000 150~000 150 The bus conductors will be designed for a temperature rise of 65°C above 40°C ambient temperature. (ii) Construction The bus will be of standard self-cooled design with conductor and tubular enclosure of aluminum. The c~rrent r?ating is such that either a self-cooled or forced cooled design will be possible. With a forced cooled design, the size and costs will be lower; however, · if the forced-cooling plant fails, the bus would be severely derated to a rating 1 ess than 50 percent of the forced cooling rating .. · The self-cooled designs are used up· t.o 30,000 amps rated current and are therefore recommended for this installation where the ratings will not exceed 18,000 amps. •. ,, 12-74 ' I I ....... I I ----~c I 1 -·., ·' I "'"' ' I .... ~~ ' I ..... ' I .... .. I 1'· ... ... I ''-" I ' __ ,-'. -- I I 1- . I~ ._., ., ' ,, ' t ' I ;...Jf ' I .. ,.. l ' _ The enclosure will be of welded construction and each bus will -be 'grounded. The constructiQl1 Js_highly reliable; will eliminate phase-to-ph-ase faults,-neutralize -the magnetic field outside the enclosure; and provide protection against contamination and moi s- ture, with consequent minimum maintenance requirements .. (e) Generator Circuit Breakers (f) The generator circuit breakers will be of the enclosed air circuit breaker design suitable far mounting in line with the generator isolated phase bus ducts. They are rated as follows: Rated Current: Voltage: Breaking capacity, symmetrical, amps 9,ooo Amps 23 kV class, 3-phase, 60 Hertz 150,000 The short circuit rating is tentative and will depend on detailed analysis in the design stage. - The breakers will be designed and constructed with a high degree of rel ia- bil ity. The phase spacing of the breakers will be generally the same as the isolated phase bus duct. The breakers will be mounted on strong foun- dations on the generator fl oar designed to absorb the reaction forces when the breaker operates. A separate compressed air plant will be provided for the high rel i ~bi 1 ity compressed air system requirements of the air circuit breakers. 345 kV Oil-Filled Cable ( i) General The recommended 345 kV connection is a 345 kV oil-filled cable. sys- tem bet·ween the high volta.ge terminals of the transformer and the surface switchyard .. The cable will be installed in a vertical cable shaft. Cables fran two transformers will be installed in a single cable shaft.. · This system of 345 kV connection was chosen after a technical and economic analysis of alternative methods of connection~ including: -SF6 isolated bus system; -High pressure oil pipe cable system; and -Solid dielectric cable system. The SF6 bus system is considered to be the best alternative to the oil-filled cable system. Its advantages are a generally better overall reliability, including a low fire nazard. However, it costs approximately 5 to 6 times that of th.e oil-filled cable install a- t ion, and ··requires almost twice. the dianeter cable shaft of the cable installation. The overall cost di·fference is approximately- $7,000,000 in direct costs. (g.) The oil pipe cable will consist of three conductors contained within an ci,l-fi'lled steel pipe~ · This system has the highest potential fire hazard of all the cable systems and is not recommended fo.r high head vertical cable instal 1 at ions. The solid dielectr·ic (pol}flleric} c~bles are still under development at the 345 kV to 500 kV voltage class. ~ . . It is reconmended that further detailed. study of the oil-filled cable in comparison with the SF6 bus and other more recent SF6 cable designs under developnent be undertaken at the design stage .. By far the greatest number of high voltage, higtT capacity install a,-. tions utilize oil-filled cables. A formidable experience recor·d·ts evident for the oil-filled cable installations associated with large power plants all over· the world. Typical installations include the 525 kV/650 MVA units at Grand Coulee III, the 345 kV/550 MVA units at Churchill Falls in Canada, the 400 kV/2640 MVA cables at Severn River crossing in Great Britain, and the 400 kV /2340 MVA cables at Di"norwic pumped storage plant in Great Britain .. (i i) Rating and Characteristics The cable will be rated for a continuou~ maximum current of 800 amps at 345 kV +5 percent. The -max imiJTl conductor temperature at the max- imum rating will be 1o•c over a maximum ambient of 35°C. This rat- ing will correspond to 115 percent of the generator overload rating .. The normal operating rating of the cable will be 87 percent, with r. corresponding lower conductor temperature which will improve the overall performance and lower cable aging over its project operating life. Depending on the ambient air temperature, a further overload emergency rating ·of about 10 to 20 percent will be available during winter conditions~ The cables will be of single-core construction with oit flow through a central oil duct within the copper conductor. Cables will have an al uminun sheath_ and PVC over sheath. No cable jointing wi 11 be re- quired for the 700 to 800 feet length cable installation. Control Systems ( i) Genera 1 A Susitna Area Control Center will be located at Watana to control both the Watana and the Devil Canyon power plants as shown in Plate • The control center will be linked through the supervisory -sy-s~t-em to the Central Dispatch Control Center at Willow'as described i~ Section 14. The supervisory control of the entire Alaska Railbelt system will be done at the Central Dispatch Center at Willow~ A high level of con- trol automat·lon with the aid of digital computers will be sought, 12-76 I I ' ,I I ...... I I -·1 .,-..,...-. t "s~.: 0 • I ' t ' I ' I ' • . w I ' I I I I ' ' ' I I ,, I I .. ' I a ' a ,, but not a complete computerized direct digital control of the ,watana and Devil Canyon power plants.. Independent operator controlled loca:l-manual and local-auto operations will still be possible at Watana and Devil Canym1 power plants for tes-ting/commissioning or during emergencies. The control system will be designed to perform the fo llo,wing functions at both power plants: -Start/stop and loading of units by operator; -Load-frequency control of units; -Re~ervoir/water flow control; -Continuous monitoring and -data logging; -Alarm ann unci at ion; and -Man-machine communication through visual display units (VOU) and console. In addition, the computer system will be capable of retrieval of technical data, design criteria, equipment characteristics and oper- ating limitations, schematic diagrams, and operating/maintenance records of the unit. The Susitna Area Control Center will be capable of completely inde- pendent control of the Central Dispatch Center in case of system emergencies. Similarly it will be possible to operate the Sus_itna units in an emergency situation from the Centra 1 Dispatch Center,. although this should be an unlikely operation considering the size, complexity, and impact of the Susitna generating plants on the sys- tem. The Watana and Devil Canyon plants will be capable of 11 black start" operation in the event of a complete b 1 ack out or co 11 apse of the power system. The control systems of the two plants and the Susitna Area Control Center complex wirJ be supplied by a non-interruptible power supply. (ii) Unit Control System The unit control system will permit the operator to initiate an en- tire sequence of· actions by pushing one button at the control con- sole, provided all preliminary plant conditions have been first checked by the operator, and system security and unit commitment have been cleared through the central dispatch control supervisor. Unit control will be designed to: -Start a unit and synchronize it with the system; -Load the unit; -Stop a unit; -Operate a unit as running spare (runner in air with water blown· down in turbine and draft tube); and -Operate as a synchronous condenser (runner in air as above) .. Unit control will be essentially possible at four different levels in a hierarchical organization cf the control system: (iii) .... -Local control at the machine floor at individual turbine-generator control boards: (primarily designed for coomissianing and recommis- sioning of units}. It will be the responsibility bf the operator for performing individual control operations in~ the correct se- quence, and monitoring instrumentation during Tocal control opera- tions. · · -Automatic or semi-automatic system for start-up and shut ... down of generating unit at the local board at the machine floor. -.Fully automatic system at Susitna Area Control (at Watana) for Watana and Devil Canyon power plants.. (This will be the normal Susitna operation.) -Fully automatic system through supervisory control from Central Dispatch tenter at Willow. (Abnormal or emergency situations only). Computer-Aided Control System Traditionally, control systems for power plants in general, and hydro plants in particular, have utilized hard-wired switchboard type equipment (such as electro-mechanical relays, instruments, alarm annunciators~ signal lamps, mimic diagrcm and control swit- ches) for the operation, indication, alarm and control of the power plant. Such equ·i(lllent was installed both at the plant local control area on the machine floor as well as in the control room, with a limited degree of miniaturization of equipment at the control desks in the control room. Whfle traditional switchboard type equi1J11ent is still utilized at the local control level, supplemented with progranmable control sys- tems at many plants, the design of control and display equipment at modern central control rooms has been rapidly moving towards computer-aided or fully computer.acontrolled systems, especially where remote control operations are contemplated. One of the prob- lems encountered by utilities is the necessity for operating person- nel familiar with the conventional control systems to adapt to the new computer-aided control systems·. In this contect, establishing a mojern computer-aided control systan in the Alaska Power Authority electrical system for the Susitna Project complex should not pose any special problems for the adapt ion and training of -operators. The computer-aided control system at the Susitna Area Control Center at Watana will provide for the following: ·-Data acquisition and monitoring of unit (MW, MVAR, speed, gate· position, temperaturess etc .. ); -Data acquisition and monitoring of reservoir headwater and tail- water levels; 12-78 I . '· ',' ' I I I I .I ' ' I . ,, ' t ' I I t ,..,.. I I . I . ' ' ···. ,f. ' I f I I e a: ·I· ,. ' ' t ; . , I I· I t I ., ... . . .. .. Data acquisition and monitoring of electrical system voltage and fre_quency; -Load-frequency control; -unit start/stop corytrol; -Unit 1 o ad in g; -Plant operation alarm and trip conditions (audible and vi sua 1 alarm on control board, full alarm details on VDU on demand); -General visual plant operation status on VDU and on giant wall mimic diagram; -Data logging, plant operation records;: -Plant abnormal operation or disturbance automatic recording; and .. Water management (reservoir control). The block diagrCJn of the computer-aided control system is shown in Plate . The supervisory control and telemetering system and central dispatch center system details are described in Section 14 .. (iv) Local Control and Relay Boards Local boards will be provided at the powerhouse floor equipped with local controls, alarms, and indications for all unit control func- tions •. These boards will be located near each unit and will be utilized mainly during testing,. conmiss ion ing, and maintenance of the turbines and generators. It wi 11 also be uti 1 ized as needed during emergencies if there is a total failure of the remote or computer-aided control systems. The unit electrical protective relays will be mounted on relay boards, with one board for each generator located near the unit. Differential protection will be provided for each generator and transformer. The differential .zones of protection overlap will in- clude all electrical equipment and connections.. The 345 kV oil- filled cable. to the surface switchyard will be protected by a p·ilot- wire differential protection·relay. The overall differential relay protects the generators. transformers, and 345 kV cable. Sensitive ground fault stator, protection will be provided for the genetator. Protection will also be provided for negative phase sequence opera- tion, loss of excitation, ov.ervoltage, and under frequency. A phase impedance relay will provide backup protection for the generator. Other protective relays are shown in Plate .. (h) {v) Load .. 'Freguency Control (Automatic ~enet·ation Control) The load frequency control system will provide remote control of the output of the generator at Watana and Devil Canyon from the -central d~ispatch control center through the supervisory and computer-aided control system at Wat~ana: The basic method .of automatic generation control (AGC) will use the plant error (differential) signals from the load dispatch center and will allocate these errors to tne power plant generators automatically through speed-level motors. Provis- ion will be made in the control system for the more advanced scheme of a closed-loop control :;ystem with digital control to control gen- erator power. The control system will be designed to take into account the digital nature of the controller-timed pulses as well as the inherent time delays caused by the speed-leve1 motor run-up and turbine-generator time-constants. The load set-point for the Susitna area generation wi 11 be set at the Central Dispatch Center.. The sumnated power will be telemetered from the Susitna Area Control center to the Central Dispatch Center, from which the required differential plant generation (15 error11 ) will be determined and transmitted by the supervisory system to Susitna Area Control Center. From this point, the remaining functions for the automatic ~eneration t.ontrol will be carried out by the plant supervisory control systems to load the individual generating_ units at Watana and Devil Canyon .. The unit will be automatically removed from load-frequency control for various conditions including failure of supervisory system, unit controller or computer system, abnormally high plant frequency, unit shut-down, and de power failure. When the unit is taken off au~o­ matic lo.ad-frequency control, it will be returned to manual load and frequency control by the operator at Watana Control room .. Station Service Auxiliary AC and DC Systems ( i) Auxi 1 i ary AC System The station service system will be designed to achieve a reliable and economic distr-ibution system for the power plant and switchyard, in order to satisfy the following requirments: -Station service power at 480 volts will be obtained fran two 2,000 kVA auxiliary transformers connected directly to the generator circuit breaker outgoing leads of Units 1 and 3; -Surface auxiliary power at 34.5 kV will be supplied by two sep- arate 7 ~5/10 MVA transformers connected to the generator leads of Units l and 3; 12-80 ' t ·I ' 11 ,, ' ' a I ~ I. ' ,- t t 8 I ' I ' t ~·' I , ' ,, '; ' ,, '\ J I ' t ' t I I . , I· ... Station service power will be maintained even when all the units are shut down and the generator circuit breakers are open; -100 percent standby transformer capacity will be avail able; - A spare auxiliary transformer wi1l be maintained, co·nnected to Unit 5; and ... 11 81 ack start" capability will be provided for the power plant in the event of total failure of the auxiliary supply system, 500 kW emergency diesel generators will be automatically started up to supply the power plant and switchyard with auxiliary power to the essential services to enable startup of the generators .. The main ac auxiliary switchboard will be provided with two bus sec- tions separ·ated by bus-tie circuit breakers. Under normal operating conditions, the station-service load is divided and connected to each of the two end incoming transformers. In the event of failure of one end supply, the tie breakers will close automatically.. If both end supplies fai 1, the energency diesel generator will be auto- matically connected to the station service bus. Each unit will be provided with a unit ~uxil iary board supplied by separate feeders from the two bus sections of the main switchboard interlocked to prevent pa.ra.llel operation. Separate ac switchboards will furnish the auxiliary power to essential and general services in the power plant. The unit auxiliary board will supply the auxi1 iaries necessary for starting, running, and stopping the generator...:turbine unitJ These supplies will include those to the governor and oil pressute system, bearing oil pumps, cooling pumps and fans, generator circuit break ... er, excitation system, and miscellaneous pumps and devices cannected with unit operation .. The station essential service supplies will include powerhouse sump pumps, drainage pumps, compressors for circuit breakers~ station air and generator brakes, de battery chargers, control and metering de- vices, communications, fire protection pumps, and other miscellan- eous. essential power requirements. The station general supplies w'll1 include powerhouse. lighting, heat- ing~ ventil xting and air-conditioninp, elevators, cranes, machine shop and trJols, and othar miscellaneous pumps and general requi$'e- men~s .. The 34.5 kv supply to the surface fac'ilities will be distributed from a 34'e 5 kV ~:~itchboard located in the surface contt~,l and admi n- istration bu11d 1 r.~. Power supplies to the switchyard power intake:t and spi 11way as well as the· 1 ight ing systems for the access roads and tunnels will be obtai.ned from the 34.5 kV switchboard • 12-81 The unit auxiliary board wfll supply the auxiliaries necessary for ,. starting, running, and stopping the generator-turbine unit. These sup·p 1 ies wi 11 include those to·. the governor a.nd ail pressure system, bearing oil_ pumps, cooling water pumps -and fans, generator circuit breaker~ excitation ·system, and miscellaneous pumps and devices con• nected with unit operation. The station essential service supplies will include powerhouse sump pumps, drainage pumps, compressors for circuit· breaker, air and gen- erator brakes, de batte,ry chargers, control and metering devices, conununications, fire protection pumps, and other miscellaneous essential power requirements. The station general supplies will include powerhouse lighting, heat- ing, ventilating and air-conditioning, elevators, cranes, machine shop and tools, and other miscellaneous pumps and general require- msnts. The 34.5 kV supply to the surface facilities will be distributed from a 34.5 kV switchboard located in the surfa.ce ~ontrol and admin- istration building.. Power supplies to the swi~~J~,;r.ard power intake~ · ar..d spillway as well as the lighting systems for the access roads and tunnels '#ill be obtained from the 34.5 kV switcnboard~ The two 2000 kVA, 15000/480 volt. stations service transformers and thle spare transformer will be of the 3-phases dry-type, sealed gas- filled design. The two 7.5/10 MVA, 15/34.5 kV transformers will be of the 3-phase oil-immersed OA/FA type. - Emergency diesel. generators, each rated 500 kW, will separately sup- ply the 480 volt and 34.5 kV auxil ia.:--y switchboards during emergen- cies. Both diesel generators will be located ·;n the surface control building. An uninteruptible high security.power supply will be provided for the computer control system. (ii) DC Auxiliary st·ation Service System The de auxiliary system will supply the protective relaying, super- visory, alarm, contr.ol, tripping and indication circuit in the power plant. The generator"' static excitation system will be started with . "flashing•i power from the de battery. It will also supply the emergency 1 ighting system at critical plant 'locations. Separate duplicate lead-acid batteries for 125 volt de will be pro- vided in the powerhouse. The 48 volt battery supply for the super~ visory and .computer aided control system and microwave communica- tions will be located in the surface control building. 12-82 I I I t ,I I I I I -' 'I ' I ' ' I I ' ' ' .I ' I I I t I· ,. ' t t ' I I t I I The main battery system will be supplied by double charging equip- ment consisting of a full wave rectifier system with regula:ted out- put va:ltage which normally will supply the continuous de lo~Q io the system. The battery capacity will be suitable for an emergency loading based on a fai 1 ure of ac station service 1 ast i ng 5 hours .. (iii) .:J31ack Start 11 Capabilit,y The Watana power plant will have a built-in capability of starting up a completely blacked-out power system in a very short time. Only a, few basic requirements will have to be satisfied: -Sufficient water will ·be available in the reservoir for the mini- mum generation required for "black start" operation; -The governor oil system will have sufficient stored energy capable of operating the turbine wicket gates to full open position; -The generators wille be equipped with static exciters capable of being flash-started from the stat ion battery system -De control power will be available for the startup circuits. -The above described emergency power requirements will not exceed about 200 kW for one unit and wi 11 be easily supplied from the emer- gency dies~l generator. With the startup of a single unit~ the com- plete power plant and switchyard auxiliary power will be immediately avail able, enabling a11 the units in the power plant to be started up sequentially within the hour. (i) Grounding System The power plant grounding system will cons it of one mat under the power plant, one mat under the transformer gallery, risers~ and connection ground wires. Grounding grids will also be included in each powerhouse floor .. The power plant grounding system will be co~nected to the swi tchyard grounding system by three 500 MCM copper ground conductors to minimize the overall resistance to ground.. The grounding system will be designed to provide. a ground resistance of 1 ohm or lower.: All exposed metal part and neutral connections of generators and transformers will be ~onnected to the grounding system for the purpose of protecting personnel and equipment. from injury or damage. · (j) Lighting System The lighting system in the powerhouse will be supplied from 480/208-120 volts lighting transformers connected to the general ac auxiliary station service system. The lighting system will be all fluorescent-and incandes- cent fixtures operating on 120 volts and all outdoor type high pressure sodium fixtures operating on 208 volts. The lighting level varies gen- erally from 20 to 50 foot candles depending upon the powerhouse area; the \ ...... ·--· . \)' . I , I I ' (k) --.(1}- ·nigher levels will be.at control areas. Adequate illumination will be provided on vertical switchboards with local 1 ighting canopies. An emergency lighting system will be provided at. the power plant and at the control room at all critical operating ·locations with an illumination level of 2 f\lOt candles. The emergency 1 ighting system will operate from a sep- arate 120 volt ac circuit whic.h, by means of automatic transfer switches, will be automatically connected to the 125 volt de system upon failure of the a~ system. · Co111nun i cat ions The power plant will be furnished with an internal communications system, including an automatic telephone switchboard system. A communication sys- tem will be provided at a11· powerhouse floors and galleries, transformer gallery, access tunnels and cable shafts, and structures at the intake, draft tube chamber, spillway, and darn. The convnunicat ions system for the central dispatch control system, tel e- rnetering, supervisory and protective relaying system is described in Section 15 .. -lnsul at ion Coordination and Lightning and Switching Surge Prot<iction I ' I . t I I I I t The electrical insulation and protective devices will be sele~ted and co-·--J ordinated to provide a safe margin of insulation strength above the maximum · abnormal. voltages permitted during 1 ightning, switching, and shOl .. t-circuit surges. The 1300 kV basic insulation level (BIL) specified for the trans- former and other BIL values stated for the electrical equipment and connec- tions are tentative and are subject to detailed study in the design stage of the project. In principle, lightning arresters will be mounted on or adjacent to all major electrical equipnent having wound-type internal construction, and will be provided at the generator 15 kV terminals and the main transformer 345 kV terminals. · 12.19 -Switchyard Structures and Equipment TO FOLLOW 12.20 -Project Lands Project 1 ands acquired for the project wi 11 be the minimum necessary to con- struct access and site facilities, construct permanent facilities, to clear the reservoir, and to operate the project. ,, ' I ' I I ' t I I I t I I I I t t I ' ' I I I t I I I Appendix C contains 1and status backgrou~d information relative to the susitna Project, together with an inventory or private and public lands required for the project. A 1 arge amount of public ·1 and in the Watana· aret\ is managed by the Bureau of L.and Management. There are large blocks of private Native Village Cor_poration Lands along the riv-er .. · Othe'!' private holdings consist of ·widely scattered remote parcels. The state har$ sel'ected much of the federal 1 and in this area and is expected to receive a patent. 1 ;::·· ... ,.... . . . . ·. (. . ' . . . . . . . . . . . '· . 12-85 ' ~· !f<t:!t~~.,,g,...,..:;~>~•,;,,....,..,".~<'""'-.. •'h''""''"'"""""'-•""r~~-,,~,. •o0 "o•·'"'•'•';.~ .... ~~<.,_,;:<"~~·><•A•~';;;>;....d';_~'•.,7...., ........ "•'=''"~~· ....... "':',....._,,~....,..,,..,=o.• ..... >•••...-.·•• ...... ~'~.::...· '"""""'"'' .... ~-~•·•.,.,...'"'""' ...... ' .. '"""'""""""''' ..... '"'•• ..... "'"' ....... ., ..... '-,..........,, ..... ._ ....... _.,, .... , ........ -····;;i;;,i•~~-· ...........,...........,_....,_...._ .... ·.-.··-.,;,.<··~--· ··-·----~-·-·••··iiilli·"-··iiiiiiiioii•••·•~•' Calertdar Year 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 TABLE 1Z.1: WATANA P£AK WORK fORCE AND CAMP/VILLAGE DESIGN POPULATitlN . Yearlx ·.Pea)(. Force _£_amp/Village Design 910 1000 1360 1500 4005 4400 5635 6200 5635 6200 5635 6200 5635 6200 4000 4000 2000 2ZOO 1090 a 1200 270 300 I I I :I I I I ' I I t I f I e I I I I I ' I I I· I I I I ' I I t I I I I I· I TABLE 12.2: ROCKF'Ilt. AND EARTH DAMS IN EXCESS OF 500 FEET Dam - Ragun .Nurek Watana Tehri · Kishaw Sulak Mica Patia. Chicoasen Chivor Oroville Esmeradla Sayansk keban Altinkaya New Melones Don Pedro Swift Portage fobuntairc New Bullarda Bar Dartmouth Okoy Ayvacik Takase Hasan Ugurlu Nader Shah Gura Apelor Retezat Hagar in Charvak Boruca Kremcu~ta Trinity Thomson Talbingo Tokujama LaG£-ande No9 2 Palo Quemado Grand Maison Sao reli:K F'ierze Cougar Yacambu Emborcacao f''instertal Cumberland Canales Narmata Goesi::heneralp Salrajina Gepatach Foz do Are1.a Tedorigawe Carter Country ·USSR USSR USA India Ir.dia - I.ESR Canada Colombia Mexico Colombia USA Colombia USSR Turkey Turkey USA USA USA Canada USA Australia Turkey Turkey Japan Turkey Iran Romania Jordan USSR Costa Rica Gi:eece USA Australia Austr.alif! Japan Canada S. America France Brazil Albania USA Venezuela Brazil Austria Australia Spain Japan Switzerland Colombia Austria Brazil Jaoan USA reet - 1,066 1,040 885 856 830 802 794. 787 787 778 771 754 738 679 640 626 614 610 600 590 590 590 587 577 574 574 568 561 5':·1 548 541 538 53{} 530 528 525 525 525 525 519 5'19 519 519 519 510 510 508 508 505 503 503 503 500 TABI,.E 12.3: SUMMARY OF' ~SlGN OATA FOR LARGE EHBANIO£NT DAMS IN SEISMICALLY ACTIVE AREAS ·--~---""""' .. ·-' •-"'+----... "' . .......,., treat Ratio of . Height Width Corel· Width -::·~-~. " .. Dae Feet FreebOard Feet . to J)am Height Watana (U) 885 25* 35 a .. so Mica (C) 794 26 111 0.45 Chicoosen {M) 787 33 az 0 .. 42 Oroville (U) 771 22 30 0.34 ·Don Pedro (U) 614 - Ayvacik (T) 587 17 50 0.34 lakase (J) 577 17 46 0.40 Tedorigawa (J) 503 13 40 0.31 Netzahualcoyotl (M) 453 18 50 0.43 Iweya (J) 413 62 33 0.33 Kazurya (J) 413 39 Narakura (J) 4'10 16 y:} 0.56 Pyramid (J) 400 36 T amahara ( J) 380 13 39 0 .. 43 Seta (J) 364 20 36 0.29 .. -.. -~-"" ----.. * Watana freeboard -normal maximum operat~ons level to nominal crest (additional heicjlt allowed for seismic slumping) Up at reM .. Slope 2.4 2.25 2.2 2~6 2.4 2.5 2.6 2 .. 6 2.0 2.5 2.6 2.7 2.5 2.7 2.5 I t DownstreBI! I .. Slope ' .. 2.0 I 2.0 z.o 2.0 I 2.1 1.8 I. 2.1 1.85 I 2.0 2.0 I 1.8 2.7 I 2.0 2.2 2 .. 0 __ , I I I I I I t ·f ~ ;I I I I ·I I I I I 'I I I I I I I t I I ,.., Narr.e Watana Mica Ciiicoasen Oroville Ayvacik Tokase falo Quemado Tedorigawa El Infiernilla larbela Netzahualcoyotl Mangla Oerbendi Khan Tseng~en Pueblo Viejo Be as Alicura Ramganga Iwaya Narakura Shimokotori Bao Tamahara Seta Guri Legend Earthquake: H -High M -Medil.lll L-Low lUi-:1 UJ.Mt:.N::J Crest Seismic Height Length Countrr Activity ( ft) (ft) USA H 885 Canada l 794 2,600 Mexico H 787 1,640 USA L-M 771 5,600 Turkey M-H 587 1,400 Japan H-H 577 1,200 S. Americ~ H 525 1,215 Japan M-H 503 1,380 ~texico H 486 1,100 Pakistan H-H 469 9,000 Mexico H 453 1,570 Pakistan H 453 3,400 Iraq H 443 1,460 Tah.an H 436 1,440 c. Arne ric~ H 436 820 India M-H 435 6,400 Argentina H 426 2,620 India M-H 413 -Japan H 413 1,200 Japan -I H-H 410 820 Japan M-H 390 915 C., America H 388 1,312 Japan H 380 2,000 Japan H 364 1,120 Venezuela l 361 1,970 " Impervious Core: CV -Central vertical C -Central very slightly sloped 5 -Sloping -~ Free,.. board (ft} 25 26 33 22 17 17 26 13 25 1.8 18 32 33 33 49 30 16 22 62 16 13 24 13 13 20 TABLE' 12.4: DAMS IN SEISMIC AREAS Crest Width Core (ft) Type 35 cv 111 s 82 CY 80 5 50 c 46 cv 40 cv 40 cv 40 cv 40 s 50 cv 41 s 56 cv :n cv 43 cv 45 cv 39 I cv 39 cv 33 c <· cv 39 36 c 26 cv 39 cv 36 c 36 cv foundation: R -Rock A -Alluvium D -Downstream U -Upstream -· .J t:ol:'e Width Ratio Core at Base Width to (ft) Dam Height 440 0.50 360 0.45 330 0.42 263 0.34 197 0.34 230 0.40 295 0.56 157 0.31 164 0.34 262 0.56 197 Oe43 230 0.51 330 0.74 410 0.94 157 0.36 '!31 0.30 275 0.65 197 0.48 138 0.33 230 0.56 151 0.39 184 0.47 164 0.43 105 0.29 180 0.50 CG . .,. Consolidation grouting ~B -Concrete block over rock flt't:.I'!!!~IJ~ ~~Ht. Slopes Unified of Core Classi- Zone fication 0.25:1 SM --0.15;1 Cl -GC -CL 0.15:1 ~· 0.25:1 Ml 0.15:1 -0.15:1 CL -GW/SH 0.15:1 ML/HH -CL o. 3:1 CH/CL 0.4:1 SM/GM 0.15:1 CL 0.1:1 CL 0.3:1 CL 0.2:=1 CL -- 0.2:1 ---0.2:1 CL ----0.2:1 Hl fJ.Ll ~I'(:;) t I,UNU/Jal .\ L}N liquid Plastic U/S Thick-0/S Thick-treat-Under Lim!t. Limit ness {ft) ness (ft) Type ment Shells 2) 8 60 60 R :i:G R/UO ----R CG A/UD 40 20 25 25 CB CG A/UO ----. --CB ---50 50 R CB R/UD ---50 R -A/UD 33 7.5 1J 13 R CG R/UD --26 26 R -A/UD 49 25 8 a R CG A IUD --_, -A -A/UD 50 20 -13 R CG A/U ---13 R CG A/UD 50 26 20 30 R -R/UD 22 8 --R CG R/UD 41 19 23 23 R CG R/UO 30 12 20 20 R -A/UO 35 15 10 10 R -A/UD --98 79 R -A/UD --33 65 R -R/UD --20 49 R -A/UO --39 39 --R/UO 40 20 20 20 R -CB & CG --79 79 R CG R/UD --39 39 R -R/UD 50 15 5 6.5 R CG R/UO 0 G_- J I I I I I I I I I ·I I I I I I · · CColumn c D E and G H I and J TABLE 12.5: GENERALIZED SURF'IC!AL STRATIGRAPHIC .COLUMN AREA "D"·ANO.RELICT CHANNEL Unit - SUrficial Alluvi\111 & Fluvial Deposit$ Outwash Till/Waterlain Till Alluvi\111 Till Alluviun Estimated Thickness o-s• 0-18' 12' average 0-15' 0-35' 15' average 2-50' 12' average 0-40' >10' to 65" 20' average to 160' .Description Boulders, organic silts .and sands. Silty sand with. sane gravel and cobbles. occasionally. Usually brown although becomes gray in limited areas. Thickest in northern portions of area,.thickening southward, often absent near Suaitna River. Sand with some silt, occasional gravel. Generally brown, found only along course of limited drainage channels formed in o~twash "E". Generally sorted. . Sand, silt, gravel and cobbles, pertly sorted, with fragment~:~ sub-angular to r:ounded. Silt and sand lenses often present. BroNn to gray brown with a cobble/boulder zor1e often present at the basa of lhit 11 F". Contac.~t bebeen ~£" and "F"'' is often poorly defined. Clayey, silty sand, usuai.J.y gray, often plastic. Contains cobble\$ and .gravel in many areas~ Occasionally present as a lacusttine deposit showing l811inatior~s. &tid/or varves .. Generally a till deposited th~~ugh or near stand!ng water. Sand, silt, gravel, partly to well sorted. Often absl!nt between Units "!.. and "G". \.bit represents period of meltirg pr,oduci~ alluvium/outwash. between these .'deposits. Appears as narrow bands representing channel fillings. Thickest in western portion t1f' the area. Poorly sorted sand, silt; gravel and cobbles, occasionally with clay. Generally gray to ·gray brown. Continuity uncertain due to lack of information at depth. Silt o~ sand layer 2 inches -6 inches thi {·1< often found in cen.ter of Unit "1". Ba~~ unit on top of bedrock, except in buried channel. Contact between "111 and "J" often poorly defined. -, Sand, gravel, cobbles, boulders, few fines, permeable. Found only in bottom of buried channel. Top at 292 feet extending to rocl<. at 454 feet. Note: Letters used to define units are arbitrarY-and were used for . correlation purposes. Two letters may define parts of the same unit. TABLE 12~6: RING FOLLtr.fit:R GATES ". PROJECT LOCATION SIZE (IN.) (1) New ~lanes i California 96 {2) Ne~ Melones California 72 (3) Portage Mountain Canada 84 (4) Hungry House Montana 96 (5) Yellowtail Dam Montana 84 (6) Trinity Dam California 84 (7) ~and Coulee · Washington 102 " (8) Glen Canyon Colorado 96 (9) Green Mountain Colorado 102 *Maximum static head; maximum oper~cing head -250 feet. "'· .. •: 'i HE' , : . ~·~T) ~~·' 607 .591 550 495 470 450 354* 337 261 _l~H Uf -c INITIAL OPERATION 1979 .1979 1967 1952 1967 1962 1940 1965 1943 I I I ,- I ' I I I- I I I I I I· I I I I I I I I I I I I I I I I I TABLE 1Z. 7: PRELIMINARY UNIT DATA 1 -.GENERAL DATA 4\iumber of Units ................. ~-. • .• • • • • • • • • •. • • • • 6 Nominal Unit Output •••••• " ~> •••.• • • ., ............... . Headwater Levels: ncrl'nlll maximum ............................ ., •••• __,_ mini~~~t~~. • •. ~ ........................................ D Tailwater levels: minirnucn ••••••••••••••••••••••• .. •••••••••••••• normal ••• ·• •••••• ~ •••••••• o .••••.••.•• o .e ••• ·• •••.•• maxifM.Iin ·~••••••••••••••~•••-••••••••••o••••• 2 TURBINE DATA 17.0 MW El •. 2185 El. 2045 El. 145Z El. 1459 El. 1465 Type · ...................................... ~......... V~ttical Francis 0 Rated Net Head • • • • • • • • • • • • • • • • • • • • • • .. • • • • • • • .. • • • 680 t ·~et Maxinalm Head ••••••••••o•••••••••••••••••o•~••••• 724 fee~· Minimum Head •••••••••••••••••••••••••••••••••••• 584 f.:tet Full Gate Output: at rated head •••••••••••••••o•••••••••• .. ••••• 250~000 hp at maximun head •••••••••••••••••••••••••••••• 275,000 hp --at minimum head .......................... ·• • • • • 200, 000 hp Best Gate Output •••••••••••••••••••••••••••••••• 85 percent full -Full Gate Dischar~e at Rated Head ••••••••••••••• 3560 cfs Speed •••••••••••.•••••• ~. :, •• ~ •••••••• o ,, •• c.,.. . • • • • ~25 rpm Specific Speed ................................... 32.4 Runner llischarge Diameter ·~····· .. •••• .......... ., •• 132 in Runaway Speed •••••••• f) • • • • • .. • • • • • •• • • • .. • • • • • • • • • • 385 rpm Center line Distributor ···············~·········· El. 1422 Cavitation Coefficient (sigma} •••••••••••••••••~ 0.081 3 -GENERATOR DATA Type ••••• fll , • ··• ••••••••• : •........................... Vertical Modified Unbrella Rated Output •••••••••••••••o•••••••••••••••••••• 190 MVA Power Factor •••••••••••••••••••~••o••••••••••a~• 0.90 \'oltagf.! .................. ·••••••••••••••••u••,.••u.;•4' 15 k\' -Syncht\'mcus Speed ,., .......................... e.. . .. 225 rpm In~rtia ~.stant (H)-~t , ••• ,. • • .. • • • • • • • • • • • • •• • • ... • 3.5 MW/sec/MVA Flywheel Effect (WR2)'1P .. • • • • • • • ..... • • • • • • • • • • • • • •• 5Z x 106 lb-ft2 Heav.iest l.if't •••••••••••&•••••••••••o••••••••••• 770,000 lb *Including turhitle J ... ';·.1 I .. TABLE 12.B: ASSUMED PROPERTIES ~~OR STATIC ANALYSES or .WATANA. DAM -~: ... R'i£erJ.ai : R Rur RF" R5 .~. ito ,s. n m. -·- CORE: --Soft(1) 140 200 300 .. a --Stiff(2) 140 700 800 .35 TRANSIT!ON(3) 145 1300 1500 .4 SHELLS (4) 145 1800 zooo .4 where: = Unit weight, pcf K = ~ulua number, ksf Kur = Elastic unloading modulus nunber, ksf n· ~ Modulus exponent Rf = Failure ratio Kb :Bulk modulus number, ksf rn ~· Bulk modulus exponent C = Cohesian, psf = Friction angle, degrees .6 60 .a .a 280 • .2 .72. 900 .22 .67 1300 .16 = Decrease in friction angle perlog cycle increase in 3' degrees Ko = Esrth pressure coefficient Q_ 35 Q. .43 0 JS 0 .43 0 3.5 6 .43 0 JS 6 .43 Note: Value~ taken from ~mean et al., 1980, "Strength, Stress-Strain and Bulk Modulus Parameters for Finite Element Analyses of Stress and Movements in Soil Masses," Report No. UCS/GT/SD-01, University of California, Berkeley. ( 1 ) Mica Creek Dam Core, 2 pe.rcen\; wet of opt inu.Jm (2) Mica Creek Da11 Core, 2 percent dry of optimum (3) Orovi!le Dam silty sandy gravel (4) Oroville Dam Shell -Amphibolite gravel e I I I I ·I I - I I I I I I I I I I I I I I ••• I. I I I ,, I I I I I I I I I I I TABLE 12 •. 9:. WATANA OAM-CR£ST EL£VATION AND.JP.EE80ARD 1 in$0 Year River Inflow Storm -· Normal max~ reservoir elevation 2185 Storra surchargo 6 Still water elevation 2191 . Wave runup allowance 6) Dry freeboard allowance 3) Elevation top of core 2200 Roadway ovGr cere 3 Minimum crest elevation 2203 Governing elevation f'or cres~ of main dam Highest still water level to be 2 feet above fuse plug pilot channel Sill of pilot channel in fuee plug 1 In 10,000 Probable Year Storm Maximum F'lood 2185 2.185 8 . 17 2193 2202 6 NIL 2199 2202 J 3 2202 2205 2205 2200 Note: The above· elevations do not include allowances for static. settlement and seismic slumping. TABLE 12.10: RECENT HIGH HEAD fRANCIS TURBINES I I', I I I I I I I I I I I I •• ,, I ,_, I~ ;, --- --· --· -... ----·-~ .. -·-- . l .. 0 . , ....... ~ 4p -~-' - IC i ' • tf! u -1&1 C!» ~ c ~ % u ., -Q ·. . -.. l470' 1480 1480 1500 tOlO tD20 HEAOWAT~R f='-t:VA"ON (FT .. ) WATANA DlVEit$10N - ' . TOTAL. FACI~ff'Y RATlNG CURVE -- '' ·; ;?' A ,,, "'"':• '. < - . ... -' -·--- ~ 1j l f f ~ j'. j ( r i , j 1 .... FIGUR~ 12.1 ') I I 01 I I I 1: I I I I I -1 I I- I I -0:1 2300 2200 2100 ~ 2000 l.L - z 0 1- <[ > LLJ ...J 1900 w ·" " ·t - -· \ \ \ ·~ ~ ~ ....... -- ·- .· 2.200 " 2100 ~, 2000 \ \ v·WEl YEAR - \ \ \ 1900 " ~ . \ .\ WET Y~ \ - AR 2300 2JOO 2000 ,- / 1900 .·~.· l ,. i ' ' --· f l ' ".j. . -- ' ~\ \ -· -· " ~ i ' ' r-WET ~ EAR ' ; ' ; cr: 0 > n: LLJ (I) "' Q!_ 1800 AVE I MON lr AGE HLY_.~ ~ ~- \\ \.\ 1800 \ 1800 \ ' ~ 1\ \ ! ' / 1\ \ \ . \ ' ' \ / 1\ ' I 1700 1700 1700 \ -- \ ' l I \ ~~ \ 1600 1600 1600 ' \ / l~ N - ~ ' 0 A J 1500 1500 150 J F M A M "' J A S 0 N D J F M A M J J A M J J A S 0 N D J F M A M J J A S 0. N 0 J F M M J 2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 2. 4 6 8 10 MONTHS MONTHS MONTHS -· JANUARY START MAY START SEPTEMBER START WATANA -RESERVOIR EMERGENCY . DRAWDOWN \ . A 12 \ --w· 1\ .I ! " ' I i £ I 4 ,, ~ i I t I i ·~ 4 -- ' 1 i' i 1 ~ s 0 -N 14 ' ' D 16. ~. ~- .-I . •• 1: ··I: I' ···.··1·· . . ' ~·, I 1: .. · .. ·1: I. I I. I. I. I I -....1 (/) :E·· - z 0 t- ct > w ..J w a:: 0 > a: laJ en w a: t-------UN!T 2000r-----~-----1·--~~------~----------~~----~~----+-----------~--~--~~--~--~--~--~ --. 2 NITS -3 U ITS 1900 '"> 1800 1700 0 1400 ~-----~~--------L-----------L-----. ......... ._. ___ _,__ _____ .-..~~.....-_____ ___., ______ ~-+' 0 5000 10000 15000 2.0000 2.5000 30000 35000 . CAPAC,T.Y (CFS} ' . ' I· ~· -------- ----'"-·--l_ . t . - I t • i 10 . I ... z • -•o Ill • ~- ; 10 -... ~ 40 • • l t \ ~- ~ ... ' u ~ ... to .. "'!'; ii I ! I I 100 f o.a O·QI ., j . WATANA COMPARISON' b.F-GRAIN Jt~E "' ~ -CURVE$ FQR VARIOUS COqE NATER.IAt.S ~,. .. . '\. · fii . ·----~. . . 1 ·' ' ~-._. l I . FIGURE 12~0 ~ ·. . ., ~-. .... .. ------------------- 90 80 - 70 ~ IW'\ - tO 0 i ' • I . a IIJI • ' I' 1~ ~ .. . . . , . ' . !c. . . :~ . ... -. • BOli.DERS OJ88LES ·coor•• M91STU8E CONTiNT l'QUIQ 1-IMlT· .. P'-ASTICITY U~D~X . .. RANGt; ,.,. ,. 9/Q o-.11 NP-~ I ~ 1/1111 4 It) II ~ . IOIDD ~(I) I'D ~ ~'"' r..r\ ~ r\ ~ . . r• I ,. ' II . ~ ~ ,...r-., ~ r\ ~ ~" ~ ·l'lllo ~ ~-l'\ ~ ~'\:.~ ~ ~ K_'\_~ [\~ ~ ~ ~'" I\ I\ ~ " ,-~ .. . ~ [\\ ' ~ ~ ~ ""' "' t\ :"\ '\ ~ ~ ~ . ' 1\.. ~ ~ ~" ~ ' t" ~ ~~ ~""~ -~~~ .. ----.. "'-~ ~" ~"-"'0~~ ., .. •GRWEL I FIM AVERAGE a2.a 0/o I? .I H.P - p . ~ ~"'"~ " ~ f;.;. "' ~"~ ~t\ ~ '''"" "'' ' ~-1\'\ •" I\. 1\, ' 't\. ' 0.1 ~ SAND Coone l Midi""' J fine WATANA.·BORROW AREA Q GRAIN SIZE QURV~S~ VNITS C,D t Clf) ~ ' 0 L'\ K' f' . . ~ ~ ~ ~ ~~ ~ ~~ K' ~' '"' ~ • ~;: ~""'!' . .. ·-·.-.i:.·' .. .. . . . •. -· ~ ~~ ~~ ~ ! ''"" 1\1\ ~ ~ ~~ " ~ ~ ~ ~ :~ ""~ ~ " ~ ~ o.ot ome FINES Silt Slrel .. ·.~· ... ·,., . "71 i. -.~ -~ :, ~1- ~ Q ~ 1\ . ... I) f ~ f . ! 2Q ! i \ $ 50 l i 1:..- ' ~ ~ l ~ 60 ;. ! ' t ~ ~ -j ' i 10 : : e -. ~ . ! .. 10 ,. ~ ~'\ ~-· leiiSiztl I .t l J i I f t I. f ~ 'l -----------------·----- 150 -~ ~ ~ !Ill>• 10 I .. I • • ' ' . . .. . . .. . l BQU..OEftS cota.£1 M01$TURE COtfTENT LIQUID LIMIT PLASTtC&TY tNDEX ,. I J 1111 a ~ UJ..III 4 J ~-tl ~ . 80100 ~Cllt I 0 ··' I' .'r' ~ ~ ~~""-" r.~-r--.~ ~ ~ ~ . . ~~~~"',·~~r-."11\~~~~ •· ' 'COOflt ftANGt: .. Ci-1$0/o O-t7 NP-1 .. '. . ~ ~"'"" I" I\ r'\ ~ ~ ~" ~~ . " ~" ~r-. ' ['., ~ l\: ~-~"'~ ~~ ''"' 1\ !" l' ~ "'" ['\._~ ['\."' "" (\I\ r\ ~~ ·" l"\ ~ ~"\, .... 0.. "" "' ~ I'. "'Il ~ ~ ~"' [\., "'"' (\ 1\ l'f\ I~ ~'\: r0..."'" ~ 1\ 1\f\ '"' ~"""' 1\ ,, ~ ~" I""" ~ "" !'-. r--~ ... D· I GRAVEL : I Flnt .. Coarse 1 · Medium AVERAGE 10.2 °/o 12.8 NP ' ' ' I' -. ~ -~-~""' ;,.;:~. -·· -~·, - ~ . [\. :·~ ~ 1\ l" R ~ [\. t\ i'\."\ ~' ~ ' ~"' ~ ~~ ~"' "" ~ l\ ~' ~" ~"""' ' ~ ~ l'.." ~~ i' 1'\. 1\ l' ~ ~ ~ ~"'~ !" " 1\ ~ ~ ~'\ "'"' "'""' " 1\ ~ ~" ~,, I"' I' . ··~ ~"""' r-. 1\ ' r-.1\ ~!" ,._. 0.1 SANO I Fine WATA~A·BORROW AREA P GRAIN $1ZE, CURV~S-U~ITS ~-,F +ElF t. .. ·~ L . .I .... ~ ~ 1"-[\ I' ~ ~ l' I' [\ .... ~ ~ 't' . ... Q .. I) .. ... •· . ' 20 . 59 .. :.&A . ..,_ I n .t eo I ~ ~ . r-r r0 l'.." . N K ~ ~-· . ' "' . I ['\."\ K" ~'~ ,_ ~ ['\." ~~ ~~ .-.... Ill. ~ ~ ~~ 1\1' r"'~ ~ ~ ~ ~'\ 1\f\ I' ~ ~ .. ~- ~ I\ I' I"" ~ ~ ~ ~ '-~" ~ o.oa o.aoe FINES Slit Sfz•-~.Sll•• .. - .... -~-. -..;::; ~ FIGURE I_ 2. 7 lil \ ' ·- 80 ,• ....... .....- to 0 - - -.. - - - - - - - - -·--- if " .......................... . I • •• • ~ Ill I .-JIII/I • 10 20 44 10100 . lQ:) I'D • .. ' ~-.-&'!..." -~ I' ' ~" ~ ~ ~'" rt:~"' ~ • . ' i' ~ ' i\, -~ ~ ~ ~ ' "' ~ ~'"" ~r-., ['\ "\ ~ ~ .. .. r-.,~ ~ f'\. i'.. ~ ~ ~ ~ 0 ~"' ~"'" ,I\ i'. ~ ~ ~ ~ ~ ... , -..,. s ~ ~~~ "' ~ ~ ~ I'.." ~"" ... ' ~ -~ ~~'\..~ """ ' [\. :\ ~ ~" ~'~ 1\ -~ ~~ "' ' ['... " 0 L'-.'\ '''" ~ ' ' ['\ ' "' ' ~ ~ ~ ~ ~'"' I\ .... .... ~ . 1\,. ' ' ~, ~ " ~ r0 ~" ~"'-."0 ,, ' ~ ~ • . ~ ~ ~ ~' ~'"' 1\,. ' ' ["\ ~ ~ K" [\...""~ ~ ' ' ~ ' ~~ ~"-" 1\ ' ' t\ .. '-~ ~""~ I" ~ ~ ' ' '"' ....... 1\ ' " ["'\ ·~ 1"\ r': ~ ' ' ~~ ' ~ ~ t'l !~ ~ ~ . ' - .. · . ·-·--· .. D. •. • 0.1 . . .Q~ . GRAVEL ~ . 'SAND 80ll.DER3 · COBILEI. 'Coorq I Fl .. eoar.. J Mldlum I Fine ( RANGE A~RAGE MOISTUR' CONT~NT LIQUID LIM.tT PLASTICJTV INDEX t-40°/o ~ . . . ·c. '..,~. ~~ JW-1$ WATANA,-tiQRROW AR~A D GRAIN SIZE C\JRV~-. UNIT$ (it F /G - .. ,, ~· -Q ., • IQ ' ; ... I rr ,. ! ' t 2Q ' } -~ '-. ' 1 !G ~ ~ 11 \.. -~ :\.."'\ ~ ~ - ~ ~ 0...'-. . ~ J '10 " ~ ~"""' ~ ~ ~"" ~""" ~ l ~ ~ ~" ~"'~ "., ~ I ~ 0 L'\.." '''""' r\ ' "' ~ i ~ ~ ~'\..~ 1\'\ "\ ~ ~ ? 1Q ' i'o. f\..' ~" '~""" "' '\[', ~ ['\ ~ . !- ~ ~ ~"' r-., '\!-\ ~ ~ ~ . . :- ' ~ ~'...~ r-..' "'' 0 ~ ~ ' ' 10 ~"' "'' "' ['.....,: ~ ......... =' --~~ .:::~ ~ I'' ,. ""' ' 0.01 00015 .. FINES SllfSiz• ICkWSiln - - --· - --- - - - -·--· - - --' - - 90 80 . 0 1 -a 'i •eo eo 40 ~ ~ .... 10 0 I • • ' l . . BOU..DERS ~ES 1 3 Ill I ilt 1/1111 [I' 1 If ·r IT ' - • t ·GRAVEL ! ;Coan• 1 Flnt • 10 ~ ~ . IQIQD ' '\ ~ L'.'-' ~ It ' J '. ,..... -:-~ ~~ Ill.. ~ ·r-. ~ ~ 1'1~ ~ ~ ·-~ I" ~ ' " ~ . ~ r--.."\ ~~ ~ ~ ~ ' ~' ~~"'' .I~"~ '"'" ~i'> '~"~ 1\ . . ~ • 0.1$ . SAND Coaru 'I MI41U~Jt I Fine WATANA-BORROW AREA . D GRAIN SIZE CURV£S -UNIT H ~ ' . tCU I fiG .. II 0 . ' 't I) ... . .. ' . . ., ... ' 20 .. ·-. ' ~ ". ' 30 l ' ,_ ~ '· ' ·' eo ' ; ~ ~ 1\. ' r-.~ ; ~i\ ~~ .. 10 ' 1\ ['\ ~ ... ~ llillllft ~~ ~ l' -,_ ..... ~ ~ ~ h.. 10 ~ f\..'~ ~ l\1' " " " o.oe oDD~ FINES Slit Sfz• ltkWSim ' "'· ftGURE l2.t . ------------------- .. u. .................... . f • ., ~ a 11 II I ~ 1/21/1 .. ·~ 21 ~ IOtQD tO ) 11'0 ~ • •r-~ ~ ~ ~ ~"-"" ~"" I' li ' ' ~ II ~ ~ . 100 ~ ~ ~"" " r' ["\ ~ " ~ ~" I' "' 1'. ~ "" l'." ~ . . ' r--" "' ["\ ~ l'\ ~ ~"" 1'\. ~ ~ " [\.." ~~"""" ~ ~ ""' ' 80 ""' ~ ~ ~"" ~ ' '" r'O ~ ~"'~ "" ~ ~ ~ ~" "t\ 1'\,r-.... ~ [': ~ ~ . ~ t\r-.. t\.. ~ ~ ~ ~~ ~"' t'-, ~ l" ~ ~ ~ 50 "'~ ~ 0 ~" ~~ "'Il ~ ~ 10..'" ~ 40 '" !'\."-.~" ~'" 1'\ l' ~ "'Il ~ "'" I" ['\ ~ ~ 30 ~ .::s: f" ~ ~ r-.i\ -I'll~ I' ['\ [\ -. r-f'_ ~ 10 . ICIO IQ D O.CJ ill Q.OI GRAJIEL ; SAND fJOlLOERS ~s. ' 'Coors• I FIM COcrM I t.tldlum I Fin• RANGE AVERAGE MOISTU~ COt·fTE"T LJQUIO LIMIT · PLASTICITY INDEX t-13 OJq 88.2-39 0.$-1~ l0.3 °/o 29.1 7.5 WATANA-BORROW AR~A ·p GRAI~ SIZE CURV~$-UNITS. l,J+ll.J . '' ~ ~ ~ "" ~" ~ l'\." ~ ~ ........ .~ ' . .. . ., ~ ~"'" ~ ..... l\.."-~ l'f' 1"1111 ~ ~ to. ~"" ,I\. 1\'-~ , ...... ~ ~~ ' " ~ .:\ ~' Q.OI FINES SUtSiz• Q 10 ' a) ~ ., ., ' 50 l ' -~' ; ' ' i eo ~ W\ ·- : 10 IAn ,-- ~ --90 ~J. -• I # J J !'\." i'l.."' : QOOIICX) (Clot SiMI FIGURE 12.10 . ·' --~-----~-.~------------;;! J,f f • ~, l . - ----··---------:-- .u. ................. . 0 19 • ' 10100 . -. \. -""r\"~~,,~,~ ~:-== ::::: .~: ::: ::m:::::m::!~'''r--r-.r--.,i'.' '-'~'','''" ....... ~-~ ... l' I '\.. '' l "-. '' ~ • • 111 • • • • • • ; ; ~ ll.r. : :: :: ::: :::: ::!:::: t:~ "'-"\ ' ~ ~ " ~ ': ' "'-.: '-..''" " -~ f "\. l' ~ ~ I 0.1 • ~\:::::::: : : :: :: :: :. ~ r\. · '·~ C! ~01 0.0' OJJOI QOOtQJ 10 ' -GRAVEL . SAND FINES BOlL~RS (X)8BLE8 Coar11 I Fine Coone I Medium I Fine Silt Slz• .ICtclfSizu ,• 1 ., .• / t WATANA REQUIRED GRAIN SIZE CURV~S 0 \ MAIN PA~ t r I l ----- - - - - - - - - - -----· -•. - - .u ....... ...,. '"'' ........ I • • • a Ilia ' !It 1/'1.1/1 .. 10 ao '44 110100 ~(I~ .. !1ft . T T ! I I" ' ' ' II II I ' ' 'I ~ Q , ~ ~ .,., 1 1'-o:o-.. ~8.50Jo L No.4 .. l 0 90 .. r-.... t--... i 80 , t 20 ..... ~ l i ! ......... i I ~ t"oo .. ~ to-.. 70 . ~ ' . ' ... 40 """ !it) ~ . "'-, 45°/o L No.'200 eo ' ~ "" . 40 . l 1--. • i' :vu ~ 30 ' 70 ~ ... j . I l. . """' -- 20 A- ) '"'t"oo ~ -· ' r-... "-' 10 10 :~ 0 100 10 I)· I 0.1 .o.oa o.oaa QOOl~ BOll..DERS 'CoesLES GRAVEL SAND FINES ·coors• I Fine Coarse I Medium I Fine Silt Siz• l~SiHI 0 · WI\TANA COMPOSITE GRAIN §IZE CURVE-BORROW AREA 0 FIGUA~ 12.lt . Iii. I I I I I I I' I I I I I I I I, I I I ·I 145 140 13 = 13 !li.~ 5 u Q. -> ... 0· 12 8.7 12 5 120 2 NOTE: . . . . .Y -/ BSOJ. ~· v 4.20~ ~ ~ l IWI"" v ' f7.5 olc. f\ MAX DRY -DENS ITY_\ \1 ~ 10.1 . ~ . \ 6 8 10 12 WATER CONTENT (0/o) MATERIAL PASSING 3/-4" SIEVE· WATANA MODIFIED PROCTOR COMPACTION COMPOSITE SAMPLE BORROW AREAD ' -., FlGURE 12.c.13 li·l· ~~~--------------------------------------------~-------r-- I I I I I. ·I I I I I I •• ;I I I I I I •.• , __ 135 130 127.6 -. -:;· 125 .Q, - -U) z LLl Q 121 .. 2 > a:: 120 Q 115 uo 4 NOTE~ . " . . ~ J . VroJ~~ I J· l .--v . \ ~ I I ~ 50fe MAX DRY DENSITY ~.0°/• -12.~0~ - v '1!1 I/ 1\ . . 1\ . 6 10 1·2 14 . WATER CONTENT (0/0 ) MATERIAL PASSING No. ·4 SIEVE WATANA STANDARD PROCTOR COMPACTION COMPOSITE. SAMPLE BORROW AREA D ' -. I ~ ~· I> 16 18 FIGURE l2.14 • •• ••• . . . . ---------~--------- 0 " 80 ~------~--------------~------~--------~~--~~------~------~ 60 --I. ...,.,. !II fJ) ' Cl) ·&1.1. 0: 40 t- Cl) . 0:: c( UJ X UJ . 20 NOTE: EFFECTIVE STRES'-CONeOt-IDATED UNDRAINED~ &HEAR TEST, . 4 INCJ-1 DIAMETER SAMPLE$, OPT+ I% AT 16%. MODIFIED PROCTOR G C0¥~~9TION. . . 40 60 80 100 l20 NORMAL STRESS (psi) WATANA CONSOl-IOAlES'tiNDRAINED TRIAXIAL TEST RESULTS . COMPOSITE SAMPLE-BORROW AREA D !!- ... 140 l60 c FIGURE 12 •. l5 ;, .•. ·-. -·~ "I I I . . . I I I I I I I I I I I I ~"I 0 -1- 4 a: Q ·-0 > .4200 ' .3800 •o =.3761 .3&00 , . ~ . . 'a__ ~ l•. .340<) .3200 . 3000 .2800 .2600 .2400 ~-2200 .2000 .1800 .1100 .1400 OJ ! ~ 'a_ "' . ~· \: ~ . " "' "' ta •t= .. 2345 -~· ~ . - 4. . ~ -. -... 0.2 0.3 0.5 0.7' t.O 2.0 4.0. 8 10 P"ESSURE ( TS F ) WATANA CONSOLIDATION TEST-BORROW AREA D STANDARD PROCTOR COMPACTION-OPT. +2 °/o 20 30 so 70 100 I 1-!'I.GURE 12.16 J11l -'~----~~~----------~--------------------~--~------------.-·~~~-.··~.· \!, . II. --- -~· 0 KlO 90 0 160 50 40 30 ~ -- 10 II I ·BOU..DERS • • ' ~"""' ~"'~ '"""' ~'\_~. ~'~ ~'\ ~"'"'!' ~~r- . ·~" ~!' \...~~ \. "" ~~ COBBLE$ . .u.a.at...., ""'~' I 2 U/1 I·~ 1/IJ/1 • 10 20 40 ., 100 ZOO IPD ~ !'. ."'' N l I I' I ' ~" r"\."'"" ~~ ~ ~ I ' . 1 r t\ I' ~ ~ ~ ~ ~"""' \ 1"\ ' ~ ~ 1 UPPER ZONE \ .... ~ !' ~ r0 l' ~"' ~ 1';. '" ~ :'\ l". ~ ~ J' I\ I\ L\ L' ~ k'\.' ~ I' f\. L'. ~ ~ ~" ~ ~ !'~ ~ ~ l'\ ~" ~~ f" ~ ~ ~ ~"' ~~ " ~ . ~I\ 1\ ['\ L\ ~ ~ ~ I "'ii ~ ~ "' ~ ~""~r-.~ I\ I' ~ ~ ~" ~'" r\ l\ K" ~'"'" I'~ " 1\ f'-I' ['\ [\ ~· ~ I'.~ ~'~ " ~' ~'"" I' r-.1\ i'. i'l' ['\ ~ ~ !:S." ~' "~ l!l'ii ~ ~"'" "" !\I' I'~" ,, ~ ~ ~ ~ &~ I' I' I' ~ "'"""~ i"'oi' ~ ~ " r--~' r'--[', [\ ~ ~ ~~ I' I' ~ [\ ~ ~ '-" r-1' r'-1" l'f' ~ r0 t\..~ l~"" ~" f' [', [\ ~ ~ "i; i" I' ~ I' ~~ I'~ L"-C'\: ['\.."-' ~~ l'r-~ ~ ~ ~"' ~"\ ~"~ I'~ r"l'o ~ """ ~" ~"" "\1' ~I' f' l'\ l"'-~ ~"" ~"" 1'1 ""'~ N ~ r0...~~1' "' r\ ~· ~" ~"~ r-.1' N ~ " I' ['\ !\ to-... ~ I'~" R' ~ K K" ~'"' 1\ I\['. r'\ ~ ~ ~ LOWER ZONE./ I"' roo "" ~ ~ ~ ~' L':."-~ r-.f" I' ~ t'\ ""' ~· I' "' --~ ~ "' ' \. " ~ '" "V ~ " ' " ' t\.." "'" "' 1). I 0.5 QOS . GRAVEL SAND Coors• 1 Fine Coarse I Medium I Fine WATANA . GRAIN SIZE CURVES -BORROW AREA E ':;) . .. . 0.01 FINES I If ' t i t ·• '~ ' ~ l' L ~ ~ 1 'f. ' .. • ·. 0. 10 . 20 ~ ~ ~ 50 Ia\ l- 70 lan ·- 90 ! -·~ ·• • ... -! l. QOOIK)O SllfSiz• ~Sizn FIGURE 12.17 -~-------, ------------------------- • / ·~~~~R~~ ~HTHH~+-4---+H~~~~~~~~~~~0~~"~~~~++~~~~+H~~r-+-~~HH~+-~--~~~~4-~~~ ~~++~-+~~"---H~+·~.~~~~~~~,~~~~~~~~~~~~~~~~~----~~~~~~~~~~~~--~~~~~-4--~ ~~~"~""' ~"' ~t-~ FINES SAND GRAVEL BOU...DERS COBBLES Slit Siza Medium Fint WATANA GRAIN SIZE CURVES~. BORROW AREA ! . FIGURE 12.18 --,,- j -0.30 ~AsLo--~--~,0--~~--2~0 --~--~3 -0 --~---4~0--~--~5-0--~---6~0--~--~70~--~--s~o--~--~9~0--~--~100) Time (sec) I I ; ~ I I ' WATANA MAIN DAM EARTHQUAKE TIME HISTORY FiGURE 12.19 •• I ._.,. I ... I I I I I I I I I I I I I I I I ,, ·r---~--------------~--------------~----------~~--~~---~~·-----~----~ 1200' 1000 800 400 200 '"' 0 0 l..EG£Nl e-FiXED CONE 0 • HC1LOW JET . . 3. at . 4. v; ·9-. . ' 22 • . --·~ ---· ·---- ' ' .• . ' . . ' - . ~WATANA 12.-.,5 le• - 14. 6. .... 17. ..--100 z·e· 7. uO . . 21·- . so 75 VALVE 01At.£TER -INCHES FREE DISCHARGE VALVE EXPERIENCE PLOT 18 • .SiTES .. t. 1<£BAN 2. ROUND BUTTE 3. TOCT RIVER 4. MAMMOTH ~- 5. 'LA AMISTAD 6. PORTAGE MTN. 1. COPETAN DAM a CAUFORNIA 9. BIG CRE~K I 0. GLEN CN..m:N t .... Kf.ERV....:e£ ow 12. VJU.ARINO 13.MANGLA 14,. llARTMOUnf DAM I e~ NEW MELONES t6.NEW~ES 11. JUNCTAN I a r.ul MOliN'miN 19. NEW EXCHEQUER 20. CAVADO 21. 'PORTAL • 22. LAKE MATTHEWS [,::IL ~ANYON •• 3 . ,. zoe! 5-. - lOO 125 FIGURE! 12'.20 .. I ...... ._ ' I 740 - :- I .. 720 I -· ... I .. 700 I I ......... · 680 . . I 1- LU I LU 660 u. I 0 ~ LJ~ % I 1- &IJ z "-. 640 . I I 620 "' I •' -. l . /, ,.-~~ BEST EFfH IEitt-1 I I 600 1\ l I . RESERVOif ; EL.. 2045 -... .,...._ 580 I I 100 120 I I I it~~ G!NERA" "OR RA D POWER RESERV<~R EL. 2!85 .·/ . c ....-.-.:. --.~ . l " ,~ I I I . v WEIGHTED fl. 'ERAGE HEAD I ~ INIMUM DECEMB ER HEAD I . f+-170 MW . I L FUU. GATE I . 160 180 200 UNIT OUTPUT-MW FIGURE 12.21 . 220 • •••• . . .. · · .. , .. , . . :;:· . ' . . . "-ll' I rei ~·­_/'"- ' ·"'·~ · .... I I I .~· •, "' I -· I I I t I I I I I I I I . ,-' #:. -,. i &lj. a ~.~ ii: 80 t--------~~----+------+------+-'-~-~~------. ~ .&&.. &II -ell Ll.. To~------~------~------~--------~-7~--~--~~s . IJ.I C) a:: ~ ~ ~ cs t-· .-----+-----+-~E:.---+----+-----+---12000! 120,000 100,000 200,000 TURBINE OUTPUT { HP) WATANA -TURBINE PERFORMANCE (AT RATED HEAD) . \' . \\ 2.4q:K)O FtGuA~ 1a;a2 .• Iil· .. · ~· .· . . 1\,. I I I I I I I I I I I I I a, I a I ·I .'~ 2 • ., • 0 ~~-------+--·------4-~----~~------~--------~~----~ .40 I 27 ... ~3 • • a ~~--------+--~----4-------~~------~--------~--·----~ c w % .... . w. z .. 3 . J ~· WATANA~0 • 15 ~~------_.·--~~~~·a~------~--------~------~------~ DEVIL CANYON-.........,;..0 ' • 22 Jl 3~ !a . • 24 • .29 13 ,.31 6 , .32 ~~----~-+------+---~"~~----4-------4------~ ~N .a J7 •21 ~·~~~ ~ 200~~-----+--------~------~~------~--~~~-'28~~~~~------~ 20 S) 60 SPECIF'IC SPEED ( N. S ) " . FRANCIS TURBINES SPECIFIC SPEED EXPERIENCE CURVE FOR RECENT UNITS ao FIGURE 12.23 [iJ .I ' ' I . a·· .. . ' I I -"' ,, I _, ' ' t I I I I ~ I 13 ··DEVIL CANYON UEVELO~MENT This section describes th¥a various comp.onents of ·the Devil Canyon development, including diversion facilities, emergencyrev~ase facilities, main dam, primary outlet facilities, res~voir, main and emergency spillway; saddle darn, the power intake, penstocks, and the powerhouse complex, including turbines, generators, mechanical and.electrical equipment, S\'litchyr d structures, and equipment and project 1 ands • A description of permanent and temporary access and support facilities is also included. 13.1 -General Arranaement The evolution of the Devil Canyon general arrangement is described in Section . 10. The De vi 1 Canyon reservoir and surrounding area is shown on Plate 64. The site layout in re.1ation to main access facilities and camp facilities is shown on Plate 66A. A more deta i 1 ed arrangement of the various site structures ; s presented in Plate 64. The Devil Canyon dam will form a reservoir approximately 31 miles long with a surface area of 7,800 acres and a total volume of 1,092,000 acre feet at a nor- mal maximum operating elevation of 1455. The operating level of the Devil Can- yon reservoir is controlled by the tailwater level of the upstream Watana devel- opment. During operation, the reservoir will be capable of being drawn down to a minimum elevation of 1405. · The dam will be a thin arch concrete structura with a crest elev.ation of 1465 and maximum height of 645 feet. The darn wi 11 be supported by mass concrete thrust blocks on each abutment. On the left bank, the generally lower ground surface ievei will requir-e a substantial thrusta Adjacent to this thrust block, an earth-and rockfi 11 saddle dam wi 1 1 pro vi de closure to the left bank. The· saddle dam will be a central core type generally simtlar in cross section to the Watana dam. The dam will have a maximum height above foundation level of approximately 260 feet •. During construction, the reservoir wi 11 be diverted around the main. construction area by means of a single concrete-lined diversion tunnel 32 feet in diameter on the left bank of the river. A power intake located on the right bank will comprise an approach channel in rock 1 eadi ng to a reinforced concrete gate· structure. From the intake structure four penstockss consisting of concrete-lined tunnels each 20 feet in diameter will lead to an underground-powerhouse complex housing four Francis turbines each with a rated capacity of 150 MW and four semi-umbrella type generations each rated at 180 MVA. Access to the powerhouse complex wi i 1 be by means· of an unlined access tunnel approximately 3,200 feet long, as well as a vertical access shaft about 950 feet deep.. Turbine discharge will be conducted to the river by means of a single 39-foot-diameter tailrace tunnel leading from a surge chamber downstream from the powerhouse cavern. Compensation flow pumps at the power plant will ensure suitable flow in the river between the dam and tailrace tunnel outlet portal. A-.separate transformer gallery just upstream from the powerhouse cavern will house six single-phase 15/345 KV transformers. The transformers will be connected by 345-KV, single-phase$ oil-filled cable through a cable shaft to the switchyard at the surface. J3-1 -· The primary outlet facility will consist of .seven individual outlet conduits located in the lower part of the main dam; it will-be de:stgned ,to discharge all floods With a frequency of 1:50 years or less. Each outlet conduit wi11 have a fixed-cone valve similar to those provided at Watana to tni'nimize undesirable nitrogen supersaturation-in the flows· downstream. Flows ·resulting from floods with a frequency areater -than 1:50 years but 'less than 1;10,000 yea~s will be discharged by a ehute spillway on the-right bank. alsQ similar in design to that provided for Watana• An emergency spillway on the left bank wi11 provide suf·· fici~nt additional capacity to permit discharge of the PMF without overtopping the dam;; An emergency-release. low-level outlet facility will allow lowering of the dam to permit emergency inspection or repair. 