Loading...
HomeMy WebLinkAboutUnAlaska Geothermal Feasibility Study Draft Rep Vol l 1-1087Alaska Energy Authority LIBRARY COPY UNALASKA GEOTHERMAL FEASIBILITY STUDY DRAFT REPORT VOLUME I January 30, 1987 Dames & Moore UNA O25 Vol. UNALASKA GEOTHERMAL FEASIBILITY STUDY DRAFT REPORT VOLUME I January 30, 1987 Job No. 12023-026-02 UNALASKA GEOTHERMAL FEASIBILITY STUDY DRAFT Prepared For ALASKA POWER AUTHORITY By Dames & Moore in association with SAI Engineers, Inc. and Mesquite Group, Inc. January 30, 1987 Contract No. 2800020 VOLUME I: 1.0 EXECUTIVE SUMMARY . . 1 2 3 eee . 14. 1.5 2.0 ENGINEERING CONSIDERATIONS 3.0 ENVIRONMENTAL CONSIDERATIONS .. 3.1 3.2 3.3 ww ee uw Www ee ona 3.9 MLF6/C1 TABLE OF CONTENTS MAIN REPORT ee ee ew ww et BACKGROUND . 2... 2 ew ww we ee ene STUDY OVERVIEW. ........ SUMMARY OF SIGNIFICANT FINDINGS . . Engineering Findings. .... Environmental Findings ....... Economic Findings ........ RECOMMENDATIONS ..... 2.2 ee ee ORGANIZATION OF THE FEASIBILITY STUDY ee ee Cr GENERAL PERMIT REQUIREMENTS .... 3.1.1 Federal Permits ....... 3.1.2 State Permits ........ 3.1.3 Permitting Status and Timing .. CONSTRUCTION ACCESS FACILITY PERMITS . . 3.2.1 Driftwood Bay Airstrip ..... 3.2.2 Driftwood Bay to Fox Canyon Road OPERATIONS ACCESS FACILITY PERMITS . .. 3.3.1 Makushin Valley Road Alternative 3.3.2 Broad Bay Boat Dock ..... TRANSMISSION LINE ....... BRINE DISPOSAL. ........ 2 eee 3.5.1 Surface Disposal ..... 3.5.2 Pipeline to Driftwood Bay... 3.5.3 Injection. ...... POWERPLANT SITE IMPACTS .. OPERATIONAL IMPACTS .... ALTERNATE USES OF WASTE HEAT... . 3.8.1 Driftwood Bay........ 3.8.2 Broad Bay ........ 3.8.3 Other Considerations .... AIR QUALITY ........ REPORT Page ~ 1 ~ ee 1 NDUMNWWHHE w nN 1 ' ~ 1 fe le bo 1 ' WOOO WMDANINYNAAHEFFWWWNHNH HL we. ae eee eo? x11 3-11 3-12 oft 4.0 ECONO 4.1 4.2 + oe Fw 4.5 5.0 CONCE 6.0 CONCLUSIONS AND RECOMMENDATIONS . DADA . Fwne VOLUME II: APPENDIX A APPENDIX B APPENDIX C APPENDIX D VOLUME III APPENDIX E MLF6/C1.1 TABLE OF CONTENTS (continued ) MIC ANALYSIS .... ee © © © ew ew ew APPROACH AND ASSUMPTIONS ....... 4.1.1 Economic Assumptions ..... 4.1.2 Generation Data and Assumptions 4.1.3 Generation Data and Assumptions THE ELFIN GENERATION MODEL ...... 4.2.1 General Overview ....... 4.2.2 Load Data. ........2.4. 4.2.3 Dispatch Order. ........ 4.2.4 ELFIN Output ......... THE ECONOMIC MODEL. .......2.2-. RESULTS AND SENSITIVITY ANALYSIS... 4.4.1 Diesel Price Sensitivity ... 4.4.2 Load Growth Sensitivity .... 4.4.3 Discount Rate Sensitivity ... 4.4.4 Project Life Sensitivity ... CONCLUSIONS AND RECOMMENDATIONS .. PTUAL DESIGN... ee © © ew ew ew ee TECHNICAL CONSIDERATIONS . . ENVIRONMENTAL AND PERMITTING CONSIDERATIONS ECONOMIC CONSIDERATIONS ....... RECOMMENDATIONS ...... APPENDICES A, B, C, D GEOTECHNICAL FIELD EXPLORATION AND LABORATORY TESTING Diesel. . Geothermal GEOTHERMAL DRILLING PROGRAM AND COST ANALYSIS AIR QUALITY ANALYSIS ECONOMIC ANALYSIS DETAILS : APPENDIX E CONSTRUCTION COST DETAILS Page + 1 _ 1 Se Siti iin a RrPODAYNNUFP WHE 1.0 EXECUTIVE SUMMARY 1.1 BACKGROUND Utilization of naturally occurring underground hot water and steam has proven to be a low cost source of electrical power. Worldwide, more than 3,000 MW (megawatts) of geothermal generating capacity have been installed. This capacity is growing at an 18 percent per year rate (Republic, 1985). The main mission of the Alaska Power Authority is to increase the use of local energy sources and to reduce reliance on imported energy sources. In keeping with this mission the Power Authority embarked on a program to identify poten- tial sources of geothermal power. This program resulted in identification of the Makushin Volcano, Unalaska Island as a possible commercial source of power for the nearby community of Dutch Harbor/Unalaska. 5 weewity de phe Aisle ayqrup eth S te He Da slen’ Pers Aactees In 1980 ,the Power Authority—initiated a $5 million study to explore for a usable geothermal resource on the Makushin Volcano. This study was conducted in enibele a three phases by Republic Geothermal, Inc. and Dames & Moore. Phase I produced promising indications of geothermal resource potential (Republic, 1983). A Phase II drilling program resulted in the discovery of a significant geothermal reservoir. Figure 1-1, from Republic, 1956, illustrates the location of geo- thermal manifestations in the Makushin area. An electrical generation analysis, also part of Phase II, indicated that the resource would likely support commer- cial scale electrical power production (Republic, 1984). Phase III activities were directed toward obtaining further data on the areal extent and technical characteristics of the reservoir discovered in Phase II (Republic, 1985a). The results of the Unalaska geothermal exploration program were summarized in an Executive Final Report (Republic, 1985b). The results of the exploration program established the existence of a geothermal resource which is technically capable of producing commercial scale power for the City of Unalaska/Dutch Harbor. The environmental and economic feasibikity had yet to be determined. Thus in 1986 the Power Authority began a simultaneous three part program, of which the present report is part. This three part program consists of an environmental study by Alaska State Agencies (DNR, 1986a,b,c; ADF&G, 1986), an electrical load forecast (R.W. Beck, 1987) and MLF6/C5 1-1 UNALASKA Geologic Map and Geothermal Manifestations GEOTHERMAL MANIFESTATIONS Ig] Fumarolc Ares 1 Hot Spring Group (8 Warm Ground GEOLOGIC UNITS Quaternary Alluvmuem & Glacier Deposits Makushin Volcancs ilq Tertiary Phutonacs Unalaska Formation FIGURE 1-1 STATUTE OR REGULATION Clean Water Act Section 404 Rivers and Harbors Act of 1899 (Section 10) Rivers and Harbors Act of 1899 (Section 9) Fish & Wildlife Coordination Act Endangered Species Act of 1973, 50 CFR 17 Endangered Plant Permit 50 CFR 17.62 Endangered Wildlife Permit 50 CFR 17.22 Marine Mammals Protection Act, Endangered Species Act, Fish & Wildlife Coordination Act of 1934 Federal Water Pollution Control Act, 40 CFR 125 Underwater Injection Control Clean Air Act, CFR 515-5] Clean Air Act, Section 160-169 Marine Protection, Research and Sanctuaries Act of 1972 Gravel Extraction Contracts (AS 38.05.110; 11 AAC 76) Tide Lands Permits (AS 38.05.330; 11 AAC 62) Anadromous Fish Protection Permit (AS 16.05.870; 5 AAC 95.010) Fishways for Obstruction to Fish P. (AS 16.05. Wastewater Disposal Permit (AS 46.03.020-100 100 18 AAC 15, 60) Solid Waste Disposal Permit (AS 46.03.020-100; 18 AAC 15, 60) Coastal Zone Consistency Determination MLF6/CT6 TABLE 1-1 PERMITS, STATUTES AND REGULATIONS AFFECTING THE DEVELOPMENT OF THE UNALASKA GEOTHERMAL PROJECT AGENCY U.S. Army Corps of Engineers U.S. Army Corps of Engineers U.S. Coast Guard U.S. Fish & Wildlife Service U.S. Fish & Wildlife Service National Marine Fisheries Service U.S. Environmental Protection Agency U.S. Environmental Protection Agency U.S. Environmental Protection Agency U.S. Environmental Protection Agency U.S. Environmental Protection Agency State of Alaska, Dept. of Natural Resources DFLWM State of Alaska, Dept. of Natural Resources DFLWM State Dept. of Alaska, of Fish & Game State Dept. of Alaska, of Fish & Game State of Alaska, Dept. of Environmental Conservation State of Alaska, Dept. of Environmental Conservation State of Alaska, Div. of Governmental Coordination DESCRIPTION Permits for discharge of dredged or filled material into navigable waters or wetlands (includes wet tundra) Permits and stipulations for any structures or work including dredging and filling, in navigable waters and adjacent wetlands Permits for construction of bridges and causeways in navigable waters Review proposed permits to be issued by Corps of Engineers or Coast Guard for any work or structures in navigable waters or adjacent wetlands Determination of threatened or endangered species presence; Stipulations on dis- turbance level near sensi- tive areas if endangered species are present Review proposed plans and permits for activities affecting nearshore and offshore marine resources NPDES permits for discharge into navigable waters Permits for various cl es of materials to be injected underground via wells - applicability to geothermal injection is questionable. PSD permits and standards for new source air quality Permits for the emissions of air pollutants and set standards Permits for ocean dumping Contracts for sale of gravel on state lands Permits for lease of state-owned tidelands Permit for activities affecting anadromous fish (salmon & arctic char) Permits that guarantee fish passage in all streams in the state Permits for wastewater discharge including geothermal waste injection Permits for disposal of all solid waste Review activities in the coastal zone for consistency with Alaska Coastal Management Program TIME FRAME 15 days after submittal there is a 30-day period for public comment. If no objections, permit issued within 90 days 15 days after submittal there is a 30-day period for public comment. If no objections, permit issues within 90 days 90 days if no objection Variable Variable Variable Must apply 180 days prior to discharge, 30 days for public comment 30 days for response from EPA. 30 days for public comment, 1 year maximum for final determination 30 days for response from EPA. 30 days for public comment, 1 year maximum for final determination 30 - 90 days Apply one season prior to construction 60 days for final action 30 days for final action 30 days for final action 60 days for final action 60 days for final action 50 days for final action, coordinated with other state permit applications 3. Under the base case assumptions the 5 and 7 MW geothermal systems offer about the same savings over the all-diesel case. Later expansion from 7 to 9.5 MW in 2003 is economic. Assuming a high diesel price trend or a high load growth case or both further increases the cost savings from geothermal development. Only in the case of a low diesel price trend and a base case load growth is geothermal development more costly than continued reliance on diesel generation. If a high load growth and low diesel prices are assumed, systems with 5 or 7 MW of geothermal capacity are roughly equal in cost to the all-diesel scenario. Increasing the present value discount rate to 4.5 percent (placing less value on future as opposed to present costs) reduces the magnitude of the cost savings from geothermal development. However the rankings of the cost for each scenario do not change. If the life of the geothermal development could be extended to 2025, even greater cost savings would result from geothermal development. Some facil- ities replacement would be required under this scenario. Table 1-2 summarizes the results of the economic analysis. RECOMMENDATIONS A financial feasibility analysis should be undertaken as soon as is prac- tical. Load growth in Unalaska is expected to increase at a rapid rate over the next five years. If a geothermal development is to be economi- cally feasible it must utilize the resource before investment in diesel capacity is committed to serve the burgeoning demand. Final siting and alignment for the Makushin Valley road, transmission lines and dock facilities should be determined in the field with input from ADF&G and other interested agencies. As soon as a decision to develop is reached, or possibly before, if devel- opment is considered likely, it would be prudent to begin the environmental MLF6/C5 1-6 TABLE 1-2 ECONOMIC COMPARISON OF ALTERNATIVE GENERATION SCENARIOS FOR UNALASKA DIESEL DIESEL DIESEL ALL- + 5 MW + 7 MW + 9.5 MW SCENARIO DIESEL GEOTHERMAL GEOTHERMAL GEOTHERMAL 1986 Present Worth of Scenario 1988-2016 (Million 1986$) BASE CASE LOAD DIESEL PRICE: MEDIUM 77.6 74.5 73.7 74.9 LOW 63.6 68.8 69.4 70.6 HIGH 91.7 80.3 77.8 79.3 HIGH CASE LOAD DIESEL PRICE: MEDIUM 99.4 91.5 86.9 90.6 LOW 81.9 83.3 81.5 85.5 HIGH 117.0 99.8 92.2 95.7 MLF7/CT1 approval process. This will avoid a situation in which the narrow develop- ment season is missed due to permitting delays. 4. An additional incentive to prompt action in geothermal development is the presently depressed level of construction and drilling costs. If a deci- sion to develop is reached during the current depressed drilling market it may be possible to obtain a drill rig at a fraction of its book value. This rig could be committed to the project offering considerable saving over day-rate rental. 1.5 ORGANIZATION OF THE FEASIBILITY STUDY REPORT The feasibility study report is bound in three volumes. Volume I contains the main report, including sections on Engineering Considerations (2.0), Environmental and Permitting Considerations (3.0), Economic Analysis (4.0), Conceptual Design (5.0) and Recommendations and Conclusions (6.0). A biblio- graphy appears at the end of Volume I. Volume II contains technical appendices on Geotechnical Field Exploration and Laboratory Testing (Appendix A), Geothermal Drilling Program and Cost Analysis (Appendix B), Air Quality Analysis (Appendix C), and Economic Analysis Details (Appendix D). Volume III contains a technical appendix on construction cost details (Appendix E). MLF6/C5 1-7 2.0 ENGINEERING CONSIDERATIONS The project considered under this feasibility study employs a geothermal hot water resource which will require the development of three major systems: geo- thermal resource extraction, power generation, and transmission. The resource extraction system consists of production wells of pipelines for transportation and collection of geothermal fluids, and of effluent injection wells. The generating system includes primary steam turbine driven generating units, and secondary use of the remaining (lower enthalpy) geothermal fluids to power binary cycle generating units. The third major system consists of transmission facilities. The power produced by either or both generation systems will be synchronized to a station bus, stepped up to transmission voltage levels. Then the power will be transmitted from the plant to the existing primary system by means of overhead, buried and submarine transmission lines linking the plant to the Unalaska distribution grid. 2.1 GEOTECHNICAL CONSIDERATIONS 2.1.1 Well Site and Plant Site Present plans are to develop the geothermal resources on Unalaska Island with a production well near the existing test well on a gently sloping bench situated on the eastern flank of the Makushin Volcano. The plant facilities would be located on two adjacent benches located on opposite sides of Fox Canyon. The two benches are situated at roughly the same elevation with gentle slopes down to the south or southeast. The perimeters of both benches to the North, east and south are steeply sloping down into ravines and gullies. The steep slopes are marked by fresh exposures of the nearly horizontal layered site subsurface profile. The freshness of these exposures is indicative of a con- tinuing failure of the gully slopes. The western margins of both benches are at the base of a hillside with relatively steep slopes. At the southwest corner of the well site, the majority of the area is occupied by a landslide deposit with a hummocky surface. Based on observations of the exposures on the steep slopes at the perime- ters of the benches and information from the Department of Natutfal Resources report (1986), the subsurface conditions of the proposed sites generally consist MLF7/G 2-1 of a surface layer of volcanic ash (tephra). The tephra consists mainly of a silty fine sand with some organic material and a trace of gravels. The layer ranges in depth from approximately 3 to 13 feet at the existing well site and was generally loose in consistency. Beneath the tephra, an ash flow tuff depos- it is encountered and consists of a medium dense to dense, silty gravelly sand. It is estimated that the ash flow tuff deposit is approximately 25 to 80 feet thick. Below the ash flow tuff, it is believed there is a relatively thick bouldery till deposit which is underlain by bedrock. To supplement the ADNR (1986) geotechnical investigation of the existing well bench, Dames & Moore obtained samples during the September 5, 1986, field investigation. The field program carried out was minimal and is described in Appendix A. The samples gathered during this investigation were tested in Dames & Moore's Anchorage laboratory. The test program included classification type testing for all the samples and shear strength determinations on the relatively undisturbed sample. A description of laboratory testing procedures and the test results are presented in Appendix A. The classification tests on the four samples obtained in the surface tephra deposit indicate a silty fine sand with some organic material and relatively high moisture contents. The strength tests performed on the silty fine sand indicate an angle of internal friction on the order of 25 degrees with a cohe- sion intercept of approximately 200 pounds per square foot. Foundation Discussion/Recommendations It should be emphasized that the discussions and recommendations, which are based on relatively little subsurface information, are intended to be used only as a guide for preliminary foundation design. A comprehensive subsurface investigation with further analysis is required prior to final foundation design. The surface silty fine sand (tephra) deposit at both the well and the plant site may be prone to substantial consolidation under additional loads imposed by the planned structures. In addition, due to the grain size distribution of the deposit, the partial saturation of the deposit and the high seismic hazard risk of the area, the tephra deposit at the site may be susceptible to liquefaction. ~ MLF7/G 2-2 Due to the aforementioned factors, we anticipate that a deep foundation system that extends through the tephra and is founded in the more competent underlying ash flow tuff deposit would be the most feasible for the main structures of the Proposed power plant facility and the proposed production well development. Driven displacement piles using locally available timber or steel pipe piles are recommended. The allowable bearing capacity for this type of foundation system would probably be in the range of 20 to 40 kips per square foot (ksf). Based on available information, it is estimated that structures supported on pile foundations would experience negligible settlement. Drilled and cast in place piles could also be used for this project. However, some difficulties may be encountered keeping the drilled excavations open and would require temporary or permanent casing. The allowable bearing capacity for this type of foundation system would be less than driven pile system and would range from approximately 10 to 30 ksf. Shallow foundations may be possible for light weight support structures. We anticipate that the allowable bearing pressure would range from 2 to 3 ksf with a minimum embedment of at least 3 feet. In addition, the structures would have to be designed to tolerate total settlements on the order of 2 to 4 inches and to mitigate the possibility of liquefaction. Further, it is recommended that care be used in the layout of the facilities and well appurtenances. It is advisable to avoid placing any structures in the area of the landslide deposit area, due to the very loose, unstable subsurface soils. In addition, the buildings should be kept away from the edge of the steep slopes at the perimeter of the benches, due to possible weakening of the bearing soils by slope erosion. It would be prudent to investigate the movement of the slopes due to undercutting by the rivers and streams, and to provide slope protection if found necessary. Alternately, a monitoring scheme could be developed to document the progression of slope migration back into the sites. Suitable borrow sources for on-site road construction and structural fill (if required) are numerous along the construction access route back to Driftwood Bay. Material sites 8, 9, 10 and 11 as identified by the ADNR-could supply suitable material. An alternate to pile foundations under major facilities MLF7/G 2-3 would be a structural fill (mat) foundation with the surficial silty soils removed and replaced. This alternative would also minimize the adverse effects of seismically loading the surficial silts. After searching Staging Area 2 on the existing road, operational access will be via the new road to the plant site. 2.2 GEOTHERMAL RESOURCE DEVELOPMENT CONSIDERATIONS Geologic drilling and testing investigations to date in the Makushin Area have amply demonstrated the existence of geothermal resources capable of pro- ducing geothermal fluids at sufficient rates to generate at least 10,000 kW of power. Resource longevity is less certain because of practical limitations to well flow test time, but it probably far exceeds thirty years, the considered useful life of the power generation equipment. To bring this resource from the reservior to the power generation facili- ties, wells need to be drilled to a depth of about 2,000 feet, in sufficient numbers and with adequate well bore diameter to provide the required flow on a long term basis. The concentrations of chemical constituents in the resource fluids par- ticularly arsenic do not permit the disposal of the plant effluents in streams discharging to the Makushin River. The minimum stream flows do not provide suf- ficient dilution of objectionable constituents to levels permitted by environ- mental agencies. Therefore, all of the resource produced must be injected into a disposal well after heat energy for power production is extracted. (See discussion in Section 3.5.) Steam from the turbines is condensed in an air cooled condenser after which it is combined with the liquid effluent from the plant for injection in an injection well. The injection well will be of the same diameter as the produc- tion wells. It will be located across a major fault thought to be sealing the resource and at a sufficient distance from the production well so as to preclude the possibility of affecting the temperature of the resource. Injection tests have not been conducted to date at this site and will be required for confir- mation of feasibility. An interval suitable for injection between 300 and 600 MLF7/G 2-4 feet deep is anticipated based on resistivity survey interpretations for the area and two nearby core holes. The geothermal resource will be conveyed from the production wells to the power plant site by means of insulated pipes of a diameter consistent with flows and allowable pressure drops. Likewise, the fluids for disposal from the power plant will be collected and conveyed into the injection well. It is expected that a single 13-3/8" production well can deliver sufficient fluid from the Unalaska geothermal resource to produce approximately 5 MW net in a single flash steam turbine driven power plant and approximately 7 MW net with the addition of binary units using heat from the unflashed hot water. The selected flash stem pressure determines the fraction of the geothermal resource which flashes to steam for use in the steam turbine. The higher the flash pressure the lower the fraction of steam produced. On the other hand, lower steam flash pressures result in larger steam volumes per pound and less energy available from each pound of steam flashed. The net result is that lower flash pressures require larger turbines. Conversely higher flash pressures require the drilling of more wells and the production and handling of more total pounds of geothermal resources. 2.3 POWER PLANT AND TRANSMISSION DESIGN CONSIDERATIONS 2.3.1 Well Considerations To generate 5,000 kW from a flash steam cycle or 7,000 kW of power from a binary or hybrid cycle, 1,100,000 lbs of fluids will be required at a turbine inlet pressure of 60 psia (well head pressure 80 psia). Three different well diameters were considered in an effort to determine which well size would result in the lowest total cost for wells. The following tabulation shows the number of wells of each size required to produce the needed resource. (See Appendix A): Well Diameter Flow/Well Wells Required Total Flow 7 130,000 1bs 8 1,040,000 lbs 9 5/8" 350,000 lbs 3 1,050,000 lbs 13 3/8" 1,100,000 1bs 1 1,100,000 lbs MLF7/G 2-5 The seven-inch (7") and nine inch (9-5/8") diameter wells were considered in an effort to optimize the well drilling cost by using a small drilling rig which would represent lower mobilization costs to the site. As shown in Appendix A, two 7" wells cost about the same as one 13-3/8" well and would have inadequate capacity (i.e., at least 8 wells required). Three 9-5/8" wells would have marginally enough capacity and would cost considerably more than one 13-3/8", Large diameter wells had been recommended in previous studies. Each production well with a 13 3/8" diameter bore will provide the fluids required for a power plant using steam turbines with a total net capacity of approximately 5,000 kW. Drilling one additional production well is recommended to assure adequate redundancy of supply and avoid costly remobilization in the event of a problem. In the case of the 13 3/8" diameter wells recommended, one such well will, in effect, provide 100% spare resource production capacity. The optimum steam turbine inlet pressure as discussed in paragraph 2.2 for this resource which has a bottom hole temperature of 382°F is approximately 60 psia, and therefore, is the selected pressure for this feasibility study. Approximately 7 MW total net electrical power can be produced by a hybrid plant using geothermal resources supplied by a large diameter single production well. Ideally, the power plant site should be located as close as possible to the production well(s), but in any event along the pipeline route between the pro- duction well and the injection well. The site should be physically capable of supporting all interrelated system components and should be readily accessible for construction and plant operation and maintenance. Considerations should also be given to the disposal of effluents. The area immediately adjacent to the existing ST-l1 Well-Test-Site (W-T-S) is sufficiently large to develop the drilling pads for multiple production wells and a possible plant site to accommodate 7,000 kW of generation capacity. Environmental requirements dictate that the plant effluent cannot be disposed in creeks adjacent to the W-T-S and reservoir management and geologic considerations indicate that the plant effluent cannot or should not be injected in the immediate vicinity of the resource. Therefore, the effluent- will have to be piped to an injection well site, preferably north of the W-T-S (across Fox Canyon), at Site "E" (Figure 2-1). MLF7/G 2-6 Construction access to the north side of Fox Canyon is much easier than access to the W-T-S on the south side. The south side is accessible only by helicopter whereas the north side can be reached by road from Driftwood Bay or Broad Bay (See Section 2-4). As long as the injection well must be located on the north side, it would be advantageous to locate the power plant and produc- tion well pad there as well. Using directional drilling techniques, it would be theoretically possible to locate the production well pad on the north side and reach a target near the W-T-S on the south side. However, this concept was rejected on practical grounds as the horizontal reach of such a well would equal or exceed the target depth, necessitating extreme deviation angles. Locating the power plant on the north side of the canyon therefore would involve pipelines from the well site accross Fox Canyon. After careful eval- uation and cost comparison, it was decided that the additional cost of these pipelines (one for water and one for steam) is less than the additional cost to construct the power plant on the south side. Therefore, it was concluded that the production well should be located on the south side of Fox Canyon near the W-T-S and the injection well and power plant should be located on the north side at or near Site "E" on Figure 2-1. 2.3.2 Flash Steam Plant The moderate geothermal resource temperature (382°F) at -Unalaska and the projected electrical load to be served ranging between 5,000 kW and 7,000 kW result in a situation where a flash steam plant of 5,000 kW and a binary cycle plant of 5,000 kW are nearly equal in capital cost to each other. Small binary cycle plants with a maximum capacity of 2,000 kW have lower cost per kW than a similar rated flash steam plant. However, when the plant rating exceeds 5,000 kW, the cost of the flashed steam turbine driven generating units is lower than a plant comprised of binary units only. A higher resource temperature would tilt the economics further in favor of a flash steam cycle if all other factors remain equal. The total flow cycle employing a rotary separator turbine followed by a condensing steam turbine is not cost competitive in the small size units which are applicable to the Unalaska electrical system. MLF7/G 2-7 2.3.3 Binary Cycle Plant The alternative to the flash turbine generating plant is a binary cycle plant consisting of modular generating units such as those manufactured by Ormat and possible others. Binary cycle plants require less pounds of resource per kWH produced, therefore, a single production well can supply the resource requirements of a larger capacity binary cycle plant than for a flash steam plant. Larger binary cycle units in the range of 2,500 kW would require field erection and therefore would be more costly due to the high labor costs for work performed on relatively complex system components at remote locations. The modular units would be installed in a power plant building to provide protection from the weather. The condenser, isopentane storage tanks and the isopentane unloading and transfer pump, would be located outside the building. 2.3.4 The Hybrid Cycle Plant The hybrid power plant cycle is comprised of two flash steam turbine units of similar design to be installed and operated as the plant conceptually described in the previous paragraph 2.3.2 and of the same rating and capacity. These two units would serve as the building blocks for the first 5,000 kW of plant capacity and would both be installed in the same plant building. The sub- sequent building block would be comprised of two binary units of total 2,000 kW Net capacity, of the same design and unit rating as those described in paragraph 2.3.3. They will be located in a separate power plant building for fire safety reasons. Since binary cycle units can use the heat remaining in the water not flashed for use in a steam turbine, a hybrid plant employing steam turbines plus binary cycle units permits maximum use of the resource available from each well thus minimizing well costs. Each steam turbine driven generating unit and each binary cycle unit can operate independently of the other units in the plant. MLF7/G 2-8 2.4 SITE ACCESS SELECTION The plant site, construction access roads, operator access roads and transmission line routings proposed are based on a site visit, aerial obser- vation and reports from the Engineering Geologic Technical Feasibility Study performed by the State of Alaska Department of National Resources in September 1986. 2.4.1 Construction Access The major access for drilling and construction equipment will be from Driftwood Bay via an existing road to the plant site. The existing road is in relatively good shape and will require repairs for use as indicated in the ADNR report. The repairs consist primarily of culvert placements and filling of minor washouts. A bridge will be required in Driftwood Valley. The airstrip at Driftwood is also in good condition and could be usable with minor upgrading. The new road required traverses terrain similar to the existing road and will be straight forward construction with only minor drainages crossed. Near the plant site, it may be possible to use a cut/fill configuration for road construction. Borrow sources for road upgrading and new construction have been identified by the ADNR. They contain good road building material with minimal haul distances. Material Source 1 near Driftwood Bay can be used for airstrip and road upgrading in the valley. Barge offloading at Driftwood Bay is possible in calm weather. The northeast end of the Bay (near the existing tanks) appears to be the quietest portion with slightly more gradual beaches and generally smaller sized beach deposits. The access roads are recommended for the plant site. The first access road extends from Driftwood Bay to the proposed plant, Site "E" following the route F-D-E, see Figure 2-2 and Figure 2-3. This access road would be used to transport the drilling equipment and materials, the power plant equipment, construction equipment and construction materials. The road, Section F-D, 5.6 miles long, is an existing road built during World War II and is to be used as access for construction purposes. It would MLF7/G 2-9 require substantial repair work and the construction of a bridge crossing, six culverts and some grading work. Driftwood Bay is excessively distant from Unalaska and either marine or air access during bad weather conditions would be too dangerous; therefore, this road is not considered suitable for access by plant operation personnel. For these reasons a road from Broad Bay to the plant site for operator and main- tenance personnel access is required, 2.4.2 Plant Site and Operator Access Sete and Vperator Access The operational access to the plant site will be by boat from Dutch Harbor to Broad Bay and then via a road up the Makushin Valley to Staging Area B. From Staging Area B to the plant site. (See Figure 2-4.) The beach at Broad Bay is composed of medium to fine sand and generally slopes at 15 to 20 degrees. Construction of a small dock and breakwater should not be problematic geotechnically. The dock can be supported on the near shore sands. The breakwater will also be constructed from timber piles. The lower one-third of the Makushin River Valley is characterized by marshy surface vegitation. Below the approximately 3-foot thick mat of vegetation, roots and organics is a relatively thick deposit (as this as 20 feet) of muck consisting of super saturated silts from a lagoonal/lacustrine origin and possibly intermixed with wind blown fine ash. Underlying the muck is a stiffer deposit of silt, fine sand and clay sediments which could provide a moderate bearing surface. This deposit is underlain at depth with coarser sediments and possibly glacial tills. The fine grained sediments pinch out and give way to surficial alluvial deposits at the 1/3 point up the valley. The old road through the lower one-third of the valley followed the river valley and utilized a veneer of alluvial material as road foundation. This road has been largely obliterated by bank migration and erosion and is visible in only minor segments. The road in the upper valley will be bearing for the most part on alluvial sands and gravels and road construction should be relatively conventional. Both MLF7/G 2-10 the south and north channels of the Makushin will have to be crossed with bridges. The bridges can be supported on either conventional pile foundations or spread footings bearing on alluvial soils. This is a more active portion of the valley and elevations should be set with this in mind. Good alluvial borrow sources have been identified by the ADNR in this area and sufficient quantities of good quality material should be available. The Makushin Valley access road would be used for transmission line construction, for underground cable installation and access of personnel during construction and operation of the