13.2 -Site Access (a) Roads At Devil Canyon the main access road will enter the site from the south. A low level bridge crossing the Susitna River will be located just upstream ·of the dam. ln addition to the main access, several ancillary roads. will be required to the camp. village, tank farmt borrow areas, and construction roads to the dam and all major structures. The~e roads, with the exception of temporary haul roads, are shown on Plate 66A~~ The construction roads will be gravel-surfaced roads 40 feet wide with small·radius curves and grades 11mited to 10 percent. Major cut and fill work will be avoided. A gravel pad approximately five feet thic ·will be required for the roads. This will provide a drivable surface and also will protect against the sporadic permafrost areas. (b) Bridges The existing low level bridge upstream of the-dam will be used to cr.oss the Susitna River during construction. This bridge will be used during abut- ment excavation. After construction of the cofferdams is complete, the crests of these structures will be used to cross the river. After construction of the main dam is completed, the crest of the main dam will prOw'ide access across the Susitna River. {c) Air-strip A permanent airstrip will be located at the Watana site, approximately 30 miles west of the Devil Canyon site. This strip will be used far the Devil Canyon development. Thet airstrip will be capable of accommodating the C-130 Hercules aircraft, and will also accommodate small jet passenger air..: craft. {d) Access Tunnel An access tunnel wi 11 be provided· to the underground pawerhous·e and associ- ated works~ The main access tunnel will be concrete-lined and will be approximately 35 feet Wide and ~8 feet high. The tunnel will a'llow perman· ent access to the operating development and will also be utilized during construction as the main construction tunnel. The tunnel will have con- struction adits branching off to the various components of the development during construction. I -1-, ... .,, ·.-.. ·. ' I I I I -...; J I ' ' I ... ' I I I t I t I .,---." -- .. _,-) ~ L , 01 I I I ... I _, ' ' I ' I I I ' ' I (e) A vertical 20--foot diameter access· shaft with an elevator will also be pro- vided for_ access to the underground facilities. The powerhouse access through this shaft ~Will be at the opposite end to the access tunnel. -.. 13.3 -Site Facilities (a) General The. construction of'the Devil Canyon development will require various facilities to support the construction activities throughout t~e entire construction period. Following construction, the planned operation and maintenance of the development will be centered at the Watana development; therefore, minimum facilities at the site will be required to maintain the power facility. As described for Watana (Sect-ion 12), a camp and construction village will be constructed and maintained at the project site. The camp/village will provide housing and living facilities for 3,200 people during construction. Other site facilities include contractor's work areas~ site power, · services, and communications. Items such as power and conmunications and hospital services will be required for construction operations independent of camp operations. It is planned to dismantle and demobilize the facility_upon completion of the project. After demobilizing the site, the area will be reclaimed. It is planned to utilize dismantled builC:·~ngs and -other items from the Watana development as much as possible in the camp/village. Since the Watana develo_pment will be in service during the construction period, electric power will be available. It is~therefore planned to meet all heating requirements with electric heat and not with fuel oil, as is planned for the Watana development. The salvaged building modules from the Watana camp/village will be retrofitted for electric heat. (b) Temporary-Camp and Village The proposed location of the camp/village is on the south bank of the Susitna River between the damsite and Portage Creek, approximately 2.5 (see Plate 66A) miles southwest of the Devil Canyon. The south side of the Susitna was chosen because the w.ain access is from the south. South-facing slopes will be used for the. camp/village location. The camp will consist of portable 'l!.'codframe dormitories for single status . workers with modu 1 ar mess ha 11 s, recreation a 1 bui 1 dings, bank, post office, fire·· stat ton, warehouses, hospita 1, offices, etc. The camp wi 11 be a single status camp for approxi111ately 2,900 workers. The village, designed for approximately 320 families, will be grouped around a service core containing a school, gymna.sium, stores, and recrea- tion area. -! '" --.''.-·: "1 The two areas wi 11 be separated by approximately 1/2 mile to provide a buffer zone between ar~as. The hospital will serve both the main camp and the village .. This ctrnp 1_ocation wi.11 separate 1i ving areas frorn the work areas by a mile or more and wi 11 keep t\t-avel t1 me to work to less than 15 minute$ for most personnel. The camp/village will be constructed in stages to accommodate the peak work force as presented in Table 13.1. Table 13.1 also presents the camp/ village facility design ntJnbers. The facilities have been designed for the peak work force plus 10 percent for "turnover 11 • The "turnover .. includes provisions or buffers for overlap of workers, vacations, and visitors. The conceptua 1 1 ayouts for the camp/vi 11 age are presented in P 1 ates . and ,. ----~ (i) Site Preparation Both the camp and the village areas will be. cleared in select areas for topsoil, and the topsoil will be stockpiled for future use in reclamation operations. At· the village site, selected areas will be left with trees and natural vegetation intact. Both the main camp and the village site have been selected to pro- vide well-drained land with natural slopes of 2 to 3 percent. A granular pad varying in thickness up to 8 feet will be placed in selected areas at the main camp. This wi 11 provide a uniform work- ing surface for erection of the high density housing and service buildings and wi 11 serve in certain areas to protect the permafrost where 1 t under 1 i es the camp. In the vi 11 age area, a granu 1 ar pad will be installed only as-necessary to support the housing units and to provide a suitable base for construction of the temporary town- center buildings~ ~'- All roadways within the camp/village areas will be flanked by road- side ditches, with CMP culverts carrying water across the intersec- tions. In general, drainage wi 11 be through construction of a sur- face network of di.tcheso Peripheral ditches will intercept overland flows from adjacent non-cleared land and carry it around the camps. Runoff will ultimately be directed to existing drainage channels leading to the. Susitna River for the village and the main camp. ( i i ) F ac i 1 it i e s Construction camp buildings will consist largely of trailer-type factory-built modules assembled at site to provide the various facilities required. The modules will be fabricated with heating, lighting, and plumbing services, interior finishes, furnishings~ and equipment. Trailer modules will be supported on timber cribbing or blocking approximately two feet above grade. - . l I ' I I I I I I ' t I I t I ' I ' I I I~ I ' 'I I I· I 'I I I t I ., I ' I I I I Larger structures such as the, central utilities building, ware-~. houses, and hospital will be pre-engineered, steel-framed structures , with metal cladding. The larger structures will be erected on concrete-slab foundations. The slab will be cast on a non-frost susceptible layer at least equal to the thickness of the annual freeze/thaw layer. Heated permawalks will connect· the majority of the buildings and dorms. The various buildings in the camp are identified on Plate 67. (c) Site Power and Utfl ities (i) Power .Electrical power will be required to maintain the camp/village and construction activities. A 345 kV transmissioh line and substation will be in service during the construction activities. Two trans- formers will be installed at the substation to reduce 1;he.line vol- tage to the distribution voltage, One of the transformers will be the same transformer used at the Watana development during its con- struction. · · Power will be sold to the contractors by APA. The peak demand during the peak camp population year is estimated at 20 MW for the camp/village and 4 MW for construction requirements, thus totaling 24 MW of peak demand. The distribution system in the camp/village will be 34.5 kV. (i i) Water The water supply system will serve the entire camp/village and s~lected contractor•s work ar·eas. The water supply system will pro- vide for potable water and fire protection. The estimated peak pop- ulation to be served will be 3,950 (2,900 in the camp and 1,050 in the village). The principal source of water will be the Susitna River, with a backup system of wells drawing on ground water. The water will be treated in accordance with the Environmental Protection Agency {.EPA) primary and secondary requirements. · A system of pumps and constructed storage reservoirs will provide the necessary system demand capacity. The water distribution system will be a ductile iron pipe sy.stem contained in utilidors as described in Section 12.3. (iii) Waste Water The waste water collection and treatment system will serve the camp/ village. One treatment plant will serve the camp/village. Gravity flow lines with lfft stations will be used to collect the waste ·water from all of the camp and village facilities. The 11 in-camp 11 0 13-5' and 11 in .... vi l'lage'' collection systems will be run through the perma- walks and uti lidors so that, the collection system wi 11 always be protected from the elements. ·At the villaget· .an aerated collection basin will fie installed to collect the sewage. The sewage will be pumped from this collection · ·basin through a force main to the,sewage treatment plant. An aerated collection basin will be needed at the village to balance out the high 1 y eye 1 i c waste water flows. - Chemical toilets located around the site will be serviced by sewage- trucks, which wi 11 discharge direct 1 y into the sewage treatment plant. The sewage treatment system will be a biological system with lagoons. The system will be designed to meet Alaskan state water 1 aw secondary treatment standards. The 1 agoons and system wi 11 be modular to allow for growth and contraction of the camp/village. The location of the treatment plant is shown on Plates . and· • The location was selected to avoid unnecessary odors in the camp as described for Watana. The sewage plant will discharge its treated effluent through a force main to the Susitna River. All treated sludge will be disposed of in a solid waste sanitary landfill. (d) Contractor's Area The contractors on the· site will require officet shop and general work areas. Office space for the contractors has been provided and ~its location is shown on Plate 66A. Partial space required by the contractors for fabrication shops, storage or warehouses, and work areas within the camp confines has been designated and is shown on Plate· 66A. Additional space required for the aforementioned i terns wi 11 be located between the main camp and the main access road. 13.4 -Diversion (a) General ,, __ Diversion of the river flow during construction will be accomplished with a ~i ngl e 32-foot diameter horseshcs shaped section diversion tunne 1. The concrete-lined tunnel will be located on the left bank of the river, and wi 11 be 1,490 feet in length.. The diversion tunne 1 p 1 an and profile is .shown on Plate 69. The tunnel is designed to pass a flood with a return frequency of 1:25 years routed through the Watana Reservoir •. The peak inflow will be 37,800 cfs. Routing effects are negligible and the peak flow that the tunne.l will discharge will be 37,800 cfs. The maximum water surface elevation upstream of the cofferdam will be E1 944. A rating curve is presented in Figure 13.1. 13-6 ••• I ' I I ·I I I I t t I I I f. t t ., .. _ .. ; . ; ·I· , . . . i ,. l . ~. 41 I I I I I I ,,~,~' I I I I I I I I I t I I I {b} {c) .. ' ,. "" ~,' -c-.. :-;-.---,~~.:::;"--':;:·4~-;··-~"'•< ~----!? . .,.. •.•• _-::.,.._ Cofferdams Due to the depth of alluvium present in the Susitna riverbed fount:iation~ a grouted zone thr-ough the alluvium material to bedrock excavation and to sound rock in the abutment areas will be installed. The depth of al1uvium material in the riverbed ranges up to a maximum of 70 feet. The alluvium material consists of open-worked gravels with numerous cobbles and boulders. Due to the coarseness of the alluvium material, a grouted zone was selected rather than a slurry wall; it will be constructed through the closure dam and alluvium material to bedrock and will minimize the amount of seepa4Je into the main dam excavation. The abutment areas will be cleared and grubbed with excavation of all material to sound rock prior to placement of any cofferdam material. .The upstream cofferdam wi 11 be a zoned embankment found on the closure dam (see Plate 69). The closure dam wil1 be constructed to Elevation 915 based on a low water level of Elevation 910 and will consist of coarse material on the upstream side grading to finer material on"the downstream side. When the closure dam is completed, the groutjng will commence and the zone will be constructed to minimize seepage into the main dam foundation excavation. The cofferdam from Elevation 915 to 947 will be a zoned embankment consist- ; ng of a centra 1 core, fine and coarse upstream and downstream filters., and rock and/or gravel shells with riprap on the upstream face. The downstream cofferdam .J~ill be a closure dam constructed from Elevation 860 to 895 (see Plate C:9). It will consist of coarse material on the downstream side grading to finer material on the upstream $ide. When the closure dam ts completed~ the grouted zone can be constrw:teci in the finer material to minimize seepage into the main dam foundation excavation. The upstream cofferdam crest elevation wi11 have a 3 foot freeboard allow- ance for settlement and wave .run up.. Thermal hydraulic studies conducted showed the discharged waters from the Wat.ana Reservoir will be 34°F when they pass through Devil Canyon. Thus, an ice cover will not form upstream of the cofferdam, and no freeboard allowance was made for ice. Tunnel Portals and Gates A reinforced concrete gate structure will be located at the upstream end of the tunnel (see Plate 70). The portal and gate for the tunnel will be designed for an external pressure (static) head of 250 feet. Two 30 feet high by 15 feet wide water passages will be located in the· gate structure with guides for the diversion closure gates separated by a center concrete pier. Each gate will be a fixed wheel vertical lift gate. operated by a wire rope hoist ·;nan enclosed housing. The gate will be designed to operate with the reservoir at Elevation 950, an 80 foot operating head. Stoplog guides will be installed in the diversion tunnel outlet portal, and . stoplogs will be provided to, permit dewatering of the diversion tunnel for plugging operations. · ·· 13-7 The. stoplogs will have a downstream skinplate and upstream seats (relative to river flow) and will be arranged in suitable sections to facilitate reJatively easy handling, with a mobile crane .using a fo.llower beam. (d) pperation During .Diver.~ion The tunnel wi 11 pass all flows from BOO cfs to the upper design flow of 37,800 cfs. The rating cur-ve for the diversion tunnel is shown in Figure 13.1. (e) Final Closure andReservoir Filling_ Upon completion of the concrete dam to an elevation sufficient enough to pass the environmental flows with the discharge va 1 ues that are i ncorpor- ated in the dam, the tunnel will be temporarily closed with the intake gates-and crynstruction of the permanent plug will commence. It is estimated it will take a year to completely place and cure the plug. During this time the upstream gate and intake structure wi 11 be designed for a reservoir elevation of 1130, which will create an external pressure of 250 feet. The filling of the reservoir will take approximately 11 days to full reservoir oper-ating elevation of 1455. 13.5 -Arch Dan · {a) General The arch dam at Devil Canyon has been selected in preference to a rockfill dam for the reasons given in Appendix 05$ The s_hape of the canyon is suit- ed to an arch dam, with a crest length-to-height ratio of .approximately 2. The height of the dam will be approximately 650 feet, \'iell within the range of heights of dams constructed elsewhere. A comparative list of some 'large arch dams constructed throughout the world is given in Table 13.2. Because Alaska is a highly seismic area., the arch dam wi 11 be designed to withstand dynamic loadings from intense seismic shaking .. Some dams con- structed throughout the world in high earthquake areas include the 741- foot high El Cajon dam in Honduras, the 696-foot high .Mohamed Reza Shah Pahlavi dam in Iran, and the 548-foot high Vidraru Arges dam in Rumania. The Vidraru Arges dam and the 372-foot high Pacoima dam in California have both withstood high earthquake loadings, with the latter experiencing a. bas,e ground acceleration of between 0.6 to 0.8 g. Green Lake dam is presently being constructed to a height of 210 feet in Sitka,. Alaska. {b) Location The arch dam will be located at the upstream end of the canyon at its narrowest point. The rock is outcropping or very close to the surface at the abutments, and the c'ontours just downstream of the left abutment swing in toward the ,river so that the left side of the dam wi 11 be founded against the upstream side of a slight promontory. 13-8 I I I I I I I I I I I I I ' ' I I I I ' ~."li··~ ~ I I I I ., I II I I I I •• I I I I I I I The rock forming the right abutment rises several hundred feet above ·the ·dam crest but on the left side the rock surface rises only to Elevation 1400. It will be necessary to construct a mass concrete thrust block at t~ds point to artifically form the bearing surface of the dam. (c) Foundations The arch dam wi 11 be founded on sound bedrock located 20 to 40 feet below the bedrock surface. The foundation will be excavated and trirrmed beneath the dam so that no abrupt irre·gul ariti es \'li 11 occur at the foundations which could cause stress concentrations within the concrete. During exca- vation the rock will also be trimmed as far as is practical, to increase the synrnetry of the-centerline profile and provide a comparatively uniform stress distribution across the dam. Areas of dikes and the local areas of poorer quality rock will be excavated and supplemented with dental con- crete. 'lne foundation will be consolidation grouted over its whole area, and a double grout curtain up to 300 feet deep will run the length of the dam and its adjacent structures as shown in Plate 75. Grouting will be done from a system of galleries which will be run through the dam and into the rock. Within the rock these galleries will also serve as collectors for drainage holes which will be drilled just downstream of the grout curtain and will collect any seepage passing the curtain. High on the left abutment open cracks are evident in the rock; these will be excavated t·o sound rock and the excavated material replaced with can ... crete in the form of a deep thrust b 1 ock. On the right abutment a mass concrete thrust block will be founded at the end of the dam to match the left block and improve the dam symmetry. (d) Arch Dam Geometry The philosophy and design of the dam is more thoroughly described in Appen- dix D, but is su11111arized herein. The dam geometry is shown in Plates 72 and 73. The crown section at the center of the river wi11 be of a double curved cupola shape inclined downstream. The static load from the reser- voir will be taken primarily in the arches; the three-dimensional stress action of the structure 'Ni 11 tend to induce tension in the downstream face of the cantilever. This will be-offset by the gravity forces of the over- hanging section, which also will counteract any seismic loadings produced by downstream ground motion. · A two-center configuration wi 11 be adopted for the arches to counteract ·. :~ sli.ght assymetry of the valley and give a more uniform stress distributio.-, across the dam. The arches ·wi 11 b·l formed by circles with c~oters located on the vertical axis plane running along the center of the canyon. The radi.i of the arches on the right and wider side of the canyon will be greater than· those on the 1 eft, and the thrust wi 11 be directed more nearly ~normal to the rock abutment rather than parallel to the face" as would occur with a smaller radius arch. The radii of the i ntrados or downstream face will be smaller than those of the extrados, producing a thicken·Ing of the dam at the abutments where stresses would tend to be highest. 13-9 '• (e) Thrust .. Blocks The· thrust blocks are shown on Plate· 7 4. The massive concrete-b 1 ock on the left abutment wi11 be formed to take the thrust from the upper part of the dam above the existing sound rock level,,. The. thrust block will also serve as a transition between the concrete dam· and the adjacent. rock fill saddle dam. Ttle incline!!.enct~~~e.nf_.the block will abut and seal against the impervious saddle dam co_re-, and it will be wrapped by the supporting .rock shell. A thrust block will also be formed high on the right.abutment at the end of the dam and adjacent to the spi 11 way control structure. The b 1 ock wi l1 improve the symmetry of the dam profile, as previously stated, and will be stable under load conditions similar to those incurr.ed by the thrust block on the left bank. (f) Construction and Schedule Construction of the dam will be completed over a five-year period as des- cribed in Section 17. Construction will take place throughout the year with cooling coils built into the concrete to dissipate the heat of hydra- tion and special heating and insulation precautions taken in the winter to prevent excessive cooling :vf concrete surfaces.. Concrete aggregates will be obtained from the alluvial deposits in the terraces upstream of the dam. · Concrete will be placed by means of three highlines strung above tha dam between the abutments. 13.6 -Saddle Dam The design philosophy for the saddle dam at Devil Canyon is essentially the same as that for the main dam at Watana described in Section 12.6. The most significant difference is the exclusive use of rockfill in the shells instead of . river gravels used a.t Watana. The use of gravels in the upstream shell at Watana is to minimize settlement of the sh.e11 on saturation during filling of the reservoir and to ensure a free draining material. These aspects of the design are not as significant for the much smaller structure at Devil Canyon. The amount of settlement will be less and the drainage paths for the dissipation of any e·xcess pore pressures will be much reducedo Many dams of equal or larger dimensions have been constructed of similar materials and the design is well within precedent. (a) Proposed Dam Cross Section [,etails of th.e proposed saddle dam are shown in Plate 76. As at Watana, the central vertical impervious core will be protected by fine and coarse filters on both upstream and downstream slopes and supported by rockfill she.lls. The core w111 have a crest width of 15 feet and side slopes of 1H:4V to provide a core thickness to dam height ratio slightly in excess of 0.5. . The wi'de filter zones will provide sufficient q}aterial for self-healing of any cracks which might occur in the core du~ tc settlement or· as the result of seismic displacement. 13-10 _, 0 I I I' I I I I I I I I I I I I I I I I .;.lJ I I I I ·.I I I I I I I I I I I :1 I I I • j .. The saturated sections .of both she.11s ~lill be constructed of compacted clean rockfill, processed to remove fine material in order to minimi.ze pore pressure generation and ensure rapid dissipation during and after a seismic event. Pore pressures cannot develop in the unsaturated .section of the downstream shell and the material in that zone will be unprocessed rockfi11 from surface or underground excavation~~. Protection on the upstream slope will consist of a 10-foot layer of rip- rap. (b) Sources of Construction Material No source of materia 1 suitable for the~ core of the saddle dam has been identified closer than the borrow areas at Watana (Areas D and H). The current proposals are to develop Area D for core material at Watana and, since access roads will be established to that area, the same source will be used for the Devil Canyon core. Investigations to date indicate that suitable material can be obtained from areas above the Watana reservoir 1evele In the unlikely event that insufficient material is available from Area D, then Area H would be de'Jeloped. The in-place volume of core material is 306,000 cubic yards. The.filter matet~ial will be obtained from the river deposits (Area G) immediately upstream of the main arch dam at Devil Canyon. This area will also be exploited for concrete aggregates. The total volume available in Area G is estimated to be 6 million cubic yards, while the concrete aggregate demand is some 2.7 million cubic yards. The estimated voltrnes required for the dam are 228,000 and 181,000 cubic yards for the fine and coarse filters, respectively. · Rockfill for the shells will be obtained primarily from the excavations for the spi 11 ways, tunne 1 s, and powerhouse camp 1 ex. The tot a 1 rock fill required will ·be approximately 1.2 million cubic yards. The proportion of sound rock sui tab 1 e for use in the dam-which can be obtai ned from the exca- vations cannot be accurately assessed at this stage, but it i.s proposed to make good any shortfall· by deepening and extending the emergency spillway ·cut. This will be more economical and environmentally acceptable than developing quarry Area K, some 2 miles from the damsite. (c) Excavation and Foundation Preparation The excavation and foundation preparation will be as for the Watana site with all alluvium and other unconsolidated deposits under' the dam removed to expose sound bedrock to eliminate any risk of 1 i quefacti.on of the dam foundation under earthquake loading. Weathered and heavily jointed rock will be removed from beneath the core~ and filters and local irregular·ities in the rock surface either_ trimmed back or concrete added to provide a suitable contact surfaca for placing the core. (d) Grouting and Pressure Re 1 i tll, '· As at Watana, the rock foundation will be improved by consolidation grout-; ng over the core contact area and by a grouted, cutoff a 1 ong the center 1 i ne of the core. ·The cutoff at any location will extend to a depth of 0.7 the water head at that , 1 ocati on as shown on Plate 75. 13-ll (e) ,, (f) (g) . •, . A grout ·ing and drainage tunnel will be excavated in bedrock. b~neath the dam along the centerline of the core and. will connect with a similar tunnel beneath the adjacent concrete; arch dame Pressute relief and drainage holes will be dr-illed from this tunnel and seepage from -the drainage system will be disehar.g~d into the arch dam drainage system and ultimately downstream be;low tailwater level. ·· · lmpervi ous Gore and Fi 1ters. The requirements for impervious core and both fine and coarse filters will be as for the Watana dam (Section 12.6). Rockfill Shells The processed rockfill to be placed in the saturated zones of the. dam will have the same grading as the processed alluvium used at Watana. The maxi- mllll size shall not exceed 12 inches and not more than 10 percent of the material shall be f·iner than 3/8 inch size. This restriction on fine material will not apply to rockfill in the unsaturated zone above Elevation 1375 in the downstream shell. All rockfill wi:l be placed and thoroughly compacted in 24-inch layers. Freeboard and Superelevation The highest reservoir level will be Elevation 1466 under maximum probable ·flood ( PMF) flows. At this e 1 evat ion the fuse p 1 ug in the emergency spi 11- way will be breached and the reservoir level will fall to the spillway sill elevation of 1434. The norma 1 maxi mum poo 1 e 1 evat ion i s 145 5. It is proposed that a minimum freeboard of three feet be provided for the PMF flood; ·hence, the crest of the saddle dam cannot be lower than Eleva- tion 1469. In addition, an allowance of one percent of the height of the dam will be made for potential slumping of the rockfill shells under seismic loading. An allowance of one foot has been made for settlement of both abutments; hence, the final crest elevations of the saddle dam will be 1470 at the abutments, rising in proportion to the total height of the dam to Elevation 1472 at the maximum section. u·nder normal operating cond1- ti ons, the freeboard will range from 15 feet at the abutments to 17 feet at the center of the dam. Further allowances must be made to compensate for . static settlement of the darn after completion due to its own weight and the effect of saturation of the upstream shell, which will tend to produce additional breakdown of the rock fill at point contacts. It is proposed · that one percent of the dam height be allowed for such settlement, giving a maximum cTest elevation on completion of the construction of 1475 at the maximum height~ and 1471 at the abutments. The a·ll owances for post-construction sett 1 ement and seismic s 1 umpi ng wi 11 be achieved by steepening both slopes of the dam above Elevation 1400. 13-12 · .. I I ' I 1: I I I I I I I I I I I I I I I I I I I I I I I I I I I :I I I I (h} Instrumentation . Instrumentiltion will be jnstalled within all partso of· the dam to provide morlitoring.during cons:truction as \vell as ·durfng operation. Instruments for measuring_-internal vertical and hortzontal displacements, stresses and strains, and total and fluid pressures, as well as surface monuments and markers, will be installed. The quantity and location will be decided dur- ing final design. Instrumentation will include the following: (i). Piezometers ... fo measure static pressure of fluid in the pore spaces of soil and rockfill. (ii) Internal Vertical Movement Devices ... Cross-arm settlement devices as developed by the USBR; -Various versions of the taut-wire devices deve 1 oped to measure. internal settlement; and -Hydraulic settlement devices of various kinds. (iii) Internal Horizontal Movement Devices -Taut wire arrangements; .. Cross-arm devices; -Inclinometers; and -Strainmeters. (iv) Other Measuring-Devices -Stress meters; -Surface monuments and-alignment ,markers; and -Seismographic recorders and seismoscopes. (i) Stability Analyses As .at Watana, special precautions have been taken to ensure stabi 1 'ity under earthquake 1 oading by the use of processed free draining rockfi 11 in t'he saturated zones of the dam, the incorporation of very wide filter zones, and the removal of all unconsolidated natural material from beneath the dam. Static and dynamic stability analyses of the upstream slopes of the Watana dam (Section 12.6), have confirmed stable slopes under all conditions for a 24H:1V upstream slope and a 2H:1V downstream slope. Since these same slopes have been used for the Watana dam and the construc• tion materials are essentially similar, it was considered unnecessary to· carry out further analysis for the specific details of the saddle dam to confirm feasibility, though such analyses will be required during the final design phase. 13-13 13 .. 7 .~ Primary Outlet ~aci lities (a) Gen.eral The p.rfme function of the outlet facilities is to provide for discharge through the main dam of routed floods with ·UP to 1 :SO years recurrence perjod at .the Devil Canyon reservoir.· .. Downstream erosion is to be mintmal and nitr·ugen supersaturation of the releases is to be restricted as much as possible as in the case of.the Watana development .. A further funct.ion of the releases is t_o provide an emergency drawdown for the reservoir, should maintenance be necessary on the main dam or low level submerged structures, and a 1 so to act. as a diversion during the 1 atter part of the construction period as described in Section 13.4. The facilities will be located in ·the lo.wer portion of the main dams as shown on Plate 76A, and will consist of seven free dischnrge valves set in the lower part of the arch dam!' (b) Outlet The discharge valves will be fixed· cone valves located at two elevations: the upper grouping, consisting of four 102-inch diameter valves~ will be set at Elevation 1050, and the lower grouping of three 90-inch diameter valves wi 11 be set at Elevation 930. The valves will be installed radially (normal to the dam centerline) with the upper set centered ott a point slightly downstream of that corresponding to the lower valves. ·rhe fixed cone valves will be installed on individual conduits passing through the dam, set from the downstre·am face, and protected by upstream ring follower gates located in separate chambers within the dam, as shown on Plate 76A. The gates wi 11 serve to i so 1 ate the va 1 ves to all ow main- tenance. Monorai 1 hoists· wi 11 be located above each valve and gate assem- bly to provide for their withdrawal and maintenance. The gates and valves will.be linked by a 20 foot high tunnel running across the dam and into the 1 eft abutment where ac~ess wi 11 be provided by means of a verti ca 1 shaft exiting through the thrust block. Although secondary access will be p~o­ vided via a similar shaft from the right abutment vehicle, access and installation· are both considered to be from the left side. The valve and gate assemblies will be protected by individual trashracks installed on the upstream face. The racks will be removable along guides running on the upstream dam face. The racks wi 11 be raised by a mobile crane normally stationed at Watana but employed for,both sites. (c) Fixed Cone Valves The 102-inch diameter valves operating at a gross level of 420 feet and the 90-inch diameter valve.s operating at a head of 525 feet have been selected to be within current precedent considering the valve size and the static . head on the valve. The valves will be located in individually heated rooms and wi 11 be provided with e 1 ect ric jacket heaters instal led around the cylindrical sleeve of each valve. The valves will be capable .of year round oper·ation~ although winter operation is not co.rltemplated.. Normally when the valves are closed, the upstream ring follower gates will also'be closed to limit leakage and freezing of water through the valve seats. 13-l4 / •... · ,.f ; I I I •• -·: I I I I I I I •• I I I I ··I, ,. .. · .•. )::: 1~--~~··~~~~--~~.~~~~~~~~--~----~--~----~~~~--~~~~~~=<~~~···~··~ .. ~.·~~~-j~= I I· I ----:---;:-~-=-:-:·;>·-~-:---·c-;--""----~--" -~ I -1 (d) •• I I I I I I I -The v~lves will be operated either by twq .:hydraulic ~operators or' by screw stef!l hoists" _ The former hav~ been a;ssumed for pre 1 imtnary design purposes. The valves will normally be -operated remotely fr-om Watana, but local opera- tion i.s also po$sible~ · Thorough research and model studies wi 11 be required for final design of the valves, particularly fn regard to preventing vibration. In $izing the· valves it has been assumed that the valve gate opening will be restricted to 80 percent full stroke to reduc.e~pJ;l$,$_i_qi_1ities of vibration., Ring Fo 11 ower Gates Ring follower· gates will be installed upstream' of each valve and will be used: -To permit inspection and maintenance of the fixed cone valves; ...; To relieve hydrostatic pressure on the valves when they are in the. cl<>sed position; and · .. To close against flowing water in the event of malfunction or failure of the valves. · The ring follower gates will have nominal diameters of 102 and 90 inches and wi 11 be of we 1 ded or cast stee 1 constructi.on. . The gates wi 11 be designed to withstand the total static head with full reservoir. Existing large diameter, high head ring follower gates are suimlarized in Table. 12.6.. . c . The ring follower gates will be designed to be lowered under flowing water conditions with maximum head, although normally they will be raised under b·alanced head conditions only. Valved bypass piping will be used to equal- ize the pressure on both sides of the ring follower gate before raising. The gates wi 11 be operated by hydraulic cylinders with a nominal operating pressure-of 2,000 p.si. Either local or remote operation of the ring follower gates will be possible. A grease system will be installed in each gate for injection of grease between the gate leaf and the gate body seats to reduce frictional fort~s when the gates are operated. I I (e) Trashracks I I I I A steel trashrack ·wi 11 be installed at the upstream entrance to each 'water .passage. The racks wi 11 prevent debris from being drawn into the d·i:scharge valves~ The bar spacing on the racks wi 11 be approximately 7_ inches, and the racks will be designed for a maximum differential head of about 40 feet. The maximum net velocity through the racks will be approximately ft/s. Provision will be made· for monit·oring h~ad loss across the racks. . 13-15 _I 0 {f} Bulkhead Gates " ,. Provision w·iJl be made for installing intake bulkhead gates at the upstream entrance to each of the conduits for the f1xed cone valves._ Embedded· guides will be ins:talled at each· conduit entrance extending to above rnaxt- rnum norma 1 water> 1 eve 1. · the bulkhead gates will be installed only under balanced head conditions using a gantry crane. The gates for the ·upper valves will be 12 feet square and wi 11· be. 10 feet square for the 1 ower va 1 ves. Each gate will have a downstream skinplate and seal and will be designed to withstand full differential head with maximum reservoir water level. One gate of each size has· been assumed and the gates will be stored at the dam crest level. · A temporary cover-will normally be p 1 aced in the bulkhead gate check at trashrack level to prevent debris from getting behind the trashracks at the front of the valve conduit. The crane' for handling the.bulkhead gates will be an electric travelling gantry type crane located on the main dam crest at Elevation 1468. The estimated crane capacity is 40 tons. The crane will have a single point lift hoist mounted on a moveable trolley. The hotst pickup will be incor- porated into a follower for handling the bulkhead gates. " 13.8 -Main Spillway (a) General The main spillway at Devil Canyon will be located on the right hand side of the e.anyon (see Plate 77). The upstream contro1 structure will ·be adjacent to the arch dam thrust block and will discharge down an inclined concrete- lined chute~ constructed down the steep face of the canyon, and then over the flip bucket which will traject flows downstream and into the river below. The right side location for the spillway facilities is considered prefer- rable to the left because of the superior quality of the rock~. with its 1 ower degree of weathering~ and the downstream a 1 i gnment of th.e river·~ which allows for spillway discharges paralleling the direction of flow. The spillway will be designed to pass the 1:10,000 year routed flood at Watana in conjunction with the outlet facilities, giving it a design capa• city of·ll5.,000 cfs, which will be discharged over a total head drop of 550 feet. No surcharge will occur above the normal maximum reservoir operating level of 1,455 feet. This will be below the discharge of 150,000 cfs over 600 feet on the Mica Project, developed for British Columbia Hydro., Canada, which has operated successfully for a number of years .. 13-16 I I I I I I I I I I I I I· I I I I I I '• 0 ·--~~ <.!:--. . .. ~ ' -:<· •l ., I I I I I I I ·•- 1 I I I I 'I I --1 I I (b) Approac)L.Channel and Control Structute The approach channel wi11 be· excav.ated. ttl a depth. of approximately 100 feet in the rock with a wi.4th of just over 130 feet and .an invert ~levation of 1375,.which will produce a flow velocity of approximately 11 ft/s under design dischar-ge. · . · "----~ --· --- The control structure wi1 1 be a three unit concr-ete structure set at the end of the channe 1. · Each unit · wi 11 ·house a· 54 foot high by 30 foot wide gate which will sit on t()p of an agee-crested weir and, in conjunction with the other gate. units, wi 11 control the f1 ows passing though the spillway. The gates wi 11 be fixed wheel gates operated. by individual rope hoists._ Eat:h gate will be contained with a separate monolith unit consisting of an ogee overflow weir, piers, and integral roadway deck. The box configur-a· tion of the unit will give the individual monoliths stability during earth- quake motion, and the "split pier" construction, with each unit having its separate piers, Will allow for some relative motion with no stress trans- ference between units during seismic events and less chance of gate seiz- ure. Model tests will be necessary during the final design stage to deter~ mine final geometry and dimensions of pier noses, crest. slope, and pier lengths. The main access route will cross the dam and the control struc- ture deck upstream of the gate hoists. The main dam grout curtain and drainage system will pass beneath the structure. · {c) Spillway Chute The spillway chute will cut across the steep face of the canyon for a dis- tance of approximately 900 feet and will 'terminate at Elevation 1000~ The chute will taper uniformly over its length from 122 feet at the upstream end to 80 feet downstream. The slope of the bedding planes on the r1ght abutment above the chute will be at an approximate angle of 55° or more. Because of_ the i nstabi 1 ity _along :these planes, the rock above the spillway will be cut back parallel ·to the bedding and the face will be reinforced with steel mesh and rode bolts. , The chute itself will" be concrete-lined W'ith invert and wall slabs anchored btack to the rock. The profi 1 e of the chute wi 11 be such that the invert s 1 abs wi 11 be founded on sound rock. Pa·rt way down the chute on the side closest to the river the depth of cut will be insufficient to provide the supporting rock to the slabs; hence, the side wall will take the form of a gravity section over approximately a 200 foot 1 ength. The velocity at the bottom of the chute wi 11 be approximately 150 ft/s. In order to prevent cavitation of the chute surfaces~ air will be introduced into the discharges. Air will be drawn in along the chute via an under- lying aeration gallery 'and offshoot ducts extendin-g to the downstream side of a raised step running transverse to the chute, as shown on Plate 77 .. The chute will be underlain by a series of box drains at the rock/concrete . interface which will drain through drilled holes to an inclined rock tunnel below running the length of the spillway. 13 ... TT . -·~1 i ' i (d) Flip Bucket The spillway chute will terminate in a mass contrete fl'fp bucket -founded on sound rock at E1 evation 1090. approximately 100 feet above the river. The · cu~ve of the flow ,surface· of the bucket will be adjusted to confine the issutng dfschargej but at present i~ is ass·utned to be<cylindrica1 and will be modified··at the fin_al d~si'gn stage following model tests-.-A grouting drainage gallery will be provided wi.thin the bucket to allow for foundation consolidation and relief of uplift pressures. The jet issuing from the bucket will be traj_ected downstream and parallel to the river below. (e) Plunge Pool The iiJ1lact area of the issuing spillway discharge will be limited to the area of the downstream river surface to prevent excessive erosion of the canyon side walls. This will be done by modification of the flow surface of the flip bucket as described above. Over this impact area the alluvial material in the riverbed will be excavated down to_ sound rock to provide a plunge pool in which most of the inherent energy of the discharges will be dissipated, although some energy will alreaqy have been dissipated by fric- tion in the chute, in dispersion, and friction through the air. It was considered necessary to excavate the river material to determine the general area of any downstream erosion and also to prevent excessive eros- ion and random downstream deposition of material which might occur if dis- charges were allowed to·excavate their own pool. 13.9 -Emergency Spillway (a) General The emergency spillway will be located on the left side of the river beyond the rockfill saddle dam. It \'lill be set within the rock underlying the left side of the saddle and will continue downstream for approximately 2,000 feet. · An erodible fuse plug, consisting of impervious material and fine gravels, will be constructed at the upstream end and will be designed to wash out. when overtopped by the reservoir, releasing floods of up to 160,000 cfs in excess of the combined main spillway and outlet capacities and thus pre- venting overtopping of-the main dam. · (b) Fuse Plug and Approach Channel ·. ; .•... ' ;"- . . ::····· .· .. .-· I I I I I I I I •• •• I I The approach channel to the fuse plug will be cut in the rock and will have I a width of 310 feet and an inv.ert elevation of 2170. The channel will be cr-ossed by the mai nc access road to the dam on a bridge consisting of can- crete piers., pref':ast beams, and an in situ concrete bridge deck. The fuse 'I• plug will close the approach channel and will have~ maximum height of 31.5 feet with a crest elevation of 1465.5. The plug will be located on top of a flat-crested weir excavated in the rock and protected with a concrete ,1· slab. Since the rock slopes quite steeply at the channel location,. it is I . 13-18 ;.. ·I I I I. I I I I I I I I I I I I I ··,· '~· desirable to "keep the. spillway chute :as narrow as possible to reduc~ the excavation quality. For this reason a drop se·ction downstream of the plug will be introduced to ·;ncrea,se the discharge coefficient'S, at the plug · sectlgns and thus enable a reduction in the length of the plug. · The plug will be-traversed by a pilot channel with an 'invert elevation of 1464, and will have a similar zoning to that described in Section 12.10 for Watana. (c) Discharge· Channel The channel will n~rrow downstream and lead to a high tributary valley above the Susitna River. This channel-wi11 rapidly erode under high flows but will serve the purpose of straining the initial f1ows in-the dir""eCtion of the valley. 13.10-Devil Canyo!l Power Facilities (a) Intake - The intake structure is located on the right bank-as shown on Plate 83. Separate intakes are provided for each of the units which will operate for reservoir levels between Elevations 1455 and 1405. Each intake has a single intake gate, a set of steel trashracks, and provision for placing a bulkhead upstream from the gate as shown on Plate 83. A trave 1 i ng gantry crane on the intake deck at Elevation 1466 w:ill service all four intakes. r) The mechanical equipment i.s described in more detail below. The intake is 1 ocated at the end of a 200-foot ... l ong un 1 i ned approach chan- nel. The structure is founded deep in the rock. The rock 'face between in..., takes will be lined with concrete to stabilize the rock surface. The. '9rOUt curtain and drainage holes will continue beyond the main dam and beneath . the structureQ The 60-foot spacing of the intakes was set similar to the spacing of tJJe turbines, which will give parallel penstocks and allow for easier setting out ofthe tunnels during construction. (b) Intake Gates The four power intakes will have a single fixed wheel intake gate with a nominal operating size of 16 feet wide by 20 feet high. The. gates \-till have an upstream skinplate and seal 'and will be operated by hydraulic or wire rope hoists located in heated enclosures immediately below deck level. The gates will normally close under balanced head conditionsto permit dewatering of the penstock and turbine water passages for t-urbine inspection and maintenance. The gates \'lill also be capable of closing under their own weight with fu11 flow conditions and maximum reservoir water level in the event of loss of control of the turbines. A heated air VP.nt will be provided at the intake deck to .satisfy air demand requirements when the intake gate is closed with flowing water conditions. 13-19 (c) lntake Bu1khead Gates . One se.t of lntake bulkhead,_ cons,isting of two ga.te sections will be provided for closing the intaKe openings. The gate will be used to permit inspec- tion and maintenance of the tnt.ake gate and intake gate guidese: ·The gates will be raised and lowered under balanced ·water conditions only. To bal- . -ance water :pressure when ra.ising the intake bulkhead, the space between the gate and the downstream control gate will be flooded by means of a · follower-operated bypass valve on the top gate section·. An. air valve will be provided in the top of the gate. The gates will have a downstream skin-- plate and seal on the downstream Side. They wi11 be designated to with- Stilnd full differential pressure. () (d) Trashracks Each of the four intakes· will incorporate trashracks just upstream from the maintenance gates. The trashrack opening will have a bar spacing of about 6 inches and designed for a maximum differential head of about 30 feet. The m~ximum gross velocity through the racks will be about 4 ft/s. Each trashr ack wi 11 be constructed in two sections for remova 1 by means of a follower suspended from the intake gantry crane• (e) Intake Gantry Crane - A 60-ton capacity (approximately) electrical traveling gantry crane will be provided on the intake deck at Elevation 1466 for handling the trashracks~ maintenance gates and intake g3tes. The crane will incorporate a double point lift hoist mountedon an enclosed trolley. The hoist pickup \'lfll be incorporated into a follower for handling the gates and trashracks. The crane will also have a grappling hoist with a grapple of approximately 5-ton capac.ity for removing debris from in front of ~he trashrack. 13.11 -Penstocks The power plant will have four penstocks, one for each unit. The maximum static head on each penstock is 638 feet, measured from normal maximum operating level (Elevation 1455) to centerline distributor level (Elevation 817). An allowance· of 35 percent has been made for pressure rise in the penstock under transient conditions, giving a normal maximum design head .of 861 feet. Maximum extreme head, corresponding to maximum reservoir flood level, is 876 feet. The penstocks have been designed as individual concrete-lined rock tunnels from the intake to the powerhouse~· The section 200 feet upstream of the powerhouse is steel lined.. The inclined sections of the concrete-lined penstocks are at ss~ to the horizontal. (a) Steel Liner It has been assumed that the rock adjacent to the powerhouse cavern wi11 be incapable of long-term restraint against the forces transmitted from the penstock hydraulic pressures. The first 50 feet of steel liner will there- fore be required to resist the design pressure without contributing from the surroundfng concrete. For the remainder of the steel liner, which extendes a further 150 feet upstream, allowance is made for partial rock 13-20 / ·; .. ~, :1-J . '1 ·.JI I I I I I I I I I I I ·' ' _. _:. I I' I I •••• •••••• ••• • j I -I . _>. •• I I I I I .I I I I· I I I I I I s.upport to .reduce the stee 1 stress. For pre 1 i mi nary design, it is ass.umed that not more than 50 per.cent of the design head is taken by the .rock support over this transition length • . ' Beyond the steel liner, the hydraulic loads are taken solely by the rock tunnel with a concrete. linE!r. . . ' ' . . The steel liner is surrounded by a concrete infi11 with a minimum thickness of 24 inches. A preliminary analyses has evaluated that the optimum inter- nal diameter of the ·stee.l lining is 15 feet, based on the minimum total cost of construction and the capitalized value of energy reduction due to head loss. A tapering steel transition has been provided at the junction between the stee 1 1 i ner and the concrete 1 i ner to increase the inter na 1 diameter from 15 feet to 20 feet. (b) Concrete Liner The penstocks are fully 1 i ned with concrete from the intake to the steel lined section adjacent to the powerhouse; the thickness of lining will vary with the design head. The minimum thickness of lining is 12 inches. Based on preliminary analyses, the optimum internal diameter of the con- crete liner is 20 feet. (c) Grouting and Pressure Relief A comprehensive pr"essure re 1 i ef system is required to protect the unde.rground caverns against seepage from the high pressure penstocks. The system will comprise of small diameter boreholes set out in patterns and curtains ,to intercept the jointing in the rock. · . Grouting round the penstocks will be provided to: -fill and seal any voids between the concrete infill and the steel liner which may be left after the concrete placing and curing; and, -fill joints or fractures in the rock surrounding the penstock;} to reduce flow into the pressure relief system and to consolidate the r.)ck. 13.12 -Powerhouse {a) General The powerhouse complex will be constructed underground in the right abut- ment. This will require the excavation in rock of three maj'or caverns (powerhouse, transformer gallery and surge chamber), with interconnecting rock tunnels for the draft tubes and isola ted phase bus ducts. An unlined rock tunnel will be required for vehicular access to the three main rock caverns. A second unlined rock tunnel will provide· access from the powerhouse-to the foot of the arch dam!.! for routine maintenance on the fixed cone valves;~this tunnel will also provide construction access to the. 1 ower section of the penstocks •. 13-21 / Verti,cal shafts wit 1 be required for personnel access by elevator to the underg,round powerhouse; for oil filled cable from the transformer gallery~ and for surge chamber ventingG The draft tube: gate gallery and cavern are 1 ocated in 'the surge chamber"" cavern, above maximum design surge level. · The general layout of the powerhouse complex is shown on Plates 85, 86 and 87. The transformer gallery is located upstream of the powerhouse cavern; the surge chamber is located downstream of the powerhouse cavern. The clear spacing between the underground caverns is at least 1.5 times the main span of the larger excavation', from geotechnical considerations. (b) Layout Considerations The powerhouse is located underground in the right abutment. ~later for power generation is taken from an intake structure to the right of the main spillway, and carried through 1 ndi vi dua 1 penstocks to the turbines. \~ater is discharged to the river by a single tailrace tunnel 6800 feet in length. The draft tubes and tailrace tunnel are protected against excessive tran- sient pressure rise by a downstream surge chamber, which also provides storage for the turbine start-up sequence~ The intake structure is dP;signed for a maximum drawdown of 50 feet and is located close to the mairt arch dam thrust block for ease of access. The powerhouse·is located t:J provide the minimum total length of penstock, as- suming an inclination of 55° to the horizontal for the sloping section of penstock~ The orientation of the powerhouse has been selected as a compro- mise between the desired orientation for power flow (E-W) and the geotech- nical data on known shear zones and joint sets. Minimum clear spacing be- tween major rock excavations is at least 1.5 times the span of the larger_ excavation. This is considered a conservative estimate for preliminary design purposes. The downstream surge chamber will· be constructed as close as possible to the powerhouse for maximum protection to the draft tubes under transient load conditions. For this reason the underground transformer gallery will be located upstream from the powerhouse" The rock around the powerhouse cavern and transformer gallery is protected against high pressure seepage from the penstocks by a 200-foot continuous steel-lining and an extensive pressure relief system. (c) Access Tunnels and Shafts Vehicle access to the underground facilities at Devil Canyon is provided by two unlined rock tunnels. The main access tunnel, 3,000 feet long, con- nects the powerhouse cavern at Elevation 852 with the canyon .access road on the right bank. A secondary access tunne 1 runs from the main powerhouse access tunnel to the foot of the arch dam, for .routine maintenance of the fixed cone valves. A branch tunnel from the secondary access tunnel will provide construction access to the lower, section of the penstocks, at Ele- vation 820. Separate branch tunnels from the main access tunnel give vehicle access to the transformer gallery at Elevation 896 and the draft tube gate .gallery at Elevation 908. The maximum gradient on the permanent 13-22 0 I I I I I I I I I I 1: I I ., I I I ••• t I I· I I I I I I I I I I I I I :1 I I (d) access tunnel is 8 percent; the maximum gradient on the secondary acces.s tunnel is 9 ·percent. · · . The <:ross section of the access tunnels is dictated by requirements for ·con'StPuction plant; for preliminary desfgn a modified horses·hoe shape35- feet wide by 28-feet high has been used .. The main access shaft is lot:ated at the north end of the powerhouse cavern, providing personnel access by elevator from the surfaceo · Horizontal tun- nels are provided from this shaft for pedestrian access to the transformer gallery and the draft tube gate gallery(> At a higher level access is also available to the fire protection head tank. Access to the upstream grouting gallery is from the transformer gallery main access tunnel, at a maximum gradient of 13.5 percent. Powerhouse Cavern _.;...a;;,..;;;.;;.;;.;;;..;;.....;;.;;...;..;;;;.;...;..;. The main powerhouse cavern is designed to accommodate four vertical shaft Francis turbines, in Jine, with direct couplitig to overhung generators. Each unit is d~signed to generate 164 MW at 575-foot head. The overall height of the cavern is governed by the physical size of the turbine and generator, space requirements for miscellaneous equipment and s~rvices, the design dimensions of the turbin~ draft tube, the overhead travelling crane clearance and size, and the rise of the roof arch. The selected unit spacing is 60 feet; in addition, a 110-foot service bas has been allowed at the south end of the powerhouse for routine maintenance and construction erection. The 1 ocal control room is 1 ocated at the nor·th end of the main powerhouse flooro The width of the cavern allows for tne physical size of the generator plus galleries for piping, air-conditioning ducts~ electrical cab1es, and isolated phase bus. The penstock steel-liner is continuous with the turbine spiral case; no penstock inlet valves al~e provided in the po~1erhouse. Continuous drainage galleries are provided to a low-level sump. Compensation flow of 500 ft3/s is required to the river immediately downstream of the arch dam, in view of the length of the tailrace. tuooel ( 6800 feet). This flow is provided by two No. 1300 h p vert i ca 1 shaft. imi xed flow pumps, installed in a gallery below the service bay. Each pump is rated at 115,000 gpm at 35-'foot h~ad. Water ·is taken from the base of the surge chamber and pumped 1000 feet to the dam through a steel pipe 1ai<'1 partly in the secondary access tunnel and partly in a separate outlet tunnel. Multiple stairway access points are aV<:tlable from the pOwerhouse main floor to each ga 11 ery 1 eve 1. Access to the transformEr ga 11 e~'Y from the powerhouse is by a tunnel from the main access shaft or by a stairwaJ through each of thr! four i so 1 a ted phase bus shafts. Access is a 1 so avai'lab.le to the dt·~ft tube gate gallery by a t::~nnel from the main access shaft. · A service elevator is provided for access from the service bay area on the main floor to the machine shop~ and the pumping and drainage ga11eries on the lower floors. Hatches have been PY"ovidtad through a11 main floors for in$t.a.11 ati on and_ routine maintenance of pum}'s, valves and other heavy • < .. equi pm~nt using the main overhead travell ino crane •. (e) Transformer Gallery The transformers are located underground in a separate unlined rock cavern, 120 feet upstream of the powerhouse cavern, with four interconnecting tunnels for the isolated phase bus. There a1re 12 single-phase transformers in four groups of 3.! one group for each generating unit. Each transformer is rated at 13/345, 70 MVA. For increased rteliability, one spare transfor- ~r and one spare HV circuit are provided. The station service transfor- r.~~~rs (2x2 MVA) and the surface faci1iti.es transformers (2 x 7.5/10 MVA) are located in the bus tunnels. Generator excitation transformers are located on the main powerhouse float"'. High voltage cables are taken to the surface in t~'lto cable shafts, each 7 feet-6 inch internal diameter; provision is made for an inspection hoist in eat;h shaft. Vehicle ac;cess to the transformer ga.ll ery is from the south end vi a tbe main powerh~use access tunnel. Personnel acc1ess is from the main access sha?,; or through each of the four isolated phi3Se bus tunnels. (f) Surge Chamber A simple surge chamber has been provided 120 feet downstream of the power- house to control pressure rise in the turbine draft tubes and tailrace tunnel under transient load conditions, and to provide storage for the machine start-up sequence. The chamber.is common to all four draft tubes and the inlet pipe to the. compensation flow pumps. The chamber design is gt>verned by an assumed full load rejection surge and the requiremen~ for ; nci pi ent stabi 1 ity under part load operation, together with estimated floor levels from the tailwater rating curve. The .:~raft tube gate gallery and crane are 1 ocated in the same cavern~ above the maximum anticipated surge level. Access to the draft tube gate gallery is by a rock tunnel from the main access tunnel. The tunnel is widened locally for storage of the draft tube gates. The chamger is generally an un 1 i ned rock ex~:avati on with 1 oca 1 i zed rack support as neces~ary for stability of the roof arch and walls. The guide b 1 ocks for the draft tube gates will be of reinforced concrete anchored to the rock excavation by rockbolts. 13.13 -Reservoir The Devil Canyon reservoir, at a normal operating level of 1455 feet, will be approximately 26 miles long with a maximum width in the order of 1/2 mile.. The total surface area at normal operating level is 7800 acres. Just upstream of the dam, the ~aximum water depth will be approximate1y _feet. The min·imum reservoir level will be 1405 fe9t during norma.l operation, resulting in a 13-24 ,, ·I" I I I -I I I I I" I I 1- ·1 I I I •••••• ••• ' I I I I I I I I' I I I I I I I . I maximl!m drawdown of 50 feet. The reservoir will have a· total capacity of 1,090,000 acre feet of whien 420,000 acre feet will be 1 ive storage. Prior to reservoir filling, 'the area below elevation 1460, five feet above maxi- mum operating level, will be cleared of all trees and brush, ·A field reconna-is- sance of the proposed reservoir area was undertaken as part of these studies. This work inc 1 uded ex.ami nation of aer i a 1 photographs and maps; an aer i a 1 over- flight of the reservoir and collection of recent {1980 field season) forest inventory data from the U~ s. Forest Service. As described for the Watana reser- voir, most of the vegetatal material within the reservoir consists of trees with very little in brush. The trees are quite small, and the stands are not very dense. In the Watana reservoir area, an es-timated 3,200,000 cubic feet of \'load exi.sts averaging approximately 500 cubic feet of low co11111ercial quality, and some very significant logging problems would be posed by the ste~p slopes and incised terrain excavated. Approximately 87 percent of the available timber are soft woods.. The results of the timber reconnaissance studies are described in more detail in Appendix C3. The combination of steep terrain, moderate-light tree stocking levels, small trees, erosive potential of the reservoir slopes, remoteness, and very restrict- ed access to the reservoirs are major factors affecting the choice of harvesting systems to be utili.zed for this project as discussed in Section 13.14. Present market d~mand for the timber at Susitna is low, however, theworldwicie demand for wood fluctuates considerably. It is anticipated that use of the har- vested material would be limited to either sale as wood-waste products and as fuel. Slash material including brush and small trees, which will be suitable for either of the above uses, will be either burned in a carefully con· ""Olled manner consistent wth applicable laws and regulations, or hauled to a dis~~sal site in and adjacent to the reservoir. Material placed in disposal areas will be cover- ed with an earthfill cover sufficient to prevent erosion and subsequent expo- sure. A number of unstable areas will undoubtedly result during reservoir operation. These areas will require remedial t~eatment depending on the nature and extent of the instab~lity. · 13.14 -Tailrace Tunnel ? The tailra~e pressure tunnel is provided at Devil Canyon to carry power plant discharge from the surge chamber to the river. The tunnel has a modified horse- shoe cross-secti.on with a major internal dimension of 38 feet, and for prelimin- ary des·fgns it is assumed to be concrete lined throughout with a minimum thick- ness of 12 inches. The length of the tunnel is '6800 feet .. The size of the tunnel was selected after an economic study of the cost of construction and the ·capitalized value of average annual energy losses caused by friction, bends and changes of section. Since the size of the surge chamber is related to the effective diameter of the tailrace tunnel, the cost of the surge chamber was also included in the optimization studies as a function of tunnel size • 13-25 " ~.-····~ w •• -• The tailrace portal site has been located at a prominent steep ;,~ock face on the right bank of the river to provide the required tunnel :cove.r (about 60 feet) in as short a di,stance as ·po.ssible. The portal provides a gradual transition from the tunnel modified horseshoe shape to a rectangular cross-section at the. outlet; it also reduces the maximum outlet velocity to 8ft/sec_, to reduce the _velocit-y head loss at exit.. Vertical stoplog· guides·-·are provided for closure of the tunnel if required for tunnel inspection and/or maintenance. 13.15 -Turbines and Generators (a) Unit Capacity The Devi 1 Canyon powerhouse wi 11 have four generating units with a nominal capacity of 150 MW. This is the available capacity with minimum December. reservo~r level (El. 1393) and a corresponding gross head of 553 fe"et in the station. The head on the plant will vary from 605 feet maximum (597 feet net head) to 550 feet minimum (538 feet net head). Because maximum turbine output varies approximately with the 3/2 power of head, the maximum unit output will change with head as shown in Figure 13.2. The rated head for the turbine has been established at 575 feet, which is the weighted average operating head on the station. Allowing for generator 1o.sses, this results in a rated turbine output of 225,000 hp {168 MW) .. The generator rating has been se 1 ected as 180 MVA with a 90 percent power factor, which corresponds to a power output of 162 MW. The ~enerators will be capable of continuous operation at 115 percent rated power. Because of the high capacity factor for the Devil Canyon station, the units will be operated at or near full load a 'I arge percentage of the time. The genera- tors have therefore been sized on the basis of maximum turbine output at maximum head, allowing for a possible 5 percent addition in power from the turbine. This maximum turbine output (250,000 hp) is within the continuous overload rating of the generator. (b) Turbines The turbines wi l1 be of the vertical shaft Francis type with steel spiral casing and a concrete elbow type draft tube. The draft tube will have a single w,ater passage (no center pier). rne rated output of the turbines wi 11 be 225,000 hp at 575 feet rated net head. Maximum and minimum heads on the units wi 11 be 597 feet and 538 feet~ respectively. The full gate output of the turbines wi 11 be about 240,000 hp at 597 feet net head and 205,000 hp at 584 feet net head. Over- gating of the turbines may be possible, providing approximately 5 perct:nt additional power. For preliminary design purposes, the best efficiency (best gate) output of the units has been assumed at 85 percent of the full gate turbine output. This wi11 be reviewed at the time of preparatiar• of bid documents for the turbines. J3-26 .·, 1 .. · ' . ' ~ -, .-· ..•. _·.-· I I •• "I I I I' I -1 I 1: I I I I I ~ I I I I I -I I I I I I •• ~· I I 'I I I I The full gate and best gate· efficiencies of the turbines will be about 91 percent and 94 percent, respectivel.y, at rated head. The efftciency wi 11 . be about 0.2 percent lower at maximum head and 0.5 percent lower at minimum head. The pre.liminary performance curve· for the turbine is shown in Figure 13.3 .. A speed of 225 rpm has been selected for the unit for preliminary design purposes. The resulting turbine specific speed (N 5 ) is 37.9. As shm·m in Figure 12.23, this is wtthi n present day_ practice for turbines operating under 575 feet head. The considerations for selection of turbine speed are briefly discussed in Section 12"16. On the basis of information from turbine manufacturers and the studies on the power plant layout, the centerline of the turbine distributor has been- set at 30 feet below minimum tailwater level. The final setting of the unit will be established in conjunction with the turbine manufacturer after the contract for the supply of the turbine equipment has been awarded. The mechanical/structural designs of the turbines wi 11 be basically the same as for Watana. Because of the relatively short penstocks and the surge tank location trrmedi ately downstream from the powerhouse, the hydrau- lic transient characteristics of the turbines are favorable. Assuming nor- mal gen-erator inertia (H = 3~5 MW-Sec/MVA), a preliminary analysis has in- dicated the following: -Water starting time (Tw) ............................. 1.2 sec. -Mechanical starting time {Tm) •• ,. •...••.•..•.••..•.. 7.6 sec. -Regulating ratio (Tm/Tw) ........................... 6.3 -Governor time.-....................................... 5.0 sec .. -Speed rise on full load rejection . • . • • • • . • .. • • . • • . . 35 percent -Penstock pres~ure use on full load rejection ••••.•• 20 percent The regulating ratio is above the minimum recorrmended by the USBR for good regulating. Also~ unit speed rise and penstock capacity pressure ri:se are within normal accepted values. Because of the relatively short distance between the turbine and the tailrace surge tank and the deep unit setting, there should not be any problems with dr·aft tube column separation. As discussed in Section 12.16 for Watana, the units will be capable of op- eration from about 50 to 100 percent load. Considerations for draft tube surges and corresponding power swings as mentioned for Wataoa also will apply to Devil Canyon. As with Watana, the relationship between generator natural frequency and the possi b 1 e draft tube surge frequency is desi rab 1 e and wi 11 require study in later design stages. 'Becaase of the high capacity factor for the. Devi 1 -Canyon units~ part load operation for these turbines is not as critical as at Watana; therefore, the possibility of problems with power swings is somewhat less of a concern than at Watana. (c) Generators The four generators in the De vi 1 Canyon powe.rhouse wi 11 be of the vertical shaft, overhung type directly connected to the vert·i ca.l Francis turbines. 13-27 (d) The generators will be similar in construction and design to the Watana g~nerators and the gener.al features described in Section 12.16 for the stator, rotor, excitation system, and other details which apply for the Devil Canyon generators. The rating and characteris:ties of the generators are as· follows: Rated CapacitY: Rated Power: Rated Voltage: Synchronous Speed: Inerti-a Constant:. Short Circuit Ratio: Efficiency -at Full Load: Governor System 180 MVA, 0.9 power factor with overload rating of 115 percent. 162 MW 15 kV» 3 phase, 60 Hertz 225 rpm 3.5 kW -Sec/kVA 1.1 (minimum) 98 percent (mini mum) A governor system with electric hydraulic governor actuators will be pro- vided for each of the Devi 1 Canyon units. The system wi 11 be the same as for Watana. 13.16 -Miscellaneous Mecryanical Equipment (a) Compensation Flow Pumps The two pumps for providing minimum discharge into the Susitna River be- tween the dam and the tailrace tunnel outlet portal wi 11 be vertical mixed flow or axial type located in the powerhouse service bay below the main a erection floor, ~s shown on Plate 87. Each pump will be rated at 250 cfs (115~000 ga 1/min} at 35 feet total head, and wi 11 be driven by 1,400-hp induction motors. The preliminary pump and motor data is summari.zed in Table 13.3. A single pump intake will be ··located in the surge chamber with an 8-foot·- dlameter intake tunnel leading to the powerhouse. The intake tunnel will bifurcate into individual pump intake conduits within the powerhouse. The pump discharges wi 11 converge into a single pump discharge tunne 1 .. Butterfly type valves wi 11 be installed in the intake-and discharge 1 ines. of e.ach pump to permit isolation of a pump for inspection or malntenance. Trash screen guides and a trash screen wi 11 be provided in the surge cham- ber at the pump intake. It will be possible to remove the trash screen us- ing the draft tube gate crane discussed below. The width of the guides . 13-28 I I I ' I' I I I ., I I I I I I! I •• I I II I I· I I I I I I I I I I I I I I •• I · .. I will be select-ed so that one of the turbine draft tube gates may be installed in the intake to permit_ d~wateri ng the pump intake tunne 1 for inspection and/or maintenance .of the tunnel or the intake butterfly valves. Stop 1 og guides and a set of stop 1 ogs will a Tso be provided at the down- stream end of the pump discharge tunnel to allow.the discharge tunnel to be dewat.ered. The stoplogs: will be handled with a mobile crane and a follower. Pumping operation wi 11 be continuous; therefore, pumping equipment wi 11 be . conservatively designed to provide efficient operation with minimal main- tenance. Crane access will be provided. for the pumps, motors, and valves to permit equipment servicing. In the detailed design stages, consideration should.also be given to turbine-driven rather than electric motor-driven pumps. A header from at least two of the main turbine penstocks would supply water to the turbines, with the turbine draft tubes connected to the pump discharge. {b) Powerhouse Cranes . Two overhead type powerhouse cranes will be provided at Devil Canyon as at Watana. The estimated crane capacity will be 200 tons. (c) Draft Tube Gates ·Draft tube gates wi 11 be provided to permit dewatering of the turbine water passages for inspection and maintenance of the turbines. The arrangement of the draft tube gates will be the same as for Watana, except that only two gates will be provided, each 21 feet by 21 feet. At the time of start- ing of Unit l, one gate will be installed in Unit 4 with the other gate available for Unit L. Bulkhead domes.wil'! te installed in Units 2 and 3. {d) Draft Tube Gate Crane A crane will be installed in the surge chamt:-er for installation and removal of the draft tube gates. The crane will be either a monorail (or twin monorail) crane or· a gantry crane~~ For the preliminary design, a twin monorail crane of approximately 25-ton capacity has been assumed. The crane will be pendant-operated and have a two point lift. A follower will be used with the crane for handling the gates. The crane runway will be located along the upstream side of the surge chamber and will extend over the intake for the compensation flow pumps, as well as a gate unloading area at one end o·f the surge chamber. (e) Miscellaneous Cranes and Hoists In addition to the powerhouse cranes and draft tube gate cranes, the fol- lowing cranes and hoists will be provided in the power plant: - A 5-ton monora.il hoist in the transformer gallery for transformer mair~­ tenance; ,_Small overhead jib, or A-frame type hoists in the machine shop for handling material; and 13-29 ,I -·A-frame or monorail hoists in other powerhouse areas for handling small ··1· equipment. (f) Elevators . Access . and service eleva tors will be provided for the power p1 ant as follows:. ... Access elevator from the control building to the powerhouse; -Service elevator in the powerhouse service bay; and -Inspection ho4sts in cable shafts. The elevators will be as discussed in Section 12.17 for Watana. (g) Power Plant Mechanical Service System~ The power plant mechanical service systems for Devil Canyon will be essen- tially the same as discussed in Section 12.17 for Watana, except for the following: -There will be no main generator breakers in the power plant; therefore, circuit breaker air wi 11 not be required'!t The high-pressure air system wi 11 be used only for governor as we 11 as instrument .air. The operating pressure will be 600 to l,OOOpsig depending on the governor system oper- ating pressure. An air-conditioning system will be installed in the powerhouse control room. -For preliminary desigrJ purposes only, one drainage and one dewatering sump have been provided in the powerhouse. ihe dewatering system wi11 also be used to dewater the intake and discharge lines for the compensa- tion flow pumps. (h) Surface Fac_ilities Mechanical Service Systems The entrance building at the top of the power plant wiil have only a heat ... ing and ventilation system. The mechanical services in the standby power building will include a heating and ventilation system, a fuel oil system, and a fire protection system, as at Watana. (i) Machine Shop Facilities A machine shop and tool room will be located in the powerhouse service bay area to take care of maintenance work at the plant. The facilities will not be as extensive as at Watana. Some of the" larger components will be transported to Watana for necessary machinery work. 13.17 -~cessory Electric.~) Equipment {a) General .' ! The accessory electrical equipment described in this section includes the following main electrical equipment: 13-30 I . I I I I I I I I I I I I I I I I ¥ ~ ~ I I I I I I I I I I •• I I I _I I I "(b) 0 .. Main generator step-up 15/345 kV t~ansformers; -Isolated phase bus connecting the generator and transformer"S; -345 kV oil-filled cables from the transformer terminals to the switch- -yard;. -Control systems; and -Station service auxiliary ac and de systems. Other equipment and systems described include grounding, lighting system and communications. The mai-n equipment and connections in the power plant .are shown in the Sing·le Line Diagram, (Plate 88}. The arrangement of equipment in the PO\'Ierhouse, transformer gallery, and cable shafts is shown in Plates 85 to 88. General Design Censiderations for Transformers and HV Connections (i) General Twelve single-phase transformers and one spare transformer will be located in the transformer gallery. Each bank of the three single- phase transformers will be connected to one generator by i so 1 a ted phase bus located in bus tunnels. The HV terminals of the transfor- mer will be connected to the 345 kV switchyard by 345 kV single- phase oil-filled cables installed in 800-foot long vertical shafts. There will be two sets of three single-phase 345 kV oil-filled cables installed in each cable shaft. One additional set will be . maintain2d as a spare three-phase cable circuit in the second cable shaft. These cable shafts will also contain the control and power cables between the powerhouse and the surface control room, as well as emergency power cables from the d·iesel generators at the surface to the underground facilities. As described·in Section 12.18 for the Watana power plant, a number of considerations led to the choice of the above optimum system of transformation and connections. Different alternative methods·and equipment designs were also considered. In surrmary, these are: -One transformer per generator versus one transformer for two gen~· erators; -Underground transformers versus surface transformers; .. Direct transformation from generator voltage to 345 kV versus intermediate step transformation to 230 kV or 161 kV3 and thence to 345 kV; -Single phase versus three-phase transf()rmers for each alternative. method considered; and -Oil-filled cable versus solid dielectric cable or.SF6 gas- insulated bu~. · Reli.ability considerations are based on the general reliability requirements for generation and transmission described in Section 15 regarding ·the forced outage of a sing)e generator, transformei"', bus or cable in addition to planned or scheduled .QUtages in a single 13-31 (c) (d) 1.. , ... , ..... . contingency_situation,~ or a subsequent outage of equipment in the double contingency sitU.ation. In the first case~ the system should be capable of readjustment after the outage for loading 'llithin normal ratings and, in the second case,-within emergency rati n~s. The one transformer per generator scheme was selected since. the __ operation of the Devil Canyon power plant wi11 e-ssentially -be a con- tinuous base-load type operation; also the smaller riu.--nber of units at Devil Canyon compared to Watana will allo~l a transformer ga11ery of reasonable length for a unit generator-transformer scheme. A~ at Watana, transport limitations for both dimensions and weight wi 11 preclude the use of the 1 arger size three-phase transformers; hence, single-phase transformers will be used. One distinct advan- tage of single-phase transformers is that a spare transformer can be provided at a fairly low incremental cost. For the same reasons as given in Section 12.18 for Watana, surface transforme\ ... S and the double-step transformation scheme {15/161 kV generator-transformer, 161 k V cab 1 e and 161/345 k V auto-transformer at the switchyard) were ruled out. The direct transformation (15/345 k\1) scheme with 345 kV oil-filled cables is considered a better over a 11 scheme. Main Transformers The transformers \'lill be of the single phase, two-winding, oil-immersed, forced-oil water-cooled (FOW) type. A total of twelve single-phase transformers and one spare transformer wi 11 be provided, with rating and characteristics as 'follows: Rated capacity: High Voltage Winding: Basic Insulation Level (BIL) of HV Winding: Low Voltage Winding: Transformer Impedance: 70 MVA 345/ 3 k V, .grounded Y 1300 kV 15 k V, Delta 15 percent The design and construction details are identical to the transformers at Watana as described in Section l2a18. Generator Isolated Phase Bus Isolated phase bus connections will be located between the generator and the main transformer. The bus will be of the self-cooled, welded aluminum tubular type with design and construction details g2nerally similar to the bus at the Watana powr~r plant. The rating of the main bus is as follows: Rated current: Short circuit current momentary: Short circuit current symmetri ca 1 : Basic Insulation Level (BIL): 13-32 9,000 amps 240,000 amps 150,000 amps 150 kV I I I I I I I I I I I •• I I I I I I I, I I ,~I a· I I I I I I I I I I I I I· I I I .,: ::.-,.·· (ia) 345 kV Oil-Filled Cable The general design considerations leading to the choice of the 345 kV Oil- filled cable for the connections between the transformer HVterminals and the 345 kV switchyardat the surface are the same as described in Section . 12.18 for the Watana plant. The cables will be rated for a continuous maximum current of 400 amps at 345 kV ±.5 percent. The cables will be of single-core construction with oil flowing through a central. oil duct within the copper conductor. The cable.s will be installed in the 800-foot cable shafts from the transformer gallery to the surface. No cable jointing wi1:1 be necessary for this installation length. (f) Control Systems ( i} Genera 1 The Devi1 Canyon power plant \<if 11 be designed to be operat@d as an unattended plant. The plant will be normally controlled through supervisory control from the Susitna ftrea Control Center at Watana~ The plant will, however, be provided with a control room with suffi- cient control, indication, and annunciation equipment to enable the p 1 ant to be operated during emergencies by ore operator in t\ie con- trol room~ In addition, for the purpose of testing and commission- ing and maintenance of the plant, local contr·ol b()ards will be mounted on the powerhouse fl oar near each un~~t. Automatic load-frequency control of the four units· at OeviT Canyon will be accomplished through the central computer-aided control system 1 ocated at the Watana Ar'~t:t ;Control Center. The power plant will be provh~t:.'.J with 11 black start11 capability similar to that provided at Watana, to enable the start of one t-.nit without any power in the powerhouse or at the switchyard·, except that provided by one emergency diesel generator. After the. start-up of one unit, auxiliary station service power will be established in the power plant and the switchyard; the remaining generators can then be started ant: after the other to bring the plant into full output within the hour. · As at the Watana powe~ plant, the control system will be designed to permit local-manual or local-automatic starting, voltage adjusting, synchronizing, and loading of the unit from the powerhouse control room at Devil Canyon. The protective relaying sy.stem is shown in the main single line diagram, Plate 88, and is generally simi 1 ar to that provided for the Watana power plant. 13-33 (g) ,;;.· I_ (h) ----..,;:-;:-:-.--;:. ,-:· --·-. Station Service Au xi. 1 i ary AC and DC Systems ( i ) AC Auxi 1 i ary System The station service system wi 11 be desi ~ o-· achieve a r·e li ab 1 e and economic distribution system for the r plant .and the switch- yard and surface facilities. The auxiliar.;' system will be similar to that in the \~at ana power plant except that the swi tchyard and surface faci 1 it i es power wi 11 be obtai ned from a 4.16 k V system supplied by two 5/7.5 MVA, OA/FA, oil-immersed transformers connect.- ed to generators Nos. 1 and 4, res;:H~ctively. The 4.16 l<V double- ended switchgear will be located in the powerhouse. It will have a normally-open tie·breaker which will prevent parallel operation of the two sections. The tie breaker will close on failure of one or the otJer ·of the incoming supplieso The 1400 hp compensation f1ow pumps \\\ill be supplied with power directly from the 4.16 kV system. Two 4· .. J5 cables installed in the cable shafts will supply power to the St'i"face facilities.. · The 480 V station service system will be exactly similar to the Watarta system described in Section 12.18, and will consist of a main 480 V switchger.r, separate avxi 1 i ary boards for each unit, an essen- tial auxiliaries board, and a general auxiliaries board. The main 480 V switchgear will be supplied by two 2000 kVA, 15,000/480 V grounded v-1ye sealed gas dry-type transformers. A third 2000 kVA transformer will be maintained as a spare. Two emergency diesel generators, each rated 500 kW, will be connected to the 480 V powerhouse main switchgear and 4.16 kV surface switchbo.ard, respectively. Both diesel generators will be located at the surface. An uninterruptib le high-security power supply wi 11 be provided for the supervisory computer-aided plant control systems. (ii) DC Auxiliary Station Service System The de auxiliary system will be similar to that provided at the \~atana plant and \'lill consist of two 125 V de lead-acid batteries. Each battery system wi11 be supplied by a double rectifier charging system. A 48 V de battery system wi 11 be provided for supplying the supervisory a!'ld communications systems. (iii) B1 ack Start Capabi 1 ity As at the Watana power plant, the Devil CatJ.YOn power plant wi 11 be provided with "black start 11 capability which wi11 enable the plant to start un in a completely "blacked out" condition of the power plant and/or the power system. Other Accessory Electrical Systems The other accessory electrical systems including the grounding system, 1 ignti ng system, and powerhouse communications system will be similar in general design and construction aspects to the: systems described in Section 12.18 for the Watana power plant. 13-34 I I. .I ;-; I I. I I I I I- I I I I I I J. I ,Ji -'--' .I I I I I I I I I I I I I I I I I I .I I 13.18 -Swi tchyard Structure.s and Equipment To follow 13.19 -Project Lands Project lands acquired for the project will be the minimum necessary to constru<. ~ ·access and site faci li ties, construct permanent faci li ties, to c 1 ear the reservoir, and to operate the project .. Appendix C contains 1 and status background information re 1 ati ve to the Susi tn a Project, together with an inventory of private and public lands required for the project. A 1 arge amount of pub 1 i c 1 and in the Devi 1 Canyon area is managed by the Bureau of Land Management. There are large blocks of private Native Village Corporation tands along the river. Other private holdings consist of widely scattered remote parcel5. The state has selected much of the federal land in this area and is expected to receive a patent. 0 13-35 :; ·, TABLE 13.1: WATANA PE!~K WORK fORCE AND CAtllP/VILLAGE DtSIGN POPULATION Calendar Year- 1992 1993 199~. 1995 1996 1997 1998 1999 2000 Yearly Peak Force 180 730 ,, ' 1635 2455 3180 3180 2000 770 455 Camp/v~Ilage Oes~gn 200 aoo 1800 2700 3.500 3500. 2200 850 500 ------------------------------------------------------------ I I I I I I I I I I 'I I I I I I I I I I I I I I I I I I I I I I I I I I I ···1· .. - Dam - loguiri (1985) Vaiont (1961) Mauvoisin (1957) Chirkei (1975) El Ca1on (1964 Contra (1965) Glen Canyon (1964) Mohamed Reza Shah Pahlavi (1963) Almendra (1910) Vidraru-Arges Gocekaya Morrow Point Pacoima TABLE 13.2: ARCH DAM EXPERitNCE Locatiort Height ft(tn) Crest length ft(m) ""' l ... Georgia, 892 2,513 USSR. (272) (766) Veneto, 858 624 Italy (262) (190) Valais, 777 1,706 Switzerland (237) (520) North Caucusas, 764 1,109 USSR (233) (338) Yoro/Cortes, 741 1,253 Honduras (226) (382) . Ticino, 722 1,246 Switzerland (220) . -{380) Arizona, J10 1,560 USA. (216) (475) Khouzestan, 666 696 Iran (2(13) (212) Salmanca, 662 1,860 Spain (202) (567) Romani·~ 548 588 (167) (292) Turkey 521 1,620 (159) (494) Colorado 465 720 California 372 589 (113) {180) TABLE 13.3: COMPENSA TIO~ FLOW PlffP DATA 'vertical, axiat1 or mi)(ed flow Rated head (total dynamic level) .,. ........................ o 35 ft Rated discharge ·······~······~····••••••••••••••••••••••• 115,000 gal/min Pump input !It ••• ~ ••••••• "' ........... c ••••••• ~--§ •••••••••••• •--• 1 ,300~ hp Speed ••••••••••••• , ••••.•••• if~ •••••••.•• ~ ••••••••••.•••••••• •~ • 400 r_pm In.,eller diameter ......................................... 51 in (approx.) H~tor Type • • • • • • • • • • • • • • • • • • • • • • • • •. • • • • • ... • • • • •. • • • • • • • • • • • • •.• • • verticaJl induction Rated power ••••••••••••••••••••••• o...................... 1.,400 hp Speed ••••• -••• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 400 .rp~n Voltage ••o•••·e••••••••••··--·•••••••••••••••••••••-.e••••••• .4,160 V ~~o. phases •••••.• ~ ••••••••• ·• o ••••••••••••••••••••••••••• c .. 3 F~equency ••••••••••••••• -....................... '81 •• ,. • • • • • • • • • 60 hz I I I I I I I I I I I I I I I I I' I I I I I .I I I I I I I TABLE 13.4: PffELIMINAAY UNIT DATA 1 GENERAL DATA 2 'Number ·of units e&e't99e•••Gee.eo•e.eO•ea•.;:,eeil!•e•4t•·••••..tf'• Nominal unit output •••••••••••••••••••••••••••••••••••• Headwater levels .. normal maxif11Jm ••••••••••••••••.••••••• , ........... ~· ..... . -miniml..ID * • ,. .•••••••.•• ~ .••••••••.•• • ••••.••••••.•.••• .-•••.•••• Tailwater levels minimum ............................. • ••••••••••••••••••• normal •• & .......... d .................... e • a •••••• 0 , •••••• maximum •••••••••• ~ .... ft •••• ·e ........................... 9 TURBINE DATA 4 150 MW El 1445 El 1390 El 860 EL 840 El 838 Type· • •••••••••••••••••••••.• ~ •••••• _. ••••.•••• q. ~~~ • • • • • • • • • vertical .·Francis Rated net head· • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • .• 5 1S ft MaxitnUm head •••••• a ...................................... ~ 597 ft Minimllll head •••••••••••••• ·• ................ c........... .. . 538 ft Full gate output: at rated level •••••••••••••••••••••• ., .................. 225,000 hp -at maximum 'head ....................... o••••••••••••••• 240,000 hp -at minimum head .......................................... 205,000 hp Best gate output ........................................... 85% full gate output Full gate discharge at rated head ........................ 3,790 cfs ~eed ............................... 10 •••.•••••••.•••• ~ • • • • • 225 Specific speed •••••a•••••••••·GJ~··••••••••••••••••••••• 37.9 Runner discharge diameter ................................. 135 in Runaway speed ••••• ~ ....................................... , ••• c. 395 rpm Centerline distributor ............................... 4 •• 820 Cavitation coefficient (sigma) •••••••••••• ,. •• o••••••••• 0.089 3 -GENERATOR DATA Type •• ~· ••••••••••••• ~ •••••••• ~ ••• ••.• •••••••••••• e •••.•.•• ve·rtical·modified l.lllbrella R~ted output •••••• Ct •••••••••••••••• e •• , • • • • • • • • • • • • .• • • • 180· ·M't*A· Power factor· •••••••.a••••••••••·•••••o•••····•••••-e••••••• 0.9 Voltage • • • • • • • .. • • • • • • • • • • •. • • • • • • • • • • • • • • • • • • • • .• • • • • • • • • 15 kV Inertia constant (H)* ••••••••••••••••••••••••••••• .. •••• 3.5 MW-sec/MVA Synchronous.speed ·~s••••••••••••••••••••••••••••••••••• 225 rpm 6 2 Flywheel effect (WR )* •••••••••••••• .. •••••••••o•••••••• 54 x 10 lb-ft Heaviest lift ••••••••••••••••••••••••••••o••••••••••••• 750,000 lb * Including turbine " 50 --. . . """""""' ~ ~· 30 u < . ~ ~ -. ~ . ,. ~ ~ ,,""' ..tt# .. -,- ~ ~""" . . ' lO • I ' eao -890 900 910 920 8!0 140 960 HEADWATER EL:EVATION (FT.) . I DEVIL ·CANYON DIVERSLON · · RATING CURVE FiGUR!:; i!.l ]i} ~--------------------------------------------------------------------------------·· I I I I I I I I I :I I I I I I I I =~· !~' ' ,_, ' 580 540 520 100 --. - usc lo GENERATOR RA1 'POWER - .0 - RESERVOIR EL. 1455 I ----;' I - I IL WEIGHl .J ED AVERAGE H ~AD BES r GATE 7-I / J ,, ~ GENERAl PR RATED POWER ...- I I BEST EFFIC 'f.NCY-j FULL GATE ! """'IR /e. 1400 MINIMUM DEC fs::ur:u::~ HEAD HI l--- -1!5 ) MW ' . 120 140 160 180 200 220 UNIT OUTPUT-MW I} DEVIL CANYON..;.. UNIT OUTPUt FlGURE' l3.2 ~''("';~ I I I I I I I I I I I I I I I I I t'"," I I :.,:· -?fl. -·~ Z" t.&J• • "· • ~ • •• .... • ' •••• > • Q . L~----~~--------~--------~------~---------r--~4000 b: 80- "-1.5.1 lU z ai § }--en "" 10 1-------+-----t-----+--~;c_.-+-----+----t 3000 8 ~~-----+--------~--------+-------~--------~--~1000 40,000 80,000 120,000 160,000 200,000 TURBINE OUTPUT {HP) DEVIL CANYON -TURBINE PERFORMANCE (AT RATED HEAD) 240,000 FlGURE · 13 .. 3 .. 1&.1 .(!) cr ·<~;· ·~ Z5 I I I ~,I I I •• I I I I ~I I I I I I ' . ....,.,.---,~.-------;:.,-:-·--:--·--:;·· •.. , ·~-' _, --;~-_,-~ ~ -.-t·-· ·_ .. - . ·' 14 -TRANSMISSION FACILITIES Th~ objectiv~ of this secti.on is to record and describe the studies performed to .select a power delivery system from the susitna River basin generating plans to the :ffiaj or load centers in Anchorage and Fatv .. ba.nks. This syst.em wi 11 be com ... prised of transmission lines, substations, a dispatch center, and means of com• munications. The major topics of the tra.nsmission studies include: -Electric system studies; -Transmission corridor selection; -Transmission detailed route selection; -Tr ansmi ss ion towers, hardware and conductors; -Substations; and -Dispatch center and communications. Further discussion of the importance of these studies in determining the method of operation of the Railbelt System is presented in Section 15., In the following text, each of the major topics will be discussed with respect to previous studies, methodology~ additional data obtained, and conclusions arising from the studies. 14.1 -llictri c. System Studies ,, Transmission planning criteria were developed to ensure th.e design of a re1i.·:able and economic electrical power system, with components rated to allow a smooth transition through early project stages to the ultimate developed potential. Strict application of optimum, long-term criteria would require the installation of equipment with ratings larger than necessary ,at excessive cost. In the · interest of economy and long-term system performance, these criteria were tem- porarily relaxed during the early development stages of the project. Although allowing for satisfactory operation during early system development, final system parameters must be based on the ultimate Susitna potential. The criteria are i·ntended to ensure maintenance of rated power flow to Anchorage and Fairbanks during the outage of any single 1 i ne or transformer element. The essential features of the criteria are: Total power output of Susitna to be delivered to one or two stations at Anchorage and one at Fairbanks; -"Breaker-and-a-half~' switching station arrangements; -Overvoltages during line energizing not to exceed specified limits; -System voltages to be within established-limits during normal operation; ... Power delivered to the loads to be maintained and system voltages to be kept within established limits for system operation under emergency conditions; 14-1 -·:..- .p-,, I . f .. ~ •.. - -Transient stability during a j-phase line fault cleared by breaker action with no r~closing; and · -\~here performance li.mfts are exceeded, the most cost effective corrective measures are to be taken. (a) Exi sti ns Ststem Data Data have been compiled in a draft report by Commonwealth Associates Inc., dated November, 1980, and entitled •tAnchorage-Fah·banks Transmission Inter- tie -Transmission System Data". The contents of this report have be.en included, with minor revisions, as Appendix B of the Susitna Hydroelectric Project Planning Memorandum-Preliminary Transmission System Analysis (1). Other system data were obtained in the form of single-line diagrams from the various utilities. (b) f._ower Transfer Requirements The·Susitna transmission system must be designed to ensure the reliable transmission of power and energy generated by the Susitna Hydroelectric Project to the load centers in the Railbelt area. The power transfer re- quirements of this transmission system are determined by the following fac- tors: -System demand .at the various load centers; -Generating capabilities at the Susitna project; and -Other genet~ation available in the Rai lbelt area system. Most. of the electric load demand in the Railbelt area is located in and around two main centers: Anchorage. and Fairbanks. The largest load center is Anchorage, with most of its load concentrated in the Anchorage urban area. The second 1 argest 1 oad center is Fai rbank.s. Two sma 11 1 oad centers (Willow and Healy) are located along the Susitna transmission route. The only other significant load center-s in the Railbelt region are Glennallen and Valdez. However, their combined demand is expected to be less than 2 percent of the total Railbelt demand in the foreseeable futureo A survey of past and present load demand levels as well as various forecasts of future trends indicates these approximate load levels at the various centers: Load Area Anchorage -Cook Inlet Fairbanks -Tanana Va 11 ey Glennallen -Valdez Percent of Total Railbelt Load 78 20 2 Considering the geographic location and the currently projected magnitude of the total load in the area, transmission to G1ennallen-Vtildez is not likely to be economical in the foreseeable future. If it is ever to be economical at all, it would likely be a direct radial extension either from 14-2 I 7: -,.. 1--·-._·. .. -~: ·. I· I I I I I I I I I I I I I :~:::.~ I I ;_:, I I I I I ;I I I a· I 'I I I I ~. I I I I I Susitna or. from Anchorage •.. In either case its relative magnitude is too small to h·ave sign<ificant influence on either the viabil_ity or-development characteristics of the Susitna project or the transmission from Susitna to the Anchorage and F'airbanKs a·reas. As a result, it .has been assumed for study purposes that approximately 80 .percent of the generation at Susitna will be transmitted to the Anchorage area and some 20 percent to Fairbanks. To account for the uncertainties in future local load growth and local generation development, the Susitfta ·transmission system was designed to be able to transmit a maximum of 85 percent of Susitna generation to Anchorage and a maximum of 25 percent to Fairbanks. The potentfal of the Susitna Hydroelectric Project is expected to be devel- oped in three or four stages as the system load grows over the next two decades. The transmission system must be designed to serve the ultimate Susitna development, but staged to provide reliable transmission at every intermediate stage. Present plans call for three stages of Susitna · development, at Watana in 1993 followed by a further ·340 MW in 1997, and 600 MW at Devil Canyon in 2002. Development of other generation resources could alter the geographic load and generation sharing in the Railbelt~ depending on the location of this development. However, currenf studies indicate that no other very large projects are likely to be developed until the full potential of the Susitna project is utilized. The proposed transmission configuration and design should, therefore, be able to satisfy the bulk transmission requirements for at least the next two decades. The next major gen~eration development after Susitna will then ·require a transmission' system determined by its own magnitude and location. The resulting power transfer requirements for the Susitna transmission sys- tem are indicated in Table 14.1. (c) Transmission Alternatives Because of the geographic location of the various centers, transmiss.ion from Susitna to Anchorage and Fairbanks will result in a radial system ,con- figuration. This allows significant freedom in the choice of transmission voltages, conductors, and other parameters for the two line sections, with only limited dependence between them. In the end, the advantages of stan- dardization for the entire system wi 11 have to be compared to the benefits of optimizing each section on its own merits. Transmission alternatives were developed for each of the two system areas, including voltage levels~ number of circuits required, and other parameters, to satisfy the necessary transmission requirements of each area. Havi_ng estab 1 ished the peak power to be de 1 i vered and the distances over which it is to be transmitted, transmission voltages_and number of circuits required were determined. To maintain a consistency with standard .ANSI volt~ges used in other parts of the United States, the following voltages were considered for Sus itna transmi ssi.on: 14-3 . ' ''(~'\ .... -c, .. . . ~ Watana to Devil Canyon and on to Anchor~tge: 500 kV or 345 kV 345 kV or 23.0 kV • Devil Canyon to Fairbanks: (i) _· Susitna to Anchorage _ ( i i) Transmission at either of two different voltage leve,ls (34'5 kV. or 500 kV) caul d reasonably· provide the necessary pawet transfer capa ... bil ity over the dista·nce of approximately ·140 miles between Devil Canyon and Anchorage. The required transfer capability of 1,377 MtJ is 85 percent of the. ultimate generating capacity of 1, 620 MW. At 500 kV, two circuits would provide more than adequate capability. ·At 345 k V, either three circuits uncompensated or two circuits \'lith series compensation are required to provide_ the necessary rei iabi 1- ity for the single contingency outage criterion. At lower voltages an excessive number of parallel circuits are required, while above 500 k V two circuits are st i 11 needed to provide service in the ever•t of a line outage. Susitna t6-Fairbanks -.~~~~~--~~ Applying the same reasoning used in choosing the transmission alter- natives to Anchorage~ two circuits of either 230 kV or 345 kV were chosen for the section from Devil Canyon to Fairbanks. The 230 kV alternative requires series compensation to satisfy the planning criteria in case of a 1 ine outage. · (iii)· Total System Alternatives The transmission section alternatives mentioned above were combined into five realistic total system alternatives. Three of the five alternatives have different voltages for the two sections. The principal param:!ters of the five transmission system alternatives analy~ed in detail are as fo'llows: Susitna to Anchorage Susitna to Fairbanks Number of Number of A 1 ternati ve · Circuits VoltaJe Circuits VoltaJe (kV (kV 1 2 345 2 345 2 3 345 2 345 3 2 345 2 230 4 3 345 2 230 5 2 500 2 230 Electric system analyses, including simulations of line energizing, load flows of normal and emergency operating conditions, and transient stability performance, were carried out to determine the technical feasibility of the various alternatives. An economic comparison of transmission system life cycle costs was carried out to evaluate the relative economic merits of _ each alternative. All five transmission alternatives were found to have acceptable performance characteristics. The most significant difference was that single-vo'ltage systems (345 l<.V, Alternatives 1 and 2) and systems without series compensation (Alternative 2) offered reduced complexity of 14-4 >I., ~-- 1. __ 1 '. -~ __ ._ } '1- '\ ," .., I I I I I I I I I I 'I I I I I I •. I I I I I I •~ I; •• ~. I I I "I I I I ···I I (d) design and operation and therefore were likely to be marginally more reliable. The present-worth life cycle. costs of Alter-nativ~s l through 4 were all within o·ne. percent of each other. Only the cost of the 500/230 kV scheme .(Alternative 5) was 14 percent above the others~-A surtmary of the lif.e cycle cost analysis for the various alternatives is shown in Table 14.2. Full details of the technical and economic comparisons are explained in Reference 1. A technical and economic comparison was also carried out to determine possible advantages and disadvantages of HVDC transmission, as compared to an ac system, for transmitting Susi tna power to Anchorage and Fairbanks. As outlined 'in detai 1 in Reference 1, HVDC transmission was found to be technically and operationally more complex as well as having··higher life cycle. costs. Configuration at Generation and Load Centers Interconnections between generation and load centers and the transmission system were developed after reviewing the existing system configurations at both Anchorage and Fairbanks as we 11 as the possi bi liti es and current development plans in the Susitna, Anchorage, Fairbanks, Willow, and Healy areas. (i) Susitna Configuration Preliminary development plans indicated that the first project to be constructed would be Watana with an initial installed capacity of 680 MW~ to be increased to approximately 1020 MW in the second development stage. The next projectt and the last to be considel ... ed in this study, would be Devi·l Canyon, with an installed capacity of 600 MW. · (ii) Switching at Willow rransmission from Susitnaoto Anchorage is facilitated by the intro- duction of an interr,~ediate s~-1itching station. This has the effect of reducing line energizing overvoltages and reducing the impact of 1 i ne outages on system st abi 1 i ty. Willow is a sui tab 1 e locat 1on 'for this intermediate switching station; in addition, it would make it possible to supply local load when this is justified by development in the area. This 1 oc a 1 1 oad is expected to be 1 ess than 10 percent of the total Railbelt area system load~ but the availability of an EHV line tap would definit~ly facilitate future power supply. (iii) Switching at Heal~ A switching station at Healy was considered early in the analysis, but was found to be unnecessary to satisfy the p 1 anni ng criteria. The predicted load at Healy is small enough to be supplied by local generation and the existing 138 kV transmission from Fairbanks. 14-5 c (i v) Anchorage Configuration In its 1975 report on the Upper Susitna River Hydroe1 ectr•i c Studies (2), the United States Department of the Interior Corps of Engineers favored a transmission route termi_nati ng at Point MacKenzie. The 1979 Economfc Feasibi 1 ity Study Report for the Anchorage-Fair-· banks Intertie by Ittternational Engineering Company, Inc. (IECO) (3) recomme_nded one circuit from Susitna terminating at Point MacKenzie and another passing through Palmer and Ek1utna substations to Anchorage along the eastern side of Knik Arm. Analysis of system configuration, distribution of loads, and dev-el- opment in the Anchorage area led to the conclusion that a transfor- mer station near Palmer would be of little benefit. Most of the major loads are concentrated in and around the urban Anchorage area at the mouth of Knik Arm. In order to reduce the length of sub- transmission feeders, the transformer stations should be located as close to Anchorage as possible. The routing of transmission into Anchorage may be chosen from the following three possible alternatives: .. Submarine Cable Crossing From Point MacKenzie to Point Woronzof This. waul d :require transmission through a very heavily developed area. It would also expose the cables to damage by ships 1 anchors, which has been the experience with existing cables, resulting in questionable transmission reliability. -Overload Route North of Knik Arm via Palmer This may be most economical in terms of capital cost in spite of the long distance i nvo 1 ved. However, approval for this route is unlikely since overhead transmission through this developed area is considered environmentally un~cceptable. A longer overland route around the developed area is considered unacceptable because of tne mountai noL~s terrain. -Submarine Cable Crossing of Knik Arm, In the Area of Lake Lorraine and Six f.ti 1 e Creek - This option, approximately parallel to the new 230 kV cable under construction for Chugach Electric Association (CEA), includes some 3 to 4 miles of submarine cable and requires a· high capital cost. Si nee the area is upstream from the shipping lanes to the port of Anchorage it will result in a reliable transmission link, and one that does not have to cross environmentally sensitive conservation areas. The third alternative is clearly the best of the three options. The details of this configuration are as follows: 14-6 0 I 10 I I I 'I I I I I ·I ·a I •• I I I I I I I I I I I I I I 1\ I I I I I I I -1' \ -' t: submarine cable crossing. To reduce cable costs the crossing could be constructed with two cable. circuits plus one spare phase. This option requires a switching station at the west terminal of Knik Arm.. A switching station at the west terminal would clearly require increased costs c:md .complications for construction and operation as a result of poor access. It would also require a separate location for the tap to supply MEA. Plans are present 1y underway for a bridge crossing at Kni k Arm for both railway and road traffic. If these plans should be realized, transmission costs and complications could be significantly reduced by routing the tables across the bridge. (v) Fairbanks Configuration Susitna power for the Fairbanks ar·ea is recommended to be delivered to a single EHV/138 kV transformer station located at Ester. (e) Recommended Transmission System The configuration of the recommended tra.nsmission system, (Alternative 2) is shown on the single-line diagram (Figure 14.1). The.main characteris- tics of the recommended systems are summarized in Table 14.3. 14~2 -Corridor Selection (a) Methodology Development of the proposed Susitna project will require a transmission mechanism to deliver electric power. to the Railbelt area. The pre-building of the Intertie system&~ill result in a corridor and route for the Susitna transmission lines between Willow and Healy. Therefore~ three areas re- quire study for corridor selection: the northern area to connect Healy with Fairbanks; the centra 1 area to connect the Watana and Devi 1 Canyon damsites with the Intertje; and the southern area to connect Willow with Anchorage. The corridor selection methodology followed the Susitna study plan formula- tion and selection methodology. Previous studies, existing data, aerial reconnaissance and limited field studies. formed the data base. Using the selection criteria discussed below, corridors 3 to 5 miles wide which met these. criteria were selected in each of the three study areas. These cor- ridors were then evaluated to determine which ones met the more spec-ific screening criteria discussed below. This screening process resulted in one corri dar in each area being des:i gnated as the recommended corri dar for the transmission line. For a more detailed discussion of study methodology and the selection and screening criteria, refer to the Transmission Line Corridor Screening Closeout Report (~eptember, 1981), hereafter referred to as the Closeout Report. 