power plant. This road would not be adequate for construction access to the plant site because it has a section of extreme high grade and sharp turns for construction equipment from the base of Makushin Valley to the proposed Staging Area "B", Also, the section built over the flood plain would be light duty, built for use by light personnel vehicles only. 2.4.3 Transmission Line Routing The transmission line will be interconnecting the geothermal power plant and the Unalaska underground primary distribution system. The line should be readily accessible for maintenance especially the overhead sections which are exposed to severe climatic conditions. The construction should require minimum construction of access roads. The overhead portion of the line should withstand wind and ice loading conditions applicable to the area. The capacity of the line should be consistent with reliability and economic parameters for electrical loads and structural requirements. The operation voltage level and Basic Insulation levels should be consistent with the load and related line losses; and existing system voltage to which it will be connected. The underground and submarine cable portion of the line should have the mechanical and insulation strengths to provide high degree of service reliabil- ity. The installation of poles for the overhead line and underground cable should be from access road built for subsequent use by plant operating and main- tenance personnel. The location of structures should not impact existing streams. MLF7/G 2-11 Two main corridors were considered. One upland route which would require large towers and a long span designated with "U" in Figure 2-3 and Figure 2-4, This corridor was rejected due to the difficulty of constructing the line but more so due to the difficulty in access for repairs. It would be exposed to high wind conditions and would require significant construction of access toads for construction even though helicopter use would be extensive. The second route, marked with "V" on Figure 2-3 and Figure 2-4 is a better solution because it would be more accessible for construction and maintenance and would permit the installation of underground cables, which have less exposure to the weather. The foundation conditions in the lower Makushin River Valley are poor, as described in the earlier section on access roads. The soft muck underlying the marsh vegitation mat are incapable of supporting loads associated with transmission line structures. Therefore, the cable will be buried adjacent to the operational access road. Placement of this cable(s) should only require a thin width trench cut through the mat. The mat will likely close naturally relatively quickly after cable placement. The route recommended would go underground from Broad Bay, from A to B in the Makushin Valley, and would parallel the proposed access road, see Figure 2-4. From point B to the power plant it would go overhead on an H frame wood pile structure as shown on Figure 2-3. The transmission line will generally follow the proposed road. From Broad Bay to Dutch Harbor the transmission will be by submarine cable, see Figure 2-5. 2.4.4 Marine Terminal Facilities At Broad Bay a dock with a small crane will be constructed at the end of the power plant access road. This dock will be protected by a solid wood piling breakwater. Operators will cross Broad Bay from Unalaska in a diesel-powered work boat (+30') dedicated to the power plant. The road will be capable of delivering small maintenance equipment, lubricants and operating supplies including isopentane. It will also be capable of transporting the all terrain vehicle, used by the operators and maintenance personnel for land transportation to and from the plant, back to the Town of Unalaska for repairs and maintenance. MLF7/G 2-12 Equipment landed at Driftwood Bay will be unloaded by cranes or over ramps carried on barges directly onto the beach or trucks and trailers on the beach. A temporary road on the beach will be constructed from steel mats laid directly on the beach. Barge unloading operations will only be conducted during calm weather so that protection from seas will not be necessary. Therefore, there will be no need for special marine facilities at Driftwood Bay. MLF7/G 2-13 3.0 ENVIRONMENTAL CONSIDERATIONS 3.1 GENERAL PERMIT REQUIREMENTS Permits and approvals potentially required for development of the Unalaska Geothermal Project are listed in Table 3-1 along with the timing of permit acquisition. Specific aspects of permitting are discussed below. 3.1.1 Federal Permits Most of Unalaska Island is federally owned and is part of the Alaska Maritime National Wildlife Refuge administered by the U.S. Fish and Wildlife Service (USFWS). However, the lands that would be potentially affected by geothermal development have been selected for ownership by the Aleut Native Corporation under the terms of the Alaska National Interest Lands Conservation Act. The process of reconveyance was initiated in 1984 but has not been com- pleted. From a permitting standpoint, it is essential for land conveyance to be completed prior to development of geothermal facilities. Federal regulations prevent the development of geothermal energy within wildlife refuges (Ziellemaker, personal communication). Contacts with the Aleut Corporation indicate that the conveyance process is proceeding with high priority placed on the lands that would be affected by the Makushin geothermal project (Cardinalli, personal communication). The Aleut Corporation is expected to receive patent to the land sometime in mid-1988 but some activities could probably occur prior to that time under an interim conveyance agreement. The U.S. Army Corps of Engineers (COE) will be heavily involved in the per- mit process because of its responsibility as permitting agency involved with dredge and fill of waters and coastal alterations. Road construction across wetlands will require COE permits as will barge docking facilities. A National Pollutant Discharge Elimination System (NPDES) permit would be required if wastewater were discharged into freshwater or marine environments. This permit, administered by the U.S. Environmental Protection Agency (EPA), can be complex and will require detailed environmental monitoring plans. In light of permitting complexity and the arsenic levels of the geothermal brine, the decision was made to inject brine back into the groundwater. This obviates the need for a NPDES permit. , MLF6/C11 3-1 STATUTE OR REGULATION Clean Water Act Section 404 Rivers and Harbors Act of 1899 (Section 10) Rivers and Harbors Act of 1899 (Section 9) Fish & Wildlife Coordination Act Endangered Species Act of 1973, 50 CFR 17 Endangered Plant Permit 50 CFR 17.62 Endangered Wildlife Permit 50 CFR 17.22 Marine Mammals Protection Act, Endangered Species Act, Fish & Wildlife Coordination Act of 1934 Federal Water Pollution Control Act, 40 CFR 125 Underwater Injection Control Clean Air Act, CFR 515-51 Clean Air Act, Section 160-169 Marine Protection, Research and Sanctuaries Act of 1972 Gravel Extraction Contracts (AS 38.05.110; 11 AAC 76) Tide Lands Permits (AS 38.05.330; 11 AAC 62) Anadromous Fish Protection Permit (AS 16.05.870; 5 AAC 95.010) Fishways for Obstruction to Fish Passage (AS 16.05.840) Wastewater Disposal Permit (AS 46.03.020-100 100 18 AAC 15, 60) Solid Waste Disposal Permit (AS 46.03.020-100; 18 AAC 15, 60) Coastal Zone Consistency Determination MLF6/CT7 TABLE 3-1 PERMITS, STATUTES AND REGULATIONS AFFECTING THE DEVELOPMENT OF THE UNALASKA GEOTHERMAL PROJECT AGENCY U.S. Army Corps of Engineers U.S. Army Corps of Engineers U.S. Coast Guard U.S. Fish & Wildlife Service U.S. Fish & Wildlife Service National Marine Fisheries Service U.S. Environmental Protection Agency U.S. Environmental Protection Agency U.S. Environmental Protection Agency U.S. Environmental Protection Agency U.S. Environmental Protection Agency State of Alaska, Dept. of Natural ources DFLWM State of Alaska, Dept. of Natural Resources DFLWM State of Alaska, Dept. of Fish & Game State of Alaska, Dept. of Fish & Game State of Alaska, Dept. of Environmental Conservation State of Alaska, Dept. of Environmental Conservation State of Alaska, Div. of Governmental Coordination DESCRIPTION Permits for discharge of dredged or filled material into navigable waters or wetlands (includes wet tundra) Permits and stipulations for any structures or work including dredging and filling, in navigable waters and adjacent wetlands Permits for construction of bridges and causeways in navigable waters Review proposed permits to be issued by Corps of Engineers or Coast Guard for any work or structures in navigable waters or adjacent wetlands Determination of threatened or endangered species presence; Stipulations on dis- turbance level near sensi- tive areas if endangered species are present Review proposed plans and permits for activities affecting nearshore and offshore marine resources NPDES permits for discharge into navigable waters Permits for various cl of materials to be injected underground via wells - applicability to geothermal injection is questionable. PSD permits and standards for new source air quality Permits for the em: ions of air pollutants and set standards Permits for ocean dumping Contracts for sale of gravel on state lands Permits for lease of state-owned tidelands Permit for activities affecting anadromous fish (salmon & arctic char) Permits that guarantee fish passage in all streams in the state Permits for wastewater discharge including geothermal waste injection Permits for disposal of all solid waste Review activities in the coastal zone for tency with Alaska tal Management Program TIME FRAME 15 days after submittal there is a 30-day period for public comment. If no objections, permit issued within 90 days 15 days after submittal there is a 30-day period for public comment. If No objections, permit issues within 90 days 90 days if no objection Variable Variable Variable Must apply 180 days prior to discharge, 30 days for public comment 30 days for response from EPA. 30 days for public comment, 1 year maximum for final determination 30 days for response from EPA. 30 days for public comment, ] year maximum for final determination 30 - 90 days Apply one s construction son prior to 60 days for final action 30 days for final action 30 days for final action 60 days for final action 60 days for final action 50 days for final action, coordinated with other state permit applications The complexity of the federal permit process will depend on the complexity of the project and its impacts. Assuming no brine discharge and Aleut Corporation ownership of project lands, then the only major federal agency involved will be the COE. It appears likely under these circumstances that the level of impact will not justify an EIS (that determination will be made by COE). With brine discharge, the EPA would be involved and the possibility of an EIS requirement would be significantly increased. Other minor federal permits such as Coast Guard bridge permits may also be needed. 3.1.2 State Permits The Alaska Department of Fish and Game (ADF&G) has taken an active interest in the project because of its mandate to protect fish and wildlife resources. ADF&G recently completed a study of the area potentially affected by geothermal development and have prepared an environmental analysis report (ADF&G, 1986). Road and transmission line development will affect streams containing signifi- cant fish resources and, thus, ADF&G "Title 16" permits will be required for bridges, culverts, buried cable stream crossings and other project activities that affect fish streams. The Alaska Department of Environmental Conservation (ADEC) will be involved in air quality permitting. ADEC will also be involved with air quality per- mitting. A tidelands lease will be required from the Alaska Department of Natural Resources for structures in the tidal zone. For a project of this magnitude, state permits will be coordinated by the Division of Governmental Coordination (DGC). DGC will, at the same time, deter- mine consistency with the Alaska Coastal Management program. 3.1.3 Permitting Status and Timing Pre-application coordination has been in progress with critical regulatory agencies, especially ADF&G. Project concepts have been developed with permit- ability in mind, and mitigation measures have been incorporated into designs to address specific agency concerns. It appears likely that permits for the project could be obtained in about six months, assuming that an EIS is not needed. The permit process would logi- cally consist of the following steps: MLF6/C11 3-2 Distribution to the agencies of a reasonably detailed project descrip-— tion. Pre-application meeting with interested agencies to allow discussion of concerns and to receive recommendations regarding project concepts. Modification of project concepts (if appropriate). Preparation and submittal of the various permit applications and required documentation (see Table 3-1). Permit follow-up and expediting. It should be emphasized that aside from the above permit process, coor- dination with the Aleut Native Corporation will be essential. As the primary landowners, the Aleuts will participate in land use decisions. 3.2 CONSTRUCTION ACCESS FACILITY PERMITS 3.2.1 Driftwood Bay Airstrip The existing airstrip will be upgraded for project use. No direct impacts beyond those already present will occur from these improvements. Air traffic and activity associated with the airstrip will create substantial disturbance to the western valley margin. This disturbance will probably reduce use of the cliffs adjacent to the airstrip by birds. Bald eagles currently roost along the bluffs and future activity could displace them to other areas. Intense dis- turbance will be intermittent and will probably not have long-term impact to the birds of Driftwood Valley. No special permits will be required for the airstrip as long as the size of the strip is not altered. 3.2.2 Driftwood Bay to Fox Canyon Road Upgrading most of the existing road from the airstrip to the Makushin/ Driftwood divide will require little effort and no disturbance to new terrain. Construction of a bridge at the crossing of the Driftwood River (where the old culvert was washed out) will be the only significant new construction. Instream work associated with bridge construction will need to be coordinated with ADF&G through its Title 16 permit system. Timing of bridge construction may be MLF6/C11 3-3 restricted to prevent disturbance to spawning or outmigration of salmon. Constraints on bridge construction will depend on the amount of instream work required. An additional permit may be required from the U.S. Coast Guard for the bridge design. A section of new road will be required from the north side of Fox Canyon to the existing roadway. Most of the terrain crossed is dry, windblown tundra intermixed with areas devoid of vegetation and a few small marshy depressions. Wildlife value is low and no environmental problems are anticipated. COE wetland permits will be required if the roadway involves filling any wetland areas. Any such areas will be very small and no special permitting difficulties are anticipated. Access from the end of the Driftwood Bay road across Fox Canyon to the geothermal well site will be by helicopter. Therefore, no terrain impact will occur in Fox Canyon. 3.3 OPERATIONS ACCESS FACILITY PERMITS 3.3.1 Makushin Valley Road Alternative Operational access to project facilities during project operation will be via a permanent road extending along the Makushin Valley bottom and up to the head of the valley to the powerplant site. This road will initially be constructed along the transmission line route to aid in construction of the line (see below). In the absence of mitigation planning a Makushin Valley roadway would have the potential to cause significant impacts to the Makushin River and its tributaries and, consequently, could affect the relatively valuable fish resources of the lower river. Detailed mitigation planning will be required during design and construction of the road in order to obtain Title 16 permits (Table 3-1) from the Alaska Department of Fish and Game (ADF&G). It will be necessary to complete applications for each stream alteration activity. It is likely that ADF&G will also require a detailed construction plan that describes equipment, techniques, timing, sequence and mitigation measures. Extensive discussions with ADF&G personnel have been held regarding location, design and construction of this road. In addition to the ADF&G permits, permits will also be required from the COE for those portions of the road that traverse wetland terrain. A MLF6/C11 3-4 substantial portion of the lower Makushin Valley is wetland. The road, as currently designed, will traverse about 5,500 feet of wetland. One of the primary impact concerns is the potential blockage of fish passage at roadway drainage structures. The Makushin River is used by adult salmon for spawning up to river mile 5.3 and possibly to the head of the valley (ADF&G 1986). Rearing salmon (primarily coho) have been found throughout the valley but especially in clear tributaries that flow across the valley sides into the main stem Makushin River. The preliminary road route has been selected to avoid sensitive stream crossings as much as possible. Nevertheless, the road will cross the main stem at least once and will cross smaller tributaries 8-13 times. Main stem crossings will be via bridges which, if designed and constructed care- fully, will not interfere with fish resources. Culverts may be permitted on smaller streams that do not support spawning but the culverts will need to be designed and constructed according to ADF&G standards to allow passage of juvenile fish. Existence of a road in the Makushin Valley may also affect hydrologic and hydraulic conditions on the valley floor and, thus, indirectly affect conditions in the lower Makushin drainage. A road could block sheet flow or shallow sub- surface flow and change drainage patterns possibly affecting flow in tributaries used by fish. Careful routing and use of drainage structures will be essential to minimize potential impacts. The Makushin River has changed channels many times in the past and road designers will need to take this into account. Portions of the road may have to be armored to prevent erosion by the river and the consequent siltation of downstream areas. The road will need to be fully protected against flood events. Some impact will be unavoidable during actual construction of the road. Work in or adjacent to streams during bridge construction or culvert installa- tion will cause introduction of sediment into streams. The extent of stream siltation will depend on the methods used and flow conditions at the time of construction. Construction methods and timing when in the vicinity of streams will need to be coordinated with ADF&G. Permit stipulations will likely limit the timing of instream work to avoid times when adult salmon are present and when eggs or fry are in the gravel, leaving a narrow construction window (mid-May to mid-July). MLF6/C11 3-5 3.3.2 Broad Bay Boat Dock Regardless of the means of access through the Makushin Valley, a pile- supported dock facility and a small breakwater will be required for docking the boats used to move personnel from Unalaska to the Makushin Valley access point. The dock will be located off the beach near the south margin of the valley in an area that is relatively rich from a biological standpoint. Some disturbance of bald eagles, seabirds, gulls and marine mammals will occur. However, docking activity will be intermittent and the breakwater may enhance habitat for some species; therefore, adverse impacts are not expected to be significant. Permits from the COE will be required for coastal structures. Design and location of facilities should be carefully considered to minimize potential impacts to sensitive species. 