14-7 (b) Previous Studies The two reports reviewed which contained the most information relevant to the transmission line .stu~n es were; -The Susitna Hydroelectric Project Interim Feasibi1ity Report, prepared by the U.S. Army Corps of Engineers. (hereafter referred to as the COE re- port); and -The Economic Feasibility Study for the Anchorage-Fairbanks Interti e by International Engineering Company, Inc./Robert W. Retherford Associates (hereafter referred to as the IECO/RWRA report). The COE report consisted primari 1 y of an eva 1 uati on of alternative corridor locations to aid in the selection of those which maximized reliability and minimized costs. Utilizing aerial photographs and existing maps, general corridors connecting the project site with Anchorage and Fairbanks were selected. This study was general in nature and was intended only to demonstrate project feasibility. The IECO/RWRA report utilized the COE report as background information for both economic feasibility determination and route selection. The corridor selected by IECO/RWRA was very similar to that selected by the COE with further definition. The route selected was based on length, accessibility and environmentai compatibility. The report also presented a detailed economic feasi bi 1 i ty study for the Anchorage-Fairbanks transmission study. (c) Selection Criteria and Selection Results ( i) Cri teri a The objective of the corridor selection conducted by Acres was to select feasible transmission line corridors in each of the three study areas: -The northern area, to connect Healy with Fairbanks; -The central area, to connect the Watana and Devfl Canyon damsites with the Intertie; and -The southern area, to connect Willow with Anchorage. Technical, economic, and environmental criteria were developed in order to select corridors within the three areas. These criteria are 1 i st ed i n Tab 1 e 14 . 4. Environmental inventory tables were then compiled for each corridor selected, listing length, number of road crossings, number of river and creek crossings, topography, soils, land ownership/status, existing and proposed development, existing rights-of-way, scenic quality/recreation, cultural resources; vegeta~ion, fish, birds, furbearers, and big game. These tables are included in the Closeout Report. ? 14-8 I I J I I I I I I I I I I' I I ~a, ·- 1 I I I I I 'I I I I 'I I •• I I I I I I I I I ( i i ) Results · UtilTf!.ing existing information, 22 ctJrridors were selected based on their· ability to meet technical, economic and environmental crit.eria as listed in Tabl~-14.(. Four of the corridors are in the southern study area, 15 ·;n the cen- tral area, and four in the northern study area. Three of the corri- dors in the southern study area run in a north-south direction while one runs northeast to Palmer, then northwest to Willow. Corridors in the central study area are in two general groups: those running from Watana Damsite west to the proposed Intertie and those running north across the Denali Highway and the Chulitna River. Corridors in the northern study area r·un either west or east to bypass the Alaskan Range, then proceed north to Fairbanks. See Figures 14.2, 14.3, and 14.4 for the location of these corridors. (d) Screening Criteria and Screening Results ( i) Criteria The objective of the screening process was to screen the previously selected corridors to determine which best meet additional techni- . cal, economic, and environmental criteria as listed in Table 14.5. The rationale for selection of these criteria is explained in Appendix E2. In addition to these criteria, each corridor w&s screened for reli- ability to determine if a line could be operated with a minimum of power operation. Six basic factors were considered in relation to r e 1 i ab i l i t y: -Elevation: Lines located at elevations below 4000 will be less exposed to severe wind and ice conditions which can interrupt service. -Aircraft: Avoidance of areas near aircraft landing and takeoff operations will minimize the risk of collisions. . -Stability: Avoidance of areas susceptible to land, ice, and snow slides will reduce the chance of power failures. -Existing Power Lines: Avoiding existing transmission lines will reduce the possibility of lines touching during f.ailures and will facilitate repairs. -Topography: Lines located in areas with gentle relief will be easier to constru~t and repair~ .. Access: Lines located in reasonable proximity to tran.sporta- tion corridors will be more quickly accessible and~ therefore, more quickly repaired if any failures occur. 14-·9 -;-;; . .:.::. ( i i) The screening criteria-and reliability factors for each corr-idor were evaluated utilizing topograpt:ic maps, aer\'al photos, aerial.· overflights~ and published materi ~ls •. Each corridor was then · assigned four -ratings (one· each for technical, economic and environ ... .mental consideratto.ns, and one overall summary rating.} Ratings wer-e defined a·s fo1lows: - A ... recommended . C -acceptable but not preferred F .oc unacceptab 1 e From the technical point of view~ rel i abi 1 i ty was the main objec- tive: An environmentally and economically sound corridor was re- jected if a line built in· the corridor would be unreliable. Thus~ any line whicn received an F technical rating was assigned a sumnary rating_ of F and eliminated from further consideration. Similarly, because of the critical importance of environmental con- siderations, any corridor which received an F rating for environmen- tal impacts was assigned a summary rating of F, and eliminated from consideration. Results Table 14,.6 summarizes the comparison of the corridor screened in the southern, centra 1 and northern study areas. One corridor in each of the three study areas received A ratings for a 11 three categories. . These three corridors and the rationale for their A ratings are dis ... cussed be low. For a description of a 11 22 corridors and the ration- ale for their ratings, see Appendix E2. -Southern· Study Area Corr.'idor Two -Willow to Point MacKenzie vi a Red Shirt Lake • Oescri pt i on Corri. dor AOFC, consisting of Segments ADF and FC (Figure 14.2), comnences at the point of intersection with the Intertie in the vicinity of Wi 11 c~w but immediately turns to the southwest, first crossing the railroad, then the Parks Highway, then Willow Creek just west of Willow. The land in thevicinity of this part of the segment is very flat, with wetlands domi·nating the terrain. Southwest of Florence Lake, the proposed corridor turns, crosses Rolly Creek, and heads nearly due south ti passing through extensive wetlands west and south of Red Shirt Lake. Th.e corridor in this area parallels existing tractor trails and crosses very flat lands with significant amounts of tall-grow- ing ve_getation in the better drained locations. 14-10 ~ l.o .," .. , -; l' 1:. " ·~l :: 1-,· .. ·~. 'r_,_ • t~-: I I I I I I I I I I I I ·-.1'-·.·.t. ' . I I I " ' 1··;·.· ' ' ' ~ " i. ·.. : I' . •, ~ ~' 1 I I I I I I ·~-.1· ' '." ./ .. I' ·- 1 I I ,I ' i; .' I I -:,·, ~'t._ ;; ",'} Northwest of Yohn Lake~ the corrid~r segment turns to the southeast, passing Yohn Lake and My Lake befor-e crossing the Little Susitna River •. Just south of My Lake~ the :Corridor turns in a southern direction, passing Mi. dd1e Lake and east of Hor·seshoe take before finally inte~se:cting ttre ,~xist1ng B~luga 230 kV transrnissi orf line at a spot just .north of Maci<enzi e Point. From here., the corridor parall~ls MacKenzie Point 1 s existing transmission facilities before crossing under l<nik Arm to emerge on the eastern shore of Knik Arm in the vicinity of Anchorage. The 1 and in the vicinity of this segment is extremely flat and wet. '!t supports stands of tall-growing vegetation on the higher or better drained areas. . Technical and Economi ca 1 Rating Corri do.r ADFC crosses the fewest number of rivers and roads in the southern study area. It has the advantage of paralleling an existing tractor trail for a good portion of its length~ thereby reducing the need for new access roads. Easy ace ass wi 11 allow m·a; ntenance and repairs to be carried out i'n minima 1 time. This corridor also occurs at low elevations and is approximately one-half the length of Corridor Onec • Environmental Rating This corridor crosses extensive wetlands from Willow to Point MacKenzie. . At higher elevations or in the better drained sites, extensive forest cover is encountered. Good agricul- ·tu~al soils have been identified in the vicinity of this corri- dor; the state plans an Agricultural Lands Sale for areas to be traversed by this corridor. The corridor also crosses the Susitna Flats Game Refuge. The presence of an existing tractor trail near considerable portions of this corridor dimiriishes the significance of some of these constraints. Furthermore, its short length and the fact that it crosses only one river and eight creeks increases its environmental acceptability. -Central Study Area Corridor One-Watana to lntertie via South Shore, Susitna River . ~script ion This corridor originates at the Watana Dam site a·nd follows the southern boundary of the river at an elevation of approximate 1 y 2,000 feet from Watana to Devil Canyon (Figure 14 .. 3). From Devil Canyon, the_~grridor continues along the southern shore of the Susitna River at an elevation of about 1,400 feet to where it connects with the lntertie, assuming the Intertie follows the railroad corrigor. The land surface in this ·area is relatively flat., though incised at a number of locations by tributaries to the Susitna River. The relatively flat hills are covered by discontinuous stands of dense, tall-growing vegetation .• 14-11 • Techn-ical and. Economical Rating Corrido,r One is one of the 'shortest corridors considered. It- is approximately 40 miles, long, making it economic-ally favor- able. No technical restrictions were obser-ved along the entire · le.ngth of this corridor. . Environmental Rati-ng Because of its short length, environmental disturbance caused by transmission line construction would be reduced. The more noteworthy constraints are those identified under the catego- ries of land use and vegetation. Corridor One would require · the development of a new right-of-way between Watana and Devil Canyon with some opportunity existing to utilize the COE- developed road for access between the In terti e and De vi 1 Canyon. The potential does exist in this cor.ridor to use the proposed access road. Wetlands and discontinuous forest cover occur in the.corridor, especially in the eastern third of the route. Access road development, and the associated vegetation clearing, present additional constraints to this corridor. -Northern Study Area Corridor One -Healy to Fairbanks vi a Parks Highway Description Corridor One (ABC) , cons 1St i ng of Segments AB and BC, statts in the victnity of the Healy Power Plant (Figure 14.4). From here, the corridor heads northwest, crossing the existing Golden Valley Electric Association Transmission Line, the railroad, and the Parks HighWay before turning to the north ,and paralleling this road to a point due west of Browne. Here, as a result of terrain features, the corridor turns northeast,. crossing the Parks Highway once again as we 11 as the existing transmission line, the Nenana River~ and the railroad, and continues to a point northeast of the Clear Mi ssi 1 e Ear 1y \~arning Station (MEWS). Continuing northward, the corridor eventually crosses the Tanana River east of Nenana, then heads northe-ast, ftrst Crtls-s-~ · ing Little Goldstream Creek, then the Parks Highway just north of the Bonanza Creek Experimental Forest. Before reachi rig the drainage of Ohio Creek, this corridor turns back to the north- east:s crossing the old Parks Highway and heading into the Ester Substation west of Fairbanks .• Terrain along this entire corridor segment is relatively flat, · with the exception o.f the foothills north of the Tan an a River.· . Much of the. route, especially that portion. between the Nenana and th·e Tanana River crossings, is very broad and flat. It has standing water during.·the summer months and, in some places, is overgrown by dense stands of tall-growing vegetation. This corridor segment crosses the heavily wooded foothills northeast of Nenana. · '14-12 ,, I· ·J--·~. . .---. I I I I I I I I I I I I t: I .<• 1_," ''. ~· ••• . .· . ·. ·•' 1·, ... . . ,' I I I >;; ., I I '• I I I I ~. ·I I I I I I I (e) An option to the ,above not shown in-the figures has been considered., closely paralleling and. sharing rights-of-way with the existing Hea1y ... .fairbanks transmission 'line. ·While it is · usually attracti v~ to parallel extsti.ng corridors wherever - .. p.osstble·, this op.tion necessitates a great number of· road crossings and results in· an extended 1 ength of 1:he eorri dor paralleling the Parks Highway .. A potentially significant amount of highway-abutting land would be usurped for contain-. ment of the right-of-way. The combination of these features precludes this corridor from further evaluation. o Technical and Economical Rating • This corridor crosses the fewest water courses in the northern study area. Although it is approximately four miles longer than Corridor Two, it is technically favored because o.f the existence: of potentia 1 access roads for almost the ·entire · length .. Envi ro~!]nta 1 Rating Because it parallels an -existing transportation corridor for much of its length, this corridor would permit line routing. that w.ould avoid most visually sensitive areas. The three pro- posed road crossings for this corridor (as-opposed to the 19 road crossings of the Healy-Fairbanks transmission line) could occur at points where roadside qevelopment,exists, in areas of· visual absorption capability, or· in areas recommended to be opened to long-distance views. Four rivers and 40 creeks with potential for impacts are crossed by this corridor. It crosses the fewest number of water courses of any route under consideration in the northern study area. In addition, the inactive nest site of a pair of peregrine falcons occurs within this proposed corridor. r.; As with visual impacts, land use, wildlife, and fishery re ... source impacts can be lessened through carefully route location and utilization of existing access. Impacts on forest clearing can be lessened through the sharing of existing transmi sslon line corridors. Cone lusi ons A review of previous reports, other existing information, and aerial over- flights was used to select corridors for consideration in this study. These corridors were screened against certain technical, economic and envi- .ronmental criteria, resulting in one recommended corridor in each of the southern, central and northern study areas. The corridors shown in Figures 7.1 through 7.8 of Appendix E2 are .believed to best meet the technical, economic and environmental criteria; therefore, these corridors are the best locations in which to p 1 ace the Susitna transmission 11 nes. .. 14-13 14.3 -Ro.ute Selection '-}::-.:::·-~-:-.:.-:....:.. ~-, (a) _Method() lQSl: After identifying the. pr:efer.red transmJsston line corridors, the next $tep in the route selection process involved the analysis o:f the data as gather- ed and presented on the base map~ overlays were compi.ted so that various cmi:itr:aints affecting construction or maintenance of a transmission facil- ity could be viewed on a single map. The map is used to select possible routes within each of the three selected corridors. By placing a 11 major constraint-s (e.g., area of high visual exposure, private lands, endangered species~ etc.) on one map, a route of least impact was gelected. Existing faciliti.~s, such as transmission lines and tractor trails within the study area, ~t:re also considered during the selection of a least impacted route. Whenever poss-ible, the routes were selected near existing or proposed access roads, sharing whenever possible existing rights-of-way. The ~ata base used in this analysis was obtained from the following sources: -An up-to-date land status study; -Existing aerial photos; -New aerial photos ·conducted for selected sections of the previously rec• ommended transmission line corridors; -Environmental studies including aesthetic considerations; -Cl imatologi ca 1 studies; -·Geotechnical exploration; -Additional field studies; and ~ Public opinions. {b) Selection Criteria {i) Criteria ·The purpose of this section is to identify three selected routes: one from Healy to Fairbanks, the second from Watana-Devil Canyon damsites to the intertie!! and the third from Willow to Anchorage. The previously chosen corridors were subject to a process of refin- ing and evaluation based on the same technical, economic, and envi- ronmental criteria used in corridor selection (see Table 14.15).. In addition, special emphasis was concentrated on the following points: -satisfy the regulatory and permit requirements; -selection of routing that provides for minimum visibility from hi-ghways and homes; dnd . . -avoidance of developed agricultural lands and dwellings .. - 14-14 . 1' . . \f.;..:,.,· ' .I• ..•. ' . ' I I I I I I I ·. I l I I I' r~ ; ., ' I I ·~ill I I ':1··. ·.. i '· ..•.. :. " .. . . « I I t· I . . '·'*' ~ ' I I 1·.·! . ' '!'>,, ., I ~ . I ...... ·., .... ~;;_ -,t '. ~ I ··''':,,. ·, ,_7- 1 I ;. Figures 1 througb 14 in J.\ppendi~< E3 show the selected trartsmission line. rout.e fo.r t~e· three areas of study; hatnely, the southern study ar,ea; the central study area; and the northern study area. · As a first step, the. 3...,to~S ... mi le-widt.h corridor previously selected for each of the three study areas was narrowed to a half-mile-width corridor based on the previous. criteria" The preliminary centerline of the right-of-way is shown in the figures. This centerl tne represents a right-of-way width of 400 feet. This width is adequate for three~ single-circuit, parallel lines with tower structures having. hori zonta 1 phase sp.aci ng of 33 feet. However, between the Devil Canyon damsite and the intertie, the width of the right-of-way is 700 feet which is needed to accommodate five sing le~ci rcuit 1 i nes. Southern Study Area -Willow to Point Mackenzie Via Knik Arm._C_r_o_ss_,_·n_g ___ ~--------- Description Evironmental Considerations . Techni ca 1 and Economic Considerations This .route crosses a very few number of rivers and roads.. It paral- lels the existing tractor trail for a considerable portion of its length, thereby reducing the need for new access roads. At Knik Arm, the route parallels a proposed 230 kV transmission line and will share (if possible) its right-of-way. This will avoid pi aneering new right-of-way especially in an area full of dwellings and other· constraints. · 14-15 Central Study Area -Wat~na to Intert ie ·oescrlptio.IJ Environmental Considerations Technical and Economic Considerations This is the shortest route among the studied ones in the central study area~ which makes it economically feasible. The route paral- lels a proposed access road almost through its entire length north of Sus itna River for the sect ion between the two d amsites and south of it between Devil Canyon and the Intertie. This wi11 add to the reliability and economical considerations. Northern Studx Area -Healy to Ester Desc~iption 14-16 . . ·~ I I I I I ..• 1·.· •. I I -~· I :I I · ......... I I I " .~ I I 1·, -, . . l c ~: . . • ""<"'~ ,.. ~ •• ._,~ _..,, : ........... ;:;·,~_.;.-~ ··•r~· ~· ~-•·-'·· ·. - I . I~"·~ .;''I I. I I I ' I I ., ..... · 1 ~ I -.,, I I ~..,.,.,..: I I .. ,.'._, ' . 'i I . I ' t\ '' . . :.::.:: .. . . . (c) . . ··Environmental Considerations Technical and .Economic Considerations The existence of access roads for almost the entire length makes this selected route technically and economically feasible. -Its short length will add to its economical considerations. Route Soil Conditions -~ (i) Description Baseline geological and geotechnical information has been compiled through photointerpretation and terrain unit mapping .(Reference 4). The general objective was to document the conditions that would significantly affect the design and construction of the transmission line towers. r~ore specifically, tpe objectives included the deline- ation of and forms of various origins, noting the occurrence and distribution of significant geologic factors such as permafrost, potentially unstable slopes, potentially erodible soils, pqssible active fault traces, potential construction materials, active floodplains, organic materials;, etc. Work on the air photointerpretation consisted of several activities culminating in a set of terrain emit maps delineating surface mater- ials and geologic features and conditions in the project area. The first activity consisted of a review o'f the 1 i terature concern- ing the geology of the Intertie corridors and transfer of the infor- mation gained to high-level photographs at a scale of 1:63~000. In• terpretation of the high-level photos·created a regional terrain framework which helped in the interpretation of the lo~1-leve1 1:30,000 project photos. Major terrain divisions identified on the high-l.evel photos werethenused as an aerial guide for delineation of more detailed terrain units on the low-leve,l photos. The primary effort of the work was the interpretation of 140-p 1 us photos cover- ing about 300 square miles of varied terrain. The land area covered in the mapping exercise is shown on map sheets and displayed in detai 1 on photo mosaics . 14-17 -' -...: ' . . . . -. ·~· . . ~·' ( i i) As part .of the terrain analysis, the. various bedrock .units.· and domt- nant lithologies· were identified usirt9·; pub1 ished u'"s. Geologjcal Survey reports •. The extent of these. units was,· .roughly delineated on the: photographs, and using exposure patterns, 'shade, texture, and other features of the rock unit as they appeared on the photographs, ~ ' 1 I t unit boundaries were drawn.. Terrain unit symgols denoting the vari .... uus lithologies were utilized on the m.aps~ · Physical characteristics and typical engineering properties of each terrain unit were considered-and a large chart for each corridor· was developed. The charts identify the terrain units. as they have been mapped and characterize their properties in numerous categories. This allows an assessment of each unit's _influence on various pro;.. ject features. Terrain Unit Analysis - The terrain unit is a special purpose term comprising the land forms expected to occur from the ground surface to a depth of about 25 feet. The terrain unit maps for the proposed Anchorage to Fairbanks trans- mission 1 ine show· the aerial extent of the specific terrain units which were identified during the air photo investigation and were corroborated in part by a limited onsite surface investigation. The units document the general geology and geotechnical characteristics of the area. The north and south corridors are separated by several hundred miles and not surprisingly encounter different geomorphic provinces and climatic conditions. Hence, while there are many la~dforms (or in- dividual terrain units) that are common to both corridors, there are also some landforms mapped in just one corridor. The 1 and forms or · individual terrain units mapped in both corridors were briefly des- cribed. Several of the landforms have not been mapped ind~pendently but rather as compound or complex terrain units. Compound terrain units result when one 1 and form over 1 ies a second recognized unit at a shallow depth (less than 25 feet), such as a thin sheet of glacial till ov~rlying bedrock or a mantle of lacustrine sediments overlying till. Complex terrain units have been mapped where the surficial exposure pattern of two 1 andforms are so intricately related that they must be mapped as a terrain unit complex, such as some areas of bedroc:k and colluvium. The compo-und and complex terrain units were . described as a composite of individual landforms comprising them .. The stratigraphy, topographic position, and aerial extent of all units, as they appear in each corridor, wer·e summarized on the terrain unit properties and engineering interpretations chQ.rt .. 14-18 ,~...._ ": I -·~ I I I I I I a+ ,.;:;._ ...... I ._~-, ..... I t~ ~~· ,;, ,__., I I I I I ' ' ,. I t I "-•-" I ... ~ I ' \[>' I 'W' I '-' I ,, ~ I "':,"' I -...~- 1 .' :.,..., I -..;.. I I r.:.l.·.· .. · ( "-··' .. ·a·.·.· __ .. < '•' t {d) Conc1uslons A study of existing information and aeri a1 overflights, together with add'i- -ti onal aeri a 1 coverage, was .used to 1 ocate the recommended route i n each ,of the southern; central, and nort-hern study are.as. Additional environmental information and land status studies made it pos- sible to align tha routes to avoid any restraints. Terrain unit_ maps describi-ng the general material expected in the area were prepar.ed specifically for transmission line studies and were-used to locate the routes away from unf avor,ab le soi 1 conditions whenever possible . The route shown in Figure 14 .. 5, represents the general location of-the -r_ecommended ·route. It also shows the existing surface transportations. Fi gores 1 through 14 of Appendix E3 are be 1 i eved to best meet the techni- cal, economical~ and environmental criteria. In these figures, a half- mile-wide corridor is located; within this width, the centerline of th~ proposed route is located. The centerline represents 400-foot-w-idth right- of...;way for the entire length of the transmission line, except in the seg- ment of the 1 i ne between Devil Canyon and the Interti e where the right-of- way is 700 feet. This segment has five single-circuit lines parallel to. each other. 14.4 -Towers~ Foundations and Conductors A transmission line intertie between Anchorage and Fairbanks is planned by APA. The intertie will consist of existing lines and a new section between Willow and Healy. The new section wi 11 be bui 1 t to 345 kV standards and wi 11 be fu 11y com-.. patible with Susitna requirements. (a) Transmission Line Structures (i) Selection of Tower Type Because of unique soil conditions in Alaska, with exten~ive regions of muskey and permafrost, conventional self-supporting or rigid towers will not provide a satisfactory performance or solution for the proposed transmission 1 i ne. · Permafrost and seas on a 1 changes in the soi 1 are known to cause 1 arge earth movements at some locations, requiring towers with a high degree of flexibility and capability for handling relatively large foundation movements without appreciable loss of structural integri- ty. The guyed tower is exceptionally suitable for these types of condi- tions. The recommended· type of structure for this study is there- fore the hinged-guyed steel x-tower (Figure 14.6). 14-19 1 I j I I : 'i J ij' I :r ''.·j' '' I, .r The design features incTi.ide hinged connections between the 1 eg members and the foundations ·which, together with the longitudinal guy system, provide for large flexibility combined with excellent stability in the direction of the line. -Transverse stability is provided by the wide leg base which also accounts for relatively srna 11 and manageable footing reactions~ -. - In add 1 ti on to the above features of the selected structure, the following are important favorable points: -The x-frame structure provides for 1 ess vi sua1 impacts than other structures. This results from the simplicity of design; -It requires little maintenance except for guy adjustment when needed; -Towers caul d be stored in remote areas \'lith out concern for replacement when needed because of vandalism or deterioration; -Easy for construction., a typical tangent structure consists of only six major components with bolted connections. All bolts are loaded in shear to eliminate any special consideration for torquing of assembly nuts during field assembly; and -Economically attractive, especially when considering all the engineering features it combines. The tangent tower shown in the figure represents the majority of structures required for the transmission line (about 90 percent of the tot a 1 structures) . However, speci a 1 types of structures. may be needed to satisfy a specific field requirement, wire stringing or line angles. Also, it is anticipated that the section between Knik Arm and University Substation will consist of double-circuit, self-supporting, single-pole type structures. (ii) Design Parameters -Clearance to Ground: A ground clearance of 32 feet may be ade- quate against expected field tolerances. -Spans: A wind span of 1400 feet, and weight span of 1600 feet is recommended for the tangent structure. This will combine economy and fl exi bi 1 i ty in spotting the structures. Longer ·spans, how- ever, may be needed to overcome specific site location. ~ Structure Height: An average height of 85 feet is expected thrOughout the transmission lines. -Insulations: The center phase is a V-string assembly to control sw1ng and to provide for smaller phase spacing .. The side strings are of the single type. 14-20 i',l· ,· ', '.:::;;,--. . :·1·.·.:;_· . , ' . . . I I '-... : ~' I ,, ... ~ I "'"" I I ' .I:- i >.,,·t:'" I ·,~~ I ·~- '" ~ \.'1."'~:/ I .a:. -. I .j,;·«.,.-. ' 1 ..... , I I I 1,•- ; ' ' l I I ..... 1'"' ' ' ' ~-1' -i I I ._,. I ...,_ I --...~ I I '· I t -Shield Wires: The shield wire when needed will be 3/8 inch by 7 strand, EHS stee 1 • (iii J .Loading .Condi tj ons~ Climatic studies for transmission lines were performed to determine likely wind and ice loads based on historical data. A more detailed study inc.orporating additional climatic data was performed to con- firm or modify the obtained data. Details of the climatic studies for· transmission lines may be found in Appendix B-6. The design loads acting on wires and structures are mainly based upon weather conditions. Four cases of loadings are thus establish- ed f?r the tower design. These are: -National Electrical Safety Code (NESC) Heavy Loading: This con- sists of 1/2-inch radial ice around the wires, 40 miles per hour wind at 0°F. This wind speed produces four pounds per square inch pressure acting on the project areas of cylindrical surfaces. -Extreme Wind Loading: Consists of 140-mph wind (produces 50 psf on cy1 i ndri ca 1 surf aces), no ice at 32° F. -Heavy_Ice: Consists of one-inch ice with no wind at 32°F. -Longitudinal Pull: No ice, no wind, 32°F. The pull pressure should be applied at any one conductor attachment. The first case~ NESC heaving loadings, will be applied to the major part of thel)transmission line; however, in certain areas where the weather conditions are more severe, the seccnd and third case may be applied. Loadings Upon Supporting Structures -Description of Loads • Vertical loads on supporting structures and foundations shall be their own weight plus the superimposed weight which they sup- port, including all wires, ice coated when specified.· The ef- fective vertical span for wires shall be determined with proper consideration of the effect of support at different elevations. The weight of ice shall be determined based on 57 pounds per -cubic foot. • Transverse Loading (perpendicular to the wires). This shall be determined from the following conditions: wind in conjunction with ice and wind without ice. The effective span of determin- ing the wind on wires shall be equal to one-half the sum of the adjacent spans between supporting structures. {b) Where a change in direction of wir£rs occurs~ .. a transverse.~loadlng~ upon the suppo~t i ng structure sha 11 be a resultant .1oad equal to · . the vector sum of the maximum transverse wind load and the resultant load imposed by,the wires because of their change in ·direction~ In obtaining these loadings, a wind direction shall· be used which will. give the maximum resultant load. -Longitudinal Loading Proper ~llowance should be made for longitudinal loads which may be produced on structures by wire stringing operations and con- struction techniques.· -Application of Loads The vertical, transverse, and longitudinal loadings p~eviously specified re 1 ate to loadings on the wires and structures. The component of these loadings should be considered acting simultane- ously. (iv) Results The hinged-guyed steel x-type tower is selected as the basic struc- ture for the project because of its flexibility and performance in withstanding the unique soil conditions in Alaska. Also, the x-tower is rated very favorably concerning reli ability, maintenance, construction, economy, and aesthetics. Tower Foundations (i) Geological Conditions A generalized terrain analysis was conducted to collect geologic and geotechnical materials data for the transmission line corridors" a re 1 ati ve ly 1 arge area. The engineering characteristics of the ter- rain units have been generalized and described qualitatively~ When evaluating the suitability of a terrain unit for a specific use, the actual properties of that unit should be verified by onsite subsur- face investigation, sampling, and laboratory testing. The three main types of materia 1 s a long the transm·i ssi on line are: -Good materia 1, which is defined as materia 1 which permits augered excavation and allows installation of concrete without special form work; -Wet 1 and and permafrost materia 1 which requires additional design details providing additional depth~ and --Rock materia 1 is defined as :nateri a 1 in which dri 1.1 ed-in anchors and concrete footings can be used. · 14-22. I I ·~· I ' ' I I ~ I ' ,, I •.._. ,, ... ~.- ' I ·.,~ I I I ·7 I « I; :~ ' -: .>~ I . •· A , . I ·,-_ ·~ t I t I . i:: ~· I ~ .... I - :,.! . ! I I ""'-' I I . 1.lo.-t I I Based on aerial, topographic., and terrain unit maps, the following is recorded~ -For the sbuthern study·· area: ·Wet 1 and and .perme}rrost m~teri a 1 s constitute the major part of this area. Some ro.ck and good foun- dation materials are present in this area in a very small propor- tion. -For the central sb.1dy area: Rock foundation and good materials. were observed· in most of this study area. For the northern study area: The major part in this area is the wetland and permafrost materi a 1 s. Some parts-'!Tave good rock materials. (ii) Types of Foundations The recommended two-1 egged x-frame tower is hinged at the foundation attachment connection for longitudinal freedom and restrained by fore and aft guying to an equa 1 i zing yoke. This arrangement wi 11 result in relatively smaller loads on the foundations. The recom- mended types of foundations are shown in Figure 14~ 7. These are: -Rock Anchor This type of footing is specified whenever good quality rock is encountered near the ground surface. The concrete piers are grouted into the rock with reinforcing bars; permissible bearing values with this type ·of footing are high. The entire hole, as shown in the figure, can be drilled using the small diameter hole size without casing. This type of hole is easy and quick to drill and presents little or no problems. The minimum depth of these holes is approximately eight feet, with the entire depth grouted to insure adequate anchoring below the maximum frost depth. -Pile Foundation Most of the t1ansmi ssi on line towers wi 11 be supported by pile- type foundations consisting of heavy H-pi1e beams driven to vari- able depths, depending upon the soil conditions. This type of footing is considered in the three study areas \vhen a good bearing stratum does not occur at normal footing depth or at a reasonable distance below. The piles driven to firm strator until the required penetratiori is reached would be less costly than other types of footings for the same type of soil. A minimum safety factor of 2 is recommended to be applied against uplift. The uplift resistance is always dependent on the skin frittion be- tween pile and soi 1. The safety factor may be varied if soi 1 tests or pile loading tests indicate that uplift resistance of the pile is greater or less • 14-23 ... :~.:·c~.~ ~-~~~~-~~~~~ •. ~~~~--~ -~-' •··~·-;0-.·~~ ~~~--~--~~~~---~~ ~-~~-~-~~. ,~. ~-~~~~--~~~~~ ••• ~.--~--~~-:~:r~~-·~-. ~: ~~~·-~-~~ ~~~-~-~~~~-:~t7~~-~;·~~ .~.~~-~r_~~---~~-a~:~~.~~]~~---~~ · , ' ,• ~ . -~ -~ ~ --. ~ ': ~. . '':i~~.:~- -~ ~ . '> . . • ·I·< ··~ (c) (iii} P .. e~i gn Crite~i a The greater part of the combined maximum ·reactions on transmission structure footing is usually from temporary load.s such as broken wire, wind, and ice. ·with the exception of. heavy-angle; dead-end, or terminal structures, only ~ part ~''f the total reaction is of a permanent nature. As a consequence, the permi ssi b 1 e soi 1 pressure as used in the design of building foundations may be considerably exceeded for footing for transmissio~ structures. The permissible values of soil pressure used in the footing design will depend on the structure and the supporting soi 1. The basic criterion is that displacement of the footing is not restricted be- cause of the flexibility of the selected x-frame structure and the hinged connection to the tower footing-~ The shape and configuration of the selected tower are important factors in foundation consider- ations. Loads on the tower consist of vertical and horizontal loads and are transmitted down to the foundation and then distributed to the soi 1. In a tower p 1 aced at an ang 1 e or used as a dead-end in the 1 i ne, the horizontal ·1 oads are responsi b 1 e for a 1 arge portion of the 1 oads on the foundation. In addition to the horizontal shear, a movement is also present at the top of the foundation, creating vertical down- load and uplift forces on the footing. To select and detai 1 design the most economical type of foundation for a specific tower location, soil conditions at the site must be known. Soi 1 i nvesti gati on wi 11 furnish this needed information. A soil boring is a guide to the type of soi 1 and its strength in re- sisting the forces on the tower. ·,he cost of soi 1 borings is sma 11 compared to the line cost per mile. The primary purpose of soil borings is to assure an adequate and safe foundation. Conductor Requirements ( i) Conductor Size Based on the transmission and power transfer requirements at the various stages of the Susitna development, economic conducfor size-s were determined. The methodology used to obtain the economic conductor size and the results obtained are outlined in Appendix El. Also included in the ,£\ppendix are the capitalized costs of transmission line losses. When determining appropriate conductor size, the economic conductor is checked for radio interference (RI) and corona performance. If RI and corona performance are within acceptable limits, then_the economic conductor size is used. However, where the RI and corona performance are found to be limiting, the conductor selection is based rin these requirements. 14~24 ...,:_ ·i ~ .. ;: I -·-•• ·• c '~ \ . ' :.;~ ; I I· I -...... I f, ' I I I "-'':; I I ~ I t ..-;.- 1 I ,....,*"J ' I • 9 I ' ... , t ' I . i " .-,.~ . ·' . \< I ~ I ,, I I I t .), .. - . ' . t. I -~, t t ... t / · ( i l) Recomme_nded S'i ze.s .· ~=~~r~h,!i's~sit~a~ ir~ansmYs~fan·~fias~--seen-'(fivlded, '·fnta . three:·=9"eo 9raphi ca1 sections each of Which has particular loading and environmental re- qui remefits. · In the section from De vi 1 Canyon to Will ow to Knik Arm, there wi 11 ~ be three circuits ultimately. Line loadings are sufficiently heavy to have a significant effect on conductor se.lection, and the eco- nomic choi~e is considered to be: · - 2 x 954 kcmil, 45/7 Alumi.num Conductor Steel Reinforced (ACSR), 11 Rail" conductor .. This conductor will also be used on the two circuits between Watana and Devil Canyon. In the section from Knik Arm to University Substation, the line loading is also heavy, but environmental considerations dictate the use of two circuits instead of three and the ~conomic conductor size is: - 2 x 1,351 kcmil, 54/19 ACSR!t "Martin 11 conductor. In the northern section from Devi 1 Canyon to Fairbanks, 1i ne load- ings are light and the economic conductor size would be smal 1 er than that allowed by RI and corona considerations. In this section the minimum conductor that can be used is: - 2 x 795 kcmi 1, 26/7 ACSR, 11 Drake 11 conductor. However, since the intertie between Hillow and Healy will be pr.e- built with 2 x 954 kcmi 1, it may be advantageous to standardize by constructing all of the remaining Susitna transmission to Fairbanks with 2 x 954 kcmil. This could be determined later. 14 .. 5 ... Substations To follow 14.6 -Dispatch Center and Communications (a) Existing Railbelt Dispatching Facilities The main generating·and load centers are located in Fairbanks and Anchor- age. Both areas operate independently of each other. It is proposed by APA to conne.ct the two systems by a tie in 1984 which wi 11 be operated at 138 kV. The power· transfer capability wi 11 be approximately 70 MW. Golden Valley Electric Association (GVEA) and Fairbanks Municipal Utility System (FMUS) constitute the two major producers of electrical power 'in Fairbanks. Although both utilities are intertied at 69 kV, they each pro- vide their own dispatching. GVEA is responsible for maintaining frequency in the Fairbanks area. 14-25 Chuga~h Electric Association (CEA), Anchorage Municipal Light and Power (AMLP) and Matanuska Electric Association (MEA) are the utilities which· serve the Anchorage area and· a11 areas north of Anchor.age, including Willow and Palmer. MEA is mainly a distributing utility and imports its powe.r from CEA and the Alaska Power Administration. --AMLP generates and - ---. distributes power in the Anc;.horage area. · CEA generates and -distributes electric pow.er in Anchorage and adjoining areas. ~ach utility provides its own control center facilities and is intertied at the 115 and 138 kV level .. Presently, CEA provides frequency control for the Anchorage are~o CEA has its own system control center· that provides dispatching and supervisory functions over its generating and substation facilities. (b) 1993 Rai lbelt Power System The introduction of Susitna hydroelectric power in the Railbelt area will require several hundred miles of transmission lines fr6m the Susitna River basin to Anchorage and Fairbanks. In fact, the ultimate development will require approximately 850 miles of transmission, 5 switchyards and 2 hydro generating stations at Watana and Devi 1 Canyon. Thermal generation at Fairbanks and Anchorage will still be in operation. The installed genera- tion capacity will be. over 2,000 MW at that time. To operate such an en 1 arged Rai 1 belt system, a contra 1 system or energy management system (EMS) wi 11 be required. This system wi 11 i flsure seeuri ty of the 345 kV transmission lines and switchyards/substations operations. The system will also exercise remote control and efficient dispatching of the generating units in the Railbelt. (c) Energy Management System Requirements To provide an efficient and secure dispatching system for the Railbelt, the following functions are proposed:· ( i) Supervisory Contra 1 and Oat a Acqui si ti on ( SCADA) Subsystem Includes real-time system data acquisition; remote control of power system devices; data base and data base management; data processing; operation data logging and report generation; and man/machine int~r­ face requirements. (i i) Generation Control Subsystem Includes automatic control of hydro and thermal units in the Rail- belt area to maintain interconnected system frequency and inter- change scheduling; economic unit operation; generation reserve eval- uation; and monitoring of system generation performance~ (iii) Power Scheduling and Load Forecasting Subsystem Includes the forecasting of system load and the scheduling of the power system generation to meet load requirements in the most eco- nomical and reliable way. .•.•... :1 '! t~ '---~·~ .·/': .i' .If :: ! i f I -~ .. t I I I I I t I f I ..... t I . .· ....... t.:: . ~~--·'-~---; ...... ,;;,_" ~~·~~.;;~·~~ ... .'<.~h.;..;...;_>••-'-·~· '""'"·"·-·'"'"'d 14-26 c_ .. t ·I·· ~--' t I t· I I t . I I - t~~-_ <' It i•. ~~ ' I ' t -1-; ' ' t .. ,._. '. ' (i v) Energy Accounti_ng Subsystem Includes collection, recprding, and processing of data power trans- actions-among various utilities in the interconnected sys-tem; .al~o the cost information and the savings/losses resulting from purchase/ sale. of power. (v) sxstem Security_Subsystem Includes the ability to evaluate system performance based on p.resent and predicted system conditions and the ability to evaluate the impact of probable contingencies (loss of generation, Toss of a transmission line, etc.). (vi) System Support Subsystem Includes on-line/off-line functions that could be performed by EMS to support engineering, accounting, and system operation organiza- tions. A more detailed description of the functional requirements is covered by a report entitled, 11 Energy Management System (EMS) ... System Requirementsn dated December, 1981 . (d) Energy Management System Alternatives An evaluation of alternative system configurations showed that two differ- ent .approaches to generation centro 1 are possi b 1 e: -Alternative I provides indirect control of generating units; and -Alternative II provides direct control of generating units. To formulate and evaluate the alternative configuration, the following cri teri a were used: -Configurations must fulfill functional requirements discussed above in paragraph (c); -Configurations must be technically, economically, and operationally main- tainable through the life of the" systems (10 to 15 years); and -Configuration must be technically feasible~ as well as proven. (i) Alternative I System Configuration The Alternative I system configuration is typical of the current offerings of several EMS equipment manufacturers (see Figure 14-12, EMS Alternative I System Configuration). The configuration is based on the assumptions that: -An in-plant, computer-based control system, located at Susitna Hydroelectric Control Center will be provided; 14-27 -The Susitna in-plant control system wi11 directly control a11 hydro gen~rating units and. the switching stations at ':Watana and Devil "CatlYQ!J:.,._::gMs will determine generation participation requirements on the unit level, but the units will be pulsed by the in-p1ant system. The supervi sary contra l _actions for Watana and Devil Canyon generating stations will be initiated at-EMS level, but the control functions wi 11 be ·imp 1 emented by the in-plant contra 1 'System; The northern and southern computer-based systems will receive gen- eration participation requirements from the EMS, but participation allocation and direct unit pulsing will be accomplished by these systems; and -EMS will directly monitor and control the following 345 kV substa~ ti ons: . Ester; . Willow; . Kni k Arm; • University; and . Others, as required. (i i) Alternative II System Configuration The Alternative II system configuration is also typical of current offerings of several EMS equipment manufacturers (see Figure 14 .. 13, EMS Alternative II Syst·em Configuration). The configuration is based on the assumptions that: -An in-plant, computer-based control system, located at Watana, will be provided to monitor generating units performance and control the units; ;. All Watana and Devil Canyon generating units will be controlled (raise and 1 ower) direct 1 y by EMS from system contra l center at w·i llow; · -All northern and southern area generating units will be dir.ettly controlled (raise and lower) by EMS, Willow Control Center; and -The switching stations at Watana and Oevi 1 Canyon and the other" four 345 kV substations will be directly monitored and controlled by the EMS Control Center. (e) Communication Requirements Effective operation of Et4S is very dependent on transfer of data and immediate response of supervisory functions such as control and telemeter1ng. Various communication systems to determine the most reliable and cost-effective communication media were evaluated. 14-28 ' t •c •' ~·< -·, . •... ·· .. -- I il t ' t I ~~-~ v - I t t t t I I l ..... ~ \ "·' 1 .• -; . t ' 1-.. -. l ' t ' t I ' ' t t I ' I I ' t ( i ) Power Line Carri. er This system is dependent on the state of the transmission line and, therefore, wi 11 not be avai 1 able when the 1i ne ts. down .. ( i i ) -Te 1 ephone Telephone companies provide data transmission services but the service is very erratic and unreliable for EMS applications. (iii) Microwave A microwave s.ystem is the most re l i ab 1 e and cost-effective sol uti on for the EMS communications. It is-highly desirable to install a· looped system for power system operation. Microwave systems are line-of-sight propagation and have an average standard transmission path of approximately 35 to 40 miles in an area of flat terrain.. The cost was estimated for appro xi mate ly 17 towers and repeater stations and without having the benefit of a detai 1 communication analysis. A microwave system is recommended for this application; (f) System Software Requirements The EMS should be provided wi 11 all the software required to satisfy the functional requirements described in paragraph (c) and all the software functions described below. The software should be the general-purpose operating system~ developed and tested by a major computer supplier and verified through many i nsta 11 ati ons in real-time applicationso It should provide a reliable, high-performance environment for the concurrent execution of multiuser, time-sharing~ batch, and time-critical applications. This software will consist of the fo1lowiftg major components: · -Executive services; -System fai lover and system restart; -Diagnostic programs; -Programming services; -Special data base, CRT display, and log/generation compilers; Engineering support; and -Special I/O handlers. Fortran compatibility of the software is essential, as most of the power application programs will be written in a high-level language. (g) Control Center Facilitx The facility wi 11 be the nerve center of the APA po.-Jer system operations of 345 kV transmission network and the electric power generation. All deci- sions concerning the operation and maintenance of the power system will be implemented through this complexo The importance of this facility dictates that its location be selected with a great deal of care. 14-·29 f f. · ( 1) Location of Site The control center.must be located on a site that provides high security against disruption of power system operations by human intervention or by acts of God, Acts of human intervention that must be considered are civil disturbances and terrorist activ"iti es. Natural disturbances that could occur are floods, fires, and land- slides. Several additional factors that have a bearing on the suitability of a site are: -Land availability; -Housing availability; -Transportation accessibility; -Educational facility availability; Climatic conditions; -Power avail-abi 11 ty; and -Centralized location in the power system. All of the above factors and the fact that a major S\.c-:itchyard is already 1 oc ated in the area make it appropriate to recommend Wi 11 ow as the location for the EMS center. Willow has additional qualifications as a possible capital site. The Willow center could also be the headquarters for the maintenance staff for the transmissio~ n~twork between·Susitna and Anchorage. The t~illow site also ha.: flat lands between it and Anchorage which also reinforces the reconu-nendation to use microwave as the communica- tion media. (ii)· Control Center Building The EMS control center building can be located on the same site as the Wi 11 ow swi tchyard. The construction of this bui 1 ding wi 11 require special facilities. This is all described-in the 11 Energy Management System (EMS) -System Requirements" report. Figure 14.4 provides a conceptual layout of the Willow Control Center. Thfs layout is based on a one-level building having a total space of 14,537 ft2. (h) Staffing Requirements The functional organization of the EMS Control Center must efficiently and cbmprehensively support all aspects of the operation and control of the Railbelt's power system. This also includes not only the day-to-day opera- tions, but also the coordination of power transmission and generation and. the ongoing training of personnel to improve efficiency and effectiveness. 14-30 I t I ' I t I t t I ·I t ' I I .'j I I ' I t f ' I ' I ' ' I I ' I; ' I , , t (i) fewer SysternOperattons Staff The fo11owing operating staff are recommended: ... ,;One <:hi ef . operator; '" -·Five ·senior operators; -Nine load operators; -One engineering technician; and -One clerk. The above organization can mai ritai n a 24-hour operation for 365 days a year. (ii) Computer Applications The computer applications section should be managed by a supervisor of software applications. Reporting to this supervisor-should be at least three additional software engineers charged with the duties of mai ntai ni ng SCADA, Generation Contra llt and System Secur- ity software programs. (iii) Power Coordination The power coordination group will be responsible for evaluat1ng unit commitment runs, preparing interchange schedules, and perform- ing after-the-fact power accounting, etc. This group will include one supervisor, one power production specialist, one budget specialist, two· power system engineers/analysts, two statisticians, and one power scheduler. · {iv) EMS Maintenance Graue The EMS maintenance group will be responsible for maintaining the EMS hardware and software. As a minimum, this group should include: -One system hardware engineer; -Two system software engineers; -Two hardware technicians; -Two RTU maintenance technicians; and -One commu·n; cation maintenance technician. {j) Budgetary Cost Estimates This paragraph provides overall budgetary cost estimates for the develop- ment procurement, system testing, and installation of EMS Alternatives I and I I. Costs for the EMS Contra 1 Center and Microwave System are also provided. These costs are representative of what Energy and Control Consultants estimate as the middle price bids of such a project .. The cost estimates for these configurations, microwave system, and EMS Con- trol Center .are given in January., 1982 dollars for a fixed-price contract that includes milestone pajfllents. Table 14.7 shows comparative cost estimates. 1'4-31 f (k) Recommendations Alternative t, shown in rigure 14-.5., is recommended for the Railbe1t Energy Management System as th.e most cost .... effecti ve and deslrab 1 e, system approach. Uniike Alternative Il, Alternative I system approach allows generation · control of the southern (Anchorage) and northern (Fairbanks} areas to remain under theit respective utilities. Alternative I also encourages the formation of regional control centers for each area. Presently, this is · the trend in power system control to decentralize in large geographical areas. Alternative I is also marginally less costly than Alternative II. Microwave is recommended as a communicating medi urn. Once provided, th i s system will perform the following additional functions: -Provide a transmission media for protective line relaying; and -,"rovide reliable voice communications between the various stations. This is very important in power system operations. · It is recommended that the El'1S Contra 1 Center be 1 ocated at Wi 11 ow within the Willow Switching Station compound. This location has many advantages and is centrally located in the southern Railbelt power system. It would also be reasonable to designate this location as a maintenance center for the transmission system. This area has lots of land for expansion. There a 1 so appear to be some p 1 ans to provide. a highway crossing at Kni k Arm.. If these plans materialize, Willow would only be one hour away by highway from Anchorage. 14-32 "· I ' ;1.·.· . . t I t I t t t t I ' I ' ' t: t I . . . I I I I t I ' I I ' a I t ' ' ' ,. I . '". ,.., LIST OF REFERENCES (1) Susitna Hydroelectric Project Planning Memorandum -Subtask 8.02 Preliminary Transmission System Analysis Acres, 1981 (2) Upper Susitna River Hydroelectric Studies Report on Transmissic..,n System U.S. Department. of Interior Corps of Engineers, 1975 (3) Anchorage-Fairbanks Tr ansmi ssi on Interti e Economic Feasibility Study Report IECo, 1979 (4) Terrain Analysis of the North and South Intertie Power Transmission Corridors, Prepared for Acres American, Inc~, and the Alaska Power Authority by R&M Consultants, Inc., Anchorage, Alaska. 14-33 TABLE 14 .. 1: POWER TRANSFER REQUIREMENTS (MW) INSTPLLED CAPACITY TRANSFER RE;QUIREMENT Susitna to Susitna to Year Watana Devil C"anyon Total Susitna Anchoraqe Fairbanks 1993 680 --680 578 170 1997 1020 -1020 867 255 Jll 2002 1020 600 1620 1377 405 TABLE 14.2: SUMMARY OF LIFE CYCLE COSTS TRANSMISSION ALTERNATIVE 1 2 3 4 5 Transmission Lines 1981 $ X 10 6 Capital $156.70 $159.51 $133.96 $140.94 $159.27 Land Acquisition 18 .. 73 20.79 18.07 20.13 18 .. 65 Capitalized Annual Charges 127.34 130.14 107.43 112.83 126.91 Capitalized Line Losses 53.07 54.50 64.51 65.82 42.82 Total Transmission Line Cost $355.84 $364.94 $323.97 $339.72 $347.65 Switching Stations Capital $114.09 $106.40 $128.:32 $120.64 $154.75 Capitalized Annual Charges 121.02 113.30 135.94 128.22 165.02 -- Total Switching Station Cost 2)5.11 219.70 264.26 248.86 319.77 --0 Susitna life Cycle Cost $590.95 $584.64 $588.23 $588.58 $667.42 I t I I t I I t I t t I t t I I ' ' I .. ... •.. ,, ... , ... .. ' '. : .. -, .. I t ' ' I I I I t ., .. t I I .. I ' .. , TABLE 14.3: .TRANSMlSSlON SYSTEM CHARACTERISTICS NUmBer or-NUmber-& !h ze Line. Section Len9th Circuits Voltage of Conductors (ml.) (kVJ (komi!) Watana to Devil Canyon 27 2 3~5 2 by 954 Devil Canyon to Fairbanks 189 2 ;345 2 by 795 Devil Canyon to Willow 90 3 345 2 by 954 Willow to Knik Arm ;;a 3 345 2 by 954 Knik Arm Crossing* 4 J 345 Krtik Arm to University Substation 18 .2 345 2 by 1351 *Submarine Cable !i TyPe 1. Technical -Primary -Secondary 2. Economical -Primary -Secondary 3. Environmental -Primary -Secondary TABLE 14e4: fECHNICA!-t ECONOtUC, AND EN.VtRONMENTAL C!HlERlA USED IN CORRIDOR SELECT ION. Criteria General Location Elevation Relief Access River Crossings Elevation Access River Crossings Tinbered Areas Wetlands Development Existing Transmission Right-of-Way Land Status Topography Vegetation Selection Connect with lntertie near Gold Creek, Willow, and Haaly. Connect Healy to fairbanks. Con- nect Willow to Anchorage. Avoid mountainous areas. Select gentle relief. Locate in .pro)(imity to existing transportation corridors to facllitate maintenance and repairs. Minimize wide crossings. Avoid mountainous areas. Locate in proximity to existing transportation corridors to reduce construction costs. Minimize wide crossings. Minimize such areas to reduce clearing costs .. Minimize crossings which require special designs. Avoid existing or proposed developed areas. Parallel. Avoid pri~ate lands, wildlife refuges, parks. Se:i.ect gentle relief. Avoid heavily timbered areas. , .. ,, I I t I t I ' I I I· I I I t I I I ' I "i ,:: I t I ·t I I I. I I I I I I· I I I I TABLE ·14. 5; TECHNICAL, ECONOMIC AND ENVIRONMENTAL CRITERIA USED IN CORRIDOR SCREENING . . Technical Primary Secondary Economic Primary Secondary Environmental Primary Topography Climate and Elevation Soils Length Vegetation and Clearing Highway and River Crossings Length Presence of Right-of-Way Presence of Access Roads TopO-Jraph}' Stream Crossings Highway and Railroad.Crossiqgs Aesthetic and Visual Land Use Presence of Existing Right-of-Way Existing and Proposed Development Secondary , ·· Length Topography · Sdils C.ultural Reservoir Vegetation Fishery Resources Wildlife Resources TABLE 14;6: SUMMARY OF SCREENING RESULTS RATINGS Corridor E:nvo . Econ. Fecfi • -Southern Study Area ······(;-] ABC' C~ c c *(2) ACfC A A A (3) AEFC F c A -Cental Study Area *(1) ABCO A A A (2) ABECO F c -C (3) AJCF c c c (4) ABCJHI F F F (5) ABECJHI f F F (6) CBAHI f' c F (7) CEBAHI F F c (8) CBAG F F c (9) CEBAG F F c (10) CJAG F F c (11) CJAHI F c c (12) JACJHI F F c (13) ABCf A c A (14) AJCO c .A A (15) ABECF F c c -Northern Study Area *(1) ABC A A A (2) ABOC c A c (3) AEOC F c F (4) AEf F c F A = recommended C = acceptable but not preferred f = unacceptable *Indicates selected corridor. I ; Summarx c A F A F c f' F F F F f F f F c c F A c F F -, I ·~·· . t I t I I I I I I I I I I I I I I I I I I I I I I I I I I I 'I I I I I I TABLE 14"' 7; EMS ALTERNATIVES I AND H COMPARATlVE COST ESTIMATES EMS Project · Hardware Software Auxiliary Internal (APA costs) Susitna In-Plant Control System Hardware Software Auxiliary Internal (APA costs) Microwave System EMS Control Center Building TOTAL Alternative I $ 2,942,000 :3,956,000 1,210,000 3,416,000 $ 11,524t000, $ 1,131 ,ooo 1,200,000 750,000 1,770,000 $ 4,851,000 $ 4;920,000 $ 3,853,140 $ 25' 148,140 - Alternab.ve 11 $. 31072,000 4,200,000 1,:350,000 3,606,000 $ 12,228,000 $ 1,094,000 1,200,000 700,000 1,875,000 . $ 4, 869 ,ooo $ 5,100,000 $ 3,853,140 $ 26,050,140 •. .···.f' .. · ·.· : ; -. ' . -... ····1. ··. -. ~> .. •. e,, -· I •• ':1 I ,, I I I -· ·I .. ,. :I ., 1- ·01. r------------, ....... ? . . I. . .,.._ SUIMMICE c:AILE ff I UNDEil KNJK ARM' . ·...--+-------( •f-n-- ....___~-Hr-- ' I . .. I ·~ L--- 75MVA 3-45-l~l<V --1~-fr-- II MI. -KNIK 'ARM --, J I I I I WILLOW UNIVERSITY (ANCHORAGE) ·v..l...v 250 MVA r~ 345- 1 ·.·'""r" 115l13.:S t(\1 ,L .. 1 .. 4y-' .,... """"-r-~ 1 I \ l l . - 1 I 4 X 150 MW UNITS • ' . "'! 26 MI. WATANA I I CHUGACH ELECTRfC -ASSOCIATION ANCHORAGE MUNICIPAL 6 x 170 MW UNITS· LtGHT 8 POWER .RAILBELT 345 K.V TRANSMISSION-SYSTEM. SINGLE LINE DIAGRAM 195ML SHUNT REACTOR ESTER {FAIRBANkS ) DEVIL CANYON - STAGING. LEGEND ---1993 ----2002. FIGURE 14.1 I I I • I I I ,. I I I I I I -.., .. " --~--"ALTERNATIVE TRANSMISSION LINE CORRIDORS SOUTHE,RN STUDY A.REA • loiT,!oltlll~~ ....... ,/- ,....."" DEHAI.f J ~Tfj· c ~~~------~ .. L E.G E _N 0 ----STUDY CORRJDOR • • • • •• • • • • • • • • I NTERTlE (HYPOTHETICAL) 0 5 ID ~~~----iiiiiiil SCALE IN MILES FIGURE .14.'2 .. I I I I I I. I I I I I· I I ,, I I I I ' ·I ALTERNATIVE ":·~NSMlSSlON LINE CORRIDORS Ctk!NRAL STUDY AREA •) LEGEND ---STUDY CORRIDOR • • • • • • ~ • • • • -• • I NTERTI E {HYPOTHETICAL) 0 5 10 ...-; d SCALE IN MILES FIGURE 14.S 1111· I I ••• . ' I •• I ··I 'I I I I I .I ;I I I :I ,. I @)· . I '; i~ ~. J t # ·~ r:/ '! -~.; • :~ ... 1 ·'-\ 1 . ~-LDCA~I(m MAP 't .,._) __ .... ~............ ..~-.,.....---., ... , ... , .... ___ .......,. LEGEND ----STUDY CORRIDOR ............... I NTERTIE (HYPOTHETICAL} 0 5 . 10 SCALE IN MILES .. II' 1 I . ··.· ....•.•.... ··· "' '·' I I. I I I I I I I' I ••• I :I .>. .,I ... ALASKA RAU .. ROAD '~ .. z:.·· ••• ••••• ••••• • ALASKA RANGE L.E G END --0--HIGHWAY · 1 f RAILRCAO -·-·-TRANSMISSION LINE ROUTE 0 P.ROPOSED SWITCHING STATION \ DENALI \ NATIONAL PARK \\ \ \ \ \ \ \ \ \ \ \ \ \ \ . WATANA GOLD CREEK THE PROPOSED INT.ERTIE 'FROM Wlt,.LOW T() HEALY WILL BE CONSTRUCTED FOR 345 KV CAPABILITY AND INITIALLY OPERATED AT 138 KV. fF SUSITNA IS PROVED FE~SISLE, THE FULL 345 t<V CAPACITY WILL BE UTlLlZED, ANCHORAGE TO PROPOSED TRA,NSMISSlON LlN ROUT£ TALKEETNA MOUNTAINS i ALASKA RAILROAD 12 24 MlLES SCALE O.!!·~~~iiiiiiiiiiiiiiiiiiiiiiiiliiiiil ·;a_ I --=1' I I -1 -I ~. I I I I I I :1 I; ·I I ••• : 0 . -I "' Q) '1][ ~--~3~3~·~~o-"~~~~~~~~3~3~·t~~~--~ ..... :t: C) Uj % ,.._-GUY l&l (!) < 0::: Ul ~ DETAIL B DETAIL A -. RECOMMENDED 345 J<V TANGENT TOWER \ ·l \ ' I ' I ,.., T ~ . . ..... ~ =====-v~~~~~ . . \ -· · · ·. · · . · ·.-. CiiAMSER HOLE TO : ACCEPT HEAVY DUTY GUY-THIMBLE WITH 1 1 GUY-WIRE, a PREFORMED GUY ATTACHMEttT DETAIL A rr---Tl l! J J ;J II 11 I I DETAIL B CLEARANCE iO PILE X -FRAME GUYED STEEL TOWER ' . ' .. I ' r SECTtON A-A. SECTION B-8 · -CHAMBER HOLE TO ACCEPT HEAVY DUTY ·GUV THIMBLE WITH (;UY WIRE 8 PRE- FORMeD -GUY ATTACH- MENT 1 '1 .. I I I I I I I I I I I I . I I I I I .I ,I < ~ \.,~ • .. • l 2 EACH, 3. CU. YD. CONCRETE 2 ANCHORS 1 H REBAR lN 4 11 9.' X e:.o" HOLES 2 .EACH t l ANCtiORt lt ~t:eAH . IN 4"0 X 8~0 u HOLE$ ROCK FOUNDATION .... -. -·. . ;~II .ST·EEL 4::: t.At.,;n·, · ' · · H-PILES, 2.5'-o" LONG 2 EACH 1 MULTI-HELIX ANCHORS 15!..0" LONG STANDARD FOUNDATION 2 EACH, MULTI-HEUX ANCHORS G 301-011 LONG 2 EACH, 12 u STEEL H-PILES, 50' LONG WET LAND FOUNDATION TRANSMISSlON TOWER FOUNDATION CONCEPTS FIGURE 14.7 [iii - - -•.. ·~-·-:····-- -·--· ·-- - - --; - - NORTHERN AREA CONTROL SYSTEM G ------RTU COMPUTER COMPUTER ~-----------------~ PERIPHERALS MAN/MACHINE INTERFACE COMMUNICATION SUBSYSTEM SOUTHERN AREA CONTROL SYSTEM G -------RTU SUSITNA HYDROELECTRIC CONTROL CENTER G ------· RTU RTU r-------RTU· SUBSTATION RTU 5 ENE_RGY fvlANAGEMENT SYSTEM, ALTERNATJVE I~ SYSTEM CONFIGURATION· FIGURE 1.4.12 ~~~~~ - - - - - - - - - - - -·-- -·c-- - - NORTHERN AREA CONTROL SYSTEM G -·-----RTU COMPUTER PERIPHERALS MAN I MACHINE INTERFACE COMPUTER COMMU~·,cATlON SUBSYSTEM SOUT!:ERN AREA CONTROL SYSTEM· G ------RTU --RTU SUSITNA HYDROELECTRlC CONTROL CENTER G I ~ RTU ---.·-.----RTU5 f--rru ~---·---RTU WATANA/DEVIL CANYDN SUBSTATION RTU SUBSTATIONS ENERGY MANAGEMENT SYSTE:M, ALTERNATIVE n, SYSTEM CONFIGURATION FIGURE 14.131 MIR I I ---· - - - ----- ---- - - - - - - - l .. ..... rE--------------------170 _______ ...__ _________ ___,~ 0 MECHANICAL AND FACILITY SUPPORT )200 SQ. FT. CONFER. · . TRAIN lPROG. ROOM ROOM 400 SQ. FT. 400 SQ. FT. OFFICE AREA 1500 SQ. FT 20 FEET 40 COMMUNICATION STORAGE ROOM 300 600 SQ. FT. SQ. FT. EMS EQUIP; MAlNT. ROOM 900 SQ. FT. HALL 7.5 FT. WIDE BATTERY ROOM 350 SQ. FT. UPS ROOM 350 SQ.FT. EMS EQUIPMENT ROOM 1500 SQ. FT. ENG. KITCHEN a MEN LAV. 450 SUPPORT LOUNGE LAV. a KITCHEN 350 SQ. FT. DISPATCH AREA 650 SQ. FT. DISPATCHING · ARENA 1500 SQ. FT. LOBBY 450 SQ .. FT. WOMEN LAV. 450 MAI-.AGEMENT AREA l t {(0 ~ ;(ij) l L• 600 SQ. FT. 900 SQ. FT. . SQ. FT. SQ. FT. 637 SQ. FT. TOTAL; 14,537.5 SO. FT. WILLOW SYSTEM CONTROL GENTER, FUNCTIONAL LAYOUT .___ ENTRANCE. FIGURE 14.14 l111R] \1 I 't· I I I I I I .,._ •• I .. I. I I -I I I I " I -~ NORTHERN AREA . ...... ~..-~ CONTROL _SYStEM .. ...,_.. ____ ............ ..,... ___ __, -~-FAIRBANKS TO ENERATORS -- ESTER SUBSTATION WILLOW SUBSTATION KNIK ARM SUBSTATION UNIVERSITY SUBSTATION SOUTHERN AREA l t----------__, ENERGY MANAGEMENT SYSTEM WILLOW CONTROL CENTER I I J • --- .. .-CONTROL SYSTEM-r------------~ ANCHORAGE TO GENERATORS WATANA SWITCHING STATION TO GENERATORS J~ j ~ H A l ~~-· ~' SUSJTNA HYDROELECTRIC CONTROL CENTER l t -' . 1( w 11' ~ TO GENERATORS DEVIL CANYON SWITCHING STATION ENERGY MANAGEMENT sYsTEM; ALTERNATIVE r, · IAPBmJ CONFlGURATION BLOCK DIAGRAM · F.lGURE 14.15 U [I) 0 I ,I I I I I I I I ·. . ••••• I 15 -PROJECT OPERATION This s..ection describes the operat·ion of the Watana anr. 'lil.Canyon power plants int.he Railbelt electrical system. Under current cona;t1ons in the Railbelt a tota,,1 of nine utilities share responsibility for generation and distribution of electric power with limited interconnections. The development in Sections 6, 8, and 14 of th~ Susitna project, size and schedule of on-1 i ne dates, and the associated transmission 1 ine requirements \'las necessarily based on the assumption that a single entity \'lould eventually be set up to optimize and control the dispatch and distribution of electric power· from all Ratlbelt sources. It is not the ptirpose of this report to discuss how thts entity should be s,'::ructured or come about. However, it is important to note that the Susitna project will be the single most significant power source in the System. Careful cr,nsideration is therefore essential of the dispatch and dis~ tribution of power from all sources by the most economical and reliable means. The general principles of reliability of plant and system operation, plant oper- ation and reserv::ir regulation, stationary and spinning reserve requirements and maintenance programming are also discussed. Estimates of dependable capacity and annual energy production for both Watana and Devil Canyon are presented .. Operating and Maintenance facilitles and procedures are described and the pro- posed performance monitoring system for the two projects is also outlined .. 15 .. 1 -Plan~ and System Operation Requirements The two plants comprising the Susitna Project wi 11 represent about 75 percent of the system capacity, having an i nstall!!d capacity of 1620 MW in a total system installed capacity of 2100 MW in the year 2010. In vis_'W of its large capacity and the extent of its influence on the operational characteristics of the power system, it is appropriate that the Susitna project operation should be discussed within the framework of general power system operation considerations of economy and security. Planning studies discussed in Section 8 were primarily concerned \vith selection of plant installed capacity such that an optimum installation could be provided to meet projected generation requirements over the life of the project, which may be. considered as 50 years or more. The main function of system planning and operation control to be discussed in this section is concerned with the allocation ·of generating plant in the system on a short-term operation basis so that the total system load demand is met by the available generation at minimum cost consistent wit~ the security of supp rY. The ger1era 1 objectives are generally the same for I ong-term planning or short- term operational load dispatching, but with important differences in the latter case. In the shor-t-term operational case the actual state of the system dic- tates system security requirements overriding economic considerations in load dispatching. An important factor arising from economic and security considera- tions in the system planning and operation is the provision of reserve capacity, both as stationary reserve as well as spinning reserve. 15-1 ~- The basis of system operation is the demand to be met at any moment and is the· aggregate of all consumers• demands in the interconnected Ra11belt System. Figure 15.1 shows the daily variation in demand during typ,;cal winter and summer wee.kdays afld the sea~onal v.ari ation in monthly peak demands for estimated loads in a typi (Ca1 year (fhe year 2000). 15.2 -Gent~ral Power Plant and System Rai lbelt GriJeria The power plants and electrical system are planned and constructed in such a manner that they can be operated so that the more probable contingencies can te sustained without less of-load. Less probable contingencies are also examined and the consequences to the system are determined and evaluated. The more prob~ able contingencies are usually defined as those that occur once in 5 or 10 years. The less probable contingencies have a probability of occurrence of once in 50 or 100 years. Automatic load relief (ancl when necessary in extreme cases, load shedding) is provided to minimize the probab;lity of total shutdown of area load which becomes isolated by multiple contingencie.: The following are basic reliability standards and criteria gene~dlly adopted in the industry for power systems. Further details are described in Section 8 for generation p~itnning, in SectiGi•S 12 and 13 for the power plants, and in Section 14 for thf! transmission system. (a) Installed Gene-rating Capacity Sufficient generating capacity is installed in the system to insure that during ee.ch year the probabi 1 i ty of occurrence of load exceeding the avai 1- able generating capacity shall not be greater than one day in ten years (LOLP of 0.1). The factors affecting the calculation of probability iryclude the ci-laracter- istics of the loads, the probability of error In load forecast, the scheduled mai ntenanc·e requirements for generating units, the forced outage rates of generating units, limited energy capacity of plants, effects of interconnections, and tr·ansmission transfer capabilities. The calculation of LOLP is done in the generation planning studies described in Section 8., (b) Transmi ssio~.System Cap:abi lity The high-vcltage transmission system in general and that associated \'lith the project in particular, should be operable at all load levels to meet the following unscheduled single or douLle contingencies without instabil- ity, cascading or interi"uption of load: -The single contingency situat~on is the loss of any single generating unit, transmission line, transformer, or bus (in addition to normal scheduled or maintenance outages) without exceeding the applicab1s ;emer- gency rating of any facility. The double contingency situation is the subsequent outage of any remain- ing equ·i pmen't, 1 i ne o;· subsystem without exceeding the short time emer- gency rat 1 ng of any faci 1 ity. 15-2 ·I·_-.- . l_ I I I I I· I aJ I I I I •• 'I II I I I I •• I I I I I I. I I I I I I '·· I .. ~·- 1 I ' •. : . ; ~, ,': . : :'' ' . ' .'~ ;-.,·.·:< ' .: . (c) ln the. s;1ngla contingen_cy situation, the power system must be capable of readjustment so that all equipment will be loaded within normal ratings, and in the double contrnuency situation within emergency ratings for the probable duration of thi:! outage. During ahy contingency: -Sufficient rt=active power (MVAR) capacity with adequate controls are installed to maintain acceptable transmission voltage profiles. Q The stability of the power system is maintained without loss of load or generation during and after a three-phase fault, cleared in normal time, at the most critical locatio~. It is impossib 1 e to anticipate or test the system for all contingencies _that can· occur in present or future confi gur at ions.. Typi ca 1 ex amp 1 es of the less probab 1e contingencies are: -Sudden loss of the entire generating capability of anypm'ler plant for any reasons. -Sudden loss of all transmission lines on a single right of way. -Sudden dropping of a very 1 arge 1 oad at a major load center. The abov~ stated general principles of reliability and security have been generally applied in the design of the project and system and are described and referred to in greater deta i 1 in the v·ari ous pertinent sections of this report. These principles constitute the basis of project operation plan- ning and criteria described in this section and else'lhere in the report. Summary Operational reliability criteria thus fall into four main categories: ( i) ( i i) (iii) Loss-of-lead probability (LOLP) of 0.1 or one day in ten years, is maintained for the recommended plan of Susitna project and,$ystem operation through the year 2010 (Section 6, Generation Planning). The single and double contingency requirements are maintained for any of the more probable outages in the plant or transmission system. System stability and voltage regulation are assured from the elec- trical studies (Section 14, Transmission Facilities). Detailed studies for load frequency control have not been done, but it is expected that the st i pu 1 a ted cri teri a wi 11 be met with the more than adequate sp·:nning re-s·erve capability with six uni.ts at VJatana and four units at Devil Canyon . 15-3 ·-· Jl .. ,, ,.~ .. ; . ·~." ·.-· (i v) ·The loss .Qf all S~~ftna transmiSsion lines on a single ri ght~'of-W(LY has a low~ level of probabtlity as described in Section-18 JJtider Rlsk Analysis •. In the event -of the loss of a1l ~1ines 1 the~hydroplant~; at Wat_ana. and De vi 1 Canyon are best su5 ted to t"estore po~Jer supply ·· qufckly :after. the first line· is reslored since th~y ate· desi'gne() for .. black start 11 oper.ation. In this t .. espect., hydrQ -plants ar~ super tor· to thermal plant.~. because of their inherent black start capabi 'Hty for restoration of supply t~ a large system. · 15.3 -Economic Operation of Units The Central Oi spatch Control engineer has the responsibility of deciding which generating units should be run at any particular time. Decisions are ma:de on the basis of a number of different pieces of inform_ation, including an ~'~order ... of-merit•' schedule, short-term demand forecasts, limits of operation of units and unit maintenance schedules. · (a) Merit-Order Schedule In order to decide which generating unit should run tomeet the system demand in the most economic manner, the Contra 1 engineer is proV'ided with information of the running cost of each unit in the form vf an uorder-of- meritu schedule. The schedule gives the capacity and fuel costs for thermal units-, and reservoir regulation limits for hydro plant'5. {b) O~~imum Load Dispatchina One of the most important functions of the Control Center is the accurate forecasting of the: 1 oad demands in the various areas of the system. Are a demand forecasts up to 8 hours ahead of unit loading are based on regional short-range weather forecasts for an estimate of heating and lighting demands plus light or heavy industry loads. Short-term forecasting up to 1 or 2 hours ahead is more difficult and remains the key factor to the secure and economic operation of the system. Based on the demand, bast c power transfers between areas and an a llawance for reserve, the tentative amount of generating plant is determined, taking into consideration the reservoir regulation plans of the hydro plants. The type and size of the units should ~lsn be taken into consideration for effective 1 oad di spat chi ng. · In a hydro-dominated power system, such as the Susitna case, the hydro unit wi 11 take up a much greater part of base load operation than in a thermal dominated power system. The hydro units at Watana typically are well suited to load following and frequency regulation of the system and providing spinning reserve. Greater flexibility of operation was a signif- icant factor in the selection of six units of 170 MW capacity at Watana, rather than fewer, larger size units. No significant load following can be done by the Devil Canyon units due to environmental constraints as described in Section 15.6. 15-4 ' ..• ,! " '• ; .... : ··I· a: I I ,>"1 II I I I' I I I I I I I .~ I ;I: , I ,,.:,.1 . .,. : . ,,. I I I I I ~I I I 'I I I I '.1 ·I ( ') -~ c·-_ (d) Operating Limits of Units . ~ Ideally; the plant having the lowest fuel costs should be. allo~ated 1oao· for as long as possible, and the most expensive p.l ant required to meet--the peak demands for as short a-time as possible. In practice, it ls not~ pos-· sible to meet this ideal situation due to security requirements for the system and the characteristics of the generation plant. There are strict constraints on the minimum load and the loading rates of machines and to dispatch load to these machines requires a system wide dispatch program taking these constraints into consiaeration. In general, hydro untts have excellent startup and load following characteristics, thermal units have good p.art-loading characteristics. Typical plant lo.ading limitations are given below: (i) Hydro Units ~ Reservoir regulation constraints resulting in not-to-exceed maxi- mum and minimum reservoir levels~ daily or seasonally. -Part loading 0f units is impossible in the rough zone of turbine operation (typically from speed-no-load to 50 percent percent load) due to vibrations, arising. from hydraulic surges. (ii) Steam Units -Loading rates are slow (10 percent per minute). -The units may not be ab 1e to meet a sudden steep rate of rise of 1 oad demand. -Usually have a minimum economic shutdown period (abou~ 3 hours). -The total cost of using c·anventional units i.nclude banking~ rais- ing~ressure and part-load opertions prior to maximum economic operation. (iii) Gas Turbines Cannot be used as spinning reserve because of very poor efficiency and reduced service life. -Requires 8 to 10 minutes for norma} start-up from cold. Emer- gency start up times are of the order of 5 to 7 minutes. Optimum Maintenance Program An important part of operational planning which can nave a significant effect on operating costs is maintenance programming. The program speci- fi. es the times in the year and the sequence in which p 1 ant is re 1 eased for !l· maintenance. Monthly, yearly, and 5-year maintenance schedules are pre- pared. In a large interconnected system with minimum reserves, optimum I I I' \ \ maintenance progratl1mipg" uses h~euristic methods.. The program planning takes into consideration the avai 1abi 1 ity of trained repair and/or maintenance personnel. ·Further details of Wat.anaqmdDevi 1 Canyon power plant mainten- ance. prpgrams are given in subsection 15 ... 8. 15.4 -Qnit Oeeration Securi~y Criteria During th~ operational load dispatching conditions of the power system, the security criteria often override the economic considerations of merit-order scheduling of the various units in the system. It is impossible to anticipate all the probable conti ngencles of operation, hence an operati anal approach im- plies the use of conttn:uous on-line data updating the state of the system for the information of the operator. Projected !=lectrical power system analyses are then carried out at frequent intervals to estimate the security of -operation. Also important in consideration of operational security are system response, load-frequency control and spinning reserve _capabilities. (a) Power System Analyses (b) During the planning stages, system stddies are carried out for all credible generation and network changes and probable contingencies. The trans- mission system studies (Section 14) were undertaken to check the more probable cases for load flow, short circuit and transient stability. The load flow studies determined the voltage levels and reactive power compensation for various plant loadings to meet the various load demands up to the year 2010. The transient stability study determined that the system was stable due to a transmission system fault resulting in the outage of a critical transmission line. Load-fr~quency response studies determine the dynamic stability of the system due to the sudden forced outage of the 1 argest unit (or generation block) in the system. The' generation and load are not balanced and if the pick-up rate of new generation is not adequate, loss of load wi 11 eventually result from under-voltage and under-frequency relay operation, or load-shedding. The aim of a well-designed high security system is to avoid load-shedding by maintaining frequency and voltage within the specified statutory limits. System Response.and Load-Frequency Control To meet the statutory frequency requirements, it is necessary that the effective capacity of generating _plant supplying the system at any givan instant should be in excess of the load demand. In the absence of detailed studies~ an empirical factor of 5/3 times the capacity of the largest unit in the system is normally taken as a design criterion to maintain s.ystem frequency within acceptable limits in the event of the instantaneo'lis loss of the largest unit. The factor 5/3 allows for the maximum dip in the fre- quency of the system. It is recommended that a factor of 1-1/2 times the lar_gest unit size be considered as a minimum for the Alaska Railbelt.sys-· tern, with 2 times the largest unit size as a fairly conservative value (i.e.~ 300 to 340 MW). 15-6 ..•. ~-"; ~ .~ ·•, -~ \ 1-.• ,, . · .. ~~ : :_·· ~ . :, c' i I I I I I .<: I I~ I I I I I I li I, I -_;;:.I~L '1 ., .• J"li-;- \; 'i{: ~~~~~"-'-'='--'-'--~--'-"----'-'--'-'---'~~-~~~~. -~-'--'"-~~--'-'-'-"_._;__""-----'~~-~~'--'---'-~-"-----'------'-~~~· ·~--~ ~.,__ ---~"-~~·~~~ .·.····, . . . I ' .. I I I I I I I I I I I, I I I .~ ,.. :. Upon sudden loss of ~eneration., .additional power may ini.tfally be derived from ,. the inertia of the rotating. machjnes (in tne first few seconds) and then from the spinning reserve depending on its governor 4ction in the next 10 to 20 seconds •. · Th.e mi nimurn system frequency should in the. meantime. be' conta:i ned ·with ... in the specified statutnry limits ti 11 ste~dy state oper.ation is reached. From preliminary studies regarding the plant response in the system, it appears that Watana is best suited for system frequency contra l and regulation. The Watana.unit being the largest in the system could be put under sensitive gover- nor control. Small generating stations and base-load thermal plant would he normally given steep governor drops so that they maintain their scheduled power output despite small changes in frequency. The quickest response in system generation will come from the h-ydro units. The 1 arge hydro units at Watana and De vi 1 Canyon on spinning reserve can respond in the turbi ni ng mode within 30 seconds. This is one of the parti cul ar1y important advantages of the Susi tna hydro units. Gas turbines can only respond in a· second stage operation within 5 to 10 minutes and would not strictly qualify as spinning reserve. If thermal units are run part-loaded (example, 75 percent), this would be another source of spinning reserve. Ideally, it would be advantageous to provide spinning reserv·e in the thermal generation as well in order to spread spinning reserves evenly in the system, with a compromise to economic loadiqg resulting from such, an operation .. · Det ai 1 ed load-frequency and spinning reserve studies should be done in the design stage of the project. (c) Protective Relaying System and D&vices The primary protective relaying systems provided for the generators and transmission system of the Susitna project are designed to disconnect the -· faulty equipment from the system in the fastest possible time. Independent protective systems are installed to the extent necessary to provide a _fast-clearing backup for the primary protective system so as to limit equipment ciamag·e, ·to 1 imi t the shock to the system and to speed restoration of service.. The relaying systems are designed not to restrict the norma 1 or necessary network transfer capabi 1 iti es ~f the. power system. 15.5 -Dispatch Con,tro 1 Centers The operation of the Watana and Devil Canyon power plant in relation to the cen- tral_ dispatch center can be considered to be the second tier of a three-tier control structure as follows: -Centra 1 Dispatch Contra 1 Center ( 345 kV network) at Will ow: manages the main system energy transfers, advises system configuration and checks overall security. -Area Contra 1 Center (Generation connected to 345 kV system, for example, Watana and Devil Canyon): deals with the loading of generators connected di r- ectly to the 345 kV n-etwork, switching and safety precautions of local sys- tems, checks security of interconnections to. main system. 15-7. ... District ·or''~ho~d Centers, (138 kV and 1 ower voltage networks): generation and distributton at lower voltage levels .. · For the Anchorage' andJ~,ai.tbanks..o-.area.~ the district center funct 1 ons. are i ncorpor·ated Jrt the respective ,area contra 1 centers. . The details of. the CentraJ Oi. spatch Control Center and of the· Watana Area Con- trol Center are given in Section 14. Each generating unit at Watana and Devi 1 Canyon is started up, loaded and operated and shut down from the Area Contra 1 Center at Watana according to the loading demands from the Central Dispatch Con- trol Center with due consideration to: c -Watana reservoir regulation criteria; -Devi1 Canyon reservoir regulation criteria; -Tu~"bine l9ading and de-loading rates; -Part loading and maximum loading characteristics of turbtnes and generators; -Hydraulic transient characteristics of waterways and turbines -Load-frequency control of demands of the system; and -Voltage re~u1ation requirements of the system. The Watana Area Control Center is equipped with a computer-aided control system to efficiently carry out these functions. The computer-aided control 'System al- lows a minimum of highly trained and skilled operators to perform the control and supervision of Watana and Devi 1 Canyon plants from a single control room. The data information and retrieval system will enable the performance and alarm ·monitoring of each unit i ndi vi dua 11 y as we 11 as the p 1 ant/reservoir and project operation as a whole. 15~6 -Susitna Project Operation A reservoir simulation model ~·1as used to evaluate the optimum method of opera- tion of the Susitna reservoirs and power plants at Watana and Devil Canyon~ Substantial seasonal· as well as over-the-year regulati.on of the river flow i:s achieved with the two reservoirs. The simulation of the reservoirs and the power facilities at the· two developments was carried out on_ a m~mthly basis to assess the energy_potential of the schemes, river flows downstream and flood control possibilities with the reservoirs. Details of the computer model are described 1n Appendix B~2. The following paragraphs summarize the main features of reservoir operation. (a) Reservoir Operation Gross storage val ume of the \~at ana reservoir at its normal maximum operat- ing level of 2185 feet is about 9.5 million ac/ft which is about 1.6 times the mean annual stream runoff (MAF) in the river at the dam site. Live storage of the reservoir is about 4.3 million .. ac!ft (75 percent of MAF). Devi 1 Canyon Reservoi.r has a gross storage of about 1.1 million ac/ft and live storage of 0.34 million ac/ft. . 15-8 .-., .. _ .. ·. ,.,-::. I ,, I I I I I I I I I I I I I I I '':1······. ' . I I I I 1·, ! I I I I I. I I I I I <I The reservolr.~imulatton model uses est'imated historical month'iy streamflows at the damsites for 32 years of avai 1 able records, reservoi ro characteristfcs~ and _power facility parameters as a basis of estimating the energy potentials of the developments it Hydrological, en vi ronmenta 1, . equipment and geotechP.:ical constraints were incorporated in the simulation to take account of varied requirements. (i) !1Ydrologica1· Constraints. The 32-year records of t)ydro1o£iy simulates seasonal and over-the- year flow characteristics of the river and as outlined in Section 7. 2., i ncl odes a ser·i es of very dry years <..nd correspondingly lower energy potential in those years~ (1 i) Environmental Constraints A variety, of environment a 1 cons tr ai nts were deve 1 oped and evaluated during the course of the study and the following have been incorpor-. ated in the reservoi.r·operations: -To reduce fluctuations in downstream river flows and water levels, no significant daily load following v-1i 11 be attempted from the Devil Canyon power station; -To maintain proposed fisherie.s mitigation efforts (further dis- cussion of flows to minimize impacts on downstream fisheries is presented in Volume 2 of this report) a minimum flow of 5~000 cfs will be maintained at Gold Creek at all times during the reservoir operation. - A min·; mum flow cf 500 cfs wi 11 be mai nt ai ned at a 11 times in the river reach between the Devil Canyon dam and the Devil Canyon tailrace outlet. · -Both Watana and Devi 1 Canyon reservoirs wi 11 be operated through- out spring and summer months to attenuate flood discharges to the extent possible. This will m;inimize potential damage to fisheries mitigation efforts due to high flood peak discharges. (iii) Equipment Constraints Generating equipment in each pCiwerhouse wi 11 be assumed to function at not less than 50 percent of maxi mum output to avoid rough opera- t ion. (iv) Geotechnical Constraints The assumed ratio of reservoir drawdown and filling have been lim- ited to ensure that no serious r~servoir slope stability problems will occur. In addition, maximum drawdown limits have been deter- mined which will produce the optimum combination of firm and average energy. The dra\-JdovJn l~1mits are discussed in more detail in .. · · · Appendix 82. (b) System Demand,and Reservoir Operating Rules Studies of reser-voir operation ~y~re based ori preliminary demand forecasts established for the ini~ial development _selecti-on studies (Section 5) and subsequently ,revised to take account of the system load forecast developed in power alternativ,e studies by Battelle (Section 5 .. 7). System reliability criteria (Section 15.2) requires a guaranteed or firm energy to be avail- ab le from the Wat ana and· De vi 1 Canyon developments. This energy is a func- tion of hydrology nf the river, reservoir storage and operating procedures. The reservoir simulation mode-l uses a procedure to maximize the firm energy potential of the developments, consistent with the various constraints listed above. An optimum reservoir opef•ation requirement was thus established by an iter- ative process to minimize net system operating costs while maximizing firm and usable energy production during the earlier years of demand growth .. Four· alternative operating rules for the Watana reservoir (A, B, C and D) were selected for study, to define the possible range of operation. Case A represents an optimum power and energy scenario, while Case D reflects a case of "no imp act on downstream fisheries" or 11 avoidance fl owsn. Gases B and C are intermediate 1 eve ls of power operation and dovmstream impact. These essentially define monthly minimum reservoir levels that should be maintained to provide firm e~ergy consistent with constraints-outlined above. For feasi bi 1 i ty report purposes, operation mode 1 "A" was adopted for· project design and approximate fisheries mit i gat i on measures developed (See Volume II). Details of the computer simulation runs for energy poten- tial and their impact on project economics may be found in Appendix B"2. Table 15.1 presents a summary of potential energy generation with different operating rules for ~Jatana and Devi 1 Canyon developments. The proposed reservoir operating rule. (Case A) is presented in Fi gur~ 15~1 .. ,This mode of operation represents target minimum levels to be strived for in the operation of the reservoir"" The target \-Jatana reservoir, level to be attained at the end of September each year is fixed at 2,190 feet. This level of 5 feet above the normal maximum operating level is designed to provide a higher level of winter energy production to meet the greater winter demand. This is consistent with river hydrology in that significant floods do not occur in the peri ad from October to early r~ay, and safety of -structures is not sacrificed. For th·is mode of operation, the average annu.al drawdown in the Watana res- ervoir is estimated to be 85, and at-Oevi 1 Canyon 55 feet. However, during the driest $equence of simu 1 ated ri verf1 ows, Watana waul d be drawndown to its optimum-minimum level of 2~045 feet (see Section 12.11). (c) Energy Potential of the Watana-Oevi 1 Canyon Developments Aver age annua 1 energy potentia 1 of Watana development is 3460 Gwh and that of Oevi 1 Canyon development is 3340 Gwh. A frequency analysis of the annua 1 energy potential has been made to derive the firm annua 1 energy potential or the dependable capacity of the hydro development . .:, . •. 15-10 I I . ·~ I I I I I I I I I I I I I I ., •• < •• . . I i i I I I I I I I I I -I I I I I I I I I The federal ~nergy Regulatory Commision. (PERC) in their publicati.on Hydroelectr1~_,,Power Evaluation (DOE/FERC-0031 of August 1979) defines the dependable caj):acity of hydroelectric plants as: ... the· capacity which, under the most adverse flow, conditions of record can be relied upon to carry ~ystem load, provtde dependable reserve capacity, and meet firm power ·obligations taking into account seasonal variations and other characteristics of the lo~d to be supplied 1•. As described in Section 7.2, the recorded 1 owest flow in the Susi tn a river at Go 1 d Creek has a recurrence frequency of the order of 1 in 10,000 years. This is considered an extreme.ly rare event to be considered in an electrical system reli abi 1ity evaluation. The critical streamflow sequence of record for the reservoir operation simulation is the 32-month peri.od between October 1967 and May 1970, resulting in extreme drawdown of reservoirs in 1970 and 1971 .. This sequence has a recurrence i nterva 1 of 1 in 300 years. .Based on the Railbelt system studies and previous experience on large hydroelectric projects~ it was assumed that a recurrence period of the order of 1:40 to 1:50 years dry hydro logi ca 1 sequence would constitute an adequate" reliability for the electrical system. An analysis of annual energy potential of the reservoirs showed that the lowest annual energy generation has a recurrence frequency of 1 in 300 years (See Figure 15.4). The second lowest annual energy of 5400 Gwh has a recurrence frequency of 1 in 70 years. This 1 atter figure has been adopted as the firm energy from the developm~nt. Expressed another way,_the firm energy as defined may fall short of its v a 1 ue by about 5 percent once in 300 years. This is again a ccnservati ve interpretation of the FERC definition. Tpe monthly distribution of firm annual energy as simulated in the reser- voir operation has been used in system generation p1anning studies. Average monthly energy based on the recorded sequence hydrology is used in the economic ana lysfs (Section 18 .1). (d) Reservoir Filling Sequence Given the relative sizes of the Watana and Devil Canyon reservoirs, 'it is apparent that the most si gni fi cant impact on the downstream fl q)w regime wi11 occur during filling of the Watana reservoir. Since this will be the first reservoir filled, careful planning is essential. · ( i) Watana Reservoir Impoundment Minimum monthly flows that must be maintained in the river below the dam during filling were established in consultation with fisheries and other environmental study groups and agencies. Table 15~2 pre- sents the minimum monthly flow that is cconsidered acceptable for · river maintenance and fisheries requirements during the fi 11 ing per- iod. With the above minimum ·flow requirement, it would take at 1 east 2-1/2 years of average stream flow to fi 11 the r~servoi r,. ,- 15-11 Other major considerations in determination of tn'e reservoir filling -sequence is the level of the fi11 dam construction~ avaiJable flood discharge facilities and tlie permi ssib 1 e risk of ove_rtopp~Jig the partially constructed dam during unusua.l floods in the river. It has been assumed that a min1murnstorage volume will be maintained behind the part1a11y constructed dam at a11:times during the filling period so that with available t:rischarge faciltties {low level o~tlets and service outlet works as they become available) a 1 in 100 year flood could be safely absorbed without overtopping-of the dam. This figure has been selected as acceptable on 'che assumption that short-term flood forecasting will be made during the filling period. Consequently, with car~ful monitoring of snow pack in the basin and storm tracking, potentially damaging streamflows could be predicted with ~ufficient warning to lower the reservoir level in time. It may be noted that the placement of the fi 11· dam critically contra ls the reservoir fi 11 i ng in average streamflow years and restricts earlier filling should wet years be experienced. The driest recorded streamflow sequence waul d extend the filling period by one year, The filling sequence in the years of average streamflow would allow first power on line by July, 1993. The units could be tested and commissioned prior to this date. A bonus in power and energy could be gained with one or two units installed by July 1992 when the power intake will be submerged sufficiently to allow power generation utilizing the minimum downstream discharge required. (ii) Devil Canyon Reservoir With Watana Reservoir in operation, the fi 11 i ng of the De vi 1 Canyon Reservoir is relatively easily accomplished. Average monthly power flows from Watana between the months October through December in a single year wi 11 fi 11 the reservoir while maintaining the minimu.-n downstream flow requirements (see Table 15.2). (e) Operating Capabi 1 iti es of Susitna Units (i) Turbine Performance The reservoir operation studies described above, show that the Watana plant output may vary anywhere from zero with the unit at standstill on spinning reserve, to 1,200 MW w~en the six units are operating under maximum output at maximum head. (Note that there is a limitation in loading of a single unit in the rough zone of tur- bine op.erati on from above speed-no-1 oad operation to about 50 per- cent load). The four units at Devil Canyon have a maximum total output of 700 MW at maximum head. The operating conditions of the turbines determining its character- istics are summarized in Tab 1 e 15.3. 15-12 ·. '1··. , -. - I I I I I I :1 I I I I I I I I ,, I I I _I I .,..,, I, ·,I; I I I ~ I I I I I •• I I I I I The turbine design-head corresponds to the ;,weighted average head-. Based on the predicted da.i ly 1oad curves through the year 2010 and expected reservoir operation, it is expected that-'each unit at . Watana is to supply a load averaging between l96.MW and 100 MW .. · This is the load which corresponds most closely to the best efficiency operation of the turbine. Similarly the Devil Canyon units will supply a load between 174 MW and 100 MW. (ii) Expected Unit Performance Characteristics The rated output of the turbine corresponds to full gate operation at the rated head. Each turbine should operate satisfactorily at the maxi mum head. The output of the generator is 1 i mited by its continuous maximum rating of 115 percent with a maximum temperature rise of 80°C. The coriti nuous. maximum rating of the generator deter- mines the maximum output of the unit and it will be necessary to limit the turbine output to this value accordingly at higher heads. The expected plant performance at varying heads between minimum and maximum heads is shown in Figures 12.22 and 12.23. The plant efficiency with different number of units in operation is shown in Figures 15.2 and 15.3. In practice, for Watana, the load following requirements of the plant results in widely varying loading and resulting efficiency of operation. (iii) Stability and Governing of Units The required flywheel effect (inertia) of the unit is governed by the stability requirements of the units, namely the stabilization of the frequency for small load fluctuations. The machine inertia also influen·ces the transient stability of the units during transmission line tripout following electrical faults in the system, a larger machine inertia decreasing the initial swing of th~ generators. On the other hand, a larger machine inertia decreases the natural fre- quency of oscillation of the machine, and increases the possibility of resonance with hydraulic surges in the draft tube and penstock. Electrical transient stability studies of the Railbelt system indi- cate that the "natural 11 inertia of 3.2 to 3.5 kW-sec/KVA for the Watana and Devil Canyon generators is adequate for electrical stab- ility of the system. The pertinent plant data for stability and governing are given in Sections 12 and 13 in Watana and Devil Canyon plant respectively. Pressure rise and speed rise are within normally acceptable limits of about 40 to 50 percent. A low ratio of the starting time of the ·-wd.ter masses to the mechanical starting time of the unit is an indi- ·cat ien of the hydrau~l i c stabi 1 i ty and acceptable response (prompt i ... tude time constant) of the governor. Good governing response and 15-.13 stability is 1nd1cated for the Watana and Devil Canyon units, and is imp-ortant from the overall considerations of system load· fo 11owi ng and load-fr.equency response of the units. -{f) Watana Plant Daily Simulati.on Studies The objectives of the plant daily simulation studies are to present perfor- mance studies of the se1ected 6-170 MW unit plant at Watana. The studies· demonstrate 1ts improved performance in comparison with a 4-250 MW plant. The si mul ati on program was arranged to~ -Study the operation and load following characteristics of the Watana powerp 1 ant with dif-ferent number and rating of units; -Determine the effect of minimum and maximum loading constraints·of the units; -Determine the effect of critical single or double contingency outages of units on the amount and type of spinning reserves available in the system; -Study the effects of maintenance outages and its impac_t on generation scheduling and system security; and --Check the operation of gas turbines and peaking plant. ( i) Computer Simulation Model To achieve t~e stated objectives, a computer simulation program was used to simulate Watana power plant and system operation. The Watana turbines and reservoir are mode led in detai 1 to simulate closely the reservoir regu1ation and load ~ following characteristics of the turbines. The model includes the following principal features: Turbine characteristics as a function of head, gate opening (flow), and efficiency are used in the mode 1. -Minimum loading limitations of the turbine due to rough zone of operation up to 50 percent of the gate -openings are constraints for turbine loading and operatio!'l~ -Maximum continuous rating (CMR) of the generators constitutes the maximum loading of the units. Higher turbine capability at higher heads is blocked at the generator CMR rating .. -Predict~d daily system load demand curves are used for two typical load shapes for winter and summer, respectively. Monthly peak load variation of the load is taken into account. -Reservoir characteristics as a function of level and storage. Average maximum and minimum reservoir levels are constraints for res.ervoir regulation and operation. 15-'14 , I I 1 .. I. I ~ I I .I I l ••• I I. I I I I I .I :.~ I I - I I I I I I I ,,,_· I I I I I I .... I I I • :r -Unit by unit 1oQding B:nd de-load1ng of Watana generators according to load demand (load~following} is done taking-into iccount all constraints mentioned above. The program 1 oads 'the units equally for maximum efficiency of operation • -.Loading steam plants as·base-load plants, and gas turbines as peaking plants. . -Maintenance scheduling of the generating units. Results of the Simulations Printouts of the results of the simulations are included in Appendix. . For each run, printouts are presented for the fo 11owi ng out---p~ut~s in a typical day in each month of the year 2000 (January to December·): -Watana plant kW out~ut; -Watana turbine kW output, with flow and efficiency for each unit; -Watana turbine utilization, showing number of units loaded; -Watana reservoir level; ~ Peaking plant kW output; -Total system load kW demand; -Total system reserve, including maintenance outage; -Watana reserve capacity; and Annual energy output of Watana, thermal plant, small hydro, gas turbine plants, and overall annual system energy. Simulation re.sult of a typical December, 2000 day is shown in Fig 15.6. The simulations indicate that the six unit Watana plant (6-170 MW) has superior overall performance in terms of load follow- ing, improved overall efficiency and minim~m loading constraints of the un1ts over the four unit plant (4-250 MW). The overall reliability of the six unit Watana plant is better than the four unit plant. During maintenance the six unit plant has a planned outage of 170 MW, as opposed to 250 MW for the four unit plant. During peak December loading, a double contingency outage of two units brings down system reserve to 107 MW for the 6-170 MW unit p 1 ant and to 1 ess than zero for the 4-250 MW unit p 1 ant for the year 2000 on study. The. simulations indicate that sufficient spinning reserve of a mlnl- mum of one Watana unit is available for all pe)ak d·ay loadings for the six unit Watana plant for the year 2000 on study . 15.7 -Performance Monitoring (a) Watana Dam Instrument<:Ltion is installed to enable the performance of the dam to. be. monitored to ensure that its behavior is within the limits assumed in the design and to enable any variations beyond those limits to be recognized quickly so that remedi·al action can be taken without delay. 15-15 lt is essential~hatcontltu.tous monitoring of the 1nstrunentation installed·' in the dam be carried out' by qualified pe~sonne1 who thoroughly un~erstand the significance of the readings and more importantly, the signific·an~e of variations in the readings. The 1 nstrumentati on is installed to monitor both short-term behavior during construction and initial filling and long-term-behavior ove-r the life of the dam .. The short-term is the most significant period when the dam is subjected to its initial loading and the responsibility for read1ng the i nst)"uments, reducing the data and eva 1 uat 1 ng the results over that period normally rests with the design engineer~ It is important that personnel who wi 11 ultimately be responsible for the monitoring are involved as early as possible i~ the development~ The most important aspects of the moni taring program and likely maintenance requirements are outlined below: (i) Foundation Abutment Pore Pressures and Discharge Frcm _Drainage Pressure Relief System Since sections of the foundation are frozen, the grouted cut-off may not be fully effective and leakage may increase a~ the rock tempera-· ture increases. This condi t; on would be ind 1 c a ted by increased discharge from the drainage system and would be remedied by additional grouting from _the grouting gallery, possibly combined with additional drainage holes. (ii) Quality of Discharge from Pressure Relief System Any discoloration nf the drainage system discharge would indicate the carry over of fine material either from the rock foundation or from the core. The problem area would be located and additional grouting carried out~ Water quality should also be monitored for any change in materia 1 content. (iii) Deformation of the Structure " Most deformation of the structure as observed by settlements and 1 at era 1 movements is expected to occur soon after construction and under i nit i a 1 fi 111 ng of the reservoir. Any excessive sett 1 ement would be made good to maintain freeboard .. Deformation records would be correlated with such data as reservoir level, heavy storms and- seismic activity. (iv) Routine Observations An essential part of any monitoring program is a regular routine visual inspection of all exposed parts of the structure and the area downstream of the dam for any unusual features such as 1 ocal settlement or other movement, zones of seepage discharge, wet areas, and changes in vegetation. All exposed concrete surfaces would also be inspected and records kept of any signs of distress, cracking or deterioration. · 15-16 I I ,, I !} I I I :1··· .. ' . I a I I I I I I I I .,.. I 'I ' ' '"'\ ' I I ,I ' I I I I I I I I I I ( v) Relict Channel~ ·Particular attention_ must be ·paid to monitoring the whole area of tt·1 r·elitt channel, .including· regular readings of piezometers and thermistors or S!lrface elevation survey monitoring and inspect1ons of. the discharge zone for changes in seep,age flo\'Js and any signs of piping fatlure. · 15.8 -Plant Operation and Maintenance The system demand varies throughout the year from a winter (December/January) peak to a summer (July/August) trough~ and from hour-to-hourthroughout the dayo The central dispatch center operates with the object of ensuring that sufficient plant is available at all times to meet the varying load in accordance with a meri.t-order schedule with due considerations to security. On the other hand, generating plant must be maintained periodically for various reasons: -Preventive maintenance, to ensure safe and reliable operation (performed either on load or shut down); -Corrective maintenance, to restore lost efficiency of plant; and -Emergency maintenance, arising from plant failure. Stationary inspection~ of plant are also ~equired. To meet the conflicting requirements of supplying load and maintenance outages, the plant maintenance program is planned to fit into the electrical system gen- eration program to determine the amount of plant which can be safely permitted to shut down for maintenance during each week of the year. Due consideration must be taken of such factors as: -Repair capacity in maintenance workshops; -Delivery of spares and materials; -Availability of specialized labor, special equipment and similar resources; -Expiration of statutory inspection periods;· -Weather conditions, for outdoor installation; and -Concurrence of outages between generating plant and transmission system. The plant at Watana and Devil Canyon are high merit-order plants and maintenance must be organized to minimize outages which affect availability. The expense and time of inspecting and maintaining the .large and important machines at Watana and Devi 1 Canyon in good condition, rather than operating the machines continuously until failure occurs in service can be justified on the basis of increased reliability and lower overall cost. With this in mind, sufficient spare capacity.must be availaqle in the Susitna plants and the system .at all times to cover planned and forced outage of the large units~ (a) Frequency of Inspections and Maintenance Th.e degree of inspection at the p 1 ants varies from frequent periodic vi sua 1 inspections to a complete major disassembly and thorough inspection at long intervals of about 10 to 15 years. Certain manufacturers and users recom- mend a major inspection after the first year of service. Factors influenc- ing the degre~ of disassembly and period between inspections include the following: 15-17 .·: ' 0 (b) >' ' ""''-"-.,......