3.4 TRANSMISSION LINE The transmission line would be elevated on poles throughout most of its length. However, the line would be buried through the lower half of the Makushin Valley; the buried portion would connect to a submarine cable that would extend from Broad Bay to Unalaska. The buried portion of the transmission line represents one of the more significant project impacts and would require special consideration relative to mitigation measures employed during construction. The cable would be laid in a trench which would be excavated through an area which is primarily wet with fine grained soils. Consequently, the potential would be high for muddy water to enter the Makushin River and/or its tributaries both during excavation of actual stream crossings and during excavation within wetlands adjacent to the various stream courses. An additional impact is associated with disturbance of stream bottom substrates at buried stream crossings. Crossings within spawning areas would probably require replacement of gravel substrate on the stream bottom to reestablish suitable spawning habitat. MLF6/C11 3-6 An indirect impact that is of concern to ADF&G is possible interference with Natural patterns of shallow subsurface water flow and consequent alteration of flow within small streams that provide rearing habitat for juvenile salmon. A backfilled trench in combination with an adjacent roadway would likely have at least some effect on drainage patterns. Because the area is relatively flat, any such effects may be hard to predict. Drainage patterns and elevations should be taken into account when routing and constructing the transmission line. The buried transition from overland to subsea cable would involve some impact to the beach and intertidal zone at the valley mouth. Because of the dynamic nature of wave-washed sand beaches there is little marine life present and impacts would be minor. All activities which require disturbance to streams in the Makushin Valley will require permits from ADF&G. It is likely that an approved operating plan describing mitigation measures would be required prior to the start of work. In addition, cable burial within wetlands and associated road or workpad construc- tion would require permits from the COE. Since much of the lower Makushin Valley is wetland, a substantial permit effort will be required. 3.5 BRINE DISPOSAL 3.5.1 Surface Disposal Disposal of waste geothermal fluid via a surface drainage leading to the Makushin River was initially considered as an option. The quality and volume of the brine in relation to that of the receiving waters indicated that problems could arise in meeting state and federal water quality standards for some brine components, especially arsenic and hydrogen sulfide. The ADF&G has also expressed concern about possible impacts to fish resources in the lower Makushin River as a result of toxic substances (arsenic and heavy metals) and temperature increases. The discharge of geothermal waste fluid at some point near the generating station would probably not be permitted without substantial treatment to remove arsenic, hydrogen sulfide, and possibly other components prior to discharge. If it were permitted, extensive monitoring would be required, Because of the cost of on-site treatment and potential permitting difficulties, surface disposal was rejected as a feasible wastewater disposal option. MLF6/C11 3-7 3.5.2 Pipeline to Driftwood Bay Another option for brine disposal that was considered is conveyance of the brine to an offshore disposal area in Driftwood Bay via a 12-inch buried pipe- line that would parallel the roadway. Offshore discharge would probably stand a better chance of being permitted than onshore discharge because of the great dilution offered by seawater. Nevertheless, an NPDES permit would be required and the permit process could be long and complex. Because of the toxic elements in the wastewater, it is likely that some study of the marine biology of the discharge area would be required to provide a baseline against which the results of long-term monitoring could be compared to detect possible impacts. The distance offshore and design of a discharge diffuser would have to be negotiated with the agencies and would involve additional engineering costs. A buried pipeline would create additional terrain impacts and trigger the need for wetland permits in the lower Driftwood Valley. The pipeline option was rejected primarily because of these regulatory complications and because of the high construction costs. 3.5.3 Injection Because of the above environmental constraints, the option selected for disposal of brine is injection into the groundwater via injection wells near the generation facilities. Such a procedure avoids the environmental impact and permitting problems that would be associated with wastewater discharge. It also simplifies permitting for the total project by eliminating the need for an NPDES permit and possibly reducing total impacts to the point where an EIS will most likely not be needed. 3.6 POWERPLANT SITE IMPACTS The powerplant and associated facilities will be enclosed within either one or two buildings. All facilities on the north side of Fox Canyon will occupy less than 3 acres, thus terrain disturbance will be minimal. Wildlife habitat value is low on the high plateau at the powerplant site and impact will not be significant. MLF6/C11 3-8 3.7 OPERATIONAL IMPACTS The generating facility will be remotely operated; therefore, human intru- sion will be minimal after construction. It is anticipated that a work crew of 2 persons will inspect the site 2-3 times per week. The crew will travel by boat to the mouth of the Makushin Valley and then by road to the powerplant site, observing the transmission line on the way. During power production, the powerplant will emit some noise, primarily from fans at the top of the building. Relative to other industrial facilities, the noise level is low and no significant disturbance of birds or mammals is antici- pated. The overland portion of the transmission line will be overhead except for the lower half of the Makushin Valley where it will be buried. Powerlines will present some hazard to birds both from electrocution and as flight obstacles. Powerpole and conductor configuration will be designed to prevent large birds (eagles) from contacting two conductors at once and, thus, will minimize the possibility of electrocution. The design proposed in Section 5.0 avoids this problem. Furthermore, the largest concentrations of birds are at the mouth of the Makushin Valley where the lines will be buried. 3.8 ALTERNATE USES OF WASTE HEAT 3.8.1 Driftwood Bay The Driftwood River lowlands near the existing airstrip contain sufficient area and suitable soil foundation conditions to support either aquiculture or agricultural developments. Either prawn or shrimp operations and commercial scale greenhouse operations require large acreage for development of facilities. Driftwood Bay is very exposed to storm events and does not lend itself to water-based transportation developments. However, the Driftwood airstrip could be upgraded to provide a convenient source of transportation to Unalaska/Dutch Harbor for transfer of product to markets in the continental United States or other locations on the Pacific rim. At the present time, fresh seafoods can be shipped for $.98 per pound from Dutch Harbor to Tokyo versus $2 per pound for transportation via Anchorage to MLF6/C11 3-9 Tokyo. Completion of the airport runway extension at Unalaska next year, will increase aircraft payload limits from 15,000 pounds to 25,000 pounds which will further reduce the costs of transporting seafoods to market (Alaska Economic Report, 1986). Weather conditions on Unalaska Island are often poor and generally un- predictable. Therefore, problems can be anticipated in transferring seafood product or agricultural products on a regular schedule from Driftwood Bay to Unalaska/Dutch Harbor. Shrimp are the primary species under consideration for on-shore mariculture operations on Unalaska (Paul Fuhs, personal communication). According to Huner and Brown (1985) "There are presently no financially successful shrimp farms within the United States". However, culture techniques have been developed to the point where there does seem to be potential for commercially viable opera- tions in the United States through the use of semi-intensive pond or raceway cultures for production of shrimp. This type of technology is being actively studied at Texas A&M University and will require several more years to perfect. Additional information is being gathered through experimental operations in Hawaii, South Carolina, Texas and Japan. Even with proven technology, shellfish operations in Alaska would require simplification of legal and permitting procedures to enable those interested in mariculture to obtain clear guidelines on how they can operate and where they can place production facilities. Essentially, all successful commercial produc- tion of shrimp has thus far occurred in third-world countries where climate, labor, legal and biological factors are favorable. Research in the United States continues to focus on developing options for aquaculture in more tem- perate climates. Therefore, Alaska's options may increase in the next decade. The best options for agriculture would require the use of greenhouses because of the relatively severe Unalaska climate and techniques such as hydro- ponics which involve the cultivation of plants in water containing dissolved organic nutrients, rather than soil. Waste heat from thermal pipelines carrying geothermal fluids would provide a relatively inexpensive heat source needed for good plant growth. Crops such as tomatoes and cucumbers have been grown commer- MLF6/C11 3-10 cially in greenhouse operations in southcentral Alaska using more conventional heat sources such as electricity or natural gas. However, it is doubtful that this type of operation could be commercially viable at Driftwood Bay, even ignoring the high costs of piping the brine from the powerplant site. 3.8.2 Broad Bay An assessment of available site data near the mouth of the Makushin River at Broad Bay revealed essentially no potential for onshore development of agri- culture or mariculture facilities because the area is dominated by wetlands underlain by thick layers of volcanic ash, resulting in extremely unstable foun- dation conditions. The engineering constraints on facilities development coupled with wetlands permitting problems would essentially exclude any options for aquiculture or agriculture development in that area. However, if electrical transmission lines were constructed through the area, they would benefit offshore mariculture operations such as pen-rearing of salmon. 3.8.3 Other Considerations The options considered for disposal of waste thermal fluids include surface release at the plant site, a pipeline to the coast connected to a subsea dispos- al site, or re-injection at the well site. Thus far only the latter option appears viable because plant effluents would contain toxic components (see Section 3.3). Therefore, surface disposal would not be acceptable to the regu- latory agencies. Transportation of waste thermal fluids via insulated pipelines from the plant site to either Broad Bay or Driftwood Bay for oceanic disposal was determined to be too costly for further consideration. The remote location of a power plant site on the slopes of Makushin Volcano and the limited space available at that site for facilities development would preclude either agriculture or aquiculture development. Although it is not feasible to utilize the waste heat, it is likely that it would be feasible to utilize off-peak electric power. The geothermal plant is designed to operate at full capacity whenever it is available. (This mode of operation is desirable to simplify unattended operation of the plant. The variable cost of additional generation is limited to the royalty paid for the MLF6/C11 3-11 steam which is 1-2 cents per KWH or less. Heating loads could be found which could use power dispatched only at off-peak periods. These loads might include electric space heating, domestic water heaters, or the high school swimming pool. 3.9 AIR QUALITY NOTE: A more complete discussion of air quality impacts appears in Appendix C of this report. After the useful energy is extracted from the geothermal fluids, the waste stream will pass through a condenser. Remaining steam and some gases will be condensed into the brine and injected. However, a portion of the gases in the waste stream is non-condensable. The primary non-condensable gas is hydrogen sulfide (Hs), As part of this study the USEPA-recommended Industrial Source Complex (ISC) model was used to estimate emissions from the proposed project. Based on data obtained in the Phase III Exploration Study (Republic, 1985), H2S concentrations in the waste stream are expected to average 1.73 parts per million (ppm). The model was run under a worst case assumption that all non-condensable gases present in the stream will be released to the atmosphere and that the full 9.5 MW capacity is in place (7.5 MW flash steam + 2.0 MW binary). The model predicts that down-wind receptors 100 meters from the source would experience maximum 1 hour H2S concentrations of 0.25 ppm from a 9.5 MW plant of 0.17 ppm from a 5.0 or 7.0 MW plant. This compares with a State of Alaska reduced sulfur compound ambient standard of 0.02 ppm and a threshold limit value for HjS of 10.0 ppm recommended by the National Institute for Occupational Safety and Health (NIOSH). Because the Alaska ambient standards are exceeded, the Alaska Department of Environmental Conservation (ADEC) could not issue a permit without granting a variance or requiring emissions controls. There is, however, a sound justifica- tion for issuing a variance in that: 1. The Alaska standard is odor-based which may not be especially important in a remote area. MLF6/C11 3-12 2. Being a volcanic area, natural emissions may already exceed the odor threshold. 3. The emissions are 40 times lower than the health-based NIOSH standards, The emissions from the plant, even assuming 100 percent venting throughout the entire year would be less than 13 tons of H2S. This amount is well below the 250 ton per year emission which triggers the need for a Prevention of Significant Deterioration (PSD) review. Thus there should be no federal air quality permitting requirement. MLF6/C11 3-13 GEOTHERMAL RESOURCE ANALYSIS & DRILLING COSTS FROM MESQUITE GEOTHERMAL COST AND PERFORMANCE DATA FROM SAI LOAD FORCASTS FROM R.W. BECK DIESEL PERFORMANCE DATA FROM CITY OF UNALASKA ECONOMIC MODEL: CAPITAL, FIXED AND VARIABLE COSTS BY UNIT ELFIN GENERATION MODEL: DISPATCH BY UNIT BY YEAR FIGURE 4-1 OVERALL ECONOMIC ANALYSIS APPROACH NET PRESENT WORTH BY SCENARIO DAMES & MOORE 4.0 ECONOMIC ANALYSIS This section describes the analysis conducted to determine the economic feasibility of the proposed project. The overall approach and assumptions are described in Section 4.1. Section 4.2 describes the ELFIN generation dispatch model. Section 4.3 describes the economic model. The cases modeled and the results are reported in Section 4.4. Economic conclusions are drawn in Section 4.5. 