---;;--" ~ .. -Findings at previous visua1 inspections; -Result$ .. ,of previous tests; -Hi~story of similar mach-ines; and .. Frequency or starts, load cycli.hg and overloading during service. Ready avai 1 ability and access of stored data on the plant computer system of the records of previous inspection and tests as well as current perfor- mance trends (such as, for example, abnormal temperature recordings) wi 11 improve over a 11 ma1 ntenance and performance of the units. Experience records from machines similar to the Watana and Devi 1 Canyon machines indicate that a minimum maintenance period of 5 to 6 days are required for each machine, resulting in an outage of 150 to 170 MW capa- city for an average period of 50 to 60 days in the years. In exceptional cases, certain machines may be do'lm for greater maintenance periods. It is therefore re.asonab1e to allow a total of 2-1/2 to 3 months planned outage as a conservative approach to system generation and maintenance planning for the Susitna units~ In principle, these outages are scheduled during the months of June to August when the lower summer load demands make it possible to release the units for maintenance. The actual outages will be coordinated on a week-to-week basis with the planned maintenance of the units in the rest of the system and will take into consideration emergency shutdowns, breakdowns, delays in construction and maintenance and other unforeseen contingencies. Access and Maintenance in the Powerhouse Techniques developed both in the design and the operation of convention?-1 - underground hydroelectric powerplants have resulted in underground facili- ties which are not significantly more difficult to maintain than surface - p 1 ants. I so 1 ati on of underground i nst a 11 at ions from both penstock water and from tailrace water is a vita 11y important factor. Downstream water conduits with manifolds require draft tube isolating devices of appropriate design. Drainage and dewatering facilities must be highly reliable and of adequate capacity. There will be situations where a decision must be made as to whether to carry out maintenance and repair work on components underground or on the surface. Many i terns are large and heavy and therefore are best hand led by the powerhouse crane. Sufficient erection bay space and 1 aydown area between the generating units are provided for all normai maintenance and overh au 1 needs . Transformers wi 11 be moved within the access tunnel and transformer ga 1- lery by means of whee 1 s mounted on the transformer base~ The greatest demand in 1 aydown space within the powerhouse cavern is 11 kely to occur during the initial equipment installation process and the 10 to 15 year major disassembly/maintenance. The working area will be siz.ed to allow the simultaneous placing of turbine and generator components. ~ 1.5-18 ,, .. 1_.· . . f. I I "I I ... I I 1.; . I I I . -:> .... ' I 1- , .. a~ I --.~ I {\ .' •. \~!:: .· j I I I I I I I I I I ·I I I I I I ,, ·I (c) '- Adequate crane facilities are ~,provided both for installation. and mainten ... ance. _ The main powerhnus'e: overhead cranes have a_ capacity of about 200 tons and will be equipped with a zo ... ton auxiliary hoist. Small monorail hoists wi11 be provided as necessary at intermediate levels w1thin powerplant caverns for handling equtpment likely to require movement for. , ·l.nspecti on, maintenance and/or replacement._ . Major Overall Activities The major activities which require special space and handling considera- tion 1n the plants include: Replacing generator stator winding coils -Rotor inspection -Replacement of thru.st-beari ng assemblies -Replacement of runner seals Cavitation damage repair to runner I i mp e 11 e r .. -Repair and refi nlshi ng of water passage stee 1 and concrete surfaces '15-19 --performed in situ, may require removal of rotor assembly~ --performed for losseness, over- heating or short circuits, specially after a major trip out causing full overspeed. --designs normally permit removal and replacem~nt of components without major di smant 1 i ng, but at major overhaul intervals removal and strip down will be ad vi sab 1 e. --requires dismantling and removal of runner component. This may be possible from below by removing the draft tube cone and bottom cover, or alternatively from above by removing the he.ad cover and runner . . --will normally involve access through draft tube for in- spection and minor repair~ but ultimately requires runner re- moval as outlined above. -~access to the penstock will occur at about 5 to 6 year intervals. Major painting or refinishing will probably not be required until 10 to 12 years after commi s 1 oni ng. More frequent access will be provid~ ed to the downstream water pas~ sages when isolated by the draft tube gates and unwatered by the station dewatering system. ,~ . ~ ~ .'I ,. "'""~'~ ':~-~~ ---1 ,di (d) . . --Generator . circuit b_f'ea:ke.r · repair ... Transformer matotenance ---particular -attenti·on-and a high l~ve1 of majntenance ·are -requir.ed for. generator circuit bl"eake.rs. -~general maintenance will be carried out in situ in the transformer gallery .. , Major overhaul or repair requiring untaki ng of transformer wind-" ings will be done in the powerhouse erection bay with_ adequate crane f aci 1 1 ties. Maintenance Workshops and Operating/Maintenance Staff ' The. Watana and Oevi 1 Canyon powerplant are each provided with workshops to facilitate the normal maintenance needs of each plant. The workshop block includes operations for fitting and machining, welding, el•ectri.cal, and relay instrumentation, with adequate stores for tools and spare parts. TheWatanapowerplant will be provided additionallywith surface maintenance and central storage facilities to cater to the needs of both p 1 ants. Maintenance operation p l anni ngs of both plants are centro 1 ized at Watana. Staff wi 11 be normally located at Watana and housed at the operators vi 1- 1 age at Watana .. With centra 1 i ~ed contra 1 of the Susi tna project 1 ocated at Watana, the Devi 1 Canyon p 1 ant wi 11 not have a resident operating . and main- tenance staff. Prop.er road and transport facilities should be maintained between Watana aqd Devil tanyon to facilitate movement of personnel and/or equipment between the plants. The central maintenance staff should include the following recommended minimum personnel: 1 -superintendent of maintenance 1 -electrical maintenance engineer 1 -mechanical (and building) maintenance engineer 1 -instrumentation maintenance engineer 6 -assistant maintenance engineers; at peak maintenance will shift basis work on a Both the Wat ana and Dev 11 Canyon power plants are desi· gned to be norma 11 y operated from the Sus 1 tn a Area Contra l Center at Watana. The oper:ating staff wi 11 be stationed at Watana and would consist of the following personnel: 15-20 '=~=-.... ~~) .... ..::.,~~··-=_.· .. ·,"'-~ .·"'"-"'-. :"""""'. -''---'~--'----'-~~~:_____J/t'--~·-'=---'-~ . . .. : ::~ 1.1 . ,~1 .i:l ,, ..• ~ t I I· I t '"" I I I I I I I· I t ···:' . . :. . ·~ ( . ' . i ·., •. · .. ....... .·'·1,: •. <' . ;;:1.·· t.r' . : I I I I I I 1 .. . : . ' .. , ... I I I I I ,I " 1 · "'l sup_er-l ntend~nt, Sasi· tna project operations . · ·1."" chief :operator 3 ... control ,room operators (on shift basis) 2 ... powerhouse oper.ators . 2: ~·assistant powerhouse operators 2 -computer s.ystem operators When necessary, operators will trave 1 to Devil Canyon to assist in oper.ation :and/or maintenance programs at the powerplant. :• .15-21 · .. ~ " . ~.,.,~ . ···~.:::: ... .._' ··~~ 7 MONTH· OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP TOTAL NOTE: ' .~ t .TABLE .15~ 1: ENERGY POTENTIAl Or WATANA .:. DEVIL CANYON -DEV£LOPMEN1'S :-:. fOR OtF'FE.Rl!Nf RESERVOIR; OPERATING RULES . .. ENERG.'f POIEN11A.L G W -If c 0 w _~~. r & ~N. p. 0 N L J ·-wArJ\NJl ~-E V i L C A N Y _Q_Jf I" lf1 .M i:.NtKEY ·. AVt; .KAl:il::.' i:.l\lt.KI.iY t KM. t.Nt.r ,\.11 .AVt.H ,Aut .t.Nt; Klil . t;ASt. A c u A c I D A c D A· r.; D 234 200 172 281 2141 178 437 399 J34 511 422 34Ci 270 235 201 348 331 271 502 463 388 543 625 506 - "'" 322 276 236 445 397 364 598 547 4.58 B17 751 68): 283 242 208 383 '}57 3.25 590 480 403 715 677 618 228 202 173 318 3.35 293 .452 395 330 599 632 561 .235 201 173 276 330 277 470 398 335 532' 629 536 199 165 142 203 214 197 460 332 280 451 419 387- 180 152. 131 _-;: 180 247 174 462 304 286 465 536 399 170 135 111 175 212 191 492 ~23 278 478 485 460 182 209 345 258 267 374 j$7 471 755 521 579 784 170 311 531 344 327 545 321 659 1095 598 679 1095 158 151 155 249 158 166 29) . :326 .:390 463 346 J95 26:32 2479 2578 )459 3389 3354 5394 5099 5332 6793 6781 6168 Cases B and C were similar and only Case C was analy?:ed in detail .. 0 ; _·;·: .• :;,.: .·;.. '~ ' ~-i,. : 'l ·I', I I I I I I I I I I I: ··,: a: II I I I c~< •• • ••••••• . . .·1 I ' I I I •• .·~·· I •• I· .... I ..• 1. I I -a·· .. I TABLE 15,.2; . MINIMUM ACCEPTASLE F'LOWS i3E:LOW __ . ' WATANA. OAM DURING RESERVOIR FILLJNG MONTH OCT NOV OE:C JAN FEB MAR APR MAY JUN JUL AUG SE:P MlNIMUM.ACCEPTA8LE.FLOW CfS 2050 900 900 900 900 900 900 4000 4000 6000 6000 4600 • ~ .·-· rABLE:·15.3: . TURBtNt oPERA.llNG coNOITlONS Maximum net head MioimuiTI net head Design head Rated head Turbine flow.at rated head Turbineefficiency at design head Turbine-generating rating at rated head Watana 72S feet 580 feet 680 .feet 680 (eet 3550 feet 91~ 181;500 kW .; Devil Canyon 59.7 feet 238 feet 575 feet 575 feet 3800 feet 91% 164,000 I<W ·- ;:; . ' .,~ "-~· I I I I I I I· I I I I ..:; I I I • I I 1 . 'l .. c-c;..>l I ~:. ~ .• ; . I I I I I I -· I I ' ••••• • 'I ••• ....... ~... l ' i-. · .. ~.·. ·.•· l>;~ : I\~ ..f..'-~;.,· 100 90 0 ao c 0 70 ..J ~. 60 c. ·w a., 50 -40 .... z. 1&1 30 0 . a: ~ 20 10 0 . · / I ~ .. I ~· v -0 .• - 0 4 < a 12 HOURS 16 WINTER WEEKDAY HOURLY LOAD VARIATION ::'. lOO ' ~ ~ ' 90 c 80 c 0 ....J 70 .~ ··"« 60 w 1L I&. 50 0 .... 40 z 1&1 30 () a: Ill 20 D. 10 0 20 24 . :.,:;· ·.~· ---... .. / ...._,_ '""' ·. I( ' '\. ~ X 0 -. I - I """ ~ 4 .. ' •· 8 12 16 HOURS SUMMER WEEKDAY HOURLY LOAO VARJATlON 0 .. 20 NOTE: PEAK MW JULY 200Q AD= 658 MW . TYPICAL LOAD VARIATION lN ALASKA RAlLBELT SYSTEM 24 .. uoo 1000 900 z 800 2 .z 700 - 0 600 c( 0 -J 500 ~ ct kl. 400 n.. 300 200 100 0 ( .... : ... ; ' t . ·" ' ' ...... ~ .. ' I r I· I; : f I J F M I ........... _..., - l I A ·M J ,,-A · MONTH LOAD VARIATION IN YEAR 2000 / .. v / v .. I ) s 0 D FIGURE 15 .. ( .......... : ·• .•. ··.····.~~-. ' ' ' F • , , ' i - : . ~ ' : : ' : -~ . . ----... ,·I _ .• I I I ,, I I I I I I I I I· ,, I I I 86~+---*---~--------~--------~-------+--------;-~ ~ 0 ->-0 z w 0 82~+-------~--------~--------~-------+--------~~ fu U) .... -z ,:::» 78r-+-------~--------~--------r--------+--------~~ ~~+-------~~-------r--------+--------+--------~~ 100 500 700 PLANT CAPACiTY ( MW} WATANA-UNIT EFFICIENCY (AT RATED HEAD) 900 liOO FIGURE 15.21 Alii l··. I~·· I c I 94 --. I 90 I' S6 I -~ 0 - I I [J >-0 I z I.IJ 2 82 u.. Lt.. I.IJ ,, . .... z :::;) I 7S I I 74 I I 'I I I I -,:1 ~..~., I . /\NIT r-2 UNI'fS r-3 UNiTS /' -\/ ·."V V" !f \ v v l/ I . ,:) 100 200 300 400 500 PLANT CAPACITY ( MW) DEVIL CANYON-UNIT EFFICIENCY (AT RATED HEAD) . I [4 UNITS ' ~"- ~, ·- 600 flml FIGURE 15.31 AIIR I "' ill[. 0 -1& HRS. 1;0625£+01 ",... l.Ol.67Ef01 <( Q 18 t.o;oSE:f1fl -l <( ~" . , 1;07SOE+01 (.) kJ t-l.0792tt01 -m 20 . a.. "~ 1oOB3jEt01 >• 1-w > 1.0875£101 0 22 z -'""-t ... --. ,. l.<>917Ef01 ~ . 1.095SEt01 24 OHRS 1~1CCOH01 1G1C42E1Q1 2 1 .. 10S3Et01 4 1.1125Et01 1.1167Ef01 6 t.120BEt01 la1250Ef01 8 1.1292£+01 1.1333Et01 )-1.1375Ef01 <( 10 c l.1417Ef01 ...t <( n: 1.145BEt01 0 l&J NOON -m 1.1500H01 0... ~ >-w 1.1542Ef01 1-0 14 - l&J 1.1SB3Et01 0 1.16Z5H01 )6 1.1667H{)1 1.1703Et01 18 1.1750[~ Crj 20 1.1792EJ01 1 lOJln(li .1. I. U . ..SJ..I ~ 22 1.1875rf01 • 1)<1](J111 .lt .1.. • IV 24 HRS. 1.1~53El01 -iii r.dttViif ' · l.OB21Ei05 UATANA 6.4956EtQS 6.84o8Ef05 "7:1935EtOS 7,0NO£f05" 6.8608Et'05 6.o276H05 6.~935Et05 5.7272E+05 5.0609E105 • • - ----...o:~_-........... ____ .... __ _... ____ __;.._ ___ ~ ____ ..;.. __ ,...._ ___ f _, ______ ~*'!"!1"--~-~---..ii..;~~-------------"--~----~------f: ---~-------------~---------------------t -------~------~----------------~------+ -------------------------------:---t ------... ---------------------t . " ' H 'GEt1Kt1 EFF · iThooonoo 1~6239Etos 9.15sor-.ol .t.OOOOEtOO .1 .• 7117EfOS 9.07uSE-01 s.ooooEtob n4JB7Etos 9it747E-01 4.0000EtOO. 1.7735£+05 fh9866E-01 4;000()tf00 t~7152EtOS 9e0b61E-Ol 4,0000Et00 · 1.6569Et05 9.H55E-01 4.0000Et00 {.5994Ef05 9.1651E-Ol 4.0000EtOO 1\431BEt05 9,16SBE-01 3 I OOOOE tOO 1.6870£ +OS'. 9.1049£-01 l. . Kir;ia«UIIl l "RESER\T, -4, 47;1EtOS j 6.'01-44Ef05 ... ..,__.__._-...f 5.6o32EtQS ___ ... .,._ ... -+ . . , .. • .. 5~.316·5·· ·.E. t. 0 .. 5 __ ..;_ .... .,. s.416oaos ------~t . x 5 • 64 92Ef05 .__ ........... -•. + REStRV verses TIHt .·. . . ' Ha,d~oor 1 ~.t42BE406 0 INSTAL KWLOAII WAIRES . ·r.S310Et06 ·9.2956Ef05 ~.2646Et0S 1t53l0Ef06 · 9l6468Et05 :319129Et&S t.S310Ef0b 9.9935Ef05 3.5656EfOS 1.53l0Et06 · 9.S940Et05 J~664SEt05 1.5310£+06 9.660BEfOS 3.S971Et&S l r S~BB24EfO:i -----·--+ 1.~310Et06 9.4276Et05 ~~129SEf05 ! 6.1165£+05 --------t 1.5310Et06. 9~1935Et05 4.3634Et05 f6.}B2SEt05 ---------------+ t.5310Et06 S.5272EtOS 5.0292Et05 j1,4491E+05 . ----------------. ..... + 1.5310Et06 7•B609H05 5.6951Etos· 4.S959Ef05 ----------+-----------3.0000E+OO 1.6320£+05 9.1560E7'"01 ! 7.6141Ef05--------+----------~.1.S310Et06 7.6959EtOS S,S598Et05 4.6429£+05 4.3900£+05 4.13B1Ef05 4.3549£+05 4.5717.£+05 ~ .• 7S97Et05 S.SB47Et05 6.3796Et05 7.1728Et05 7.172SH05 7.172BEt05 7 .1729Ef05 7.2090[105 7.2451£-1·05 7.2S21Ef05 -. 6' .. 0 ... +/\1:' I• t.~ t. v.J a /\"lo9::-~or-t V;JU L I .J 7 °7 10noc-I I..S L I .J 7,67B1Et05 7.4252[{05 7 .1711[·105 t..44C5[~05 5.7250Et05 . -~----------M~----------t ___________ ,_;,. __________ + ---------------------~ J " -----------------------+ -------------------------1 -----------~--------------+ --------------------------------+ -----------------· ---------------------+ ------------~-----------------~------------+ ----------=------·-------------------------+ --------... --------------.-.-------------------+ -~----------~-----------------------------+ -~---------------~~-------------------------~ J ---------------------·--------~-----------~-~ " ' -______ _..,.,._...,_.,.,~_,. __________________ :--------------+ ______ ...., ___ .., ________________ ""'" ____________________ J. . ~ ---·----~----------..... ---______ ,...,. _____ ---oii---~..,--'"'"'!-----.l ' -----... -~--------------------~---'--------------------~ " I --------------------------------------.. ----~---+ ---~----"""1~---------------.... ----~--ooo! .. ,.-...,. _____ ,..._.._ ___ + ------------------------------------------+ ------------------~------------.aot---------t ' --------------------------------+ 3.0000E+u0 1•5476£+05 9.J7B9E-01 3.0000HOO · 1.4633Et05 9.1944£-01 3.0000£+00 1a379~£+05 9e125BE-01 3.0000£+00 1.4516£+05 9.1847£-01 3.0000E+OO 1.5239£+05 9.!B54E-01 3.0000Ef00 1.5966Et05 9.1657£-01 4.0000HOO 1.3962Et05 911393£-01 4.0000Et00 h5949H05 S\1662E-01 S.OOOOEtOO h4346Et05 9.1706£-01 5.0000Et00 S.OOOOE+OO 5.0000Et00 S,O@OE+OO 5.0000£t00 S.OOOOEtOO S.OvGOEtOO 5,0000EtOO S.uOOOEfOO S.OOOOHOO S,OOOOE+OO S,OOOOEtOO 4.0000Ef00 4,0000E+OO la.4346Ef05 1.4346£+05 1.4346£+05 1.4'41BEt05 1 • .1\490Et05 1.4564Et05 1 r ... ,6,..f0r ...... ~ .. t.. J 1 d.07BE+OS l.5862E+05 1.5356Et05 1.4S50Et05 1.4342Et05 L6121Et05 1.4315Et05 -~· ---;......,. .. ""*"" ~-•.-· !=. 9.1705£-01 9.1704E-Ql S\1704£-01 9.1'}62~-01 9.1S20E-01 9.18BOE--01~ 9 .1B33E-01 9.1629E-01 9 •. 16BBE--Ol 9.1826E-01 9.1963E-01 9 1'9_,.. .. ~ •.. 0 --~t:-~.l 9.1619£ -(}1 9.1671£-01 >-(.) z La.J o- i:i:t- LLZ La.JW 0 t-Ct: -W zo.. ::::::> _. i 7.8611Et05 --------.:.-.-· ------.------+ . . h5310Et0o 7.~429Et05 6.1123EtC-5 . S~1200Et05 _____ _;_ ______________ .-.-+ 1.5310Ef06 7.1900H05-6.3649EtOS · B·37l9Et05 -------------------------t t.5310Et06 6. 93B1EtOS 6.6165EtOS 8.1551Et05 · -------------------------+ 1.5310Et06 ].1549£+05 &..3994Et05 ; 7 • 93S3Et05 -~----------~----------+ 1.5310£+06 7 .3717Ef0S &.1822E1-05 f7.7203Ef05 ---... --------------------+ 1.3310Et06 7.5B97EtOS S.'9638Ef05 ·.L~ .• ·9 1 2 3 5 0 3E 4 · E+t. 0 05 5 .. -_-_--_-_-_ .... -_-. ---+-----+ 1.5310Et06 a.3B47HOS S.1684EtOS , ·o 1.5310E+Ob 9.1796EtQS -~.3730£+0.5 · S.3372Et05 -----t 1,5310EHl6 9,972BEtOS ~.S792805i 5.3372£+05 5,3372Ef05 · 5.3371E+05 ~ 5.3010Et05 . jS.2649£t05 : 5.2279Et05 4,8470HOS . 4t4711E+05 : ~I S790Et05 4.8319£+05 ' 5.0B48EtOS • -----r ---+ ------+ ---+ -----+ -----+ --t t i --t ----t 5.3389Et05 ------t 6.0615Et05 ---------+ 6.7B42Et05 --------------·-+ 0 1.5310Et06 9.972SE+OS 3.S7B6Et05i 1.5310EfM 9.972BEfOS ~.S78QEtOS 1.5310Et06 9t9729HOS J.5773E+OS 1.5310Et06 1.0009Et0.0.." 3.:S405Et05 ·t,5310Et06 1.0045£+0& S,.SQ3BEtOS h53H)Et06 1.00S2Et0S 3.~662Et05 1.5310Ef06 h0463Et06 3.vS47Et05 1 t5310£t06 l.OS39Ef06. 2,~70S1E+OS 1.:5310£+06 l.0731Ef0b 2,.S153Et05 1. 531 0Et06 1 t 04 78£+06 3. 0675EfQ5 1.5310£+06 t.0225Ef06 ~ .. 3l98Et0$ 1.5310H06 9 .• 9711E+OS 3.5732HOS h5310Et06 9.2485EtOS ~~2953£+05 ! .S310Ef06 8.525BEtOS .S.:01.74Et05: .. '·~, ··-·--,~ ,,.,.,. .. _ .... _ ~ .... ,~..; "' ...J ~-r o- 1-·3=. ~ ~-­w f-0 (f)·<t >-a (f) ..:.I WATANA PLANT SIMULATION . DECEMBER 2000 t I I , FIGURE 15.4 . liiJ •• I I I I I I I I I I I I I I I -I I I 16.-ESTIMATES OF.COST Thi-s section pre.sents estimates of capital and operating costs for the Susitna Hydroe 1ectri c Proj.Q ··\_,.. comprising the Wa.tana and Devi 1 canyon developments and associated transmilS\..,,~t and access faci liti·es. The costs of design features and facilities--incorporated into the project to mitigate environmental impa€ts dur- ing construction and operation ar·e identified. A cash flow schedule, outlining capital requirements during planning, construction, and start-up is presented. The section also includes estimates of the cost of capital funds required during construction, based on alternative financing scenarios. The approach to the derivation of the capital and operating costs estimates is described. The total cost of the Watana .and Devil Canyon projects is summarized in Table 16.1. A more detailed breakdown of cost for each development is presented in Tab 1 e s 16 . 2 and 16 . 3. 16.1 -Construction Costs This section describes the process used for derivation of construction costs and discusses the Code of Accounts established, the basis for the estimates and the various assumptions mad.e in arriving at the estimates .. For general consistency with planning studies, all costs developed for the project are in January, 1982 do 11 ars. (a) Code of Accounts Estimates of construction costs were deve 1 oped using the FERC format as outlined in the Federal Code of Regulations, Title 18 (1). The estimates have been subdivided into the following main cost groupings: Group Production Plant Transmission Plant General Plant Indirect Costs 16·--1 Description Costs for structures, equipment~ and facilities necessary to produce power. Costs for structures, equipment, and f aci 1 iti es necessary to trans- mit power from the sites to 1 oad centers. Costs for equipment and facilities required for the operation and maintenance of the production and transmission plant. Costs that are common to a number of construction activities. For this e.stimate only camps• and electric power costs have been included in this group~ Other indirect costs have been included i~ the costs under production, transmission, and genera~ plant costs. · (b) (c) Ov.erhead Construction Costs Costs for engineering and. adminis- tration. Further subd·ivision within these groupings was made on the basis of the various types of work 1flvalved, as typically shown tn the following ex amp 1 e: -Group: -Account 332: -Main Structure 332.3: -.Element 332.31: . -Work Item 332.311: -Type of ~lark: Production Plant Reservoir, Dam, and vlaterways Main Dam Main Dam Structure Excavation Rock The detailed schedule of account i,tems is presented in Appendix F. Approach to Cost Estimating The estimating process used generally included the following steps: -Collection and assembly of detailed cost data for labor, material, and equipment. as well as information on productivity, climatic conditions, and other related items; -Review of engineering drawings and technical information with regard to construction methodology and feasibility; -Production of detailed quantity takeoffs fr·om drawings in accordance with the previously developed Code of Accounts and item listing; -Determination of direct unit costs for each major type of work by devel- ·opment of labor, material, and equipment requirements; development of other costs by use of estimating guides, quotations from vendors, and other information as appropriate; -Development of construction indirect costs by review of labor, mater1 a 1 equipment, supporting facilities, and overheads; and -Development ·of construction camp size and support requirements from the 1 abor demand generated by the construction direct and indirect costs. The above steps are discussed in detail in the following: Cost Data 0 Cost information was obtained from standard estimating sources, from sources in A 1 ask a, from q·uotes by major equipment supp 1i ers and vendors, and from representat) ve recent hydroelectric projects. Labor and equipment costs for 1982 were developed from a number of sources (1,2,3) and from an analysis of costs for recent projects performed in the Alaska envi~onment. It has been assumed that contractors wi 11 work an average of two 9-hour shifts per day, 6 days per week, with an expected range as follows: 16 ... 2 -·· I ·..-· ... •• I I I I I I I I I I I I I I I ·I I I I I I I I I I I I I I I I I I (d) Mechan1 cal/Electrical ·work Formwork/Concrete Work Excavation/Fi 11 Work B~hoUr shifts 9 t..: .·· ·· h ~ rt . _ .. ,,our s ,." ·s 10-hour shifts These assumptions provide-for high utilizatio.!l of construction equipment and reasonable levels of overtime earnings to attract workers. The two- shift basis generally achieves the most economical balance between lat )r and camp costs. Construction equipment costs were obtained from vendors on an FOBAnchorage basis with an appropriate allowance included-for transportation to site. A representative list of construction equipment required for the project was assembled as a basis for the estimate. It has been assumed that most equipment would be fully depreciated over the 1 i fe of the project. For some activities such as construction of the Watana ma'in dam, an allowance ·tor major overhaul was included rather than fleet replacement. Equipment operating costs were estimated from industry source data, with appropriate modifications for the remote nature and extreme climatic environment·of the site. Fuel and oil prices have also been included based upon FOB site prices. Information for permanent mechanical and electrical equipment was obtained from vendors and manufacturers' who provided guideline costs o"n major power plant equipment. The costs of materi~ls required for site construction were estimated on the basis of suppliers' quotations, adjusted for Alaskan conditionsc;;. Seasonal Influences on Productivity A review of climatic conditions, together with an analysis of experience in Alaska and in Northern Canada on large construction projects was under- taken to determine the average duration for various key activities. The re.su 1 ts of this eva 1 uati on are presented in Tab 1 e 16.4 and these durati orfs have been used to develop the construction cost estimates. In general, it has been conservatively assumed for current study purposes that a 10-month construction season is the maximum feasible duration and that most work operations will cease during December and January because of the extreme cold weather and the associated lower productivity. Productiv .. ity is assumed to decrease by 30 percent during the November, February, and March time periods. This productivity decrease results from a combination of low temperatures, reduced daylight hours, precipitation, and soil condi- tions. Studies by others (4} have indicated a 60 percent or greater decrease in. efficiency in earthwork operations under such adverse conditions. Typical efficiency curves for the Fairbanks locat'ion, which were used as a guide, are shown in Figure 16.1. Although this curve cannot be used directly for the Susitna Hydroelectric Project, it illustrates the relative seasonal effect on manual labor, hauling, and earth .excavation that is likely to be experienced. 16-3 .... . . ' • ·: ..... J . ,' . . Studies performed as part of this work program indicate that,the general construction productivity at the Susitna damsite during the·months of April through September• would b~ comparable with that in the northern sections of the Western United States. Rainfall in the general region of the site .. , s moderate b.e.tween mid ... Apri1 and mid-October ranging from a low of·o.75 inches precipitation in April to a high of 5.33 inches in August. This moderate amount of rainfall should not create significant problems during fill placement activities because of the good quality river run borrow material used in the haul road and dam embankment. Temperatures in this period range from 33° to 66° for a twenty-year average. In the five-month period from November through March, · the temperature r•anges from 9.4°F to 20.3°F with sno~tJfall of 10 inches per month. Exc av at ion of grave 1 from· the river or be 1 ow ground \'later leve 1 and p 1 acing the m~teri a 1 in a fi 11 wou 1 d have to be di S'cont i nued during these months because of the snow and ice intrusion and the inability to obtain satisfactory compaction with frozen material. However, other construction activities could continue during this period (consideration being given to the·cost of snow removal) with possibly the exception of the two coldest months pf December (9.6°F) and January (9.4°F). Productivity would decrease 10 percent to 30 percent from the normal or base production rates during the periods of mid-October through November~ and February through mi d-Apri 1, and up to 50 percent during December and January depending upon the activity and the exposure. (e) Construction Methods The construction methods assumed for development of the estimate and . construction schedule, are gene.rally considered as "normal", tn line with the available level of technical information. A conservative approach has been taken in those areas where more detailed information will be developed during subsequent investigation and engineering programs. For example, normal drilling, blasting, and mucking methods have been assumed for all underground excavation. Also conventional equipment has been considered for major fill and concrete work. Various construction methods were considered for several of the major work items to determine the most economicallypractical method. For example, a comprehensive evaluation was made of the means of exc av at i ng materia 1 from Borrow Area E and the downstream river for the Watana dam shells. A comparison of excavation by dragline, dredge, backhoe, and "sauerman'' (scraper) bucket methods was c made, with consideration given to the quantity of material available, distance from the dam, and location in the river or adjacent terraces. (f) Quantity Takeoffs Detailed quantity takeoffs were produced from the engineering drawings using methods normal to the industry. The quantities deve 1 oped are those 1 i sted i ~ the detailed summary estimates in Appendix F. (g) Indirect Construction Costs ·- Indirect construction costs were estimated in detail for the civil con- struction activities. A more general evaluation was used for the mechani- cal an~ electrical work. 16-4 ·····.\ . . . . ;··-.·.: ... ' • J .I··••·. ',. •' I I I I I I I I I I •• I I I" I .I I "' I I 'I I I I I I I ·.·1 I I I I I I 'I " . . •• Indir·ect costs included the following: -Mobilization; -Techni:eaT and supervisory personnel above the level of trades foremen; -All vehicle costs for .supervisory personnel; -Fixed offices, mobile offices, workshops, storage facilities, and lay- down areas, including all services; -General transportation for wo.rkmen on site and off site; -Yard cranes and floats; Utilities including electrical power, heat, water, and compressed air; Small tools; -Safety program and equipment -Financing; -Bonds and securities; Insurance; -Taxes; -Permits; -Head office overhead; -Contingency allowance; and -Profit. 16.2 -!:!]tigation Costs As discussed in previous sections, the project arrangement includes a number of features designed to mitigate potential impacts on the natural environment and on residents and communities in the vicinity of the project. In addition. a number of measures are planned during construction of the project to mitigate similar impacts caused by construction activities. The measures and facilities represent additional costs to the project than would be normally required for safe and efficient operation of a hydroelectric .development. A summary of these mitigation costs is presented in Table 16.5. The costs include direct and indirect costs, engineering, administration, and contingencies. A number of mitigation costs are associated with facilities~ improvements or other programs not directly related to the project or located outside the project boundaries. These would include the following items: -Caribou barriers;· -Fish channels; -Fish hatcheries; -Stream improvements; -Salt licks; -Recreational facilities; -Habitat management for moose; -Fish stocking program in reservoirs; and -Land acqui st ion cost for _recreation. It is anticipated that some of these features or programs wi 11 not be required during or after construction of the project. In this regard a probabi 1 i ty fac- tor has been assigned to each of the above items, and the estimated cost of each reduced accordingly. Th.e estimated cost of .these measur:es, based on this proce- . dure, is approxi~ately $9 million. These costs have been assumed to be covered by the construction contingency. 16-5 A number of stud.ies and programs wi 11 be required to,.monitor the impacts of the project. on--the environment and to develop and record various data rluring projec.t construction and operation1l These include the following:. -Archaeological studies; Fisheries and wfldl i fe studies; .... Right-of-way studies; and -Socioeconomic planning studies. The costs for the above work have been estimated to be approximately $ -------and included in the owner•s costs under project overheads. 16.3 -Operation, Maintenance~ and Replacemeiit Cost.s The f aci 1 i ties and procedures for operation and mai nten~nce of the project are described in Section 15. Assumptions for the size and extent of these facili- ties have been conservatively made on the basis of experience at large hydro- electric developments in northern climates, noteably Canada. The annual costs developed for operation, maintenance, and interim replacement for the Watana and De vi 1 Canyon projects and the transmi ssi or~ faci 1 it i es are summarized in Tab 1es 16 ·~' 16 ._ and 16. . 16.4 -Engineering and Administration Costs Engineering has been subdivided into the following accounts for the purposes of the cost esttmat-es: -Account 71 . Engineering and Pro3ect Management . Construction t4anagement . Procurement -Account 76 . Owner • s Costs Thrc total cost of engineering and administrative activities has been estimated . at 12.5 percent of the total construction costs, including contingencies. This is in general agreement with experience on projects similar in scope and com- plexity. A detailed breakdown of these costs is dependent on the organizational structure established to undertake design and management of the project~ as well as more definitive data relating to the scope anti nature of the various project components. However~ the main elements of cost included are as follows: (a) Engineering and Project Management Costs These costs include allowances for: -Feasibility studies, including site surveys and investigations and logistics support; 16-6 -·-·· •. ·-. '. '·- 1 I·· I- I I :1 I- I I I I I I I I I- I" •. I; -I I -- I I I I I I I I I I I I ·I I I I I ... Preparation of a license application to the FERC; - -Techn·i cal and administrative input for other federal, state and 1oca 1 permit and license applicatlons; -Ov~rall coordination and administration of engineering, construction ~ management, and procurement activities;_. ·-" :::~~="'"-:'c::;: ~~ >-· -Overall p·lanning, coordination, and monitoring activities related to cost and schedule of the project; -Coordination with APA and reporting to APA regarding all aspects of the project; -Preliminary and detailed design; Technic a 1 input to procurement of construction services, support services, and equipment; -Monitoring of construction to ensure conformance to design requirements; -Preparation of start-up and acceptance test procedures; and -Preparation of project operating and maintenance manuals. (b) Co~struction Management Costs -· Construction management costs have been assumed to include: -Initial planning and scheduling and establishment of project procedures and organization; -Coordination of onsite contractors and construction management activities; -Administration of onsite contractors to ensure harmony of trades, comp 1 i ance with app 1 i cable regulations, and maintenance of adequate stte security and safety requirements; -Development, coordination, and monitoring of construction schedules; -Construction cost control; -Materia 1, equipment and drawing centro 1; -Inspection of construction and survey control; -Measurement for pa.J111ent; -Start-up and acceptance test for· equipment and systems; -Compilation of as ... constructed records; and -Final acceptance. ,. (c) Procurement Costs -· Procurement·costs have been assumed to include: -Estab 1i shment of project procurement procedures; -Preparation of non-techni ca 1 procurement documents; -Solicitation and review of bids for construction services, support services, permanent equipment, and other items required to complete the project; -Cost administration and control for procurement contracts; and -Quality assurance services during fabrication or manufacture of equipment and other purchased items. 16-7 (d) Owner's Costs Owner's costs have been assumed to include the fq1Towing: ... Administration and coordination of project management and engineering organizations; -Coordination with other state, local,_anrJ _te_deral agencies and groups having jurisdiction or interest in the project; -Coordination with interested public groups and individuals; -Reporting to legislature. and the public on the progress of the project; and -Legal costs (Account 7?). 16.5 -Allowance for Funds Used During Construction At current high leve 1 s of interest rates in the fi nanc i a 1 market-place, AFDC wi 11 amount to a significant element of financing cost for the lengthy periods required for construction of the Watana and Devil Canyon projects. However, in economic eva 1 uati ons of the Susi tn a project, the 1 ow real rates of i ntere:st assumed would have a much reduced impact on assumed project development costs. Furthermore, as discussed in Section 18, direct state involvement in financing of the Susitna project wi 11 also have a significant impact. on the amount, if any, of AFDC. For purposes of the current feasibility ~tudy, therefore, the convent1onal practice of calculating AFDC as a separate line item for inclusion as part of project construction cost, has not been followed. Provisions for AFDC at appropriate rates of interest are made in the economic and fi nanc i a 1 analyses described in Section 18. 16.6 -Escalation As noted, all costs presented in thi~ Section are at January~ 1982 levels, and consequently include no allowance for future cost escalation. Thus, these costs waul~ not be truly representative of construction and procur~ment bid prices. This is because provision must be made in such bids for continuing escalation of costs, and the extent and variation of escalation \'Jhich might take place over the l~ngthy construction periods involved. Economic and financial evaluations discussed in Section 18 take full account of such escalation at appropriately assumed rates. 16.7 -Cash Flow arid Manpower Loading Requirements The cash flow requirements for construction of Watana and Devil Canyon are an essential input to economic and financial planning studies discussed in Section 18. The basis for th·e cash flow are the construction cost estimates in January, 1982 dollars and the construction schedules presented in SectiDn 17, with no provision being made as such for escalat1on. The results are presented in Figures 16.2 and 16.3. Similarly, the corresponding manpow·er loading require- ments are shown in Figures 16.4 and 1.6 .5. These curves were used as the basis for camp loading and associated socioeconomic impact studies. 16-8 I :1/ •• :I I I :I I I' I I I I I ·I I I I ,I I I I I I I I I 'I I I I I I I I :I I· LIST OF REFERENCES (1) Code of Federal Regulations, Title 18,. Conservation of Power and Water·· Resources, Parts 1 and 2, Washington, D. c .. , Government Pri-nting · · --··· · ~-~--'~-·-{Jf-f-1-ee-; ·-l-9£'1;·-·c-"·· -· ------·· · --· --· ·· (2) (3) (4) Handbook_ of Wages and Benefits for Construction Unions, January 1981_, U. s. Department of Labor, Office of Construction Industry Services, 1981. Caterpillar Performance Handbook, Caterpillar Tractor Co., Peoria, Illinois, October 1981. Roberts, WilliamS.~~ Regionalized Feasibility Study of Cold Weather Earthwork, Cold Regions Research and Engineering Laboratory, July 1976, Special ~eport 76-2. iABLE 16.1: SUMMARY OF COST ESTIMATE Januer1 1982 Dollat·s $ X 106 Cate9ory Watana Devil Can~on Total Production Plant $1,969 $ 766 $2,745 Transmission Plant 388 91 479 General Plant 5 .5 10 Indirect 449 222 671 Subtotal $2,811 $ 1,094 $3,905 Contingency 17 .. 5% 492 191 683 Total Construction $3,303 $ 1,285 $4,588 Overhead Construction 413 161 574 TOTAL PROJECT $3,716 $1,446 . $5,162 I I I I I I I I I l"c.j I I I I .~ I I .,, -· - - - - - - - - - - -... - - - --- TABLE 16.2 ESTIMATE SUMMAR'( WATANA P57UOl .. Ul0 110 ALASKA POWER AUTHORHY TYPE OF ESTIMATE Preliminary CLIENT PROJECT ____ s_US_.I_T_NA __ I1_YD_R_O_E~LE_C_T_RI_C __ PR_O_J_EC_T_. ____________ __ APPROVED BY ---'------------ JOB NUMBER ----:---------FILE NUMBER _P_5.,...70_m_ .• _m_'b __ _ SHEET 1 QW. __ 5 __ BY -------O#TJ":£ __ _ CHKD ~£ No. DESCRIPTION AMOUNT TOTALS REMARKS PRODUCTION PLANf 330 L:and & Land Rights ...................................................... ,. $ 51 ~ 331 Powerplant Structures & Impro\lements ~··••••••••••••••••••••••••••••••• 74 332 Reser11oir, Dams & Waterways .............................................. . 1,519 333 W~terwheels, Turbines & Generators ..................................... . 65 334 Accesso£'y Electrical Equipment •••••••• ? •••.•••••.••••••••..•••••.••••.• 21 335 Miscellaneous Powerplant Equipment (Mechanical) ••••••••••••••••••••••• 14 336 Roads & Rail.roads ............... -••••••...••••...•••.•.••• -· • f:l •• • , ••••••••• 225 TOTAL PRODUCTION PLANT ............ , ................................... . $ 1,969 A~IR No. 3~0 352 353 354 356 }59 I TABLE 16.2 \~AT ANA ESTIMATE SUMMARY , CLIENT ALASKA POWER AUTHORirY TYPE OF ESTIMATE Preliminary PROJECT ____ s_U_S_IT_N_A_H~Y_OO_O_E_L_E_CT_R_l_C_P_R_O_JE_C_T ____________ __ APPROVED BY ______________ __ DESCRIPTION TOI AL BROUG~IT FORWARD ••••••• ~ •••••••••.•••.••••.•••• ,., :o ••••• • •• ·•• 61 ••••• ~ •• TRANSMISSION PLANT land & l .. a11d Rights ......................................... ~ . ,. •. ,. .•..•.. Substation & Switching Station Structures & lnprovellients ••••••8••••••• Substation & Switching Station Equipment ••••••••••••••••••e••••••••••• Steel Towers. & fixtures •.•....••.••..••.•.... ~ ..••••...••..•• " ....••..• Overhead Conductors & De"' ices ........................................... " • Roads & Trails ., .................................................. • ••.•••• fOTAL TRANSMISSION PLANT •••••••••••••••••••••••••••.••••••••••••••••••• AMOUNT $ 8 12 129 130 99 10 TOTALS $ 1,969 $ 388 $ 2,357 JOB NUMBER P57Qttl..OO FILE NUMBER P~FlU10 .. U6 -~-- SHEET 2 ~ 5 ----- BY ____ _ IDlTE --- CHKO IDATE - - --· .. -·-- -·"-- - - - - - - - -- .. 1~111 No. 389 390 391 392 393 394 395 396 397 398 399 ----------- ESTIMATE SUMMARY WATANA CLIENT ALASKA POWER AUTHORITY TYPE OF ESTIMATE __ P_re~l_i_m_i_n_ar_y_ PROJECT ____ s_us_I_T_N_A_H_Y_DR_O_E_LE_C_T_R_IC __ P_RO_J_E_c.T _______ · ______ __ APPROVED BY----------- DESCRIPTION TOTAL BROUGHT FORWARD GENERAL PLANT ·•···········•···················•···•·•····•···· land & Land Rights ···················~···················••···~~t~·~~· Structures & lrnpro'lements •...••••.••••••.•.• ·· ••• -. •..•.•..••••••.•.•••••. Office Furniture/£quip..-ner1t ••••••••••••••••••..••••••••••••• ~ •••••••• ~ o. Transportation Equipment •••..••••.•• * •••• ·o ••••••••••.•••.•• a ............. . Stores Equipment •••••••••••••••••••••••••••••••••••••••••••••••••••••• Tools Shop & Garage Equipment .......................................... . Labot·ator~ Equipment ................................................... . Power-Operated Equipment Communications Equipment Miscellaneous Equi~~nt Olher Tangible Propert~ •.......•...............•......••.•..•........ ••.•••••....••••..........•................... ·················~····························· ···············~······························· TOTAL GENERAL PLANT ••••••••••••••••••••••••••••••••••••••••••••••••••• AMOUNT TOTALS $ 2, 35"1 5 $ 5 $ 2,362 ---- JOB NUMBER P57om.:no FILE NUMBER P570Jll/ll6 SHEET 3 QfF 5 BY O..t.LW:£ CHKD OfA;T£ REMAR:~ Included under 3Jo· Included under .331' TABLE 16 .. 2 WATANA JOB-NUMBER P57rnin.IDO FILE NUMBER P57f!liD..:06 ESTIMATE SUMMARY CLIENT ALASKA POWER AUlHORUY TYPE OF ESTIMATE Preliminary SHEET 4 QiF 5 BY ~£ ----PROJECT ____ S~U_S_IT~N_A_~~lV_D_RO_E_L_E_CT_R_l_C_P_R_O_JE_C_T ____________ __ APPROVED BY ______________ __ CHKD No. DESCRIPTION AMOUNT TOTALS REMAR!itS --------+--------------------------------------------------~------------r-----------~----------------------- 61 . 62 63 64 65 66 69 T 0 I Al BROUGHT FORWA.RD ............................... c •••••••••••.•••••• ~ • INDIRECT COSTS Temporary Construction facilities ....... , •••••••••••••••••••••••• , •••••• Construction EquiJ>illent .••.•.• -.. ~ ...... -. •••••.••.• e ••••••••••••• •. • •••••••• Camp & Commi-ssary .................................................. C! ••• ·• ~ • Labor Expense •••••••••••••••••••••••••••••••••••••••••••••••••••••••••• Superintendence ................................................... , •••.••.. Insurance ·····························~································ r ees •• ~ ..................... 0 •••••••••••••••••••••• 0 • ·• ................... . Note: Costs under accounts 61, 62, 64, 65, 66, and 69 are included in the appropriate direct costs listed above. TOTAL INDIRECT COSTS ••••••••••••••••••·•••••••••••••••••••••••••······· $ 449 --:-- - - ------- - - $ 2,362 See Note See Note See Note See Note See Note $ 449 $ 2,811 ------------------- TABLE 16.2 WATANA ESTIMATE SUMMARY JOB NUMBER P5713JD .. \OO FILE NUMBER P57001.!.nb CLIENT ALASKA POWER AUTiiORI TV TYPE OF ESTIMATE Preliminary SHEET 5 QJF 5 A~IU PROJECT _ _____;:S=U=S.::;.;lT:..:...N:.:....:A_;H=Y=OR..;..:..O=E=L=EC='T:..;.;R=IC=......:..P..;.;R=OJ=E=C;,.;_T ______ _ BY O'Am":.t:. ---APPROVED BY ___ "'--------- IT4mt. CHKD No, DESCRIPTION TOTAL BROUGHT FORWARD (Construction Costs) ...•.•.....•.••....•..•.•..• Contingency 17. ·s~ .. ~ ...................................................... . TOTAL CONSTRUCTION COSTS ••••••••••••••••••••-,•••••••••••••••••••&••••• OVERHEAD CONSTRUCTION COSTS (PROJECT lNDlRECTS) 71 Engineering/ Administration ••••••••••••••• •.e................... ... . . . .. . $ 72 -Legal Expenses •• ·• •••••••.••••.••••••••••.•••••••••••••••••.••••.••••••••••• 75 Taxes ...................................................... _ .••••••••••••••• 76. Administrative & General Expenses •••.••••••••• "........................ G 11 Interest ··~···········~······················~························ 80 Earnings/Expenses IAJring Construction .................................. . Total Overhead ••••••••••••••••••••••••4••••••••••••••••••••••••••••••· TOTAL PROJECr COST •••••••••••••••••••••••••••• ~ ••••••••••••••••••••••• AMOUNT 413 - - - - - TOTALS $ $ " $ 2,811 492 3,303 413 3,716 REMAR;~ Included in 71 Not applicable Included in 71 Not included Not included TABLE 16.3 ESTIMATE SUMMARY DEV ll CANYON AIIU AlASKA POWER AUTHOR! fY CUE NT PROJECT ____ s_u_S_lT_N_A_H_Y_D_RO_E_l_EC_T_R_IC __ P_RO_J_E_C_T ____________ __ APPROVED BY-------------..,...- TYPE OF EST I MATE Preliminary JOB NUMBER P5'7f.lt;t..l!ID FILE NUMBER P57Uet ... !06 SHEET 1 Q,f= 5 BY 0./A.ll'E CHKD fl/Al"E No. DESCRIPTION AMOUNT TOTALS. R.EMAB.~'S PRODUCTION PLANT JJO -land & Land Rights ••••••••••••••oae••••••••••••e•••••••••••••••••••o•• $ 22 331 Powetplant Structures & liJllrovements •••••••e•••••••••••••••••••••••••• 72 J32 Reser~oir, Dams & Waterways ···~············••••••••••••••••••e••·••••• 571 JJJ \~alerwheels, Turbines & Generators •••••••••••••••••••• o ••••••••• • ••••• 42 334 Accessory Electrical Equipment •o•••••••••·•••••••••••••••••••••••••••• 14 335 Miscellaneous Powerplant Equipment (Mechanical) •••••••••6••••••••••••• 12 JJ6 noads & Railroads ·······•o••••·········~······························ 43 lOTAL PRODUCTION PlANT •• :II" •••••••••••••••••• " o •••••• ~ ................. , •• o ••• $ 776 --------~----------------~--------~------------------------------L-------------L-------------~~--------------------·--- ·-----------------· -------''-~----... : -·· -----'_ . -,. . : ':-. --- --~---------------·- 350 352 35} 354 356 359 lABLE 16., ESTIMATE SUMMARY OEV ll CANYON CLIENT ALASKA POWER AUTHORITY TYPE Of ESTIMATE PROJECT __ s_u_sl_T_N_A_H_Y_D_RO_E_L_E_C_T_R_IC_P_R_OJ_E_C_T --------APPROVED BY __________ ~-- DESCRIPTION TOTAL BROUGHT FORWARD ••••••••••••••••••••••••••••••••••••••••••••••••• TRANSMISSION PLANT land & land "Rights· ............ ._ ........................................ e •••••• Sllbstation & Switching Station Structures &. lnprovements ................. . . SUbstation & Switching Stat ion Equipment ............................... . -;' ~ - Steel Towers & Fixtures ...• ._ ..•. il .............................. ~ ••.•• : ••••••.• Overhead Conductors & Devices ......................................... . Roads & Trails .••.••••..••••.••.•.• o •••.••••••.••••••• -....................... o TOTAL TRANSM!SSION PLANT ···~···········~····•••••••••••••••••••••••••• AMOUNT $ 7 21 29 34 TOTALS $ 776 $ 91 $ 867 JOB NUMBER P5700"..1liD CHKO REMARKS Included in Watana. ~timat.e Included in Watana, ftlS'ti·mate TABLE: 16.3 JOB NUMBER - ESTIMATE SUMMARY DEVIl~ CANYON P5;zum.oo ~~----------~· Fl LE NUMBER _P...,..5:_:mJ_n_._0_6 __ _,:_ CLIENT ALASKA POWER AUTHORITY TYPE OF ESTI MAlE Preliminar~ sHEET 3 J.O'F ----5 A~ Ill PROJECT SUSHNA HYDROELECTRiC PROJECT 8'1 ---------ID~TE __ _ !DATE APPROVED BY--------CHKD No. 389 390 ' -; DESCRIPTION TOTAL ·BROUGHT fORWARD. • ••••••••••••••••••••••• ~ ••.•••••••••••••••••••••• ·GENERAL PLANT land &: land Rigl1ts lr ., ....... ·t!• ••••••••••••••••••••••••••••••••••.•• o ••••••• St.ructures & lnpro"ements ••.•• o ..... _ •••••••••••••••••••••••••••••••••.•• ·• 391 Office Furniture/Equipment· ........ ~ ....................................... . 392 Transpnrtation Equipment ••••••••••••••• " ............................ • ••• 393 Stores· Equipment ••••• o •••••• ~ •••••••••.•• ~. ·• •. • ............. ._ •••• ·• ••• • ••• 394 Tools Shop & Garage Equipment ••••••••••••••••e•••••••••••••••••••••••• 395 Laboratory Equiprnen·t •••• ·• ................................................. . 396 Po\oter (\>erated Equipment •••••••••••• ;, ................................... . 397 Communi·cat ions Equipment .................................................... . 398 Miscella1-.eous Equipment ................ ~ ................................ . 399 Other Tangible Property ........................................... , ••••• TOTAL GENERAL PLANT ······················~···························· --- AMOUNT TOTALS $ 867 $ 5 $ 5 $ 872 REM~-KS Included under ~~ Included under-J;J51t ---- ---- A~IU 61 62 63 64 65 66 69 No. TABLE 16.3 ESTIMATE SUMMARY/ DEVIL CANYON CLIENT .ALASKA POWER AUTH!lRI rY TYPE OF ESTIMATE __ Pr_e_l_im_i_n_a_,ry'-- PROJECT ____ S_U~SI_T_N_A'--H'--Y_DR_O_E_LE_C_T_R_Ic_·~p_RO~_J_E_CT ____________ ___ APPROVED BY ______________ _ DESCR!PTION TOT.Al BROUGHT FORWARD ..................................................... . INDIRECT COSTS Temporary Construction Facilities ............................. ~ ••••••• •.• Construction 'Equipment ..................................................... . Caflll & Commissary ••• ·• ................................ _ •••••••••• 111: ......... . labor Exp~~ae ••••••••••••••••••••••••.••••••••••••••••••••••••••••••••• Superintendence •••••••••··~·o·~··••••••••••••~w•••••••••••••~·~••••••6 Insu~ance ····················~··~·····•••••••••••••••••••••••••••••••• fees ................................................. o ••• • •••• e •••• c.. ••••• Note: Costs under accounts 61, 62, 64, 65, 66, and 69 are included in the appropriate direct costs listed abo11e. TOTAL INDIRECT COSTS-············•·······•o•••••••••••••••••••••••••••• AMOUNT TOTALS $ 872 $ 222 $ 222 $ 1,094 - JOB NUMBER P$71Jltl •. OO ----~~~------ FILE NUMBER _P_5_7,QID,;,.....;.. • .__o6 _____ ,....._ SHEET 4 {Q:ft= __ 5 _ BY -------llTJATE __ _ CHKD ~ATE See. Note See Note· See Note See Note See Note See Note AIIR 71 72 75 76 . 77 BO No, TABLE 16.3 ESTIMATE SUMM~o\RY DEVIL CANYON CLl~NT ALASKA POWER AUTHO.:.:.r<I::..;l:...'t: _________ _ TYPE OF ESTIMATE ·PROJECT __ __;;;;,S;:;.;US:;.:l:..:.l:.;.:NA~H..:..:VDR=O;.=:.EL::.:E::;:;C:..:.T::.:.;Rl=-=C:;....·.:...:PR~O:.::J;.=:.EC;;;..;T;...__ _____ _ APPROVED BY DESCRIPTION -AMOUNT TOTAL BROUGHT FORWARD (Construction Costs) ···-·······-···••4!1•••·,······· Contingency 17. ·s-% •• _ ••••••••• -: ......................... .t: ••••.•••••••••• -a •••••••• fOTAL CONSTRUCTION COSfS •••••••••••••••·····•••••••••••••••••••••••••• OVERHEAD CONSTRUCTION COSTS (PROJECT lNOlREClS) :Engineering ....... ,.. ..•.•. , ....•.•••......•••.. o .................................. . $ 161 legal Ex·penses ..•.•.•.......•....•..••..••• ·· .••.• , •••••.•.•••.•.•.•••.•••• - Taxes ·····~········•················~·······•·········~····•••••••••• - Administrative & General Expenses ·•·•••••••••••"'•·········~·· .. ••••••• -. Interest •••••••••···········~~··············•·•·•·····~·••••••••••••• - Earnings/Expenses IAJring Construction ·············~·················· - Total Overhead Costs •.•.•••• , ............. • ••••.••..•.•....••....•....•.•.•• fOTAL PROJECT COSl ••••••••••••••················•··•••••••••••••••••• ---·",--·,-·· ~ . 0 Preliminary TOTALS $ $ 1,094 191 1,285 161 1,446 JOB NUMaER P51UOl.!ii0 FiLE NUMBER P570m.J.!$ SHEET 5 O:!F BY -DAIJE CHKO QA"'i£ REMARJ§ Included in 71 Not Applicable Included in 71 NOt lncluded Not Included . 5 " I I I I I I I I I I I I I I I I 'I I t¥JW···~ ., iABlE 16.4: CONSTRUCTION SEASONS DURAtiON ·OORAfldN WORK ITEM. (MONTHS) (WORKING DAYS} Granular Fill Placement 6.0 150 Impervious Fill Placement 5.0 100 Rock Fill Placement 1.0 170 Underground Work '10.0 250 Ctincrete ·?lacement a.o 200 (Generally) Concrete Placement 10.0 250 (Devil Canyon Dam) Other Abo"eground Work · 10.0 .250 SEASoN START/FINISH Apr 15-(lct 15 May 1-0ct 1 Apr 1-0ec 1 Feb 15-Dec 15 Apr 1-0ec 1 Feb 15-Dec 15 Feb 15-Dec 15 ~~~.,..-·';'· -. ' '· TAat..E 16 .. 5: MITI.tAtHlN MEASURES SlJ.1MARY Of COST$ INCORPORAT.ED IN CONSTRtiCTlON COST .ESTIMATES --~---....;;.;.;:.....;;;.;;;.,;.;;;;;.;..;o;;.;;,.;.;;;.;;.;.;.....;;.;;.;;;,;....,;;:;.;;..;..;;.:..~;;.;;..-------........ ~ ........ · ... -- COSTS INCORPORATED IN CONSTRUCTION ESTIMAYES Spillway·Valves in Dam Main Dam at Devil Canyon Service Spill~ay at Watana Restoration of Barra~ Area D Restoration of Bo~row Area F Restoration of Cafll> Restoration of Construction Sites fencing around Caill's Fencing around Garbage Disposal Area Multileiel Intake Structure Camp Facilities Associated with trying to Keep Workers out of Local Communities Restoration of Haul Roads: SU8TOTAt Contingency 17.5% TOTAL CONSTRUCTION Engineering 12.5% TOTAL PROJECT · .$ X 10 3 . WATANA 48;500 1,617 551 1,054 4,050 414 125 18,000 10,156 756. 85,221 14,2.1_4, 100,135 12,517 112,542 $X 103 D£YIL CANYON 14,510 700 2,016 305. 125 9,000 505 27,161 ' 4, 753 31,914 3,989 35,903 148"555 .... •. _'""·:~ 1'1 ~. . ~ .: j ,, i ,~ I 1'. .~.: I I I I I I I I I I I I I - -----·-.. : ····'----· --- ~--~-------------------~~-----·----~--------------------·.~~~· ------------------------------~-------. -,..e 0 -(f) 1aJ -0 z 1aJ 0 -LL (1. 1LI ·Z. - LOCATlON-FAIRBANKS Ai<. LEGEND --·------MANUAL --------EXCAVATION ----HAULING 80 ----------~--------~------~~~~----~----~~~~------~ so.~--------~----------~------~~---------+-----~--~~-------~ 40 ', ' ' ,, ', ' ' ' ', ' ', . "' 20 ~ ', ' '?-_ -- OL' ·-~~--~-,~~-_. ~----~--~~--~ ........ ........ _ .. ____ _ JAN MAR FEB MAY APR JUN MONTHS JUL SEP NOV AUG OCT DEC EARTHWORK INEFFICIENCIES BASE.D ON MONTHLY TEMPERATURE ,LIGHTING AND 'PRECIPlTATlON FfGURE IGJ . ; I I I ' ~· . ' 1 t ·~ DEVELOPMENT. SCHEDtlLE · ... · to. ·F.o.llow ' \' \l '. .; ,_ : . 'I .I I I I I I I I I I I 1 I I I 1 I -~·""\ ,, . ·,1·· ·--,7 ·a ,.' ~ ',1 ···-.• : ·I, I I I ••• I I I I :-., I il·. ' 18' ..; ECONOMIC AND FINANCIAL EVALUATION , lS.l -EGonomic Evaluation This· section provides a discussion of the kfJy economic parameters used in the study and develops the net ~conomic benefits stemming from the Susitna Hydro- electric Project. Section 18.l(a} deals with those economic principles relevant to the analysis of _net economic benefits and develops inflation and discount rates and the Alaskan opportunity values {shadow prices) of oil, natural gas and coal. In particu1 ar the analysis -is focused on the longer term prospects for coal markets and prices. This follows from the evaluation that in the absence_ of Susitna, the next best. thermal generation plan would rely on exploitation of Alaska coal. ·The future coal price is therefore considered in detail to provide rigorous estimates of prices in the most 1 ikely alternative markets-and hence the market price of coal at the mine-head within the state, · Section l8.l(c) presents the net economic benefits of the proposed hydroelectric power investments compared with this thermal alternative. These are measured. in terms of-present valued diff~rences between benefi-ts and costs. Recognizing that even the most careful estimates will be surrounded by a degree of uncer;- tainty, the benefit-cost assessments are carried out in a probabilistic frame- work. The analysis therefore ·provides both a most likely estimate of net eco- nomic benefits _accruing to the state and a range of net economic benefits that ·can be expected with a 1ikelihood (confidence leve1) of 95 percent or more .. {a) .§.conomic Principles and Parameters (:i} _Economic Principles -Concept of Net Economic Benefits The concept of net economic benefits has a profound importance to the State of Alaska. A necessary condition for max.imizing the . increase in state income and economic growth is the select i.on of public or private investments with the highest present valued net benefits to the state. In the context 0f Alaskan electric power investments, the net benefits are defined as the difference between the costs of optimal Susitna-inclusive and Susitna-exclusive (all- thermal) generation plans. The energy costs of po~er generation are initially measured in terms of opportunity values or shadow prices which may differ from · accounting or market prices currently prevailing in the state. The concept and use of opportunity values is fundamental to the optimal allocation of scarce resources. Energy investment decisions should not be made solely on the basis of accounting prices in the state if the international value of traded energy commodities such as coal and gas diverge from local market prices .. The.choic,e of a time horizon is also cruci.al. If a too short-term · planning period is se 1 ected, the investment rankings and cho ice.s will differ markedly from those obtained through a more appropriate long-term perspective. In other words, the benefit-cost analysis , would point to different generation expansion plans depending on the selected planning period. A short-run optimization of state incomes would, at best, allow only a moderate growth in fixed capital 18-1 . formati.on, at worst, it would lead to underinvestment in not only the energy sector but also in other i nfrastruct.ure faci1 ities such as roads, airports, hospitals, schools., and communications. It therefore follows that the Susitna project, as other Alaskan ~investments, should be apprais~d on the basis of long-run optimiza.- tion, where the 1ong-run is defined as the expected economic li.fe of the facility. For hydroelectric projects, this service life is typi ca 11y 50 years or more. · The costs of a Susi tna-1 ncl us i ve gener..: ation plan have therefore been compared with the costs of the next- best a 1 ternat i ve which is the a 11-therma 1 generation plan and assessed over a planning peri,od extending from 1982 to 2040, using internally consistent sets of economic scenarios and appropriate opportunity values of Alaskan energy. Throughout the analysis, all costs and prices are expressed in real {inflation-adjusted) terms using JQnuary 1982 dollars. Hence, the results of the economic calculations are not sensitive to modified assumptions concerning the rates of general price inflation. In contrast, the financial and market analyses, conducted in nominal (inflation-inclusive) terms, will be influenced by the rate of general price inflation from 198.2 to 2040. (ii) Price Inflation and Discount Rates -General Price Inflation Despite the fact that price 1 eve 1 s are generally higher in Alaska. than in the Lo\~er 48, there is 1 ittle difference in the compara- tive rates of price changes; i.e. price inflation. Between 1970 and 1978, for example, the u.s. and Anchorage consumer price indexes rose at annua 1 rates of 6. 9 and 7.1 percent, respectively. From 1977 to 1978, the differential was even smaller: the con- sumer prices increased by 8.8 percent and 8.7 percent in the U.S. and Anchorage (1). Forecasts of Alaskan prices extend only to 1986 {2). These indi- cate an average rate of increase of ~· 7 percent from 1980 to 1986. For the lange~ period between 1986 and 2010, it is assumed that Alaskan prices will escalate at the overall u.s. rate, or at 5 to 7 percent compounded annually.. The average annual rate of pr·i ce inflation is therefore about 7 percent between 1982 and 2010~ As this ""'is consistent with 1 ong-term forecasts of the CPI advanced by leading economic consulting organizations, 7 percent has been adopted as the study value (3~4). -Discount Rates I) I I I I I I, ,I ,I I I I~ I Discount rates are required to compare and aggregate cash f:l ows I occurring in different time periods of the planning horizon. In essence, the discount rate is a weighting factor reflecting that a .1. _ d9llar received tomorrow is h·orth less than a dollar received 18-2 I I ' (l ... .> • . .., ... ' ·~ ~ • ~· .. . ~ • ... '.,,.~ -, • • ..r ~ ... • .~ 4 ,.. .. • , • • .. , ,p ~ i,J; • -• \: llo ;ae~~.~. ¥ , -G u ' • • I • • • "' ... ..-I' o "'' .. :· >v! -,•~ • ,"" •\Jd "\. .. '"• • p , • •'f 'f~"~ • ' ~' ~:1 • :. ~ "f#. fO • '!'!fr~ • ,.,.If; • • ,, • I) . ' • "' .. :'.I ' I I I I I I; I: I I I - I I I I I I I ·. ~• '-· "J ,. today.. This he 1 ds even .in an i nfl at ion-free economy as long · as ·the producti~ity of capital is positive. In other words, the value of a dollar received in the future must be deflated to· ~eflect its earning power foregone by not receiving it today.. The use of discount rates extends to both real dollar {e,conomic) and escalated dollar (financial) evaluations, with corresponding inflation-adjusted (real) and inflation-inclusive (nominal) values .. • Real Discount and Interest Rates Severa'l approaches have been suggested for estimating the real discount rate applicable to public projects (or to private pro- jects from the public perspective). Three common alternatives include: •• the social opportunity cost (SOC) rate, ee the social time preference (STP) rate, and •• the government • s real borrowing rate or the real cost of debt capital {5,6,7). The SOC rate measures the rea 1 soci a1 return (before taxes and subsidies) that capital funds could earn in alternative invest- ments. If, for example, the marginal capital investment in Alaska has an estimated social yield of X percent, the Susitna Hydroelectric Project should be appraised using the X percent measure of 11 foregone returns 11 or opportunity costs.. A short- coming of this concept is the di ffi cul ty inherent in determi n- ing the nature and yields of the foregone investments .. The STP rate measures society's preferences for allocating resource~ between investment and consumption. This approach is also fraught with practical measurement difficulties since a wide range of STP rates may be infer:"ed from market interest rates and socially desirable rates of investment. A sub-set of STP rates used in project evaluations is the 0\-Jner 1 s rea 1 cost of borrowing; that is, the rea 1 cost of debt capital .. This industrial-or government borrowing rate may be readily measured and provides a starting point for determining project-specific discount rates. For example, long-term indus- trial bond rates have averaged about 2 to 3 percent in the U.S. in real (inflation-adjusted) terms (3,8). Forecasts of real interest rates show average values of about 3 percent and 2 ·percent in the periods of 1985 to 1990 and 1990 to 2000, res- pectively. The u.s. Nuclear Regulatory Commission has also analyzed the choice of discount rates for investment appraisal in the electric utility industry and has recommended a 3 per- cent real rate .(24). Therefore, a real rate of 3.0 percent has been adopted as the base case discount and interest rat(~ for the period 1982 to 2040. 18-3 , ,=-"=~,_ I, c (iii) Nominal Discount and Interest Rates - The nominal discount and interest ·rates are derived from the real values ·and the anticipated rate of genera1 price infla- tion. Given a 3-percent real discount rate and a 7-percent rate of price inflation,_ the nominal discount rate is deter-- mined as 10.2 percent or about 10 percent*. Oil and Gas Prices - -Oil Prices In the base period (January 1982), the Alaskan 1982 dollar price of No. 2 fuel oil is estimated at $6.5u/MMBtu. Long-term trends in oil prices will be influenced by events that are economic, political and technologtcal in nature_, and are therefore estimated within a probabilistic framework. As shown in Table 18.1, the base case (most likely escalation. rate) is estimated to be 2 percent (to 2000) and 1 percent from 2000 to 2040. To be consistent with Battelle forecasts, a 2-percent rate was used throughout the OGP p 1 ann ing period 1982 to 2010 and 0 p~rcent thereafter. In the low and high scenarios the growth rates were estimated at 0 percent (1982-2051), and 4 per- cent (to 2000); and 2 percent (beyond 2000), respectively. These projections are also consistent with those recently advanced by such organizations as ORI (9), World Bank (10), U.S. DOE (11)~ Canadian National Energy Board (12). -Gas Prices Alaskan gas prices have been forecast using both export opportun- ity values (netting back CIF prices from tlapan to Cook Inlet) and domestic market prices as 1 ikely to be faced in the future by Alaskan electr"ic uti 1 it ies. The OGP analysis used market prices as estimated by Batte 11 e, s i nee there are ind ic at ions that Cook Inlet reserves may remain insufficient to serve new export markets~ Domestic Market Prices Table 18~2 depicts the low, medium and high aomestic market prices used in the OGP analysis. In the medium (most likely) case, prices escalate at real rates of 2.5 percent (1982 to 2000) J.nd 2 percent (beyond 2000). In the low case, there is zero escalation and in the high case, gas prices grow at 4. percent (to 2000) and 2 percent (beyond 2000). * (1 +the nominal rate) = (1 +the real rate) x (1 +the inflation rate) = 1.03 x 1.07, or 1.102 18-4 I I" .. I I I. I I I I I ,I I I I I I I I I ,I I I I I I I I I I l --· I I I I I I I I a II .. Export Opportunity V~ues Taule l8 .. 2 also shows the current and projected opportunity value of Cook Inlet gas in a scenario where the Japanese export market for LNG continues to be·, the alternative to domestic de- . mand._ From a base period plant gate price of $4:.69 MMBtu (CIF Japan), low_, medium and high price escalation rates have been estimated for the interv a 1 s 1982 to 2000 and 2000 ' J 2040. The cost of liquefaction and shipping (assumed to be constant in real terms) was subtracted from the escalated CIF prices to derive the Cook Inlet plant-gate prices and their growth rates. These Alaskan opportunity values are projected to escalate at 2.7 percent and 1.2 percent in the medium (most likely) case. Note that the export opportunity values consistently exceed the domestic prices. In the year 2000, for example., the opportun- ity value is nearly double the domestic price estimated by Battelle. (iv) Coal Prices The shadow price or opportunity value of Beluga and Healy coal is the delivered price in alternative markets less the cost of trans- portation to those markets. The most 1 ike ly alternative demand for thermal coal is the East Asian market, principally Japan, South Korea, and Taiwan. The development of 60-year forecasts of coal prices in these marke~s is conditional on the procurement policies of the importing nations~ These factors, in turn, are influenced to a large extent by the price movements of crude oil. Historical Trends Examination of historical coal price tre.nds reveals that FOB and CIF prices have escalated at annual real rates of 1.5 percent to 6.3 percent as shown below: . Coal prices {bituminous, export unit value, FOB U.S. ports) grew at real annu:tl rates of 1.5 percent (195U to 1979) and 2 ... 3 percent (1972 ~~ 1979) (11) . . In Alaska, the price of thermal coal sold to th·e GVEA utility advanced at real rates of 2.2 percent (1965 to 1978) and 2.3 percent (1970 to 1978) . . In Japan, the \1verage CIF prices of steam coal experienced real escalation rates of 6.3 percent per year in the period 1977 to 1981 (20,21). This repre·sents an increase in the ;;verage price from approximate'ly $35.22 per metric ton {mt) in 1977 to about $76.63/mt in 1981. 18-5 As shown below, export. prices of coal are highly corre1 ated with ·on prices, and an analysi;s of production costs has not predicted accur,ately the level of coal prices.. Even if the production: cost forecast itself ·;s accurate, it will establish a minimum coal prlce, rather than the market clearing price set by both supp-ly and demand conditions. ··· . In real terms export prices of U.S. coal showed a 94-percent and 92-percent correlation with oil prices (195G to 1979 and 1972 to 1979J .. * . . -Supply· function (production cost) analysis, has estimated Canadian coal at a pr1ce of $23 .. 70 (1980 US$/ton) for S.E. ijritish Columbia (B.C .. ) coking coal, FOB Roberts Bank, B.C., Canada (18,23).. In fact, Kaiser Resources (now B.C. Coal Ltd~) has signed ag~"eements with Japan at an FOB Price of about $47 ~50 (1980 US$/ton (19). This is 100 percent more than the price estimate based on production costs .. .. The same comparison for Canadian B.C. thermal coal indicates that the expected price of $55.00 (1981 Canadian~$) per metric ton (2,200 pounds) or about $37.00 (1980 U~S. $) per ton would be 60 percent above estimates founded on product ion costs (18, 19,23). 0 . In ltinger-term coal export contracts, there has been provision for reviewing the base price (regardless of escalation clauses) if significant developments occur in pricing or markets. That is, prices may respond to market conditions even before the ex- piration of the contract.** . Energy-importing nations in Asia, especially Japan, have a stated pol icy of diversified procurement for their coal sup- plies. They will not buy only from the lowest-cost supplier {as would be the case in a perfectly competitive model of coal trade) but instead will pay a risk premium to ensure security of supply (see Battelle 18,23). -Survey of Forecasts Uata Resources Inc. is projecting an average annual real growth rate of ?.6 percent for U.S. coal prices in the period 1981 to 20UO (9). 7he World Bank has forecast that the real price of st.·:am coal would advance at approximately the same rate as oil prices (3 percent/a) in the period 1980 to 1990 (10). Canadian Resourcecon Ltd .. has recently forecast growth rates of 2 percent to 4 percent (1980 to 2010) for subbituminous and bituminous steam coal (22). *Analysis is based on data from the JAotld Bank. **This clause forms part of the recently concluded agreement between Denison .Mines and Ter:k Corporation and Japanese steel makers. 18-6 'I .1'' . I :_ ,.,.· 1,1 . ~ . l ~ ' ~ ~! I I I I I I I I I I I I I I I I ,I I I I "I .. I < I I I I I I ·I I I I I I I -Opportunity Value of Alaskan Coal Delivered_ Prices, CIF Japan Based on these considerations, the shadow price of coal (CIF price in Japan) was forecast using conditio:na1 probabilities given low, medium and high oil price scenarios. lable 18 .. 3 de- picts the estimated coal price growth rates and their associated probabilities, given the three sets of oil prices. Combining these probabilities with those attached to the oil price c~ses yields the following coal price scenarios, CIF Japan. Scenario rvted i urn (most likely) Low High Probabi 1 ity 49 percent 24 percent 27 percent Real Price Growth ·- 2 percent {1982-2000) 1 percent (2000-2040) 0 percent (1982-2040) 4 percent (1982-2000) 2 percent (2000-2040) The 1982 base period price was initially estimated using the data from the Battelle Beluga Market Study (18). Based an this study., a sample of 1980 spot prices (averaging $1.66/MMB.tru) was escalated to January 1982 to provide a stat4 ting value of $1.95/MMBtu in January 1982 dollars.* As more recent and more complete coal import price statistics became available,· this extrapolation of the 19 sample was. found to. .give a significar~t underestimate of actual CIF prices,., By late 1981, Japan's a·1erage imJ:,Jrt price of steam coal reached $2 .. 96/ MMBtu .. ** An important sensitivity case was the.refore developed reflecting these updated actual CIF priceso The ;Up- dated base period vaiue of $2.96 was reduced by 10 percent to $2.66 to recognize the price discount dictated by quality dif- ferentia 1 s bet\'.!een A 1 ask a co a 1 and other sources of Japaru:s'e coal imports, as estimated by Battelle (18). *The es.calation factor was 1.03 x 1.14, where 3 perce1t is the forecast real growth in prices (mid-1980 to January, 1982) at an annual rate of 2 per·cent, and 14 percent is the ld-month incre·ase if the CP1 is used to convert from mia-1980 dollars to January, 1982 dollars. ** As reported by Coal Week International in October, 1981; the average Clf value of steam coal was $75 .. 50 per metric ton. At an average heat value of 11,500 Btu/lb, this is equivalent to $2.96/MMBtu. 18-7 Opportunity Values in.Alaska .• ·_Base Case-Battelle-Based CIF Prices, -~ No ,Export Potential for rfea1y Coal· · · - Transportation costs of $0.52/MMBtu were subtracted from the init-ially est iroated CJF pr tee of $1.95 to determine the-op- portunity val~~ of Beluga coal at Anchorage. In January 1982 dollars,. this base period net-back price is therefore $1.43.. In subsequent years, the opportunity value is de- rived as the difference between the escalated CIF price and the transport cost (estimated to be constant in real terms)~ The real growth rate in these FOB prices is determined residually from the forecast opportunity values.. In the medium (most likely) case, the Beluga opportunity values as~ calate at annual rates of 2.6 percent and 1.2 percent during the intervals 1982 to 2000 and 2000 to 2040, respB~tively~ For Healy coal, it was estimated that the base pe~iod price of $1.75/MMBtu (at Healy) would also escalate at 2.6 percent (to 2000) and 1.2 percent (2000 to 2040). Adding the escal- ated cost of transportation from Healy to Nenana results in a January 1982 price of $1.75/MMBtu.* In subsequent yParsr the cost of transportation of which 30 percent is repre- sented by fuel cost (which escalates at 2 percent) is added to the Healy price resulting in Nenana prices-tha·l;-grow at real rates of 2.3 percent (1982 to 2000) and 1.1 percent (2000 to 2040). Table 18.3 summarizes the real escalation rates app1 icable to Nenana and Beluga coal in the low, medium and high price scenarios. . - Sensitivity Case ... Update~ CIF Prices, Export Potential fm .. -~~~ ~f.Y Coal The updated CIF price of steam coal (2.66/MMBtu after ad- justing for quality differentials) was reduced by shipping costs from Healy and Bel_uga to Japan to yield Alaskan oppor- tunity values. In January 1982, prices are $2.08 and $1.74 at Anchorag<a and Nenana, respectively. The differences be- " tween esc a 1 a ted CIF prices and shipping costs result in FOB prices that have real growth rates of 2.5 percent and 1.2 percent for Be~ uga coal and 2. 7 percent and 1. 2 percent for Healy coal (at Nenana). Table 18.3 shows escalation rates for the opportunity value of Alaskan coal in the low)c mediu!ll and high price scenarios, using updated ,_ase period values. * Transportation costs are based on Battelle (18,23). 18-8 I I ·I I I ~- 1 I I I I I I I I ., I '"'\\ .I \\: I .... I ·~ I I I I I . ~· I I I I I I I I I I I I I (v) Generation PlanningpAnalysis ,.,. ltase Case Study Val~ Based on the considerations pr·esenter.: ":.:ct1ons (i) through (iv) above, a consistent set of fue 1 price assembled for the base ca·se probab i1 is tic generation planning · .1P) analysis, as shown in Table 18.4~ The study values include pr·obabilities for the lowl medium and high fuel price scenarios. The probabilities are common for the three fuels (oil, gas and coal) within each !:iCenario in order to keep. the number of generation planning runs to manageable size. In the case of the natur-al gas prices, domestic market prices were selected for the ·base case analysis with the export opportunity values used in sensitivity runs. The base period value of $3 was derived by deflating the 1996 Battelle prices to 1982 by 2.5 percent per year. Coal prices were. also selected from the base case using tla,7te11e's 1980 sample of prices as the starting point, with the updated CIF prices of coal reserved for sensitivity runs. Oil pr-·-ites have been escalated by 2 percent (1982 to 2040). (b) Analysis of Net Economic Benefits (i) Modeling Approach Given the econom·ic parameters discussed in the previous section,~ the alternatives for electrical energy generation in the Railbelt w~re analyzed by comparing the production costs of electricity with and without the Susitna project. The primary tool for the benefit cost analysis was a generation planning model (OGP) which simulates pro- duction costs over a planning period extending from 1982 t~ 2051. The method of comparing the '1with 11 and 11 without 11 Susitna scenarios is based on total system costs. The planning model determines the total production costs of alternative plan? on a year-by-year basis. These total costs for the period of modeling include all costs of fuel and operation and maintenance (O&M) for all generating units included as part of the system, as well as the annualized investment costs of any production plants added during the period of study. Facto~~·s which Cuntribute to the u·ltimate consumer cost of power which are not included in this model are: all investment cost for plants in service prior to 1993, costs of the transmission and dis- tribution faciJ iti~s already in service, and administrative CO$tS of utilities. These costs are common to all scenarios and therefore have been omitted ftom the study, as they have no differential impact on alternative generation plans. In order to aggregate and compare costs, all annual costs frotn the 1993 to 2051 production simulations have been converted to a 1982 present worth (PW). These PWs are computed as the sum of two com- ponents. The first is the 1982 PW of the first 18 years of model study from 1993 to 2010. The second component is the est imatea PW . of 1 ong-term systen costs from 2011 to 205 L 18-9 To model the system for an additional 40 years would r·equire the developnent of further load fo-recasts and generatiNl alt-ernatives that are beyond the rea 1m of any prudent pr:oJect ions. for-this reason, the final study year (2010) production costs were assumed to simply recur for an add it tonal 41 years, and aoded to the 18-year PW, to establish the long-term cost differences between a1t.ernat ive methods bf power generation. (ii) Base Case Analysis -Pattern of Investments With and Without Susitna The base case comparison of the with and without Susitna plans is based on production cost simulation for the period 19Y3 to 2051~ using mid-range va·lues for the load forecast, base period fuel prices, fuel price escalation, base period capital costs and capi- tal cost escalation. Load forecasts, fuel prices and construction- costs are analyzed in Chapters 5, 18.l(b) and 16, respectively. As discussed in Section 18.1(b), a real interest and discount rate of 3 percent is used. The with-Sus itn a p 1 an ca 11 s for" 680 MW of' capacity at Watana to be avail able to the system 1n 1993. Although the project can come on-line in stages durir.g that year, for modeling purposes full- load generating cJpability is assumed to be avai1abie for the whole year. The second stage of Susitna, the Devil Canyon pro- ject~ is scheduled to come on-line in 2002. The running of ~he project was tested for earlier and later dates and selection of the year 2002 found to result in the iowest long-term cost. Devil Canyon wi 11 have 600 MW of insta 11 ed capacity. The without-Susitna plan is discussed in Section 6.7. It includes 3,200·-M~J coal-fired plants added in Beluga in 1993, 1994, and 2007. A 200-MW unit at Nenana -{s added in 1996. In addition, 970-MW gas-fired gas turbines (GTs) are added during tbe 1997 to 2009 period. Base Case Net Economic Benefits The economic comparison of these plans is shown in Table 18~5. During the 1993 to 2010 study period, the 1982 !'W cost for the Sus~tna plan is l3,119 million. The annual production cost is $385.3 million in 2010. The present worth of this level cost -;:or a period extending to the end of the 1 ife of the Uev il Canyon pr·o- ject (2u51) is $3,943 million. The resulting total cost of tte Susitna ... inclusive. plan is $7.06 bi1lion in real (1982) dollar~~~ pre.~ent ly v a 1 ued to 1982. The non-Susitna plan modelled has a 19~2-PW cost of $3,213 mill ion for the 1993 to 2010 period. With a 2010 annual cost of $491 million, the total long-term cost has a PW of $8.24 billion. Therefore.~ the net economic benefit of adopting the Susitna plan is ·l:.la bill ion. 18-10 I ·1·.· . . ,, /. I I. I I I I ,, I 1: .I·: . ' ·, I I' I I I•. I, ~.- t I I I I I I I I I ••• I I I I o~ •• t :I '.,·.·.···· . . , . ··~ " !n other words, the pres~nt valued cost difference .between the Susitna plan and the next ... best thermal expansion plan is $1 .. 2 billion in 1982 dollars~ The 1982 present valued cost advantage of the Susitna plan {$1 .. 2 bill ion) is equivc1ent to a 1982 per captta net economic benefit of $2;700-.in A1aska. Expressed in 1993 dollars (the on-line d;1te of Watana), this cost saving would have a levelized value of $2.5 billion.* It is noted that the · magnitude of net economic benefits ($1.2 billion) is not sensitive to alternative assumptions concerning the overall rate of price inflation as measured by the CPI. The analysis nas been carried out in real (inflation-adjusted) terms and therefor:e the present valued cost savings-will remain at $1.2 billion regardless of CPI movements, as long as the real (infl ation-adjustea) discount and interest rates are maintained at 3 percent. The Susitna projectts internal rate of return (fKKJ has also been determined. This is the real (inflation-adjusted) discount rate at which the with-Susitna plan has zero net economic benefits" that is the discount rate at which the costs of the with Susitna and the 11 alte.rnate 11 plans have equal costs. This IRR is about 4.1 percent in real terms, and 11.4 percent in nominal (inflation- inclusive) terms. Therefore, the S~sitna investments would sig- nificantly exceed the 5 percent nominal rate of return test pro- posed by the State of Alaska in cases where state appropriations may be involved.** It is emphasized that these· net economic benefits and the rate of return stemming from the Susitna project are inherently conserva- tive estimates due to several assumptions used in the OGP analy- sis . . Zero Gro'?Jth in Far-Term Costs From 2010 to 2051, the OGP analysis assumed constant annual pro- duction costs in the Susitna and the non-Susitna plans. This has the effect of excluding real escalation in fuel prices and the (replacement) capital costs of thermal plants" and ther~by underestimating the long-term costs of thermal generation plants . . Loss of Load Probabilities The loss of load probability in the non-Susitna plan is cal- culated at 0.09Y. This means that the system in 2010 is on the verge of adding an additional plant~ and \vould do so in 2011. These costs are however, not included in the anlaysis which is cut off at 2011. On the other hand, the Susitna plan has a loss of load probability of 0.025, .. anci may not require additional capacity for several years beyond.2b10. *$1.2 bill ion x 2.105, where 2.105 is the general price inflation index for the period 1982 to 1993. **See State of Alaska's SB 25, Section 44.83.470. 18-11 . I I ~....,..---,--.,-,--.. -_-;;:--<--·. ~--· .~. ~-,--,--......,,...,. ··~---~-:,....,:-.. ~."'""":'.~ .• =';"-~,.,.-,·~·''"'7· ,., ..,.,.,. _. . ..,..,. ·'"""··=·-.-~"""'·"~. ~. __,....~,..........~-~~!!!!!!1 :-•. Far...-Term En erg~ from Susitna Some of the ~susitna energy output (about 344 GWh} is still not used by 2010. This energy output would oe avail.abla to meet future increases in projected demand in the summer months... No benefit is attribut.ed to this energy in the: analysis. Egua1.Environmental Costs The OGP analysis has implicitly assumed. equal environmental costs for both the Susitna and the non-Susitna plans. To the extent that the thermal generation expansioo plan is expected to carry greater environmental costs than the Susitna plan, the economic cost savings from the Sus itna project are understated. It is conc'eivable that these so-called negative exter~nalaties from coal-fired electricity generation will have been mitigated by 1993 and beyond, from the enactment of new environmental legislation." However, such government action action would / simply internalize the externality by forcing up the productiorr market costs of therma 1 power. . (iii) Sensitivity Analysi~ A sensitivity analysis has been carried out to identify the impact of modified assumptions on net benefits. The analysis ~<~as directed at the following variables: -load forecast. -real interest and discount rate -construction period period of analysis capital costs ~ Sus itna . Alternatives ~ O&i~ costs ~·base period coal price -real escalation in capital and O&M costs and fuel prices -system r e 1 i ab il it y -Chackachamna included in non-Susitna plan. Tables 18.6 to 18.13 depict the results of the sensitivity analysis. In particular, Table 18.13 summarizes the net economic benefits of the Susitna project associated with each sensitivity test. The net benefits have been compared using indexes relative to the base case value ($1.2 billion} which is set to 100. The greatest variability in results occurs in sensitivity tests per- taining to fuel escalation rates~ dis~ount rates, and base period coal prices. For example, a scenario with high fuel price escala- tion results in net benefits that have a value of 253 relative to the base case. In other words, the high case provides 253 percent of the base case net benefits. In general, the Susitna plan m~in­ tains its positive net benefits over a reasonably wide range of values assigned to the key-variables. 18-12 ··I· ' . : ,, II I I I I I I I I I I I I I I I 1·· .. I I. I I I I I I I I ~I· I .I I -.~t I •• . I ~.· ...•. ~.·~.~ .. .-~,'~ •, Li£ JA multi~variate analysis 1n the form of probability trees is also currently being undertaken lo test the joint effects Of varying sev- eral assumptions in combinatiqn rather than individually. This · probabilistic analysi-s will provide a range of expected net economic bene.fits and probability distributions that identify the chances of exceeding particular values of net benefits at given levels of con-. fidence. ·· · · · 18.2 -Risk Analysis A risk analysts was undertaken to provide a basis for determining the extent to which perceived risks are likely to influence capital costs and schzdule. In addition, because a mature Susitna project would represent a major portion of the tott.l-generation system, a further risk analysis was made to assess the pro- bability and consequences of a long-term outage of the proposed transmission system. This section summarizes both risk analyses. A more detailed report is included in the rroject documentation for Subtask 11.03, Risk Analysis. (a) Approach Any major construction effort is inevitably exposed to a large number of risks. Low probabi'lity magnitude floods may occur at critical periods of construction: accidents may happen: sub-surface investigations, no matter how thorough, cannot always predict actual conditions uncovered when the major excavations are undertaken: the normal estimating process impl ic ity assumes a set of reasonably "normal 11 expectations as direct costs are deve1oped, add·ing a contingency to the directly computed total on the grounds that problems usually do occur even though their specific nature may not be accurately foreseen at the outset. The Susitna risk analysis took explicit account of 21 different risks~ apP.lying them, as appropriate, to each major construction activity. The effort involved combining reasonably precise data (e.g. the probability that a particular flood crest will occur in any given year can be deter- .mined from analysis of hydrologic records) with numerous subjective judge- ments (e.g. until a particular flood crest does occur, it cannot be known with any degree of certainty what damage it will cause. The overall meth- odology is illustrated in Figure 18.4-1. (b) Elements of the Analysis Figure 18.4-2 graphically depicts important questions which were addressed at the start and relates them to elements of the analysis~ Each element is further subdivided as follows: ( i) Configurations Three primary configurations were considered: -The Watana hydroelectric project (with trans;mission); -The Devil Canydlf-hydroelectric project (with transmission); and -The Sus itna transmission system alone . 18-13 (ii) Separ,ate risk studies of these configurations permitted the produc"!! t ion of data which can be aggregate;d in various ways to accommodate alternative npo~ler"!"on-:line 11 dates which differ according to the various demand ~forecasts·~ ConsisuratJon States Two configuration states were considered: -Construction Period --applicable to Watana and Devil Canyon -Operation Period -applied only to the transmission system config- uration. (iii) Risks Twenty-one risks were identified for consideration ip the analysis and were grouped as fo 11 ows: -Natural Risks • flood . ice wind seismic . permafrost detflt' i oration . geologic conditions . low str"eamflow -Design Controlled Risks . seepage piping erosion • ground water -Construction Risks· . equipment availability . labor availability strikes mater i ai avail ab i1 i ty . equipment breakdown matedrial deliveries . weather -Human Risks . contractor capability construction quality control accidents . sabotage .vandalism -Sp·ecial Risks . regulatory de·l ay . estimating variance 18-14 ·I'; "' . .•. :···· .. ~ :. ~ ! .I. : ,' .. ·<' I I I I I I I I I I I I I. I I I· I i'l I I I I ;''I I I I I I I I I I .:1 I I (iv1 Activities For each configuration state involving constructi-on, up to 22 activ- ities ·were considered, For Watana, for example, these included: -main access -site fac i 1 it ies diversion tunnels cofferdams main dam excavation -main dam fill initial portion -main dam fill final portion -relict channel protect~on -chute spillway -emergency spillway -service spillway tunnels -intake . -penstock -powerhouse transformer gallery -tailrace and surge chambers -turbine-generators -mechanical electrical equipment ... switchyard -transmission -impoundment -test and commission. (v) Damage Scenarios Up to ten .different damage scenarios were associated with each logi- cal risk-activity combination. While these varied significantly from one risk-activity combination to another, they generally des- cribed a range of possibilities which accounted for discrete i.ncre- ments extending from "no damage 11 to •!catastrophic loss". (vi) Criteria The consequences of realizing particular risk magnitudes for each activity were measured in terms of the following criteria: -cost implications -schedule implications -manpower requirements (vii) Boundary Conditions The following assumptions and limitations were established t6 permit . a • a reasanable and consistent analys1s of the problem: 18-15 .. -All cost -estimates were made in terms of January 19B2 dollars. Thus, results are presented in this report in terms only of real potential cost variations~ exclusive of inflati~n. -The analysis was 1 imited only to the construction periods for Watana and Devil Canyon since the greatest cpotent i al cost and schedule variance would be possible during these periods. The risk analysis for the operating ·period was associated solely with the transmission system since that cor figuration represents the most 1 ikely source of a major system outage during the project operation. -The risk analysis was accomplished concurrently with finalization t ion of the total project cost estimate and was necessarily asso- ciated with the feasibility level design. There is clearly some potential foi design change as the project proceeds; a further risk analysis should be undertaken coincident with completion of final detailed design and prior to committment-to major ~onstruc­ tion actiyities. Even so, the "estimating variance" risk takes into account the fact that some design changes are likely to appear as detailed design effort proceeds. - A great deal of subjective judgement was necessarily involved in assessing certain probab il it ies and in predicting pass ib 1 e damage scenarios. This effort was accomplished initially by individual qualified professionals in the variou~ disciplines and was sub- jected tQ iterative group review and feedback efforts. To the ex- tent that individual biases entered the analysis, their effects were probably mutually offsetting. Even so,-sensitivity tests were made for risk_s \-Jhich were important contributors to the final results. · -The risk list do~s not include the important possibility of fund- ing delays or of financing problems. These issues were dealt with in a separate financial risk analysis as discussed in paragraph 18.5 below. (c) Risk Assessments For each of the risks identified in paragraph 18.2(b) {iii) above, the assessment commenced with detailed definition of credible. events. Where flood was identified as a risk, for example, the potential magnitudes and associated probabilities of the floods was estimated. Data sources ranged from reasonably accurate scientific data {particularly applicable to the natural risk category), historical experience on water resources projectss to subjective group judgements where data gaps existed. I •••• ••• . . I I ·I I I I I I I I I •• I In each case, the maximum cr--edible .. event was first established. This •... choice set an upper limit on a scale of possible events starting at "no damage" situation. Continuing with flood as an example, the m,aximum credi- ble event was considered to be the probable maximum flood which had been -,computed in ·the hydrologic studies (corresponding to a return period of I ' I ' '~ 18-16 I. I I I I I I. I I I I I I more than 10,000 years and an annual probabll ity of-occurrence of less tha.l'l---,"'_:c· ~..,,::­ . 0001)'. - Once risks were defined and logical risk~activity combinations were re- viewed, the consequences of r_eal izing each selected risk magnitude were considered (lf this risk magpitudejs realized, will a partially completed structure· be damaged? Will it fatl? If it fails, is some other work in progress disrupted?). Because of the uncertainties associated with these projections, a range of damage scenarios and asso<;iated probabilities of them occuring was established. Even if a particular risk level is realized and a particular damage is suf .... fered, the cost and schedule of restoring the activity are difficult to precisely establish. Each of the risk analysts therefore provided three values for each criterion: -·a m~,.,imum value corresponding to the one time in twenty that the weather is particularly good, materials are readily available, no accidents occur, etc. - a modal value associated with the most 1 ikely expectation of the analyst; -a maximum value corresponding to the one time in twenty that everything is more difficult than expected. In the computerized calculation process, the three criterion values sup- plied by the risk analyst were fitted to a triangular distribution, which approximated the beta distribution illustrated at the bottom of Figure 18.2-3~ In effect, then, designation of the three conceptual criterion values led to generation of a histogram with relatively narrow intervals and a nearly continuous range of possible values over a relatively wide spectrum. Figure 18.2-3 illustrates the structural relationship for handling risk- activity combinations, damage scenar·~os, and criterion values. While the procedure described above is generally applicable, some commen- tary on particular aspects of its application and on certain unique risks is appropriate~ (i) Tht1 terminology "damage scenario" has been used fo.r convenience s;i; 'Ce most identified risks will normally be thought of as reasons that the cost will be higher than had been estimated or that the schedule will be exceeded. I~ fact, however, the proces~ does per- mit consideration of what might be regarded as a "negative 11 damage scenario. The geologic conditions risk is an excellent example. The cost estimate was produced on the basis of estimates of require- memnts for some concrete lining in the penstocks, extensive grout- ing, a certain level of rock bolting, and the like. If geologic· conditions are found to be better than currently assumed, the costs could be less and the schedule might be accelerated. 18-17 .(ii) { i ; i) The estimating __ variance risk was ·treated in a.sPHcial ,:11ay .because it cannot easily be conceptualized i.n physical ter!ll~L --rc-·acc6-unts for inevitable differences which do occur between ·eshhuates and actua1 · bids, and between bids and actual activity costs -even in the ab- sence of an_y other identified risks .. Its probability of o~currence and -associated range (fractions or multiples of the basic estimate) were determined from historical data on water resources projects. It includes, but is not necessarily limited to; such considerations as: ~ the influence of competition and market pressures -estimating discrepancies or errors in unit quantities on the parts of both owner's estimator and bidder -particular contract forms and the owner's ~cceptance non- acceptance of certain risks; -labor market conditions and the nature of project labor agree- ments -productivity and efficiency changes over time; -the cost implications of variances between activity schedules and actual activity durations; -the potential for scope changes over time; -extraordinary escalation of. construction costs above the underly- ing inflation rate. In addition to estimating variance, a second special risk is asso- ciated with regulatory matters. Various legislated controls will most certainly be applied to the Susitna project and it is a a rel a- tively simple matter to compute the minimum time in which regulatory requirements could be satisfied. It is a far more difficult task indeed to estimate the precise nature and duration of poss·Ib1e future regulatory delays .. It would also clearly be inappropriate to attempt to app.ly regulatory risks at the activity level. This risk was handled by developing a separate distribution for a range of periods necessary for .satisfaction of important licensing and permitting requirements~ Data used in arriving at a distribution were based on recent experi- ences on other water resources projects as well as on distussions with staff members of the Federal Energy Regulatory Commission.. The effect of applying the regulatory risk is prl~arily one of shifting the starting time for commencement of construction activitiess lead- ing to corresponding change in the projected completion time... A lesser effect of. the r~gulatory risk was to introduce delays dur\ng construction. · Regulatory requirements have been an important influence during the past decade on major construction costs and schedules, though it is difficult to isolate their effects. In order to separately consider estimating variance risks and regulatory risks, "estimating vari- ance" probability determination relied heavily upon water resources 18-18 I I • •• I I I I I I I I I- I I I I I -,, I ••. ' . . I I I I I I I I I -• .. , ) :I I --· I ,n -----~~-.,~-. .,.,, ·~·"tonsttuct ic.\{1 -data ~dev~1oped for pr-ojects essentially completed prior · to the pa'ssage of the National Environmental Pol icy Act_ (NEPA) .-As noted .above~-regulatory risk probahn ity distributions \vere derived from-more recent projects .. . ( iv) Each of the various risk magnitude probabilities was originally calculated as an annual value. On a risk-activity.by risk-activity basis; these _annual values wer·e then converted by standard computa- tional procedures to provide a probability of occurrence during the duration of the acitivty. (v) The concept of 11 response is particularly important in the formal risk analysis process. As the terminology suggests, a 11 response 11 represents the action-to be taken if a particular event occurs .. There are two kinds of "response11 • The first -and most often used -is an expected react ion to the occurrence of a particular damage level (i.e. if tt,l;is damage le.vel is incurred, then what action must be taken to restore the activity to its pre-damage status? ·And what cost, schedule, and manpower implications (consequences) will result?). A second kind of response can a1 so be considered ~nd ·Jt provides an important link between the design tean and the risk analysis team. This latter type is the "preventive response" (i,e. what changes might reasonably be made in the design and or construc- tion procedures \vh i chwoul d permit us to avoid or reduce a particular damage level? Is the cost and schedule change which might ensue worthwhile when com::Ja~ed to the probability and magnitude of the consequences which would otherwise be incurred?} A number .of pre- ventive responses ·were identified.by risk analysts during the risk study and several of these weri incorporated into the project design and dedsi~1n criteria. There may be further opportunities for pre- ventive re'sponse. Since none would be chosen unless it offered a net benefit to cost and or schedule, it may reasonably be concluded that as dHtailed design rJrocaeds and as subsequent risk analysis updates :areaccomplished, a gradual reduction in the spread of possible values can be expected. (d) Interpretation of Results ( i) Presentation of Data Minor variations in activity costs were generated by the estimating team concurrent with development of the risk analysis. In addition, account was taken of the expectation that construction costs will escalate at a fa.ster rate than normal inflation -both in the eco- nomic analyses and the risk analyses. To avoid confusion regarding absolute cost values, the results of the risk analysis are presented in this section as percentages of the estimated project cost or as ratios between actual costs and estimated costs. (ii) Watana Cost-Probability Distribution Figure 18.2-5 illustrates the cumulative distribution of total 18-19 direct costs and their rel.ated non .... ectceedance 1Jrobabi 1 itfes as de- termined~in the risk analysis$ Certain important points noted on the figure ar-e inter.preted as follows: ... The project direct cost estimate, including contingencies, ,was -presented in Chapter 16. Point 11 A11 ott Figure 18.2-5 corresponds to this project esti.mate: the analysis suggests that the probabil- ity of completing Watana for less than the project estimate which includes a 17.5 percent contingency allowance. -Point 11 811 corresponds to the ,.low" cost estimate which was teste.d for sensitivity in the OGPS system cost analysis. The probability that Watana will be completed for less than this cost estimate is about 46 percent. -Point "C 11 on Figure 18.2-5 corresponds to a cost equal to the 11 high 11 estimate tested in the OGPS analysis to determine the effect of such a cost on total proje_ct economics. The risk analysis suggests that there is a ~0 percent probability that this cost will not be exceeded. , -As wi 11 be noted from Figure 18. 2 ... 5 t there remains a sma 11 but measu;--able possibility that the project costs will exceed even the 11 high 11 ~stimate value at Point 11 C11 • It can be argued that the degree of conservatism which was used in the analysis has over- stated the possibility of extreme upper limits on total cost. Paragraph (v) below addresses this issue, comparing these results with historical data. The expected value of the actual cost is 90.25 percent of the project estimate. (iii) Devil Canyon -Probability Distributions Figure 18.2-6 provides the cumulative probability distribution for Devi 1 Canyon costs. Points A, B, and C on the curve corresponj to those discussed above for Watana and are associated with probabil- ities of 74 percent, 47 percent, and 90 percent, respectively, for actual percentages of the project estimate being less than indicated values. Once again, a not insignificant long "tailu in the extreme upper righthand portion of the distribution provides a measure of the potential exposure to large overruns. The expected value of the actual cost is 91.5 percent of the project estimate. (iv) Total Project Distribution Figure 18.2-7 combines the separate Watana and Devil Canyon proj- ects, providing a cumulative distribution for the Susitna Hydro- electric Project as a whole. Points A, B, and C now have associated probabilities of non-exceedance of 73 percent, 47 percent and 18-20 I I .I I. I I I I'' 1. I • ••••• •• . . I I· I, 1: 1 ,~ '~ .:_1 ,, I I ·I I I ••• ·I I ... il "I ··I I I I I ,.··I· . ' I . I tv) 90 percent, respectively, suggesting that a broad range of total project cost ratios. are possible. In the 10 percent range and 90 percent prob·abil tty interva1, the cost ·range spans nearly three billion dollars. If the project follows historical patterns, i.t may be expected· that this ran_ge will narrow over time as detailed design and constructi.on proceed .. Note that the cost distributions are in every case based upon ·January 1982 do dollars and do not account for the effects of inflation. Interest during construction or finance cha-rges are not included. Only the potential for extraordinar·y con- struction cost escalation (over and above inflation) has been taken into account. It follows that if the project 1s completed in the next several decades, the final 11 actua1" cost will have to be ad- justed to equivalent 1982 dollars ·if it is to be compared with ri.sk analysis results presented herein. Comparison with Available Data r During the assessment of the important "estimating variance 11 risk (see paragr-aph 18~2(c} {ii) above), historical data for 49 federal water resources projects completed prior to passage of NEPA were considered. Figure 18.2-8 offers a cumulative probability histori- cal program for various cost ratios. In each case!1 the cost ratio reflects the actual pr~oject cost (after adjustment for-inflation) divided by the 11 initia1 11 estimated cost. It will be seen that rela- tively large overruns have occurred in the past, while there is also evidence that a substantial number of water resources projects have been accomplished for less than the originally estimated costs. In order to compare this. information with the Susitna Risk Analysis results, it is necessary to determine the meaning of 11 initial'1 .esti- mate in terms of the historical data. In each case, the 11 initia1 11 estimate is the estimate presented to the Congress at the time that a request vJas made for projtSct authorization. Thus, it would be in- appropriate to regard the current Susitna estimate (as discussed in Chapter 16) as an "initial" estimate in the federal sense .. Fortun- ately, however, the Susitha project does have a long history of fed- eral involvement. Indeed, the Corps of Engineers p~ovided a de- tailed 11 initial" estimate in 1975 as the basis for seeking authori- zation for important design activities. This uinitial 11 estimate was further updated by a second uinitial" estimate in 1979 after some additional exp1oratory~.work and further analysis were requested by the Office of Management and Budget. Inclusive of contingencies and excluding lands, the direct cost 11 initia1" Corps of Engineerst esti- mate (from the 1979 report) in January 1982 dollars for the Watana IJevi 1 Canyon (thin. arch dam) project was used as the denominator for display of possible Susitna cost ratios. Figure 18.2-9 overlays the results of the Susitna risk analysis on the historical data. Note that the cost ratios differ on this display from those on Figure 18.2-7 because of the necessity to use the 11 initial'' estimate for comparison purposes. ~18-21 As may be s~en fronfF'l'gtire 18.2-9, the Susitrta risk analysis results reflect a more ·pessimistic expectation at low cost leve1s than ~the hi.storica1 data would appear to indicate is appropriate. (vi) Schedule Risks At the same time that minimum, rnoda1, and maximum cast val.ue·s were. estimated for each damage scenario in each risk-activity set, esti- mates were also made o.f similar values for potential schedule changes. As a result, schedule probability distributions were gen- erated for each major activity. HovJever, these individual distri- butions could not be combined in the same manner in which the. cost data were handled. A critica-l path network was prepared for the entire set of activ- ities for each configuration .. Individual probability distributions for critical activities were then combined tr yield a distribution for t~e total project schedule. . . Several critical paths were identified in theprocess, since a long delay on a non-cirtical activity can, of course, place that activity on a new critical path. The 11 raw 11 schedule delay distribution .was then considered in the context of a one-year schedule contingency which had been suklt into the original estimate and in light of regulatory delay risks. The resulting distributions are discussed and interpretBd as follows: -Figure 18 .. 2-12 provides a cumulative probability distribution in months from the scheduled completion data for the Watana project .. It reflects all risk contributions except those posed by regula- .tory requirements. It is based upon a critical path through the main dam construction and takes into account the one-year schedule cont.ingency. The indicated probability of completing the project ahead of schedule or on time is about 65 percent. There is only a 17-percent chance of completing the project a year early (i ,e. in 1992)~ . . -Figure 18 .. 2-13 provides a similar distribution after regulatory risks are accounted for. Two components are included: (1) prior to the start of construction, a 1 icense must be issued by the Federal Energy Regulatory Commission. There is a small chance (estimated to be 25 percent) that the license will be issued a year earlier than the current 30-month licensing schedule antici- pates. The probabi 1 ity of meeting or improving upon the 30-month estimate is about 72 percent and there is a 90-percent probability that not more than 38 months wi11 be required; (2) during the con- struction period, regulatory delays may be imposed as a result of various permitting requirements, injunctions, etc. These delays yield only increases in schedule and range from a 50-percent prob- ability of delays of a month· or less to a 95 percent probabi 1 ity that regulatory delays c!rrring construction will not exceed 12 months. 18-22 _I I: I~ ~~ :-.• ' ·.· .. {l! -~ 1··--.'".' .. : '~ :' .-: i I I' I I I I I I- I I I ,. I I I I I I I I =• I I I I I I I I I I I ,, I :~ ··'· '. r . ~ . t.' -. 1.-' As may be seen from Figure 18 .. 2-13, the net effect of the re,gul ato"ry risks is ,to broaden the range of possible values. At the 1ower end of the distribution, it will be~noted that the chances of completing at least a year early will have increased to near1y 40 percent -· primarily because of· the chance of getting a 1 icense early and therefore, starting early. ·No significant change appears .for the probability of meeting or improving upon the schedule. A substan- tial effect is evident in the upper portion of the curve where the chances of long regulatory delays have pushed out the ~5 percent confidence level to an expect at ion of no more than three months attributable to risks other th&n regulatory, as may be seen on Figure 18.2-12. While similar distributions cna be plotted for Devil Canyon, they are less meaningful since there is flexibility associated with its starting date. (vii) Transmission Line Risks The separate risk analysis of the Susitna transmission system was conducted to determine the probability of significant power supply interruptions at the two major load centers in Anchorage. and Fairbanks. The methodology was generally similar to that described in preceding paragraphs. Recogn·; zing that the system is assumed to be be in an operating mode, those risks which had applied only for construction in the preceding analysis (e.g. contractor capability) were eliminated from the risk list. Additions to the list were made to account for the potential effects<>of lighting, aircraft, collis- ions, and anchor-dragging in Knik Arm {.applicable to the submarine cable segment). Account was taken of re.dundancies designed into the system (e.g~ a 1oss of one line in the three-line system extending south toward Anchorage can be tolerated with no loss of energy delivery capability), In addition, special attention was given to dependencies (e.g~ an earthquake which causes the loss of two lines will very 1ike1y knock out the third. On the other hand, vandalism which.causes an outage on one line is only infrequently expectt.ri to extend to all lines). Important assumptions included the availaiJility of well-trained re- pair c·rews and equipment, and a reasonable supply of spare compon- ents. ·The results of the analysis provide the cumulative probability of not exceeding a given number of days of reducE:d energy delivery capability. Figures 18.2-14 and 18.2-15 display this information for Anchorage and Fairbanks, respectively. Interpretations are as follows: -In the particular case of Anchorage igure 18.2-14), it will first be noted that the probability scale includes only the extreme upper.range of non-exceedance probabilities. The intersection of 18-23 ·:-~:'!·~ (.e) --.> -.,._ ~ , ••• •,.·-"~ ' • ".-• .,-.;; I th·e d-istribution curves on the pr·obabil ity axis indicates that the :··· ·•.· probability of no lost energy del ivety capah·ility in a given year · is 0.958 and of not having 50 percent reduction of 0.955. Beyond these points the curves rise sharply, indicating that outages be----~-~.---·,'-, yond five days are extremely unlikely. The "expected" annual . _. . value of 0 .. 06961 da days for a total delivery loss may be compared· with the "loss. of load probabil ity11 of 0.1 (one day in ten years) . -whi~h had ~een used in the gener:ation pla~ning e~forts in the eco-·I• nom1c stud1es. In short, tne r1sk analys1s conf1rms that the · reliability of the transmission system for energy delivery to Anchorage is consistent with the requirements of_ the overall Rail-1.: belt generation system. The 11 expected" annual value of 0.09171 · .· days for a 50-percent reduction in energy delivery a appears to be similarly acceptable when compared to assumed loss of load 1. probability. 