4.1 APPROACH AND ASSUMPTIONS The purpose of the economic analysis reported here is to compare the life cycle costs of electric generation using diesel generation facilities with the costs using a generating system which includes a geothermal component. Alternative geothermal configurations were compared in order to determine the optimal size and timing of geothermal capacity additions. The best geothermal system was then compared with the Base Case All-Diesel scenario to determine whether geothermal resource development would be economically feasible. The working definition of "economic feasibility" is that the discounted life-cycle costs for systems including geothermal development are less than the costs for the all-diesel scenario with all other factors held constant. It is important to stress the distinction between financial and economic analyses. Economic analysis, as represented by the analysis reported herein, is intended to compare alternative systems without regard to the timing and mech- anisms for repayment of costs. By contrast, a financial analysis takes into account the repayment timing and mechanisms. Financial analyses include such considerations as interest rate and the method of repayment of capital costs (grants, revenue bonds, etc.). Logically, it is first necessary to establish the economic feasibility of a project before investigating the financial feasibility. In general no project which is economically infeasible can be financially feasible. Once economic feasibility has been demonstrated it remains to investigate the financial feasibility. This latter step has not yet been undertaken. The overall approach used in the economic analysis is diagrammed in Figure 4-1. The approach is summarized here and discussed in more detail in Sections 4.2 and 4.3. MLF6/C30 4-1 The economic analysis relies on two separate models, the ELFIN Generation model and a discounted cash flow type economic model. The ELFIN model (copyright by the Environmental Defense Fund) uses inputs regarding available generation capacity and performance to determine which units would be most eco- nomically dispatched at any time. Performance data for diesel units were obtained from the City of Unalaska (Burton, 11/86) and N C Engine Power (Shultz, 11/86). Data on geothermal performance was developed by SAI Engineers and the Mesquite Group. Load data forecasts were developed by R.W. Beck (1987) under a separate contract to the Power Authority. The annual summary results of the ELFIN model consists of load, capacity factor and variable cost by generating unit. These results are input to a spreadsheet economic model which adds in capital and fixed operating costs to determine the total annual generation costs. The economic model also discounts and sums the annual cost to find the net present value of each scenario over the life cycle of the geothermal plant. These net present values constitute a figure of merit by which alternative geothermal configurations can be compared with the All-Diesel (No Project) alternative. The lower the net present value (for meeting an assumed load fore- cast scenario) the more desirable the alternative. This analysis was used during the feasibility study to define the relevant alternatives for con- sideration and to determine the preferred alternative(s). The economic analysis depends on numerous technical and economic assump- tions. It will be seen that the economic feasibility of the proposed geothermal development rests on two critical assumptions about unknown future conditions; i.e., oil prices and utility loads. Sensitivity analyses were conducted for those two assumptions as well as all other significant assumptions. The eco- Nomic assumptions are summarized in Table 4-1. Electrical generation assump- tions are shown in Table 4-2. Both sets of assumptions are discussed in more detail below. MLF6/C30 4-2 TABLE 4-1 ECONOMIC ASSUMPTIONS PARAMETER | BASE CASE ASSUMPTIONS SENSITIVITY ASSUMPTIONS o Inflation Rate Zero percent: Analyses conducted None in real terms. Assumes that inflation is a wash among cases. All costs in 1986 $, o Present Value Discount Rate 3.5 percent real 4.5 percent real o Diesel Escalation Average of APA (Emmerman 11/86) APA (Emmerman 11/86) Upper Rate Lower bound and upper bound bound and lower bound price price trends applied to 1986 trends diesel prices. o Unalaska City R.W. Beck (1987) base case fore- R.W. Beck (1987) high case Utility Load cast. forecast. o Interest Rate(s) Not applicable to economic analy- Not applicable sis. o Period of Analysis 25 years after first geothermal Through 2025 operations (through 2016) o Geothermal Royalty Based on conditional Land and None and Resource Agreement between Power Authority and Aleut Corp. 6/86. Assumes no further cost for multiple wells in same target MLF6/CT4 PARAMETER Orv Diesel: Existing Capacity and Efficiencies Additional Capacity Forced Outage Rate (Diesel) Major Maintenance (Planned Outage Rate) Service Life Heat Rates Geothermal: Capacity and Timing Forced Outage Rate Maintenance (Planned Outage Rate) Service Life Diesel and Geothermal: System Efficiency MLF6/CT5 T 4- LEC /AL RA BASE CASE ASSUMPTIONS As per data provided by City of Unalaska Utilities (R. Burton, personal communication, 11/86) Assume additional 855 KW units as needed to follow base load forecast case. 400 hours/unit/year or 4-6 percent Every 700,000 gallons used (4.7 percent outage) 2.1 million gallons or 30,000 MWH for 855 KW units. Existing units 12-14 KWH/gal (As per Burton 11/86) 5.0 MW Flash steam (Available 1991) 8 percent 17 percent 25 years 95.4 percent AS TIC SENSITIVITY ASSUMPTIONS +— None Assume additional units as needed to follow high load forecast. None None None None Additions to 7.0 MW in 1995 and 9.5 MW 2001 None None 34 years None 4.1.1 Economic Assumptions INFLATION: The entire economic analysis utilizes a zero inflation assump- tion. With the exception of diesel prices (which are discussed below) all prices are assumed to remain at their 1986 levels. This assumption does not imply that there will be zero inflation, but is intended to simplify the analy- sis. The relative merit of the alternatives for supplying Unalaska's electrical needs can be compared without invoking inflation forecasts. If assumptions were made about inflation trends, these same factors would be used to deflate each year's nominal dollar costs back to 1986 dollars. To avoid this unnecessary complication it is assumed that inflation is zero. This assumption, while appropriate to the economic analysis presented here, would not be appropriate for a financial analysis. PRESENT VALUE DISCOUNT RATE: As directed by the Power Authority, this anal- ysis assumes a 3.5 percent present value discount rate to reflect the real (inflation-free) cost of funds. Alternatively, the discount rate can be viewed as the State of Alaska's time preference for funds. As a sensitivity case a 4.5 percent real rate is used. This sensitivity case places greater relative empha- sis on near-term costs. Thus geothermal alternatives, which substitute up-front capital costs for ongoing diesel fuel costs, are less desirable under the higher discount rate. DIESEL PRICES: In late 1986 the City of Unalaska was paying $0.70 per gallon for diesel fuel used in its generating facilities (Burton, 11/86). For the economic analysis this price was projected to escalate in real (inflation free) terms to $0.90 per gallon by 1996, then to remain constant under the Power Authority's Low Price trend. Under the High Price trend the cost rises to $0.90 by 1987 then escalates 3.5 percent per year until leveling off at $1.73 in 2006. These trends are intended to bracket the upper and lower range of oil prices for the next 20 years (Emerman, Memo, 11/26/86 and Denig-Chakroff, letter to M. Feldman, 11/26/86). A Medium Price trend, which is the simple average of the high and low price in each year, was used as a Base Case. Sensitivity analyses were performed on the High and the Low Price trends. LOAD FORECASTS: Load forecasts (annual peak and energy) through 2006 were projected by R. W. Beck and Associates (1987) under a separate contract with the MLF6/C30 4-3 Power Authority. The Base Case and High Case load forecasts were used in this analysis. In addition to the published study, Beck provided Dames & Moore with hourly loads for typical weeks. These typical hourly data are an important input to the ELFIN generation model. They are described in Section 4.2. INTEREST RATE: There is is no need to consider the interest rate in an eco- nomic feasibility analysis. The time value of resources consumed is captured by the net present value analysis. Capital and operating costs are simply assumed to be available in each year in which they are needed. In the financial feasibility analysis subsequent to this analysis it will be important to con- sider the sources of funds and consequent interest rates, but these are not Necessary in comparing among generation alternatives from an economic stand- point. PERIOD OF ANALYSIS: In order to reflect the relatively long life of geo- thermal development investments the life cycle costs of the various generation scenarios were compared over a period from 1988 through 2016, which represents a 25 year life of the initial geothermal installation (assumed to be operative in 1991). For sensitivity comparison, results for 1988 through 2025 were also com- pared to reflect the full utilization of additional geothermal capacity added in 2001 for the 9.5 MW (megawatt) case. 4.1.2 Generation Data and Assumptions--Diesel EXISTING DIESEL: Performance data on the six existing diesel units were ob- tained from the City of Unalaska (R. Burton, personal communication to M. Feldman, 11/86). The heat rates for the 1,450, 855, and 620 KW units is 14 KWH per gallon (9,907 Btu per KWH). The older 600 unit and the two 300 KW units operate at a heat rate of 12 KWH per gallon or 11,558 Btu per KWH. Forced outage rates are assumed to be 400 hours per year or 4.6 percent. Major main- tenance occurs at intervals of about 15,000 hours, and requires about 1,440 hours per unit to complete since units must be sent to Anchorage or Seattle. This results in an estimated maintenance rate of 4.7 percent (R. Schultz, NC Engine Power, personal communication to M. Feldman 11/86). ADDITIONAL CAPACITY: As load growth dictates the need for new capacity, 855 KW diesel units are added. Diesel capacity is added in order to maintain a MLF6/C30 4-4 loss of load probability (LOLP) of 3 days per year or less. The characteristics and heat rates of these units are assumed to be the same as the existing 855 KW unit. The installed cost for each of these units is assumed to be $145,000. Additional costs for switching gear ($50,000) and cooling capacity and fuel lines ($30,000) raise the total installed cost per unit to $225,000 (R. Burton, 11/86). This auxiliary equipment is assumed to last through three generating unit replacements (about 21 years). OPERATION & MAINTENANCE AND REPLACEMENT: The capital cost of an operating 855 KW diesel unit is $225,000. This includes a cost of $80,000 for cooling, fuel storage and switch gear. The life cycle cost (exclusive of fuel and routine maintenance) is $262,000. This includes depreciation on switching, cooling, fuel storage, and overhauls. The salvage value for the diesel engine itself is presumed to equal the disposal cost. The switching, cooling and fuel gear is assumed to last through three engines. The $262,000 life cycle cost, divided by the 30,000 MWH that each engine is expected to generate, results in a cost of $0.0087 per KWH. Routine service and maintenance adds an additional $0.008 per KWH (Burton, 11/86) resulting in a total variable cost of $0.0167 per KWH plus fuel. The fixed 0 & M cost is estimated to be $280,000 per year. According to R. Burton, this cost would remain constant regardless of the number of diesel units on line. 4.1.3 Generation Data and Assumptions - Geothermal GEOTHERMAL CAPACITY: The proposed geothermal concept is described in detail in Section 5.0. Briefly, the concept includes an initial installation of 5.0 MW of flash steam with an additional 2.0 MW of binary capacity. If diesel costs and load growth warrant, an additional 2.5 MW unit may be added in 2001, yielding a net capacity of 9.5 MW. OPERATION & MAINTENANCE: Operation of a 5, 7, or 9.5 MW geothermal plant including site access and transmission facilities associated with the geothermal development will cost $172,000 per year. Maintenance of these facilities will cost $181,000, $225,000, and $305,000 per year for 5, 7, and 9.5 MW plants, respectively. These costs are fixed in that they do not change as a function of the level of generation. Variable O & M is negligible. MLF6/C30 4-5 AVAILABILITY: The geothermal plant is expected to be availible 75 percent of the time in any year. This includes a forced outage rate of 8 percent or about 30 days per year. Planned outages for scheduled maintenance (17 percent of the time or 62 days per year) can be scheduled during the off-peak seasons. GEOTHERMAL ROYALTY: A royalty agreement between the Power Authority and the Aleut Corporation was executed on 17 June, 1986. This agreement provides the Power Authority with rights of access, lands for development and steam for use in geothermal power production. The royalty rate is a function of the amount of electricity sold by the utility and the busbar cost of electric power production in Unalaska exclusive of the geothermal development. The formula for the "Power Share" is: Power Share = KWH sold x royalty rate x City Busbar Cost The royalty rates range from 0.0325 to 0.063 per KWH depending on the energy sold by the utility in any quarter. The busbar costs used in this analysis are based on an electric rate study conducted by R.W. Beck (1985). This study reports that the busbar cost for fiscal year 1986-87 will be $0.13 per KWH exclusive of fuel costs. The busbar costs (exclusive of fuel) are not expected to change over the period of analysis (R. Burton, 1986). The royalty was calcu- lated using the projected diesel fuel costs in each year plus the $0.13 non-fuel busbar cost. In addition to the Power Share the agreement provides for an annual easement fee of $5,000 per year. It should be noted that the terms of the agreement permit only a single well to be used. It is assumed that these terms could be modified to reflect the use of more than one production well and one or more injection wells at the same rate per KWH generated. An Energy Share and By-product Share royalty agreements were also included in the same royalty agreement. These rates, which would apply to non-electric heat uses were not included in the economic feasibility analysis. GEOTHERMAL SERVICE LIFE: The geothermal wells, power plant equipment, and transmission equipment is all expected to have a useful service life of 25 years. Salvage value at the end of the service life is assumed to be exactly offset by the cost of removal. MLF6/C30 4-6 4.2 THE ELFIN GENERATION MODEL The ELFIN Model description which follows is based on EDF, 1986. 4.2.1 General Overview The ELFIN generation model is a production simulation model which determines the least cost manner of dispatching the available generating units within a electrical generating system. It was developed by the Environmental Defense Fund (EDF), which holds the copyright. Version 1.30 was used by Dames & Moore to perform the economic analysis under a lease agreement with EDF. ELFIN was run on a VAX minicomputer. The model simulates the decisions which would be made by a utility by meeting demand within each specified "season" by bringing on line the available generating units in increasing order of variable cost. Fixed costs and capital investment decisions are not made by ELFIN. These decisions are exogenous to the model and are specified as input (insofar as the capacity is specified). The model allocates scheduled maintenance for each unit on a yearly basis in such a way as to maximize system reliability. Forced outages are allocated on a stochastic basis. 4.2.2 Load Data The load forecasts (peak and energy) anticipated for the City Utility are an important input to the ELFIN model. These forecasts were obtained from the study conducted by R.W. Beck under separate contract to the Power Authority. Appendix D reproduces the ELFIN input file. Peak and energy inputs used are shown in the variables ANPEAK and SALES. For easy reference, the peak and energy forecasts are also shown in Table 4-3. In addition to the annual peak and energy forecasts, ELFIN requires specifi- cation of the weekly load durations for typical weeks within each specified "season", These seasons do not correspond to the conventional four seasons but are specified to reflect distinct differences in electrical load within the year. Because the load in Unalaska is dominated by industrial users, these seasons correspond to the activities of fish processors to a large extent. At MLF6/C30 4-7 TABLE 4-3 LOAD FORECAST FOR CITY OF UNALASKA BASE CASE HIGH CASE YEAR PEAK ENERGY PEAK ENERGY MW MWH MW MWH 1987 2,981 14,119 3,021 14,262 1988 5,157 21,311 6,283 21,617 1989 6,316 24,612 8,528 29,854 1990 7,488 29,834 9,740 35,286 1991 7,622 30,324 9,922 40,767 1992 7,708 30,615 10,034 41,199 are 7,493 30,909 10,148 41,643 1994 7,880 31,208 one 42,103 1995 9,017 36,261 11,435 47,327 1996 9,105 35,568 11,558 47,820 1997 9,194 36,881 11,684 48,329 1998 9,284 37,198 11,814 48,857 2999 9,365 37,468 11,934 49,332 2000 10,496 42,486 13,106 54,561 2001 10,578 42,756 13,231 55,056 2002 10,660 43,026 13,338 55,564 2003 10,743 43,311 13,489 56,087 2004 10,827 43,600 13,624 56,648 2005 10,910 43,889 13,761 $7,213 2006 10,996 44,192 13,902 57,808 Source: R. W. Beck, 1987 Tables -- V-3, V-4,.