0 The cumulative probability distribution for-Fairbanks (Figure 18.2-15) has a slightly different intercept on the probability axis and its shape is also slightly different from those for Anchorage. These differences stan from the fact that delivery to Fairbanks requires no suomer'ged crossing and certain other risks (e.g. flood, temperature extremes) would be expected to have dif- ferent probab i1 ities for northern and southern segments of the system. In spite of the absolute differences, it may be s~n from the display that 11 expected 11 annual value of .08116 does not exceed the loss of load probability criterion of 0.1 day per year.. No 50-percent loss for Fairbanks is shown since the loss of one of two lines causes no reduction in delivery capability. Two 1 ines lost is, of course, a 100-percent loss. (viii) Emergency Response I I I I I I I I I .:, In spite of the apparent reliability of the transmission ~ystem, it is nonetheless true that a small but finite chance of relativ·ely long-term outages does exist. It is also unfortunately true that certain extreme risk magnitudes (e.g. combination of extreme loss temperature 3 wind, and ice) which could lead to an outage also tend to coincide with high demands by users on the generating system. The 11 response" in this case is extremely important. The final re- port for Subtask 11.03, Risk Analysis, provides such a response in the form of a preliminary emergency plan which includes such meas- ures as shedding non-essential loads, putting reserve capacity on 1 ine, and energy transfers from military generation systems. Prior to the time that the Susitna Hydroelectric Project begins operation~ this plan should be updated and occasional tests should be made~ Conclusions Based upon the risk analysis, it ·is concluded that: ( i) The probabilities that actual costs will not exceed values subjected to sensitivity tests in the economic anlaysis are are as follows: 'li 18-24 I I I > :...,• I I 'I I I I I I I I I I ··I I ·,· I Value Project Estimate Low Capital Cost Tested in the Economic Analysis High Capital Cost Tested in the Economic Analysis -.. Probability That Value Will Not Be Exceeded 73 % 47% 90% . . (i i) Exposure to potential_tosts above the project estimates does exist and there is about a one 1 percent chance that an overrun of 40 percent or more (in 1982 dollars) will occur. (iii) The annual probability that no interrupt ion in energy delivery to major load centers will occur as a result of transmission line failures is in excess of 95 percent. Expected values of energy delivery interruptions are less· than one day in ten years and are consistent with loss of load probabilities. asstiTled in the generation planning efforts. ( iv) There is a 65-percent probability that the ~~atana rpoject will be completed pri.or to the scheduled time "in 1993. Exposure to schedule delays is heavily influenced byregulatoryrequirements and there is a 10-percent probability that the Watana project will not be camp 1 eted unt i 1 1995 or later. · 18.3 -Marketing This section presents an assessment of the market in the Railbe1t Region for the energy and capacity of the Susitna development. A t"ange of rates at which this power caul d be priced is presented together with a proposed basis for contract- ing for the supply of Susitna energy. (a) The Railbelt Power System Susitna capacity and energy will be delivered to the "Railbelt Region Interconnected System" which will result from the 1 ink age of the Anchor-age and Fairbanks systems by an intertie to be completed in the mid-1980's. The Railbelt Region covers the Anchorage-Cook Inlet area, the Fairbanks- Tenana Valley area, and the Glennallen-Valdez area (Figure 18.14). The utilities, military installations and universities within this area \\*hich own electric generating facilities are 1 is ted in Table 18.14. The service area of these utilities is shown i'n Figure ltL15 and the generating plants serving the region are 1 isted in Table 18.15. The Railbelt Region is currently served by nine maJor utility systems; five are rural electric cooperatives, three are municipally owned and operated, and one is a federal wholesaler. The relative mix of electric generating technologies and types of fuel used by the Kailbelt utilities in 1980 are summarized in Figure 18.16. 18-25 In 1980, the Anchorage ... cook Inlet area had Bl percent, the Fairbank·.)-Tenana Valley area 17 percent, and the Glennallen-Valdez area 2 percent of the total energy sales ic the Railbelt Region~ If the recommend at ions of the May 1981 Gi.lbert/Gommonwealth Report are im ... · plemented" toe Anchorage. and Fairb_anks_Jlmier .. systems will be intertied be- fore the Susitna project·comes into operation. The propo.sed intertie will allow a capaci.ty transfer of up to 70 MW in either direction. The proposed plan of interconnect ion envisages initial operation at 138 kV with subse- quent uprating ·to 345 kV allowing the line to be integrated into the Susitna transmission facilities. .. (b) Regional Electric Power Demand and Supply A review of the socioeconomic scenarios upon which forecasts of.electric power demand were oased is presented in Section 5 of this report. The forecasts adopted here. are the mid-range levels presented by Battelle Northwest in December 1981.. Subsequent forecasts which introduce price demand considerations have not been considered at this stage. The results of studies presented in Section 5 call for Watana to come into operation in 1993 and to deliver a full year's energy generation in 1994. Uevil Canyon wi 11 come into operation in 2002 and de 1 iver a full year • s energy in 2003. Energy demand in the Rail belt Region and the deliveries 'from Susitna are shown in Figure 18.17. (c) Market and Price for Watana Output in 1994 It has been assumed that Watana energy will be supplied at a single whole- sale rate on a free market bas is. This requires in effect that Susitna energy be priced so that it is attractive even to utilities with a low cost alternative source of energy. On this basis it is estimated that for the 3315 GWh of energy generated by Watana in 1994 to be attractive, a price of 140 mils per kWh in 1994 dollars is required. Justification for this price is illustrated in Figure 18.18. Note that the assumption is made that the only capital costs which would be avoided in the early 90s would be due to new coal-fired generating plants (i.e. the 2 x 200 MW coal-fired Beluga station). The financing considerations under which it would be appropriate for Watana energy to be sold at approximately 14~ mils kWh price are considered in Section 18.4 of this report; however, it should be noted that some of the energy which would be displaced by Watana' s 3315 GvJ~ would have been gen ... erated at a·lower costthan 145 mils, and utilities might wish to delay accepting it at this price until the escalating cost of natural gas or other fuels made it more attractive. A number of approdches to the resolu- tion of this problem can be postulated. (d) Market Price for Watana Output 1995-2001 After its initial entry into the market in 1994, the price and market for the 3387 MWh of Watana output is· consistently upheld over the years to 2001 by the projected 2U-percent iricrease in total demand over this period. 18-26 I, I I I I I I I I I I I I '-1 ·•I; I I I I I I --I I 'I I I I ,: •. I nl .. .. .. 0 • . . . There would,· as a result, be-a 70~percent increase in cost savtngs compared wi-th the best thermal altern:ative: these savings per unit ·of -output are_ illu$trated i·n Figure-18.19 .. (e) Market ·and-Price f.or Watana and Devi 1 -Canyon. Output in __ 2003 A diagramatic analysts of the~ total co·st savings which the combined Watana and Devil Canyon output wi 11 . confer on the ·system compared with the present thermal option in the year 2003 is shown in Figure 18.20. Dividing these; total savings by the energy contributed by Susitna indicates a price of 250 mils per kWh would be the maximum price which .can be charged for Susitna output. Here again, the problem of competing with lower cost combined cycle, gas turbines, etc., wil1 have to be addressed; however, this problem is likely to b.e short term in nature, as by this time period these thermal power ·faci 1 it ies wi 11 be approaching retirement. Only about 85 percent of the total Susitna output will be absorbed by the system in 2002, the balance of the, output being progressively absorbed over the following decade. This will provide increasing total savings to the system from Susitna, with no associated increase in costs. (f) Potentia 1 Impact of S~ate Appropriations In the preceding paragraphs the maximum price at which Susitna energy could· be sold has been identified. Sale of the energy at these prices wjll depend upon the magnitude of any proposed state appropriation designed to reduce the cost of Susitna energy in the earlier years, At significantly lower prices it is likely that the total system demand will be higher than assumed. This, combined with a state appropriation to reduce the energy cost of Watana energy, would make it correspondingly easier to mar~et the output from the Susitna development;. however, as the preceding analysis shows, a viable and strengthening market exists for the energy from the development even when the output is priced up to the cost of the best therma 1 a1 tern at i ve. , 18.4 -Financial Evaluation (a) Forecast Financial Parameters The financial, economic, and engineering estimates used in the financial analysis are summarized in Table 18.16.. The interest rates and far-ec.ast rates of inflation ·(in the CPR} are of especial importance. They-have been based on the forecast inflation rates in CPR and forecast interest ~ates on industrial bonds, as given by Data Resources Incorporated, and conform to a range of other authoritative forecasts. To allow for the factors which have brought about a narrow·ing of the differential between tax-exempt and non-tax-exempt securities, it ·has been assumed that any tax-exempt finan- cing would be at a rate of 80 percent rather than the historical 75 percent or so of the non-exemr;:»t interest rate. This identifies the forecast inter- est rates in the fina:ncing periods from 1985 in successive five year per- iods as 8.6 percent, ·;.a percent, and7 .. 4 percent .. The accompanying rate lB-27 {b) (c) of. inflation· is around 7 percent.. In view of the uncertainty attaching to such forecasts and in the interest of consE!rvati.sm, the following financial projections have been based upon the assumption of a 10-percent rate in- terest-for tax~exempt bonds r-t;~d an· ongoing inflation t•ate of 7 percent .. The Inflationary Financing Deficit ··:·.. . .. . . !2'. ~ The basic financing problem of Susitna is the magnitude .of its .. inflation- ary deficit." Under inflationary conditions these deficits (early year losses) are an inherent characteristic of almost al 1 debt financed long 1 i-fe, capita] intensive projects (see Figure 18.21). As such, -they are entirely compatible (as in the Susitna case) with project showing a good economic. rate of return. Although only a financing characteristic brought about by the project being h-eavily financed by debt under infl ai:ionary con- ditions, this characteristic makes it possible for the project to proce~ed without unacceptable burden of early year costs on consumers. The Has ic Financial Upt ions A range of financing options compatible with the conditions laid down in Senate Bill 25 have been considered as means of meeting the inflationary financing deficit. The -financial parameters used in these plans are as given in subsection (a) above .. The options basically consist of a range of preappropr iations (in 1982} by the State of Alaska with the balance of the project financing made up by a combination of G.O. bonds and 35 year· Rev- enue bonds, with G.O. bonds refinanced into Revenue bonds at the earliest opportunity., (i) State ~f Alaska Legislative Appropriation of 100 Percent of Cost ($4.5 billion) This conforms to the possible outcome legislated by Senate Bill 25 and represents the simplest financing option. It could take the form of the state meeting capital costs as incurred over the 15 year schedule.. Alternatively, it could take the fonn of 11 preappropri- at ion 11 where such a sum would be appropriated in; say 1982, as tak- ing into account interest accumulated, would totally finance the .pro-ject. For simplicity of interpretation of the options involving state appropriation all are assumed to take the "preappropriation11 form. A preappropriation of $4.5 biilion in 1982 would wholly complete Susitna (on the basis of centr.al estimates). On the basis of the present wholesale energy rate sett~ng requirement incorporated in Senate Bill 25, the APA would, h~wever, not be able to charge more than the actual costs incurred. Given that in this case the only costs would be the very small year-to-year operation costs, this op- t ion waul d invo 1 ve the output from Sus itna being supplied only at a fraction of the price of electricity from the best thermal option. 18-28 ·.' ·.-.··"'1!'·.-·.·j ... ·. < '·_.)~~~~ ; . . -. ·~"<' .• ,1 I~ I I I :.~-·. 1 .. ·~· I I I~ I I I' I I I. I .~ I ·. I< _) 1·- 1>· . .-· ~:: \". -.(. , • .:..:-o;r. I ·I ·':1: . -·-· I I I I I ~.1 I ~I ,I •• ··.1 I I ·J I :-I (d) (ij) 50 Percent State Pr~~4;ppropriqtion {$2~5 bill ion) wit~· Residual Bon~ F1nancimr -Tbe;Foutcome fo.r this opt ion-is summarized tn terms of figure 18-22 .. It is· seen that it would still enab1e Susitrra energy to be produced at a price 37 percent less than that of the best thermal option.. It would also enable the project to be completed with on·ly $1 billion (in l982·dollars) of G.O~ bonds (see below) over the period 1990-93. The Devil Canyon stage could then be completed with a further $2.4 billion (in 1982 dollars) of Revenue bonds over_the period 1994 to 2002 .. This level of appropriation would enable Susitna energy prices to be held virtually constant at their initial levels for nearly a decade. A temporary step-up in price to the cost of the electricity from the best thermal opt ian would be required when Devil Canyon was com- pleted an the basis of its lQQ..;.percent Revenue bond financing .. Thereafter, however, the cost of the Sus itna energy would again stabil jze and give ever increasing savings compare.,d with cost of the best thermal opt ion. (iii) 11Minimum .. State Preappropriation {$2 bill ion) with Residual Bond Financing The "minimum" State appropriation is taken as the minimum amount re- quired to meet debt service cover of L.25 an the residual debt ser-· v-ice cover of 1. 25 on the residual debt financing by Revenue bonds and make Susitna's wholesale· energy price competitive with the best thermal option in its first normal cost year (1994). This level of appropriation would require $1 .. 8 billion (in 1982 dollars) of bond financing 1990-93 and a further $2.2 billion (in 1982 dollars) over the period 1994 to 2002 to camp 1 ete Oev i 1 Canyon. (Figure 18.,23)- These levels of State appropriation would all therefore eliminate Susitna's 11 inflationary financing deficit." . . Issues Arising fromthe Basic -Financing Options ( i) Tax-Exempt Bond Financing In the $2 bill ion State apprapri at ion case interest costs an the basis of tax-exempt financing accounts for 88 percent of the unit price of Susitna output in 1994. Failure to obtain tax-exempt bond financing would increase these interest costs by approximately one third. Ensuring tax-exempt status far the Susitna bond issues is therefore of fundamental importance to the economics of the project under these apt ions. Difficulties could arise in obtaining tax-exempt bond financing if the financing entailed (as would probably be the case with Revenue bond financing at the precompletion stage) contracts of the take- or-pay or take--Jnd-pay type.. This is because the bulk of the 18-29 ~ Sus itna output wou1 <:1 be taken by non-tax ... exempt uti 1 ities and contracts of this type_ with non ... tax-exempt entities would, under certain general ·conditions laid down in Section 103 o.f the IRS code~ lead to the ·band issues being classified as i"ndustrial development funds and foregoing their tax:-~x:empt status" lt is also questionable whether contracts with the Railbelt ut·nities as current1y financed(·wou1d canst itute adequate security in the eyes of bond hold.ers. Both these considerati.ons indicate the need for soma fornr of ·;nd·ep-endent financing guarantee reducing dependence on the contractual relationship with the utilities., · This might take the form of the initial financing being G~O. bonds or by a State guarantee being given to the Revenue bonds. Given that either represents the same .effective burden to the State, it is cone 1 uded . that G. 0. bonds are to be preferred on grounds of flex i- bil ity and administrative simplicity. (ii) Refinancing Watana and the Financing of bevil Canyon ,. Earl-y refinancing of any G.O. bonds used to finance Watana and the financing of Devil Canyon by Revenue bonds is taken to be an impor- tant financing objective. The main factors determining the date at which such refinancing will be possible ts the magnitude of the ini- tial state appropriation.. This is dealt with in detail terms of the risk analysis in 18 .. 5 below. The basic conclusion from this analysis is that with a state appro- priation of $2.5 billion there is a very high degree of certainty that refinancing into Revenue bonds waul d occur by 1994 and that the r·ema·inder of the project could be financed by Revenue bonds. (iii) Importance of Adequate Preappropriation Funding to Subsequent Financing The principal effect of preappropriations significantly less than $2 bill ion would be a possible delay in refinancing of the G.O. bonds issued to finance Watana and possibly the need for additional G .. O .. bon.d financing for Devil Canyon. This is because the impact of such lesser preappropriation would (as illustrated in Figure 18 .. 24) give r·ise to inadequate earnings covered in the early years of Watana and subsequently Devil Canyon so that the raising of Revenue .Jonds requiring such cover had to be delayed. In addition, such inadequate funding waul d force the Susitna price to 11 track11 the cost of energy from the best thermal opt ion until adequate revenue had been obtained for such refinancing. (iv} Impas.t on State Credit Rating of Susitna G.O .. Bond Financing The impact on state credit rafing of G,O. bond financing of the order of $1..8 bill iorl (in 1982 dollars) in the $2 bill ion state ap- propriation case has been assessed by the APAs financial. advisors•, First Boston Corporation and First Southwest Corporation. They have I~ I I I I I I I I •• I I I I ') I I .I" I 1··. . ;"; I I 'I I I I I I cl ~~, .• I I I I . rJ/ (, . . •• II 'I i~ :·· i ;:;. ' . ~~~ concurred (as fully stated in sub-s'ection (c), (i'f',) of the main report) in:· the statement that uon the assumption that the State of JUaska•s bond· rating at that time is unchanged from today's level and that norma_lity prevails in the bond market, the effect on the credit rating of the Statr; of Alaska would not be perceptible." (e) Conclusion . The principal conclusion of the financial evaluation is that with a state appropriation of not less than $2 billion and consent for G.O~ bond financing of $1.8 billion (in 1982 dollars), Susitna would be financially viable. It would also be able to market its output at an initial price competitive with the most efficient thermal options and produce very substantial long-term savings compared with this option. The evaluation, however, stressed the importance of establishing the pro~ ject on a strong financial basis that wou1d enable it to secure conversion of the :G.O. bonds to Revenue bonds in 1994 and obtain a highly competitive rate of interest without jeopardizing the tax-exempt status of the bond issues. These objec.tives (together with the marketing of the Watana output in 1994 and price some 37 percent bel ow that of· the most efficient thermal option), .cou1d be secured by state appropriation of $2.5 bill i-on-~--- Methods by which the state appropriation could be recovered have not been considered since recovery is not required by existing legislation. It should be noted, however, that the cost benefit analysis shows that full recovery long-term would be possible with a better than 10-percent rate of return. Meeting the Susitna inflationary financing deficit can therefore be considered as a separate issue from subsidization of electricity prices by foregoing recovery of all or part of the State appropriation designed to meet this deficit. · 18.5 -Financial Risk The financial risks considered are those arising to the State of Alaska and to Alaskan consl.lllers. The analysis of these risks is restricted to the period up to 2001 covering the completion of Watana and its first nine years of operation. (a) Precompletion Risk The major precompletion risk is the risk that the project will not be com- pleted. The possibility of this arising owing to natural hazard is dealt with in Section , and on the basis of this analysis this possibility is assigned a vanishingly small probability. The risk of non-completion owing to capital overrun is also assessed to have negligible probability. This is on the grounds that the project only ·involves well established techno_logy, has been extensively assessed and surveyed, and has been assessed independently by estimators and formal probabi1 ity analysis as having only a 27-to 20-percent probaoil ity of any real capital overrun. 18-31 ',.,;,,: '...,.,_.,.../ I J I I (b) Post..:comp1e~ion R is~~.· ( i} The l1ener·ation of Post ... comp-letion Risk~ A probabilistic financial model was developed taking into account the probability distri_butions of the major engineering and financial variables on which the financial outcome for Susitna depends. This mode1 was then used to consider in detail critical specific risks and the aggregative risk posed by the project. (ii) Specific Risks -Specific Risk I; Risk of G.O. Bond Financing Overrun (Figure 18.25) Extensive analysis was undertaken to assess the probability that the G.O. bond financing requirements would overrun the forecast values as a result of capital costs, inflation, interest rates~ etc., being less favorable than forecast. In the $2.5 billion state appropriation case it was found that the probabi1 ity of the G.OQ bond requirement exceeding the forecast of $1 bill ion {in 1982 dollars) by more than 50 percent was only 0.15. This implies that there is less than one in six chance of the G.O .. bond overrun exceeding $1.5 billion. The probability of its exceeding $2 bil- l ion was only 0.03. There is also a significant probability that the bond financing requirements will b~ less than forecast. -Specific Risk II; Delayed Conversion of G.O. Bonds Minimization of the magnitude and duration of G.O. bond require- ments is taken as an important financial objective. Evaluation of this specific risk in the $2.5 bill ion appropriation case indi- cates that: . The probability of any delay compared with the forecast date of 1994 for the conversion from G.O. bonds to Revenue bonds is 0.05. . The 1 atest date at which complete conversion to Revenue bonds occurs in any outcome is 1996 (3 years after completjon of Watana}. .. Specific Risk III: Early Year Non-viability {Figurel8 .. 26) The measure of financi a1 non-vi ab i1 ity in the early years is taken as the ratio of Wat an a • s unit cost to the costs of the best ther- mal option in Watana•s third year (1996). (For comparability debt service excess cover was excluded). This· analysis indicates that there is only a 0.15 chance of the Susitna costs exceeding their forecast value (30 percent of the best thermal) by more than 15 percentage points. · 18-32 I I I I I I •• I I I I I I I I I I·· ••••• ' ': el I I I I 'I I I I I I I I I I I .. • •••• I -I f iii)--I_he ·Aggregate __ Risk While specific risks of the type considered above are of importance basic concern must center on the aggregate risk. In long-term eco- nomics this is measured by the risk attaching to the rate of return. · For the purpose of the financial risk, however, it is taken as rep- resented by accumul at'lve net operating earnings at the end of the first nine years of operation of Watana. Since this statistic is. net of interest and debt repa}ment, if effectively subsumes all the· risks involved in capital expenditure, inf1 at ion, interest rates, revenue, etc .. , deviating from their forecast values. This statistic was also adjusted to allow the pricing up of Watana energy to the cost of the best thermal option so that statistic affects the 11 UpSideH risk as well. as the 11 d0W0Side.11 On this basis the statistic (see Figure 18~27) was found to have only a 0.15 chance.of being below forecast level of $1.35 billion (in 1982 dollars) by more than $.35 billion. There is also a 0~34 probability of the statistic exceeding $1.5 billion and thus creating greater savings for the Alaskan consu.mer. (c) Conclusions The analysis shows the exposure of the project either to critical s-pecific risks or to aggregative risk is relatively limited. The qualifi.cation attaching to this analysis is that the estimates and probabilities used are free from any systematic biases~ The structure of the plan of the overall plan of study for Susitna and analysis of its alternatives has however been specifically desi9.·.~d to take every reasonable precaution against this pos- sibility by seeking extensive independent verification of the key variables by Batelle and Ebasco operating wholly as independent consultants. 18-33 . LIST OF. REFERENCES - (1)·· U.s. Department of Labor, Mont~lY, Labor ~ReviJW, various issues .• (2) Alaska Department .of Commerce and Economic Development, The Alaska Economic Information and Reporting System, Ju 1y 1980. (3} Data Resources Inc., U.S~ Long-Term Review, "Fall .1980, Lexington, ,Mass., 1980. (4) Wharton Econometric Forecasting Associates, Fall 1981, Philadelphia, Pa. (reported in Economic Council of Canada CANDIDE Model 2-0 Run, dated · December 18, 1981.} (5) Baumel, W.J., "On the Social Rate of Discount", American Economic Review, Vol. s·a, September 1968. (6) Mishan, E.J., Cost-Benefit Analysis, George Allen and Unwin, London, 1975. (7) Prest, A.R. and R. Turvey, 11 Cost-Benefit Analysts: A Survey 11 , Economic Journal, Vo 1. 75, 1965. · ( 8) U.S. Department of Commerce, Survey of Curre!lt Business, various issues. (9) Data Resources, Inc., personal comnunication, .November 1981. (10) World Bank, personal communication, January 1981. (11) u.s. Department of Energy, Energy Information Administration, Annual Report to Cong~ess, Washington, D~C. (12) National Energy Board of Canada, Ottawa, Canada, personal communication;, October 1.981. {13) Noroil, 11 Natural Gas and International LNG Trade 11 , Vol. 9, October 1981. (14) Segal, J. 11 Slower Growth for the 1980•su, Petroleum Economists 80 December 1980. (15) Segal, J. and F. Niering, 11 Special Report on World Natural Gas Pricinga•, Petroleum Economist, September 1980. (16) SRI International, personal communication, October 1981. (17) World Bank, CoiTUllodity Trade and Price Trends, Washington 1980. (18) Battelle Pacific Northwest Laboratories, Beluga Coal Market Study, Final Report, Richland, Washington, 1980. {19) B.C. Business, August 1981. (20) Coal Week International, various. issues. I I ~--~,~-----···.,_ I •• I I I I I I : •... I I I'. I I :··\' I .,· . ' I I I . ';::• I I I "1. I I I I I I ,I .I ~.~ ,~.: :1 .· . ·t ·LIST. 0~ REFERENCES (Continued) {21) Japanese Ministry of International.Trade and Industry, persona;1 communication, January 1982. · (22) Canadian Resourcecon Limited, Industrial Thermal Coal Use in Canada, 1980 to 2010, May 1980. {23) Battelle Pacific Northwest Laboratories, Alaska Coal Future Availability (24) and Price Forecast~ May 1981. · · Roberts, J.Q. et al, .Ireatment of Inflation in the Development of Discount Rates and Levelized Costs in NEPA Analyses for the Electric Utility Industry, u.s. Nuclear Regulatory Commission, Washington, D. c., · January 1980. · : ''l . . .. TABLE. 18.1: REAL (lNF'LATION-AOJUSTE.D) ANNUAL .. JlROWUf·lN OIL PRICES Low Case Hsdi.un (most likely case) High Casa Base Period (Jam~ary 1982) Growth Rates (Percent) 1982-2000 -- 0 2.0 4.0 2000-2040 0 1.0 2.0 Pric:e of No. 2 Fuel Oil -$6 .. 50/MMBtu. . Probability O.J 0.5 0.2 ; ,, ···,,.~ -~· ;· ·,. ~~~ .. :~ . -::: I ., I I I I I I I •• I I I I ..•. ·.< ( . -~ I I I I I I I I I I I I I I I. I TABLE t810Z: DOMESTIC MARKET PRiCES AND EXPORT OPPORTUNITY VALUES:"OfcNil.lURAL.GAS Domestic Market Price1 Exeort Oe§ortunit~ Value Cow Rea1um Hl.9F\ Cow Me 10m igh · P-robability of Occurr&nce N.A. N .. A. N.A. 2.7% Base Period Value $3.00/MHBtu Real Escalat~on ClF Price, Japan '1982 -2000 N.A. ~ 2000 -2040 i)' G% Real Escalation Alaska Price 1982 -2000 0% 2.5% 5.0% 0% 2000 -2040 0% 2.0% 2.0% 0% 1 OGP analysis used domestic market prices with zero escalation beyond 2010. (Source: Battelle) 46% $4.65/MMBtu2 2% 1% 2.7% 1.2~ 2 Based on CIF ptce in Japan ($6.75) iess estimated cost of liqllefaction and shipping ($2.10). (Source: 13, 14, 15). 3 Source: (9), (16). 4 Alaska opportunity value escalates more rapidly than . CIF prices as liquefaction and shipping costs are estimated to remain constant in real terms. 27% 4% 2% 5&2% 2.2% <1- tABlE' 1S.Jt._ SUMMARY OF .COAL OPPORTUNlTY_VALUES -" Base :Perio~, Annual Real Grwoth Rate (Jan.-. 19.82} Value 1980 -2000 ' (%) ' 2000 .. 2040 ( $/f+tBt u) . . (%) Base Case Battelle Base Period ·crF Price Medium Scem .. rio -CIF Japan 1.95 2.0 1 .. 0 -fOB Beluga 1.43 2.6 1.2 -Nenana 1 .. 75 2.3 1.1 Low Scenario -Clf Japan 1.95 0 0 · -FOB Beluga 1 .. 43 0 0 -Nenana 1.75 0.1 0.1 High Scenario -CIF Jspan 1.95 4.0 2.0 -fOB Beluga 1.43 5.0 2.2 -Nenana 1.75, 4.5 1. 9 Sensitivit~ Case Updated Base · Period CIF Price1 Medium Scenario -CIF Japan 2.66 2.0 1.0 ... FOB Beluga 2.08 2.5 1.2 -FOB Nenana 1. 74 2.7 1.2 Low Scenario -CIF Japan 2.66 0 0 -FOB Beluga 2.08 0 0 -FOB Nenana 1. 74 -0.2 -0.1 High Scenario _, Clf' Japan 2.66 4.0 2.0 . -FOB Beluga 2.08 4.8 2.2 -FOB Nenana 1.74 5.3 2.3 1 Assuming a 10 percent discount for Alaskan coal due to quality differentials, and export potential for Heal coal. 0 Probability of Occurrence 49 49 49 24 24 24 27 2.7 27 49 49 49 24 24 24 27 27 27 ' ' I I I ' ' · ..•... _ I I I I I I I •• I I I I -1· '_a . ' :12 _· .. "' I I I I I I I •• I I .I I I I I I I I TABLE 1S.4: . Sl.ttMARV Of F'UEL PRICES USED IN THE , OGP PROBABILITY TREE ANAL VSIS a Fuel ~rice Scenario Probability of occurrence Base period Jan1,1ary '1982 prices (1982$/MMBtu) fuel Oil Natural Gas Coal -Beluga -Nenana Real esca.lationrates per year (percent) Fuel Oil -1982 -2000 -2000 -2040 Natural Gas -1982 -2000 -2000 -2040 Beluga Coal -1982 -2000 -2000 -2040 Nenana Coal -1982 ... 2000 -2000 -2040 Low 25% 6.50 ).00 1.43 1.75 0 0 0 0 0 0 -0.1 0.1 1 Beyond 2010, the OGP analysis has.used zero real escalation in all cases. ,_ Medium 50% 6.50 3.00 1.43 1.75 2.0 2.0 2.5 .2.0 2. 6 1.2 2. 3 1.1 25% 6.50 3.00 1.43 1. 75 4.0 2.0 5.0 2.0 s.o 2.2 4.5 1.9 .-"'--· .. . J;f .. TABtf 18,.5:· tCONOHIC ANALYSIS Plan 10 Non Susitna A Susit~'l c· Net Economic Benefit of Susit.na Plan _ SUSllNA PROJECr -BASE PLAN 1982 Present Worth gr .System Costs . $ X 10 · . 199:3-Estimated 1993- Components 2010 2010 2011.;.2051 2051 - 600 MW Coal-Beluga 3,213 491 5,025 8,238 200 MW Coal-Nenana 630 MW GT 680 MW Watana 3,119 385 3,943 7,062 600 MW Devil Canyon 180 MW GT 1,716 • I -·" ·-' "· 1•···.· .. · ... ' . ., ···:···!; .. . f: .. ·. ; . . . I I I I I I I I I -··· .·_ .. I :1-.. . . .. 1· . . _,. \_:_ '· I :t":':'?'' ~~T·--•,.,. ,. -.•...• ·-, '• : .. 1·----c:· . .. . ·· '· I I I I I I -· I ;'" ,, I I I I I 'I ,, ~1i\: <·: - .< ~ .. ·. ;:: : '..!.. . MW 1990 892 2000 1,084 2010 1,537 --- TABLE 18.6: SUMMARY OF LOAD FORECASTS USED fOR SENSITIVITY ANALYSIS Medium Low High GWh MW GWh l~W - 4,456 802 3,999 1,098 5,469 921 4,641 1,439 7,791 1,245 6,303 2,165 -::- GWh 5,703 7,457 11,435 TABLE 18.7: LOAD fORECAST SENSUIVUY ANALYSIS 1982 Present Worth of S~stem Costs ($ x 106) Net 1993-Estimated 1993-Economic Plan 10 Components 2010 2010 2011-2051 2051 Benefit Non-Susitna K 1. 1400 MW Coal-Beluga 2,640 404 4,238 6,878 Low forecast 200 HW Coal-Nenana 560 MW GT Susitna K2 680 MW Watana (1995) 2,882 360 3,768 6,650 228 Low forecast 600 MW Devil Canyon (2004) Non-Susitna J1 800 HW Coal-Beluga 4,176 700 6,683 10,8591 1 High forecast 200 HW Coal-Nenana 700 MW GT 430 MW Pre-1993 Susitna J2 680 HW Watana (1993) 3,867 564 5,380 9,2471 1 1,612 High forecast 600 MW De~il Canyon (1997) 350 MW GT 430 HW Pre-1993 1 from 1993 to 2040 !,._--;-.·--~'-~ ..... -.;.:. _ _,' ·~ ,,.. ~ ......... __ ._ ~~-~-.. ""' -·-•• -· -·-··-. -J -· - ------------------- TABLE 18.8: DISCOUNT RATE SENSITIVITY ANALYSIS 1982 Pr-esent Worth.of S~stem Costs($ Xc106) Real Net Discount Rate 1993-Estimated 1993-Economic Plan ID (Percent) 2010 2010 2011-2051 2051 Benefit Non-Susitna Q1 2 3, 701 465 7,766 11,167 Susitna Qz 2 3,156 32J 5,394 8,550 2,617 Non-Susitna A 3 3,213 491 5,025 0·,328 Susitna c 3 3,119 385 3,943 7,062 1,176 Non-Susitna 51 4 2,791 517 3,444 6,235 Susitna s2 4 3,080 457 3,046 6,126 109 Non-Susitna p1 5 2~468 550 2,478 -!·~946 Susitna Pz 5 3,032 539 2,426 5,459 (513) TABLE 18.9: CAPITAL COST SENSITIVITY ANALYSIS 1982 Present_Worth of S~stem Fasts-$ x 1993-. Estimated 1993- Plan ID 2010 2010 2011 ... 2051 2051 Non-Susitna. Capital Costs Ue 20 Percent Non-Susitna G 3,460 528 5,398 8,858 Susitna c1 3~119 385 3,943 7,062 Non-Susitna Capital Costs Down 10 Percent Non-Susitna G 3,084 472 4,831 7,915 Susitna c1 3,119 385 3,943 7,062 Susitna Capital Costs Less Contingenc~ Non-Susitna A 3,213 491 5,025 8,238 Susitna x2 2,710 336 3,441 6,151 Susitna Capital Costs _ Plus __ Doubled Contingen~ Non-Susitna A 3,213 491 5,025 8,238 Susitna v2 3,529 434 4,445 7,974 1 An adjustment calculation was made regarding the + capital costs of the 3GT 6nits added in 2007-2010 since the difference ~as less than $10 x 10 • Beyond 2010, this effect was not included. ---------~___,....,.,~ 106 Net Economic Benefit 1,976 853 2,087 264 -a_--__ -- I I , ...• I •• I I I I I I I I I •• I --' I .•. -. ., -- -; .>:' ,:.::/ I I ~· I I I I I I I I I I I I • ~=I 'I, ~,-J . ' .... · ' ·-""" . ' ' ' <' Base Case Sensitivity TABLE 18.10: SENSITIVITY ANALYSIS -UPDATED BASE PLAN (JANUARY 1982) COAL PRICES Base Period Beluga Coal Price (1982 $/MMBtu) 1.43 PW in 1982 ($ X 106) ------ Costs of Non-Susitna Plan 8,238 Costs of Susitna Plan 7,062 Net Economic Benefits 1,176 (Updated) Case 2.08 0 TABLE 18.12: S~NSITlVITY. ANALYSIS -REAL COST ESCALATION 1982 Present. Worth gf System Coat-s·· · ($ X 10 ) ·1993-Est1mated 1993-Net Plan ID 2010 2010 20.11-2051 2051 Benent -- Zero-Escalation in Caeitsl ~nd O&M Costs • Non,...Sus1tna 01 2,838 422 4,319 7,157 • Susitna o2 2,525 299 3,060 5,585 1,572 Double Escalation Capit~ and O&M Costs . Non-Susitna p1 3,650 602 6,161 9,811 • Susitna Pz 3,881 503 5,148 9,029 782 Zero-Escalation in Fuel Prices • Non-Susitna v, 2,233 335 3,427 5,660 • Susitna v2 3,002 365 3,736 6,738 (1,078) High Escalation ln Fuel Prices • Nan-Susitna w.t 4,063 643 6,574 10,367 • Su~Htna w2 3,267 403 4,121 7,)88 2,979 TABLE 18. 12 {a): SENSITIVITY ANALYSIS -NON-SUSITNA PLAN WITH CHACKACHAMNA Plan • Non-Susitna with Chackachamna ·• Susitna 1993 1982 Present Worth of System CQsts ($ X 10b) Estimated 1993-Net ID Components 2010 2010 2011-2051 2051 Benefit c -- B 330 MW Chackachamna 2,038 475 400 MW Coal-Beluga 200 MW Coal-Nenana 440 MW GT C 680 MW Watana 3,119 385 600 MW Dev1l Canyon 180 MW GT 4,861 7,899 3,943 7,062 837 I , •.. . ',::,_, ·-.; ••• •• I • • 'I I I I I I I I I I I I I' I I I .I I I •• I I I I I I I ••••••••• ' TABLE 18.13: SUMMARY OF SENSHIVITY ANALYSIS INDEXES Of NET ECONOMIC BENEt: ITS . BASt CASE . ($1, 176 MILLION} = 100 Fuel Escalation -High -Low Discount Rates -High-High (5%) -High (4%) -Low (2%) Susitna Cap1tal Cost -High -Low Load Forecast -High -Low Non-Sus~tna (Thermal) Capital Costs -High -Low Capital and O&M Cost Es~alat ion -H1gh -Low Chackachamna (included 1n Non-Susitna Plan) Updated Base Coal Price -44 9 223 23 178 137 19 168 73 67 134 71 0 t High fuel escalation case provides net benef1ts equal to 253 percent of the base value, 2.53 .x 1, 176, or 2,975. 2 Low fuel escalat1on case provides minus 92 pe·rcent of the base case net benefits, -.92 x 1, 176, .or -1,082. 'I· ---·- . Generating Purchases Utility Annu.Ait Capacity 1981 Predominmt Tax Status Wholesale .Provides Energy Dem~Uldf MWat0°F Type of Re: IRS Electrical Wholesale 1.980 UTILITY Rating Generation Section 103 Energy Supply GWh -·' ' IN ANCHORAGE-COOK INLET AREA Anchorage Municipal Light and Power 221.6 SCCT Exempt • -585.8 '· i Chugach Electric Association 395.1 SCCT t~on-Exempt • • 941.3 Matanuska Electric Association 0.9 Die~tl Non-Exempt • -268.0 j' Homer Electric Association 2.6 Diesel Non-Exempt • 284.8 i -\ Seward Electric System 5.5 Diesel Non-Exempt • 26.4 1 -r Alaska Power Administration 30.0 Hydro Non-Exempt -• -l National Defense 58.8 ST Non-Exempt I ---k Industrial -Kenai 25.0 SCCT Non-Exempt ---! i • IN FAIRBANKS-TANANA VALLEY ' ~ Fairbanks Municipal Utility Syatem1 68.6 ST/DieseJ Exempt -· -116.'1 Golden Valley Electric Aasociatcon 1 221.6 SCCT/Diesel Non-Exempt --316.7 University of Alaska 18.6 ST Hon-Exompt --·-; National ~ef&nsa1 46.5 ST Non-Exempt ---. IN GLENALLENNALDEZ AREA Copper Villey Electric Association 19.6 SCCT N6n:-Exempt --37.4 TOTAL 1114.3 2577.1 1 Pooling Arrangements in Force TABLE 18.14 _ RAlLBE~T UTILITIES PROVIDif~G MARKET POTENTIAL I ~~~m I · ..... •' .. I I ,I I I I I 'I •• I I I I I· I· I P~ANT No. 2 3 6 7 10 2? ~· 32 34 35 36 37 38 47 55 58 59 75 80 81 82 83 84 PLANT LIST TYPE OF NAME OF PLANT UTILITY OWNERSHIP Anchorage .No. 1 Anchorage Municipal Light and Power Mun3cipal Anchorage Anchorage Municipal Light and Power Munici.,af Eklutna Alaska Power Administration. Federal Chen a Fairbanks Municipal Utilities System Municipal KnikArm Chugach Electric Association, Inc. Cooperative Elm•ndorf-West United States Air Force Federal Fairbanks Golden Valley Electric Association, Jne. Cooperative Cooper a..ke Chugach Electric Association, Inc. Cooperat:ve Eimendorf-East United States Air Force Fideraf Ft Richardson United States Arrny Federal Ft. Wainright United States Air force Federal Eielson United States Air Force ·Fedtral Ft. Gnttlty United States Army Federal Bemice Lake Chugach Electric Association, Inc. Cooperative International Station Chugach Electric Association, Inc. Cooperative Healy Golden Valley Electric Aaociation, Inc.. Coopemive Beluga Chugach Electric Association, Inc. Cooperative Cle~rAFB United States Air Force Federal Collitr·Kenai Collier-Kenai Municipal . Eyak Cordova Public Utilities Municipal North Pole Golden Valley Electric Association, .Inc. Coopenmve Valdez Golden Valley Electric Association, Inc. Cooperative Glennallen Golden Valley Electric Association, Inc. Cooperativ-e TA81.E 18.15 -LIST OF GENERATING Pl.ANT SUPPL!NG RAI LBE1.T REGION ~~~~ l I I I ., .... . ' I I I I I ·I I ••• I I I I I I TABLE 18.16: ESTIMATED FINANCIAL PARAMETERS Project Complet~on -Year Energy Level -1993 .. 2002 -2010 Costs ~n January 1982 Dollars Capital Costs Operating Costs -per annun Prov~s~on for Capital Renewals -per ar.num (0.3 percent of Cap1tal Costs) Operat1ng Working Capital Wstana - 1993 $ 3.647 bl.llion $10..0 million $10.94 million -1'5 percent of Operating Costs -10 percent of Revenue Reserved Contlngency Fund Devil· Canyon 2002 $1.470 bilhon $5.42 million $4.41 million -100 percent of Operat~ng Costs -.100 percent of Provision for Capital Renewals Interest Rate -10 percent per annum Debt Repayment Period -35 years Inflation Rate - 7 percent per annum Real Increase in Operat1ng Costs -1982 to 1987 -1.7 percent per annum -1988 on -2.G percent per annum Real Increase in Capital Costs -1982 to 1985 ... 1.1 percent per annum -1986 to 1992 . -1.0 percent per annum -1993 on -2.0 percent per annum Total J :387 GW·h 5 721 If 6 616 " $ 5.117 bllhon $15.42 million $15.35 million • - ----..... -J - --; - - --· ... :a a & I. REVIEW _ BASE COST .--.... _...,..AND ~TART -- SCHEDULE ESTIMATE i i I SUMMARY NOTES .____, I I _, n. RISK LlST DEVELOPMENT RISK LISTS I I Ill. VI. .,.. METHODOLOGY _;_-._ SOFTWARE -REVIEW -- v. TRANSFORMATION ASSESSMENTS IV. RISK ASSESSMENTS I ' ASSESSMENT DOCUMENTS -REVISIONS VII. CONSEQUENCE /RESPONSE CRITERION ASSESSMENTS I I DOCUMENTATION .x. INITIAL COMPUTATION AND INTERPRETATION VUI. REVIEW AND REVISE RISK ANALYSIS STUDY METHODOLOGY .. )(. TRANSMISSION SYSTEM AND EMERGENCY GENERATION X f. UPDATE ...... AND -FEEDBACK XII. FINAL COMPUTATION AND INTERPRETATION ; ~ RISK ANALYSIS REPORT FIGURE 18.1 I 1" ·-~ ~ I I '1 I I I I '. ~I I :I I 1. I I I I QUESTION: -' WHAT MAJOR CONSTRUCTION PROJECTS ARE INVOLVED? WHAT KIND OF WORK IS GOING ON FOR A GIVEN . CONFIGURATION? WHAT ARE THE POSSIBLE INITlATING MECHANISMS WHICH COULD -INFLUENCE ESTIMATED COSTS OR COMPLETION TIMES'? . WHAT MAJOR PORTIONS OF ANY GIVEN CONFIGURATION ARE SUBJECT TO RISK REALIZATION? IF A PARTICULAR RISK MAGNITUDE IS REALIZED, WHAT POSSIBLE CONSEQUENCES CAN OCCUR ? HOW CAN THESE CONSEQUENCES BE MEASURED ? WHAT IMPORTANT ASSUMPTIONS AND .Ll MITATIONS MUST BE ESTABLISHED TO PERMIT A REASONABLE ANALYSIS AND TO DRAW IMPORTANT CONCLUSIONS 7 ELEMENTS OF THE RISK ANALYSIS -~----.... -.--:-~-~- RISKS FIGURE 18~2 •• ........... ~.,. el. I ·····;: t ;~~- 1· I I I •• I. I. I I I. I I ~, . ·;: ; PROBABtLlTV OF A, PARTICULAR RISK MAGNITUDE t ® PROBABl LITY OF A PARTICULAR DAMAGE LEVEL IF A PARTICULAR RISK MAGNITUDE lS REALIZED t PROBABILITY OFA PARTICULAR CRlTERION RISK NONE LIGHT MODERATE MAJOR CATASTROPHIC MI.N MODE ;)lb INCREASING CRITERION VALVE MAX CD A SERlE$ OF OlSCREJE .. . RISK PROBASlLlTY LEVELS EX1STS FOR EACH ~lSI( .. ACTJ VlfY . COMSlNATJON~-- THE ANNUAL PROSABJLlTY.· O_F. EACH IS OETERMIN.EO. @IF A RlSK EVENT OCCURS, · IT CAN CAUSE A NUMBER OF POSSIBLE DAMAGE . LEVELS, EACH WITH A PARTICULAR PROBABl LITY OF OCCURENCE. lF RlSK MAGNITUDE @ c;>CCURS, THE PROBABILITY IT WILL CAUSE MODERATE DAMAGE lS THE VALUE OF ® ON THE DIAGRAM. @FOR ANY GIVEN DAMAGE LEVEL, THREE CRITERION VALVE:S ARE ESTlMATED AND FIT TO A MODlfTED BETA OISTRl BUTtON~ STRUCTU9AL RELATIONSHIP FOR HANDLING -RISK ACTIVITY COMBINATIONS, DAMAGE SCENARiOS ' \. .r.lll.· , \ · . . FIGURE 18.3 liiiJ I ;'·....... . &.............._........_. ________ ..._ ______ .......___......_ ........... ...___ ..... AND CRITERION VALUES . .. . -. {~ -----_· .. ~-{-· -__ .· -.-' ---·-···· ___ · _: __ .t· ..... · -.. --. . i .• . . .I -- - ----~ --~-'i >-t- -m <( m 0 n: n. .8 .7 .6 I I I I I I l I I ; oT-._----~----------------­c CR I TERlON VALUE (!) CUMULATIVE DISTRIBUTION ANY POINT ON 1"'HE CURVE INDICATES THE 'PROBABlllTY I P) THAT THE CRITERION· VALUE (C) WILL NOT BE EXCEEDED. .5 .4 .3 . >- t- :i.2 -m <( en o I Q! ·n.. p I 0 ~--+--------------c CRITERION VALUE @ DENSITY FORM ANY POINT ON THE CURVE INDICATES THE -PftOBABILITY (?)THAT A PARTICULAR CRITERION VALUE ( Cl WILL BE INCURRED. ALTERNATIVE FORMATS FOR PRESENTING THE ANALYTICAL RESULTS ---m ct m 0 0: .n. 1.01 .3 .2 p .I CRITERION VAtJ.IJE @ REVERSE CUMULATIVE ANY POINT ON THE CURVE INDICATES THE PROBABll..lTY ( P) THAT THE CRTtER lON VALUE (Cl WILL BE EXCEEDED. FIGURE 18.4 1.0 0 w .9 LL.W 0~ ww ~8 (!)t-~0 .7 zZ w_. u...J .s o::_ w:= a. .5 t-w- <ti-LIJ :r: <( .::> .4 ...... ~_. - --~ >-~::> . 3 t-w --0 _., liJ _....., .2 mo·!;t <two m-,· oo-• t 0::0::~ Q..Q.._ 0 . lt '. l . ··- I EXPECTED VALUE I ~~~~ESTIMATE 90.25°.4-~ I I ~ -it ~ROJECT ESTIMATE) I _l ~L· L ~ t foe= "LOW" ESTIMATE . / . I I J . v I ~ I / I I -• I I 70 80 90 100 110 120 130 PERCENTAGE OF FINAL DIRECT COST ESTIMATE WITH CONTINGENCIES CUMULATIVE PROBABILITY DIST,RIBUTION FOR WATANA PROJECT COST ' • FIGURE 18.5 . -I ~ ! I --+ ~·~ ! f 1 ! 140 ·[iJ _J 1.0 _. -~·. .9 w 1-(!) (/)~ .a Oz ow 0 .7 _JO:: <[W ::>·0... ~6 ~ oo <(W .I-! .5 ·~<( <to :t:o t-z .4 ->-o !::w .3 _.w -o .2 mx <(l..LJ m ot--. I o:O o..z / .L' ~ v , .. 60 I -l --~~ EXPECTED VALUE~ ! ~C= •• HIGH .. ESTIM 91.5°k '~l ~ ·~~ ATE I ~~PROJECT ESTIMATE) I I . v _z i I . [_)!'B ~ 11LOW" 1 ESTIMATE ~ / l t v I L I I I I I . 70 80 90 100 PERCENTAGE OF PROJECT ESTIMATE CUMULATIVE OiSTRlBUTION OF DEV.IL CJ,\JNYON COSTS I 110 120 130 FIGURE 18.6 ,, fill -~) ' - 1.0 .9 (!) z .a a IJJ .7 w 0 w X w => .6 ..J I-~· 0 .5 z 0 lL w 0 .... 4 <( >-0 1--.3 0 ..J z -m .2 <( m 0 .I a= 0.. 0 -··} ----., -.. --" -._,, -·- I : I . -c--~ I EXPECTEO VALUE~ ~C:: ''HIGH11ESTIMATE ' 9.06% -I ~· / I 1V . ~· :/ A•(PROJECT ESTIMATE) ' - / 1 - I ,fiJB '· ., B •''LOW 1 " ESTIMATE / I I v I . I ""' I / I I ' / -:t I Y! it 70 eo 90 100 110 120 130 PERCENTAGE OF FINAL DIRECT COST ESTIMATE WITH CONTINGENCIES CUMULATIVE PROBABILITY DISTRIBUTlON FOR SUSITNA HYOROELFtr:TRIC PROJECT - £ ' ~ ' ' t ~ \ -· .' . - ; .) i ; '.~~" 1 < ~ 1 rl !I ' i ~ I I ' ' r 140 FIGURE 18.7 -· -. . " 0.9 l.t t.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 RATlO OF ACTUAL COST TO 11 fNITIAL" ESTIMATE' HISTORICAL WATER R.ESOURCES - • PROJECT COST PERFORMANCE {48 PROJECTS l FIGURE 18.8. ----~--- 11.., .eo ....----+-----t----+-~-·~---- 0 .70 ........ ---+----+-----1 .20 .10 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 RATIO OF ACTUAL COST TO "INITIAL .. ESTIMATE COMPAaiSON OF SUSITNA RISK RESULTS WITH HISTORICAL WATER RESOURCES PROJECT . COST PER~RMANCE (48 PROJECT} 2.5 f-··.:·· ? . DATA FOR 48 WATER RESOURCES PROJECTS 2.7 2,9 FIGURE 18.9 3.1 -. . . -···· ~~--------------------~----------------------------~~~ .. -·-------------~--------~--- 1.0 .9 C) z 8 0 . laJ g LIJ .7 ::l ~ ..J .• 6 0.""""" z~ LLQ ow .5 >-~ .4 t-o :iS m~ .3 '(( al ~ .2 a. .I 0 ~----- I .. ,__, ·~ ~ ·--· -/ / ~ ----/ )ifh = SCHEDULE ESTIMATE _,£ -----INCLUDING A ONE YEAR / CONTINGENCY / 0 / .. / -./ v B= SCHEDULE ESTIMATE 7 :::::::1 WITHOUT CONTINGEI~CY . . I -15 -10 -5 0 5 10 MONTHS FROM SCHEDULED COMPLETION WATANA s.~CHEDULE OISTRIBUTIO~ EXCLUSIVE OF REGULATORY RISKS ~ ~-·"- 'I '· ,. i [ ? l I ! ' ' ~ ' ;> ' ! ~ '" ' j I 1 15 FIGURE 18.10 J sia I ,,"<_:l(.._,.~--..... -------~ ...... ---...... ---..... --..... -----...-.----llllillli.--..:~......-----......-....... ~~;;;;;;;;.a ,..-i.~ ,; ,., 1.0 .9 .8 (!) z -. 7 0 i1J w 0 .6 X Ww ::l .5 ..... ..J ~~ lL 0 ~4 ow >-ti "2 1-0 ·'-~ :::io ffi z .2 <t- m ~ .I D... 0 ;f. ' ·~---, . ~ ~ / lo-" ....(o~ SCHEDULE ESTIMATE -/ INCLUDING A ONE YEAR CONTINGENCY ~ ~ . / B= SCHEDULE ESTIMATE WITHOUT CONTINGENCY ~ / . 7 __.,., v -30 . " . -20 -10 0 10 MONTHS. FR0~1 SCHEDULED COMPLETION WATANA SCHEDUL:E DISTRIBUTION INCLUDIN.G THE EFFECT OF REGULATORY RISKS \, ,,· 20 ~.·· ' ~ 1 , 1 ' I ' ! ( i1 If . . 1 . I r t I l i l ' 1 l ' ) ; ! J I ' I ' i " ' I l I . 30 FIGURE 18.ll ill 1.0 (!) z· -.99 0 wo: W<( ow X>-w a: .98 sw 0.. z l.L(/) 0~ .97 >-0 !::a -'w m~ .96 <( ,~ m-o-a:::· a a..~ • T -~o-----~~ ~ ....... ss .,.. ~ ,,P "',o~ · .r ~~v '\0 / ~XPECTED VALUES: / o\- ./ ';:)0 TOTAL LOSS 0.06961 /,jv 50°/o REDUCTION o.o917t I I I ll_ v r 0 r .. 2 3 4 5 DAYS PER YEAR OF REDUCED ENERGY· DELIV~RY CUMULATIVE PROBABILITY DISTRIBUTION FOR DAYS OF REDUCED ENERGY DELIVERY TO ANCHORAGE 1i jl .. ~ n ~~ ~ i1 n u n li p ·{ t H fi 1! ~ I; I ANCHORAGE I i ' 6 7 (j FIGURE 18.12 ~~~~~J 1.0 (!) z c ~~ o<t xw . 99 W> .,_a:: ow zO... .98 LLcn 0~ <( >-0 1-.97 :Jo _w ml-~<( mo o-0::0 .96 .a.~ T r 0 / ~ v EXPECTED VALUE: .08116 . . --~· --- 2 4 5 DAYS PER YEAR WITH NO ENERGY DELIVERY CUMULATIVE . PROBABILITY DISTRIBUTION FOR DAYS PER YEAR WITH NO SUSITNA ENERGY DELIVERY TO FAiRBANKS [ . 1 l . I ' 1 J I ! I I . I FAIRBANKS I 6 7 FIGURE 18J3 [iJ 1- )' ' ' ,1·· __ _ ' # o ••• ~ ,,._,_---._---__ . -.~ i I ~-... 1-- ' I I I I I '. I '~; I I I t ' I I RAILBELT .REGION GENERATING AND TRANSMISS.ION FAClLlTJES 0 65 .130 lC1\.0._.£TERS I :1 -- 1!0 MI\.ES !Ail 0 65 FIGURE 18.14 · · ::.·~ ,·: ,. ' 1·:. .. · I I ... < .. _,. I ""' ,, LOCATION MAP J..EG£NO \f PROPOSED OAM SlTES ----,ftOPOs.al 13$ ICV UNE I EXISTING LINES I • ......: ' I I •• •.· I I I '"l./ t ;.1 c~~~ FIGUR.E JiiJ .I I I I ·II ..,. a ·-· I I I I :I ""'' ., ' ... I I .., ' i: ·~ I . ; ~·,.· t '::'" . l : r , I Oil 2% -·~- Rural Electric Cooper1tives .10% MunicSp1l Sy,ternr.. 23% U.S. Gcv•rnment Alatka Power Adnaini!tration -Eklutna 1. Don Not Include Sttf Supplied Energy from Mili".ary tnstallatians and The UnivlfSity oi A Ink a A ENERGY SUPPLY (Based on Net G~neration 1~0) Gas 76% 1. Does Not Include GeMration by Military lnstt!latiQns and Thi UnivGnity .of Alaska C NET GENERATION SV TYPES OF FUEL (Eased on .Net Generation 1980) Rural Electric Cooperativoes 58% Municipal Systtrns 27% ' . Urliversity of ...J Afuka Administration -Eklutna B GENERATING FACILITIES (Based on Namepla~e Genera~h\l Capacit'/1980} Dinll (60.6 MW -6%) Combined Cycle Combustion Turbine (139 MW-14%) A-v-n•rative Cycle Combustion 'Turbine ~~t111MW- Simple Cycle Combustion Turb~n• (520 MW-55%) 12%l 0 RELATIVE MIX OF ELECTRICAL GENERATiNG TECHNOLOGY~· RAILBELT UTILiTIES --1980 FIGURE 18,1(1. t .. t 10,000• I I 9,000 I ~·..,_~~ s.ooo I ' "'~· I :' 7,000· t 6,000· li I I 4,000 I 3,000 I -.,. 1992 .. / Entfgy Oe!lverits From Susitna ~/ - 1•--------Watana Alone -----~--"14------Wauna And Devil Canyon------ I I 1 ' I I I ., I ' ' • I 1995 2000 2005 2010 v ...... FIGURE 1lt17 _ ENERGY DEMAND AND DEUVERIES FROM SUSITNA r~~~r~ 1 . CL 9 2 9' 250 200 i ,.JIC ~ :i t; 160 n u ~ .., c w 100 50 0 N§ lEGEND ·-~---· ••••••••• • - Area Under This line It Annual Cost of Best Thermal Option r. (Including Investment Costs) I WATANA ONLY IN 19941 . Energy Coat of Otat Thermal 0 ptlon Energy Coat of Sualtna Option Operating Coats of 'fttermal Pla nt fn U111 in 1993 btended to 1994 Shaded Ana Represtntc Plant Operating In 1992 ililpllleed by Watene ~I H:~ -!~:~.~~mt·········· .•........• ., •..... ; ... ~ .....•. , ... J ... :~~Undcr This Line Is Annual Cost of Susitna Option : Are1 Under This lines is Annual :;;n~~11:: ::::: ::::::::: ::::; .:::;: ;;;;; .•:;:; ·~ :::::: •:•: :•:~: 1,000 .----w-·.-Operaiifig Cost of Extsting ~apacity 1993/4 (Avoided Costs of Fuel and O&M Only) Ar~a Represents Annt~al Operating Costs from Existing Generating Plant .-Common to Both Susitna and , Thermal Options '" ..... , .. , ... -!I!IJ:::: .. :::::~~f:lfii!'I:~::~::::::;~~~;W::;t¥.1.~; ~:;:_:: :;~ ~:::;~§:;] ~ '::: ::·· ~ II ::::: Medium Growth System Energy :::~::: .//f.h, ,/f./.;. ::::::~:~:~imu ~·-"~:~ , · • 3.000 4.000 5,000 Annual En~'DY Output GWh FIGURE 18.18-ENERGY PRICtNG COMPARISONS-1994 - -- Rw. 1 380 360 340 320 300 -.z: ~ ~ 280 ·-~ 260. -M u u ·-.. a. 240 , c ., :: M 220 8 >-Cit .. u 200 c: w 180 160 140 120 100 ~-~-----------~~~--------- SYSTEM COST SAVINGS PRODUCED BY SUSITNA COMPARED WITH BEST THERMAL OPTION IN MILLS· PER UNIT OF SUSITNA OUTPUT IN CURRENT DOLLARS ~' # I 'I ·' I •• COST SAVINGS FROM SUSITNA INCRr!ASING # OVER WHOLE LIFE OF PROJECT •• •• ,. ., .... I !ncreaing Thermal Fuel / ~ Costs Avoided s ... . . ~ . •• ~~·· ·---·-····# ~­----:;' • /-Avolds Cost of 1 Further 200 MW Coal Find Generating Unit • • • • .,_.L. Avoids Cost of 2 x 2f.l1' ~i!W CaJI Fired Gener~ting Units Watana on Stream in 1993 Devil Canyon on Stra1m in 2002 9~4----5-----6--~-7----8--~9---~--00---0-t----~-~y-,-.~,~~4--~0~5~~06~~0~7--~0~8---0~9~~2~0~10~~,~.1---.1~2~FI_G_U~t:~E-1-8-.1-9 [i~~mJ I - 400 I"" 300 200 100 0 r-l ' t-l I I Area Under This LIM Is Annual Cost of Bast Thermal Option H \ • (lncluC::~!J lnvestmen.t Cost) ' r-l: '------------. t I r-~~~i~: I ~~~~~l:.l ~ ·····~·· - I WATANA & DEVIL CANYON ~~:~03 J LEGEND •••• . Energy Cost af Best The.rmaJ CbJ:Ith>n 11::111111 - ~ ~ Energy_ Cost of Susitna Optio.ru Operating Com of. Thermal' P.taottJn Usa in 199~ Ext1nded to 1994 Shaded Portion Rtprtstnts Plard(lQ~rating tn 2001 Obplac:ed by Susitna r-~:i:l:i:l:"'·=·:·l!l!l!:·:·:~·:·:·"'=·:·:!l!l!·:·:·.. I I ~mi11fm1l~lllllll~ ·--[111.. . r Are.• Under Thislinels Annual Cost of Sus~tn~ Option r-:;:;:::::;:;:;:;:;:;:;::::~::::;:; · 1 ~ (Including Investment Costs of Watana and Devd Canyon) 1~4:~~~~~~i~~ ~llllllltlllllllllllllllllllllltlttlllllltlllltlllllllllllallllllltB\11111ililtiJIIIIIIIIIIItii11111111111UIIIIDIIIIIIIIIIiiiiiiiiiJ :::::::::::~:::::::::::::::;:;:::: I . ; • r-~:;:::::::::::::::::::::;:::::::: .. : : ~=l=~=~:l:l:l:l:l:l:l:l:l:i:l:l:l: • : : ::::::::;:::~::::::::::::::::::::: l••••••••••alliilmt••-<•••~r.t••u•m•••••••••••a..s "l~IIRIIII.III ! -= ,...·.······················································································ I E • I ' I I ' I ' I %.\;W}fi1\®~l11Wlt€i.J.@!f§i.~{m!~ Energy 0ll1put Wallna --j Watana and Devil Canyon-~ :rlllllllllltl l i ~~=t~:~~~~~4E~~1v r-i~1~~f~~~;;~~~l~l11i1~~~11i~l~i~I~l~i~I~l;I~~~~;1~1~1~~;~lm;!~l~~~ll~~i~i!lll~I~lill~~~lllli~::;~:~~;~~~ i ~~4. N~~;~!~~~l~ijlji~l~f~lj~j1jl1ll~~l~;jijl~lj!jljljljlilillfi~jijljljil1lll1ll~lli~lllli~~i:::li~~ 1 ; i 1 1 ~Other Hydro 1,000 2,000 3,000 4,000 s.ooo 6,000 [iU Ann'Ual EHergy Output GWh ~PO[~ FIGURE 18.20-ENEr1GV PRICiNG COMPARISONS HUU[d -•~J· J.J· •••• -- Rev, 1 r---------------------~------------------·-------------------------------------------------------------------------------------------------=----------------------------------~------------. :~o 360 340 320 300 -.c ~ 280 -~ ._, ~ 260 -ait 3 ·a: c. 240 "0 c • :1 " 220 0 u > r ' GJ 200 c w 180 160 140 120 100 94 ~STATE APPROPR!ATION.· SCENA.RIO L___ 100% DEBT FINANCING Susitna Mill Rata Cost With 7% hdlatlon" 10% I ntt'J'elt COST SAVINGS 5 6 '7 8 2000 01 02 03· 04 Yean COST SAVINGS GROWING OVER WHOLI? OF SUSITt"A Ll FE Mill Rate Cost Best Thermal Option 7% Inflation, 10% Interest 05 06 07 Mill Rate Cost Beat Thermal Option 0% Inflation. 3% Interest 08 09 2010 11 12 13 FIGURE 18:21-ENERGY COST COMPARiSON 100%'DEBT FINANCiNG 0 AND 7% INFLATION '---------------------------------------------------------------------------------------------------------~--------------------------------------~------~--------------~------.-m•------------a. .. .-.......... _._. .. _.._._._._._ .. nJ -.c ~ .:r: -s :E -M CP .!:! .. 0.. "C c: ftJ "' ... 1111 0 0 > 0) .. Q) c: w "· 380 360 340 32Q 300 280 260 240 ~20 200 180 160 140 120 -- SOOA, STATE APPROPRIATION SCENARIO ($2.5 BILLION) 7% INFLATION AND 10% INTEREST Watana Completed with $2.1 billion ($1.0 bn 1982) of GO Bonds 1991 -93; Cover of 1.25 at 89 Mills/kWh and Allows Rtvenue Bond ReflnL,cing ln.1994 Susitna wholesale energy price {alJb.lts anorgy increase= to 2009 and ris~ slowly thereafter .Devil Canyon Completed with $8.0 billion ($2.4 bn 1982) of Revenue Bonds1994-2002 100 Ll/._,..~-':__;s~u:si:tn:a~W~h:_:o:te:••;Je:.E;n;e;.:rgy~P-rliB~·--~-~-~ .. 94 5 6 8 9 2000 01 02 03 .04 05. 06 07 08 09 2010 11 12 13 Yean FIGURE 18.22 _,ENERGY COST COMPARISON 50% STATE APPROPRIATION SCENARIO Rev •. 1 --L!!l -.. -· -[8J! ------- 380 360 ~ 340 320 300 -.c a: u. 280 -"" -·-:E -260 .... 4) u ·.:: Q.. "C 240 c tV Ill .... ... 0 220 u >-en .. 'CU c 200 w 'i80 160. 140 120 100 - MINIMUM STATE APPROPRIATION SCENARIO ($2.0 BILLION) 7% INFLATION AND 100,.{, INTEREST 5 Mill Rate Colt Btst Thermal Option Susitna Pricing Reatdcted to Maximum of Best Thermal Cost Susitna Wholesale EnergY Price Watana Complated with $3.4 billion ($1.8 bn 1982) of GO Bonds 1990-93; Cover of 1.25 at 142 Mills/kWh and Allows Revenue Bond Refinil~cing 1994 6 7 8 9 2000 01 02 03 04 05 Years COST SAVINGS GROWING OVER. WHOLE OF SUSITNA LIFE Susitna wholesale energy price:hlls as energy increases to 2010 and t~ ~lowly thereaftElr Devil Canyon Completed with $7.6 billion ($2.2 bn 191~2) of Revenue Bonds1994-2002 06 07 08 09 2010 11 12 13 FIGURE 18.23 ·-ENERGY COST COMPARISON MINIMUM STATE APPROPRIATION - - Rev. 1 - 380 360 340 320 300 -.c ~ 280 '"iit -~ 260 .,. _, u ·;:: 0.. 240 "'D c: 11:1 :s C). 220 o· ~ &I 200 c w 180 160 140 120 100 94 ------ $1.6 BILLION STATE APPROPRIATION SCENARIO 7% INFLATION AND 10% INTEREST Mill Rate Cost · Bast Thermal Option ' --' .. ---- -· ,- I • I I • I COST SAVINGS GROW.ING OV.ER I WHOLE OF SUSITNA LIFE ## ,# ~-~ .tit' ~,..----~~~--------.. ~~. ~ ....... « \ ........ ---··· __ .... ·-. * .~. •• •* I I '---------1 • I I Watana Completed with $4.6 billion.($2.4 bn 1982)Qf GO Bonds 1990 -93. Inadequate Covef for Revenue Bond Refinancing Until 1995 7 8 9 2000 01 02 03 Susitna Price Tracks Coat of Best Thermal Option Until Complete Conversion of GO Into Revenue Bonds pevil Canyon Complet~ with $7.5 billion ($2.3 bn in 1982) of GO Bends Converting to Revenue Bonds 2002 -2004 05 06 07 08 09 2010 11 1.2 13 6 Year~· FIGURE 18.24 -ENERGY COST COMPARISON WITH FINANCING flESTt:UCTED 94/95 AND 03/04 I I I I I I I· I I ·I I I I I I I I I <. SPECIFIC FUSfS 1: RISK OF· GO BOND FINANCING OVERRUN 0 Probability .5 .2 Foracast ($ bn) .1 ~ 0.5 . 1.0 Probability of GO Bond Requirement Exceeding $1.5.bn • .15 1.5 GO Bond Financing Requirement Probability of Exceeding $2 bn 2.0 F=GURE 18.25 -GO BOND REQUIREMENTS IN 1982 DOLLARS BILLION SPECIFIC FINANCING RISK II: EARLY YEAR NONVIAB!LITY Probability .5 .4 \ .3 0 10 20 30 Probability of Watana Unit Costs Exceeding 45% of Best Thermal Option 60 70 Watana Unit Cost as% Best Thermal FIGURE 18.26-WATANA UNIT COSTS AS PERCENT OF BEST THERMAL OPTION IN 1996 [ AGGREGATE RISK I .Probability of Cumulative Net Operating Earnings Falling Below $1 bn 0 0.5 Probability .5 .4 .3 .2 Forecast ~ .1 $1.35 bn 1.0 1.5 Cumulative Net Operatinq Earnings $ bn 2.0 2.5 FIGURE 18.27 -CUMULATIVE NET OPERATING EARNINGS BY .2001