¥-211, V-32 the Power Authority's request, R.W. Beck provided Dames & Moore with their esti- mates of the seasonal variations and the hourly loads for one typical week within each season. These hourly loads provide ELFIN with information on the load necessary to allow it to economically dispatch generation units and to calculate the system reliability. In effect, the hourly data tell ELFIN the shape of the load duration curve. The actual raw data provided by Beck (simply reformatted for ELFIN) appear in Appendix D in the ELFIN input file as a set of variables called "WKxxxxx". For better presentation, Dames & Moore constructed load duration curves using the same data weighted by season length and compiled annually in rank order of frequency. For the Base Case forecast, these curves are shown as Figures 4-2 through 4-4 for years 1990, 1995, and 2000, respec- tively. Figures 4-5 through 4-7 show the same years under the High Case fore- cast. Note that these load shape data are specified only for selected years. Load shape data for the years 1987, 1988, and 1989 were also input to ELFIN. These are the years in which there was a marked change in the load. ELFIN scales these loads up or down based on the specified peaks and energy sales for each year. 4.2.3 Dispatch Order As mentioned above, ELFIN dispatches generation capacity available in order of increasing variable cost. Thus geothermal capacity, whose only variable cost is the power share royalty for geothermal steam, is dispatched whenever available in preference to diesel. Based on fuel cost alone, even the most efficient diesel unit costs upward of 6 cents per KWH versus 1 to 2 cents for geothermal. In addition, diesel has a variable 0 & M cost of about 1.7 per KWH versus zero for geothermal. When diesel capacity is used, it is dispatched in order of fuel efficiency, as variable 0 & M is essentially the same for all units. The availability of any unit is determined by the specified planned and forced outage rate. The model assigns the planned outages to the season(s) in which they have least impact on reliability, i.e., the slower season(s). The forced outages occur randomly according to a method called the "Baleriaux-Booth" algorithm. As programmed, the model assumes that the geothermal unit can be run at reduced loads as well as at full capacity. MLF6/C30 4-8 MEGAWATTS 8.0 7.0 6.0 5.0 Hetil 4.0 3.0 4 2.074 1.0 0.0 10% RW BECK BASE CASE FORECAST UNALASKA LOAD DURATION CURVE 20% YEAR 1990 30% 40% 50% 60% PERCENT OF YEAR (8,760 HOURS) 70% 80% 90% 100% Dames & Moore FIGURE 4-2 MEGAWATTS RW BECK BASE CASE FORECAST UNALASKA LOAD DURATION CURVE YEAR 1995 2.04 atel4 0.0 ee ieee 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% PERCENT OF YEAR (8,760 HOURS) Dames & Moore FIGURE 4-3 MEGAWATTS RW BECK BASE CASE FORECAST UNALASKA LOAD DURATION CURVE YEAR 2000 11.0 10.074 0.0 | 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% PERCENT OF YEAR (8,760 HOURS) Dames & Moore FIGURE 4-4 MEGAWATTS RW BECK HIGH CASE FORECAST UNALASKA LOAD DURATION CURVE YEAR 1990 10.0 ] 9.0 8.0.4 7.07 6.074 5.07 wy 2.04 1.074 0.0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% PERCENT OF YEAR (8,760 HOURS) Dames & Moore FIGURE 4-5 MEGAWATTS RW BECK HIGH CASE FORECAST UNALASKA LOAD DURATION CURVE YEAR 1995 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% PERCENT OF YEAR (8,760 HOURS) Demes & Moore FIGURE 4-6 MEGAWATTS RW BECK HIGH CASE FORECAST UNALASKA LOAD DURATION CURVE YEAR 2000 30% 40% 50% + 60% PERCENT OF YEAR (8,760 HOURS) 70% 80% | 90% 100% Dames & Moore FIGURE 4-7 4.2.4 ELFIN Output ELFIN can provide very detailed output showing operation and cost by indi- vidual generating unit by week. For the purpose of economic feasibility analy- sis for which it is here applied, annual totals by generating type (existing diesel, additional diesel, geothermal) are sufficient. A sample annual summary output (for the 5.0 MW geothermal capacity under the high demand case) is shown as Table 4-4. The only values which are actually used as input to the economic model are the total annual load for diesel and geothermal and the fuel and variable 0 & M cost per KWH for diesel generation. Other output which is of interest but does not directly become input to the economic model includes the capacity factors and the LOLP (loss of load probability). The capacity factor indicates the percentage of time that the unit is used. This factor was used to indicate whether the geothermal capacity is being sufficiently utilized. The LOLP is an indication of reliability. If this factor exceeded 1.5 days, it was taken as an indication that installed capacity was insufficient, and there- fore additional diesel units were added. In most years LOLP was held below 1 day. 4.3 THE ECONOMIC MODEL The economic model is a discounted cash flow spreadsheet model which is used to generate the present value life cycle cost of each generation scenario. By comparing the life cycle cost among the scenerios, alternative geothermal con- figurations could be compared to all diesel scenarios. This comparison per- mitted us to determine the conditions under which geothermal development would be feasible. By iterating between the economic model and the generation model it was possible to determine the optimal size and deployment timing of geo- thermal capacity. The economic model was implemented on a Lotus 1-2-3 version 2.01 software package (copyright 1985 by the Lotus Development Company) run on an IBM-com- patible personal computer. A sample spreadsheet showing the results using 5.0 MW of geothermal capacity under the high demand medium fuel price scenario is reproduced as Table 4-5. The spreadsheet is divided into four sections. At the top are the assump- tions which describe the scenario--the geothermal capacity, the diesel price MLF6/C30 4-9 TABLE 4-4 RUN NUMHER 1 -- 1991 -- M408 UNALASKA 5 Ml) GEOTHERMAL NEW HIGH CASE DEMAND J/) 0/117 PAGE NO. i ANNUAL GENERATION 1991 PEAK WeEK NO. OUTAGE -- ENERGY -— FURL CAPACITY -B/MHIU Brus > --MILLS/KWH--- RANK*® TIME TOTAL CAMACITY OF (PERCENT) (GWH) CTHIT bACHON ne KW rUEL O+M TOTAL MARGINAL COST (MW) UNITS FORCED MAINT. LOAD PUMPING TOTAL Bi) CeCn) MILL G/KWH ANN. AV. (PCT) (M$) wild 1440 KW DIESEL UNIT 1420. 1 4.6 4.7 5577. oO. 5577. fete, TR AMAL “6 9907. m0 16.3 74.3 2 14.76 415 GBU0 KW DIESEL UNIT Mee 1 4.6 a. 7 2590 oO 2590. a Ala 1G 9907. 40 16.3 74.3 3 7.17 193 6°0 KW DIESEL UNIT Ot, 1 4.6 4.7 1519. 0. 1519. W.0 se a 46 9907, w0 16.3 74.3 4 6.28 113 600 KW DIESEL UNIT 400 1 4.6 1.9 oO. 0 oO. ag Oo U6 115500 47.7 16.3 84.0 6 0.00 ° 300 KW DIESEL UNIT HOO, 1 4.6 1g oO. 0 0 ag Og u 46 11550. “7.7 16.3 84.0 7 0.00 0 300 KW DIESEL UNIT 00, 1 4.6 179 0 ° oO. ag ag U6 11550, &7.7 16.3 84.0 8 0.00 0 tA ADDITIONAL DIESEL UV 75‘/u. 9 4.6 4.7 4325. 0 4325. AY “4 46 9907. W400 16.3 74.3 S 29.53 322 GEO FIRGT GEOTHERMAL vO0O 4 8.0 17.0 26739. 0 28739. ao &. 00 J. "0 00 8.0 1 42.26 230 TOTAL OK AVERAGE 19 5.6 8.2 42751 oO. $2751. Line ere Ah 29.7 100.00 1271 ENERGY NUT SERVED 0.00 GWH COST (8 ING UGE) IN AVERAGE MAKGINAL COST 112.4 0. 00 0 (ANNUAL AVI-KAGE MAKGINAL COST 46. 3) * ANNUAL VAL.UES~-- PEAK 9920, MW RESERVE MAKGIN 69.5 PCT LOAD 42751. GWH LOLP 0.00000) ( 0. 000 DAYS) TOTAL COST (MILLION $) 1271 ( ® INDICATES VALUES FROM | Ati TYPICAL WEEK) YEAR 1988 “969 "990 1993 1992 1893 "994 1995 1996 1997 1998 "999 2000 2001 2002 2003 2008 2005 2006 2007 2008 2009 2010 2011 2012 2013 2018 2015 2016 2017 2018 2019 2920 202) 2022 2023 2026 2025 TOTAL ECONOMIC ANALYSIS 5 MW GEOTHERMAL-~BASE CASE DEMAND MEDIUM DIESE. PRICE TREND DISC COSTS MILLION 1986§ LEVELIZED COST ($/Kwn) DISC RATE CAPITAL ADDED 3.58 =-GEOTHERS 0 & MH ANNUAL FUEL VARIABLE COST $74.5 THRU 20:6 $6.107 THRU 2016 $0.107 THRU 2025 TOTAL ANNUAL cost $87.7 THRU 2025 ANNUAL LOAD AV. ANN COST/Kwn CAPITAL 5,08 29,43 0 $ 2 5 5 $ 358 358 358 458 458 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 358 338 358 358° 356 358 358 358 °2, 545 0 0 0 0 0 0 0 0 0 0 a 0 a 0 0 0 0 a 0 0 0 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0 0 197 223 226 229 287 291 294 298 301 367 312 376 38) 385 389 394 394 394 394 394 394 394 394 394 394 394 394 394 394 394 394 394 396 394 398 12,498 § $5,090 29,437 $55 561 584 587 645 649 652 636 659 125 730 134 139 143 147 152 152 152 152 152 182 152 152 182 152 152 182 152 182 192 152 152 182 152 152 59,560 (mat ) 24,810 26,903 25,008 25,118 26,893 26,985 27,08! 27,169 27,233 28,81 28,917 28,973 29,079 29,131 29,216 29,280 29,280 29,280 29,280 29,280 25,280 29,280 29, 280 29,280 29,280 29, 280 29, 280 29,280 29, 260 29,280 29,280 29,280 29,280 29,260 29,280 995 Sw ($/KWH) — (x$1000) (xi 0. 0. 0. Q. 0. 023 ee eo coo 090 000 022 023 023 026 O24 026 024 024 -025 025 2025 025 025 026 -026 026 026 -026 -026 026 -026 026 026 -026 -026 026 026 -026 026 026 -926 026 226 026 927 AVG ADOED 675 225 225 0 0 0 0 0 0 6 a 0 a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a 0 0 0 a o 35 780 280 280 280 280 289 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 280 2€0 280 280 280 280 280 280 260 2ae 10,640 TABLE 4-5 358 43 $07 4 119 123 128 189 195 201 207 213 282 287 296 301 309 316 322 322 322 322 322 322 322 322 322 322 322 222 322 322 322 322 322 322 322 322 11,001 0 & M ANNUAL FUEL VARIABLE COST 1,355 1,609 2,008 460 488 516 545 818 860 896 934 971 1,295 1,338 1,30 1,435 1,490 1,50 1,608 1,606 1,604 1,608 1,604 1,608 1,604 1,604 1,606 1,608 1,608 1,608 1,604 1,506 1,604 1,606 1,504 1,608 1,504 1,608 52,026 TOTAL ANNUAL cost 00) (x$1000) (x$1000) (x$1000) 2,667 2,526 3,016 854 887 99 953 1,287 1,336 1,377 1,421 1,465 1,897 1,905 1,96! 2,016 2,079 2,140 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2,206 2, 206 2,206 2,206 2,206 2,206 2,206 2,206 14,192 ANNUAL, LOAD (nwt) 22,346 25,805 31,216 6,989 7,202 7,404 7,609 11,133 11,362 11,596 11,939 12,058 15,740 15,917 16,145 16,337 16,587 16, 805 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 17,058 695 Gai a 9 098 096 122 123 1b 125 16 118 V9 120 121 18 120 Vt 123 125 127 129 129 129 129 129 129 N29 2129 129 2129 129 129 129 129 129 129 V9 129 2129 129 148 AVG -- SYSTEM AV. ANN TOTAL ANN PHANN ANNUAL COST/MWH SYSTEM DISC ANN DISC ANN SYSTEM SYSTEM SYSTEM COST cost cost LOAD LOAD PER KWH ($/miH) ($1009) (1986 $M) (MWK) = (MWK) ('§/ KH) 2,672 2,492 22,346 = 20,835 0.120 7,616 6,857 25,805 23,233 0.295 32,453 28,219 31,278 27,188 = 1.038 1,409 1,183 31,799 26,698 0. 04d 1,468 = 1,190 32,105 26,024 = 0.046 1,503 1,176 32,412 25,369 0.046 1,540 1,164 32,727 24,735 0.067 1,932 1,410 38,026 = 27,751 0.081 1,985 1,399 38,347 27,023 0.082 2,029 1,381 38,675 26,316 = 0.082 2,077 1,365 39,008 25,630 0.053 2,124 1,347 39,291 24,928 0.054 2,582 1,982 64,951 27,293, 0.058 2,635 «1,959 44,834 26,522 0.059 2,695 1,939 45,118 25,772 0.060 2,754 1,519 45,416 25,050 0.061 2,821 1,503 45,718 24,389 0.062 2,888 «1,485 46,021 23,667 0.063 2,958 1,469 46,338 23,011 0.064 2,958 1,469 46,338 22,219 0.066 2,958 1,469 46,338 21,455 0.068 2,958 1,469 46,338 © 26,717 0.071 2,958 1,469 46,338 20,005 = 0.073 2,958 = 1,469 46,338 19,317 0.076 2,958 1,469 46,338 «18,652 (0.079 2,958 = 1,469 46,338 18,011 = 0.082 2,958 «1,469 46,338 17,391 0.084 2,958 1,469 46,338 16,793 0.087 2,958 1,669 46,338 «16,215 0.091 2,958 «1,469 46,338 «15,658 = 0.094 2,958 1,469 46,338 «15,119 0.097 2,958 = 1,469 46,338 14,5990. 101 2,958 1,469 46,338 = 14,097 0.104 2,958 1,469 46,338 13,612 0.108 2,958 1,469 46,338 13,148 0.112 2,958 1,469 46,338 12,692 0.116 2,958 = 1,469 46,338 12,255 0.120 2,958 1,469 46,338 11,834 0.1248 134,352 87,746 = 1,600 198 0.127 Gis Git ANG trend, and the discount rate. Changing the latter two parameters causes the spreadsheet to recalculate for the values entered. The discounted present value over the time period indicated is the figure of merit which is used to compare the alternative scenarios. This figure is the sum of the discounted annual costs for the entire system (both diesel and geothermal) for the indicated time period, expressed on constant 1986 dollars. The lower left side of the spreadsheet shows the costs for the geothermal portion of the system and the annual loads. Years 1988 through 1991] are all zeros in this part because 1991 is the earliest time that geothermal capacity could come on line (see Figure 5-2). Variable 0 & M is zero in all years reflecting the fact that geothermal 0 & M cost is completely fixed depending only on the installed capacity. The fuel cost reflects the royalty rate to be paid to the Aleut Corporation as described in Section 4.1.3 above. The geo- thermal royalty is calculated in the economic model rather than using the result of the ELFIN run, as this approach allows the economic model to calculate the present value using either the high, medium, or low diesel price trend without reference to ELFIN. (The royalty depends in part on the non-geothermal busbar cost which in turn depends on the diesel price.) Capital and fixed O & M costs were provided by SAI. The load for each scenario is read from the ELFIN run. The total geothermal cost in any year is the sum of fuel (royalty) and 0 & M costs plus the capital additions in any given year. Note that the capital costs are not levelized, but are simply summed into the year in which they are incurred. In effect 100 percent equity financing is implied. This is appropriate for economic analysis. Because of these capital costs, the cost per KWH is much higher in years in which capital costs are incurred. In reality a utility would spread these costs over the life of the capacity, using some bonded debt mechanism. The lower right section of Table 4-5 shows the diesel generation costs and loads. The fixed 0 & M and fuel costs and load are read from the ELFIN annual generation summary for each year. The fuel cost is modified (increased by the appropriate factor) if the high diesel price trend is specified. The costs are summed (as with geothermal) including the full cost of the capital additions in the year in which they are incurred. MLF6/C30 4-10 Finally the far right section of the spreadsheet shows the annual sums for the entire utility system. The total costs for both the geothermal and the diesel portions of the system are added then discounted by the appropriate fac- tor back to 1986. The sum of these present values constitutes the life cycle present value referred to above. 4.4 RESULTS AND SENSITIVITY ANALYSIS The results obtained from the base case analysis are presented in Table 4-6, The entries in this table are the net present values of the life cycle costs of the entire Unalaska electrical generation system. Thus each entry in the table represents the figure of merit for an entire scenario. The All Diesel scenarios include the capital costs for adding new diesel generation capacity as required by increases in the load over time. All fuel and operating costs are also included. The costs for each year from 1988 to 2016 are discounted (at 3.5 per- cent real discount rate) back to 1986. The geothermal scenarios include the capital and operating costs for generating systems which include both geothermal and diesel capacity. Thus comparison of the entries within each row of Table 4-6 provides a convenient means of determining the merits of the scenarios under identical assumptions. Within each row the only differences are the installed capacities of geothermal generation (0, 5, 7, 9.5 MW). Table 4-6 shows the results of each of the geothermal size classes under the three diesel cost trends and for the two load growth trends. The group of scenarios utilizing’ the Base Case load growth is shown in the upper portion of the table. The High Case load growth trend is shown at the bottom. As can be seen from this Table 4-6, the scenarios which contain geothermal capacity almost all have a lower net present value than the comparable All Diesel scenario. The only exception is under the low diesel price trend . Utilizing the Base Case assumptions of medium diesel price trend, base case load growth, as well as all the other assumptions detailed in Section 4.1, the 5 and 7 MW Geothermal Scenarios are lower in cost than the All Diesel scenario. In net present value terms, the least cost geothermal scenario, the 7 MW case is $3.9 million (or 5 percent) less expensive than continued reliance on diesel generation. MLF6/C30 4-11 TABLE 4-6 ECONOMIC COMPARISON OF ALTERNATIVE GENERATION SCENARIOS FOR UNALASKA BASE CASE LOAD VS. HIGH CASE LOAD DIESEL DIESEL DIESEL ALL- + 5 MW + 7 MW + 9.5 MW SCENARIO DIESEL GEOTHERMAL GEOTHERMAL GEOTHERMAL 1986 Present Worth of Scenario 1988-2016 (Million 1986$) BASE CASE LOAD GROWTH DIESEL PRICE: MEDIUM 77.6 74.5 73.7 78.1 LOW 63.6 68.8 69.5 75.1 HIGH 91.7 80.3 77.8 81.0 HIGH CASE LOAD GROWTH DIESEL PRICE: MEDIUM 99.4 91.5 86.9 90.6 LOW 81.9 83.3 81.5 85.5 HIGH 117.0 99.8 92.2 95.7 MLF7/CT9 The economic feasibility of geothermal development is quite robust, as revealed by sensitivity analyses conducted with respect to diesel prices, load forecast, planning horizon, discount rate, and other variables. With the excep- tion of the low diesel price trend assumption, all other sensitivity tests indi- cate that some geothermal development is economically feasible. 4.4.1 Diesel Price Sensitivity As noted in Section 4.1.1, the low diesel price trend is not a most likely case, but rather a lower bound estimate of future price trends. Recall that this assumption has diesel prices rising gradually to $0.90 per gallon by 2006 and remaining constant thereafter. It is not surprising to find that geothermal development cannot compete under this assumption. If the medium or high trend (high diesel prices) is assumed, geothermal development appears more economic than the All Diesel scenario. Under the base case load forecast and low diesel trend, the 7 MW geothermal development has a $5.9 million (9.3 percent) higher net present value. Of course, assuming the high diesel price trend further adds to the favorability of the geothermal alternatives. Under the Base Case load and the high diesel price trend, the 7 MW scenario saves 21 percent compared with the All Diesel scenario. 4.4.2 Load Growth Sensitivity Generally, higher loads favor geothermal development in that they provide greater utilization of the fixed costs of geothermal generation resulting in marked economies of scale. By comparison, diesel generation costs exhibit less economies of scale because they are tied to fuel costs per KWH which remain fairly constant. This tendency is apparent from Table 4-6 in that the high load growth case further increases the the gap between the Geothermal scenarios and the All Diesel scenarios. The 7 MW Geothermal scenario has a slightly lower present value than All Diesel, even under the low diesel price cases. Under the medium and high diesel price high load growth scenarios, the geothermal scenarios appear to have an even stronger advantage over the All Diesel scenario. MLF6/C30 4-12 4.4.3 Discount Rate Sensitivity een Table 4-7 shows the same scenarios as Table 4-6 recalculated to show the effects of a high (4.5 as opposed to a 3.5 percent) present value discount rate. Because the costs for the development of geothermal are front-end costs as com- pared with the diesel fuel costs which are spread out over time, the higher discount rate favors the All Diesel scenarios. However, it is interesting to note that the ranking of the scenarios within each row remains almost constant over the change in discount rates. Because of increased front-end weighting on capital expenditures, the larger scale of the geothermal scenarios (7 and 9.5 MW) development offer less savings relative to the 5 MW scenarios. The low diesel price scenarios more strongly favor all diesel cases. The high and medium diesel trends still favor geothermal with the higher discount rate, although the cost savings are reduced. 4.4.4 Project Life Sensitivity Extending the useful life of the geothermal plants would improve their advantage over all diesel scenarios. A comparison between Tables 4-6 and 4-8 illustrates the effects of extending the project life for 34 years or through 2025 (instead of 2016). This comparison is biased in favor of geothermal devel- opment insofar as the cost of replacement of wells and generating equipment is not included. This assumption favors geothermal development, particularly the 9.5 MW development. Under the base case project life, some $13 million dollars of investment, which represent the last 2.5 MW of capacity added, is only utilized for 15 years (from 2001 to 2016). Under the extended project life assumption, this capacity would be utilized for the full 25 years. This increases the cost savings for the 9.5 MW scenarios to the point where they, rather than the 7 MW scenarios are preferred under the Base Case load forecast. 4.5 ECONOMIC CONCLUSIONS AND RECOMMENDATIONS Because the scope of the present study is limited to an economic feasibility and does not encompass tests of financial feasiblity, a firm recommendation cannot be made with respect to the desirability of geothermal development at Unalaska. Within this constraint, however, it may be concluded that geothermal development does pass a reasonable economic feasibility test and that financial feasibility investigation is definitely warranted. MLF6/C30 4-13 TABLE 4-7 SENSITIVITY ANALYSIS PRESENT VALUE DISCOUNT RATE 4.5 PERCENT DIESEL DIESEL DIESEL ALL- +5 MW + 7 MW + 9.5 MW SCENARIO DIESEL GEOTHERMAL GEOTHERMAL GEOTHERMAL 1986 Present Worth of Scenario 1988-2016 (Million 1986$) BASE CASE LOAD DIESEL PRICE: MEDIUM 67.4 67.8 67.7 68.6 LOW 55.4 62.9 64.2 66.0 HIGH 79.3 72.7 71.2 71.2 HIGH CASE LOAD DIESEL PRICE: MEDIUM 86.3 82.6 79.3 82.7 LOW 71.4 75.6 74.7 78.2 HIGH 101.3 89.6 83.8 87.1 MLF7/CT2 TABLE 4-8 SENSITIVITY ANALYSIS EXTENDED PROJECT LIFE DIESEL DIESEL DIESEL DIESEL + UNPHASED ALL- +5 MW + 7 MW + 9.5 MW 7 MW SCENARIO DIESEL GEOTHERMAL GEOTHERMAL GEOTHERMAL GEOTHERMAL 1986 Present Worth of Scenario 1988-2025 (Million 1986$) BASE CASE LOAD DIESEL PRICE: MEDIUM 102.3 87.7 85.8 83.9 84.5 LOW 82.0 79.5 79.6 79.9 78.6 HIGH 122.6 96.0 91.9 88.0 90.4 HIGH CASE LOAD DIESEL PRICE: MEDIUM 130.8 110.2 102.9 102.0 101.5 LOW 105.6 98.4 95.69 95.1 94.1 HIGH 156.0 122.0 109.9 108.9 108.8 MLF6/CT3 The test of economic feasibility which the proposed geothermal development passed was rigorous. The comparison with the all diesel alternative was rigorous in that a high diesel efficiency is assumed, and that even the medium diesel price trend only results in a diesel prices rising to $1.32 per gallon (2006 through the end of the analysis). Furthermore, the diesel costs assume no additional structures or fixed operating costs despite a projected fivefold increase in installed capacity. The geothermal cost estimates used are conser- vative in that conservative design criteria are used, and cost estimates assume full union scale (Davis-Bacon Act) wages. In addition, the drilling cost esti- mates allow for a 20 percent contingency allowance. Although the 7 MW geothermal capacity appears to offer the greatest savings, the choice between the 5 and 7 MW geothermal development is too close to call based on the economic analysis alone. That choice should be based on the finan- cial analysis as well. The 9.5 MW capacity appears to be less desirable than the 7 MW plant under the most likely conditions. However, because the 9.5 MW plant is incremental to the 7 MW development, because it occurs later in time and involves a separate mobilization, it is reasonable to defer this decision until the first phase of geothermal development is in place. If, toward the end of the century, diesel prices have risen to the high trend levels or demand has increased to the high case loads, it would be reasonable to consider this additional level of development. Additional reservoir performance data will also be available at that time. It is therefore recommended that the financial analysis of the 5 and 7 MW geothermal alternatives be undertaken. Time is of the essence in that the load on the Unalaska utility is forecast to rise dramatically in the next few years, and that construction and especially drilling prices are currently at a very low point. If a decision to proceed is reached within the next year it may be possible to purchase rather than rent a drill rig at considerable cost savings. MLF6/C30 4-14 5.0 CONCEPTUAL DESIGN 5.1 PROJECT DESCRIPTION Geothermal power plants must be located at or very near their respective geothermal resources. Transporting the geothermal fluids, of medium or low enthalpy, for extended distances of more than one mile or so, becomes tech- nically difficult and expensive. In almost all cases, geothermal resources are found in remote locations from populated areas. The Makushin geothermal resource, however, is relatively close to and accessible from Unalaska/Dutch Harbor. The geothermal power plant in this project can be constructed for unattended operation and therefore provisions are made for its monitor and control from the existing power plant. The roads provided to the site make it accessible for periodic maintenance and to manually re-start the plant after automatic shutdown in the event of system outages. The plant's cycle using flash turbines and binary units as shown in Figure 5-1 and Figure 5-2* is selected with the objective of optimizing the economic use of the resource. The selection of materials, systems equipment rating and redundancies is consistent with the unattended operating criteria and in accordance with plant availability objectives in the range of 75%. Where geothermal power plants provide the base load block of the system demand, other types of generation generally supply intermediate and peak demands. The base load generated by geothermal plants permits the desired constant well flow of geothermal fluids. Substantial variations in well flow rates affect the long-term integrity of geothermal wells. In this project, the geothermal power plant may be operated to supply a large proportion of the system demands and thus follow the load. Therefore, to maintain constant well flow while the load varies (reduces from maximum demand) the surplus steam will be vented. *Figures 5-1 to 5-5 are oversize drawings located in an envelope at the back cover of Volume I. MLF7/G 5-1 The initial major construction phase of the project will be the drilling of production wells. Please refer to the project development schedule (Figure 5-6). Two production wells and one injection well will be drilled before pro- ceeding with work on the power plant. Two production wells are considered Necessary to insure that the plant can continue in operation even if one well becomes inoperative for any reason. The production wells are to be drilled on a plateau located on the south side of Fox Canyon and the injection well is to be located on a similar plateau on the north side of the canyon. The pipeline from the production well to the plant site must go down the side of the canyon and back up the other side to reach the power plant site. Since steam and water mixtures cannot readily be made to flow down steeply inclined pipes, it is necessary to separate the steam and water and deliver each to the power plant in separate pipelines. The power plant should be located in a suitable site anywhere along the pipeline's route leading from the production well to the injection well. A plant site close to the injection well site will have vehicular access at all times and is, there- fore, the preferred site location. The most economically attractive geothermal power plant is a hybrid plant comprised of two 2,500 kW steam turbine driven generating units and two 1,000 kW binary cycle electric power generating units designed to produce 7,000 kW from a single production well. After the resource is used in the power plant all of the effluent is injected in a well drilled for this specific purpose. Equipment will be delivered to a Driftwood Bay landing site by a barge which will be equipped to unload the materials and equipment directly onto trucks at the beach. These trucks will then deliver the materials to the plant site via the road previously constructed for delivery of the drill rig. The two flash steam turbine driven electric generating units will be constructed as the first phase of power plant construction. The binary cycle units could be built during the initial construction phase or added when electric loads have increased sufficiently to justify their ‘installation. Figure 5-1 shows the equipment arrangement for the flash steam plant and Figure MLF7/G 3-2 -¢ ein8Ty PROJECT DEVELOPMENT SCHEDULE 1987 1988 1989 1990 \ 1991 Project Apprv. jae EIR ss a a Engineering | Civil Works Driftwood Bay Staging Area Construction mune Driftwood Bay Access Road Construction — Drill Pad Construction — Well Drilling (3W) ea Broad Bay Access Road Construction ~ —— Power Plant Site Work -_—— Equipment Manufacturing and Shipping -~--- a el Equipment Installation — at Transmission Line Construction — OH, UG and SM ‘ COMMERCIAL OPERATION 5-2 shows the equipment arrangement for the binary cycle portion of the plant, Figure 5-3 is a P&ID for a flash steam generating unit and Figure 5-4 is a P&ID for a binary cycle unit. Figure 5-5 is a single line electrical diagram showing the generation and transmission systems. 5.2 BASIS FOR DESIGN 5.2.1 Buildings The power plant buildings will be of the preengineered type. Building framing will be structural steel and the outside walls and roof will be in- sulated sandwich-type panels. Fiberglass skylight panels will be installed in the roof and will provide some natural lighting during daylight hours. The buildings will be designed to withstand wind velocities of 110 MPH and in accordance with UBC requirements for Seismic Zone 4. The building housing the turbine generator units will be approximately 40 feet by 120 feet in length. An overhead bridge crane will be provided to facilitate maintenance of the turbine generator sets and other equipment. The building housing the binary cycle units will be of the same type of construction but will be slightly smaller. This building will be approximately 30 feet wide by 130 feet long. 5.2.2 Major Equipment 5.2.2.1 Steam Turbine Generator Set The steam turbine driven electric power generator set will be mounted on a fabricated steel base which will serve as a reservoir for its lubricating oil. Lube oil pumps, strainers and piping will all be factory installed on the tur- bine generator set, minimizing field labor. The turbine will be a multistage machine with an inlet pressure of 60 PSIA and a design exhaust pressure or 3.5 inches of Hg. Abs. The exhaust will face vertically upward to minimize foun- dation and piping costs. The turbine may be a high speed machine employing a speed reduction gear or it may be a 3,600 RPM machine directly driving a two pole generator depending upon which type is less expensive to manufacture. If a high speed geared turbine is used to drive the generator, the generator will be a less expensive four pole machine. In either case, the generator will employ a brushless exciter and will be of the air cooled type. Generator voltage will be MLF7/G 5-3 4,160. The turbine generator set lube oil coolers will be located outside of the building and will be air cooled, fin tube type units. 5.2.2.2 Steam Condenser The steam condensers will be air cooled, fin tube type. Thermostatically controlled motor driven fans will blow air across the tubes to condense the steam exhausted from the turbine. The condenser will be of the A frame type with steam entering a header at the top. Condenser tubes will form the legs of the A and steam and condensate will flow down the tubes to condensate collection headers at the bottom. Two stage steam jet ejectors will be used to remove non- condensable gas from the condenser. The ejector interstage cooler will be of the air cooled fin tube type. All wetted parts of the condenser and nonconden- sable gas removal system will be constructed of 304 stainless steel. 5.2.2.3 Condensate Pumps Condensate pumps will be vertical can type pumps. Motor drivers will be TEFC enclosure. All wetted parts of the condensate pipe will be 304 stainless steel. 5.2.2.4 Instrument Air Compressors The instrument air compressors will be of the single cylinder water cooled nonlubricated type. Cooling water will be pumped through an air cooled water cooler to dissipate the heat of compression to the atmopshere outside the power plant building. 5.2.2.5 Air Dryer An instrument air dryer of the desiccant type will be provided. The dryer will be of the dual tower heatless regenerative type. One tower will be in ser- vice while the other is regenerated. Automatic controls will switch the air flow from the active drying tower to the regenerated tower as the desiccant in the active tower becomes saturated with moisture. MLF7/G 5-4 5.2.2.6 Turbine Drain System A condensate drain tank and two condensate drain pumps will be provided to pump turbine drains to the condensate receiver from which they will in turn be pumped by the condensate pumps to the injection system. The drain pumps will be automatically started and stopped by a level control switch. 5.2.2.7 Binary Cycle Generating Units As discussed in Section 2, Ormat manufactures a modular binary generator unit which would be well suited to the proposed project. The secondary fluid used by Ormat is isopentane, a flammable hydrocarbon. Each generating unit would produce approximately 1,100 kW gross which would result in a net output of approximately 950 kW per unit. These modular units contain the isopentane vaporizer and turbine-generator unit mounted on a commmon base. The base is constructed in such a way that with exterior panels in place, an envelope is formed which serves as a container for ocean transport of the units. The air cooled condensers will be shipped and installed separately because they are much larger than the water cooled condensers normally employed in the design of these package units. The vaporizer feed pump is also shipped separately and mounted alongside the module served. With the exception of the air cooled condenser, the use of which is necesary for this plant application due to possible freezing conditions, the binary cycle plant is of standard modular design. Thus, the installation costs are minimized because most of the components are assembled into modular units at the factory. Piping from the evaporator to the turbine is also factory fabricated and installed. The complete lube oil system, except for air cooled lube oil coolers, is factory fabricated and installed. The lube oil coolers are remotely installed outside of the power plant building. The genera- tor is air cooled and employs a brushless excitation system. The binary cycle power fluid feed pumps are vertical can type and employ explosion proof motors as drivers. The binary cycle fluid condensers are air cooled, fin tube type, and employ fans to blow cooling air across the tubes. The condensers are located outside of the power plant building and dissipate the latent heat of condensation to the atmosphere. MLF7/G 5-5 Materials of construction for the binary cycle units are carbon steel because no oxygen will be present in the geothermal water used as a heat source and because the binary fluid (isopentane) is not corrosive. Fire protection for the binary cycle units will be provided by an automatic halon system. All electrical equipment in the binary cycle power plant building will be in explosion proof enclosures. Generator switchgear and motor control centers for auxiliaries employed in the binary cycle plant will be located in the building housing the steam turbine driven generating units. 5.2.2.8 Flash Generating Units The flash steam plant considered appropriate for the Unalaska electrical system is comprised of 2,750 kW multistage condensing steam turbine generating units. Each steam turbine generating unit produces approximately 2,500 kW net out of the plant. All vital auxiliaries such as condensate pumps, turbine lube oil pumps and instrument air compressors are provided with 100% capacity backup systems which will start automatically upon failure of the operating unit. This will minimize the number of outages experienced as a result of the failure of a plant auxiliary. The steam is condensed in an air cooled finned tube condenser. Fans are used to move the air across the tubes in order to achieve the optimum heat transfer rate. Condensate pumps remove the steam condensate from the con- denser and deliver it to the injection well. Noncondensable gases are removed from the condenser by steam jet ejectors and the gases are vented to the atmosphere or compressed to a pressure sufficient to permit their disposal in the injection well if disposal in the atmopshere is unacceptable. An air cooled condenser is recommended because of the difficulty of operating an unattended cooling tower in a climate where high winds and ice for- mation due to low winter temperatures would likely cause frequent damage to the cooling tower. Instrument and control air is provided by one of two 100% capacity nonlubri- cated water cooled reciprocating air compressors. Instrument ait is dried in one of two 100% capacity air dryers. Electric power is generated at 4,160 volts MLF7/G 5-6 and stepped up to 34.5 kV for transmission to the electric distribution system in the town of Unalaska. The plant is designed for automatic unattended operation. Normal startup and shutdown of the plant are done manually by the plant operating staff. Malfunctions of plant equipment result in an automatic shutdown of the turbine generator set affected or the entire plant depending upon the nature of the problem. When an automatic shutdown occurs it is necessary for the operators to go to the plant, correct the malfunction and restart and load the generator unit affected or the entire plant if both units are forced out by the malfunction. If the plant or a generating unit shuts down, an annunciator indicates which item of equipment failed first in order to assist the operators in locating the problem and correcting the trouble. The plant electrical output capacity is varied by bypassing steam and/or hot water around the generating unit and sending it to the injection well. In the case of a flash steam plant the steam is bypassed to the condenser and the condensate is injected along with the unflashed geothermal water. By varying the plant output capacity it will be able to match the demand on the electrical system. Since isopentane is flammable, the electrical switch gear and electrical motor control centers are located in a separate building. Thus these sources of ignition would be kept away from any isopentane which might escape from a generating unit. The generator employs a brushless exciter and a local control panel. The control panel enclosure is explosion proof. Any motors or other electrical equipment in the generator building also have explosion proof en- closures. Explosive mixture detectors are also provided and arranged to register an alarm and shut the plant down if a dangerous leak is detected. Exhaust fans are provided to remove any small leakage which may escape from the turbine shaft seals or governor valve stem seals. Isopentane piping is welded and flanged joints are kept to a minimum in order to reduce to the greatest extent possible the potential for leaks. MLF7/G 5-7 5.2.3 Access Roads Access for operations would be provided by an access road which would extend from Broad Bay via Makushin Valley, see Figures 2-1, 2-3 and 2-4. The route is indicated through points A, B, C, D, E. Route Section AB, about 3.6 miles long (Figure 2-4), would be built on flood basin alluvium that is underlain by lacustrine/lagunal sediments. The flood plain in this area is flat and marshy. The road will be built using light-weight aggregate encapsulated in a geologic membrane (Figure 2-6). There is about one mile of existing road at Broad Bay that could be improved and used as the first part of this section of the road. The proposed road will roughly parallel the existing road west to a construction point between the river and the valley slopes, then continue west but away from the river up the valley. The road will be severely constrained by the poor subsurface conditions. Present plans call for use of a light volcanic borrow material encapsulated in geomembranes and placed in a thin section (approximately 2 feet). The idea is to float the road and treat the surface course with a Percol additive for strength and durability. This embankment design is described in a later section of this report. The performance of this road will require monitoring during operations, and repairs will have to be made with care not to overload the poor bearing material. The small creek crossings along this segment will have to be bridged with open bottomed (arched) culverts. These are requirements of the Alaska Department of Fish and Game. Several borrow sources have been identified by the ADNR on the south valley wall, but they have not been evaluated in terms of a light-weight (floated) road system. It may be necessary to use an upland material source west of the valley. After crossing the north channel of the Makushin River the operational access road joins the old road as it leaves the valley bottom and switches back up the valley wall. This section of the old road has washed out in several sections and will require significant upgrading and repair. The ADNR report describes the repairs necessary in sufficient detail for planning purposes. Repairs entail placement of culverts and the addition of rockfill to washed out or slumped areas. No major redesigns are suggested for this section but some MLF7/G 5-8 maintenance should be factored into long-term operations. Material Source 12 is located within the switchback section and it will provide adequate rockfill. Section BC, 2.7 miles long (Figure 2-4), is easier to build than Section AB except that it will require two bridges for the two major river crossings in the area. Good aggregate materials for road construction are available in this section. Parts of road sections AB and BC are shown as close as 100 feet from the river. This alignment could be modified if it were necessary for environmental reasons, It will require, however, about four additional bridge river crossings. Section CD, 2.4 miles long (see Figure 2-2), follows the existing road and it will require major repair work. The road ascends a low alluvial fan before it goes up into Driftwood Bay Valley with an average slope of 10 percent. Section DE, as noted above, would be all new construction. All roads may be built 20 feet wide and the Section FDE should be designed for AASHTO Truck Load HS-20, limiting the maximum grade to 10 per- cent. The roads will have no special pavement and would be built from the aggregate existing at the route site except for road Section AB, which will have a special Percol pavement and light-weight aggregate imported from the Driftwood Bay area. Section DE, 3.0 miles long (Figure 2-2), would be all new road construction and it would go north of Sugarloaf Cone. Road construction would progress rapidly across this area. The road would have to cross several, perennially snow filled ravines. Bridges will be required at these crossings. Extensive maintenance work is expected in this area due to snowfall. Bridge abutments may be constructed with reinforced concrete supported on wood pile, as required. The selection of the crossings will be made in accordance with environmental constraints. All bridges could be constructed with steel deck and reinforced concrete abutments supported on wood piles, if necessary. MLF7/G 5-9 5.2.4 Transmission Facilities ee ane The routing proposed through the Makushin Valley was chosen after aerial reconnaissance of two other overland routes. The severe wind conditions of the area coupled with the rough terrain makes access to the line with men and equip- ment a primary consideration in the selection of the transmission line route, The use of helicopters for line maintenance and repair was ruled out as imprac- tical (especially during severe weather conditions, which are the conditions more likely to cause damage to the line). The selected line route through the Makushin Valley will be maintained from the same access road built for use by the power plant operation and maintenance crews. The proposed overhead line will have conductors capable of carrying up to 10,000 kW of capacity. The generation voltage (4,160 V) will be stepped up by means of a station transformer to 34 kV. The structures would be of the "H" frame type and of approximately 600 foot ruling spans. The overhead portion of the line will be built as far into the Makushin Valley from the plant site as would be permitted by soil conditions to support the wood piles of the "H" fra- mes, At this point, the overhead line would be connected to the second section of the line which is comprised of an underground cable rated at 60 amps and 34 kV. A preliminary calculation was performed to select the typical span using a wind pressure of 31 pounds per square foot (PSF) produced by a wind velocity of 110 miles per hour and ice load of one inch radius on a conductor "CATBIRD" ACSR/AW of 954 KcMILLs. Under these conditions the line could be built using H-1 75 foot poles with 600 foot spans. Capacitors and lighting arresters will be installed at the point of tran- sition from overhead line to underground cable to protect the underground cable. The underground cable would be three conductor 750 MCM of the direct burial type and would be installed adjacent to the Makushin Valley access road. The third section of the 34 kV line will be comprised on the submarine por- tion. This cable will be 3 conductor, 750 MCM, 34 kV insulated cable that will MLF7/G 5-10 be laid on the bottom of Broad Bay and will be terminated at Dutch Harbor at a new switch and thence to the existing 34 kV underground cable. 5.3 PLANT CONSTRUCTION COST Table 5-1 is a summary of the capital cost for the various alternative plants considered. The costs shown are in 1986 dollars. Appendix E contains the detail costs for materials and labor required for construction of each alternative power plant. 5.4 OPERATION AND MAINTENANCE 5.4.1 Operation The power plant will be designed for automatic unattended operation. Plant start-up will be done by operators who will parallel the generators with the Unalaska distribution system. The geothermal turbine generator sets will take increased load as the electric system load increases. As electric system load decreases the geothermal units will shed load. Steam will be bypassed to the condensers and hot water will be bypassed to the injection well. Condensate from the steam condenser will be pumped into the pipe conveying the geothermal water to the injection well. In the event of a failure of a plant auxiliary such as a condensate pump the standby pump will be automatically started. The standby auxiliary will be of equal capacity to the unit for which it backs up so there will be no reduction in electrical output of the plants when a spared auxiliary fails. In the event that a necessary standby auxiliary fails to start the turbine generator unit will be automatically shut down. The reduction in electrical generating capac- ity will be made up by the remaining geothermal generating units if sufficient unused capacity is available. If the load exceeds the remaining geothermal plant capacity, diesel engine driven generators in the Unalaska power plant will pick up the additional electrical load. If the entire geothermal power plant is shut down, diesel engine driven generators located in the power station in Unalaska/Dutch Harbor will carry the electric load until the geothermal plant can be repaired and returned to operation. MLF7/G 5-11 TABLE 5-1 UNALASKA GEOTHERMAL POWER PROJECT CAPITAL COST COMPARISONS (THOUSANDS OF 1986 DOLLARS) DESCRIPTION Field Development Costs: Production Wells (1) ) Reinjection Wells (1) ) Spare Production Well ) Access Roads: Broad Bay to Wells Site Driftwood Bay To Power Plant Pier at Broad Bay Transmission Line: Overhead Underground Submarine Site Prep Foundations Facilities: Buildings Substation and Switchyard Power Plant Equipment and Installation Spare Parts Engineering and Procurement Construction Management PROJECT COST 5,000 kW FLASH STEAM 5,085 1,993 911 687 2,289 1,739 1,739 122 600 223 1,099 15,871 250 1,364 545 34,517 5,000 kW BINARY CYCLE 5,085 1,993 911 687 2,289 1,739 1,739 122 739 223 1,099 16,227 150 1,388 555 35,946 7,000 KW HYBRID 5,085 1,993 911 687 2,289 2,138 2,138 122 624 264 1,149 20,787 300 1,655 662 40,804 9,500 KW HYBRID 8,918* 1,993 911 687 2,881 2,287 2,287 122 765 411 1,297 28,114 350 2,088 835 53,943 *Requires one additional production well and one additional injection well. The failure of the plant auxiliary will be registered on an annunciator located in the geothermal power plant. The failure will also be transmitted from the geothermal power plant to a data collection center in Unalaska where the geothermal power plant operation will be monitored. It will, however, be Necessary to send an operator or maintenance person to the geothermal power plant in order to determine the reason for the equipment failure and to make any needed repairs. Operators will need to visit the plant two or three times each week, weather permitting, to check the operation of all of the plant equipment. During these visits the plant operators will also perform routine maintenance and service on the power plant equipment. A dock with a small crane will be constructed at the end of the power plant access road. This dock will be protected by a solid piling breakwater. Operators will cross Broad Bay from Unalaska in a diesel powered work boat (+30') dedicated to the power plant. The boat will be capable of delivering small maintenance equipment, lubricants and operating supplies. Access to the power plant will be by road up the Makushin Valley. An all terrain vehicle will be used to negotiate the road which will be snow covered during most of the winter months. The vehicle will be stored in a small garage located on the shore at Broad Bay. The work boat will also be capable of transporting the all terrain vehicle used for land transport to and from the power plant to the Town of Unalaska for repairs and maintenance. Access to the well head and the steam-water separator which will be located on the south side of Fox Canyon will be by a foot bridge built on and supported by the pipe supplying steam to the power plant. Since the well can operate con- tinuously even if the power plants shuts down, there will only need to be oc- casional visits to the steam water separator and well head. These inspections will assure that the level control valve on the steam water separator is func- tioning and that the piping systems have not developed any leaks. 5.4.2 Maintenance Routine maintenance on spare auxiliaries will not require the plant to be shut down and can be scheduled at any time of year. Turbine and generator main- tenance, however, should be scheduled to coincide with periods of minimum MLF7/G “15-12 electrical loads. It is anticipated that each turbine will require an annual overhaul which will mainly consist of cleaning the stationary and rotating blades. Governor valves will also require annual maintenance. Turbine maintenance will be facilitated by an overhead crane installed in the power plant building. The power plant building will be large enough to house the vehicle used by the operators and maintenance personnel. Also included in the power plant building are sleeping quarters which will be used by operators or maintenance personnel when weather precludes returning to Unalaska after work at the plant. Maintenance for all equipment will be done in the power plant building, therefore sufficient floor space is included for the maintenance activity. 5.5 PROJECT CAPITAL COST ESTIMATE A comparison of Capital Costs for each of those alternatives selected for the purpose of this study is provided in Table 5-1 and summarized below: The three alternatives are: (1) 5,000 kW Flash Steam Geothermal Plant $34,517,000 (2) 5,000 kW Binary Cycle Geothermal Plant $35,946,000 (3) 7,000 kW Hyrbid Geothermal Plant $40 , 804,000 (5,000 kW Flash Steam + 2,000 kW Binary) (4) 9,500 kW Hyrbid Geothermal Plant $53,943,000 (7,500 kW Flash Steam + 2,000 kW Binary) The cost estimates include power plant costs, infrastructure such as access roads and transmission line costs, and a docking facility at Broad Bay. Also included in the estimate are engineering, procurement, construction management and the cost of a 45 man construction camp. Unit-rates used in this study estimate reflect the material and labor costs of the fourth quarter of 1986. No escalation is added. A labor productivity factor of 30% was added to the craft manhours in order to account for the decreased productivity expected at remote construction sites. MLF7/G 5-13 TABLE 5-2 ANNUAL OPERATION AND MAINTENANCE COST SUMMARY DESCRIPTION PLANT TYPE AND CAPACITY FLASH STEAM HYBRID 2.5 MW 5 MW 7.0 MW 9.5 MW Labor $ 30,000 $ 60,000 $ 72,000 $102,000 Parts 60,000 100,000 132,000 182,000 Transportation 20,000 20,000 20,000 20,000 Dock Maintenance 1,000 1,000 1,000 1,000 Total Maintenance $111,000 $181,000 $225,000 $305,000 Operating Cost! 172,000 172,000 172,000 172,000 Total Operation & Maintenance Cost $283,000 $353,000 $397,000 $477,000 1. Includes operators' wages and fringe benefits, consumable supplies, land vehicle repairs, work boat repairs as well as fuel and lubricants for both the land vehicle and work boat. MLF7/GT Labor rates used in the estimate represent composite crew rates which also include fringes, payroll taxes, and overtime premiums. A 5% contingency was used on all costs except for drilling costs which include a 20% contingency. 5.5.1 Additional Assumptions It was assumed that Contractor labor will be drawn from Anchorage, Alaska union halls. Accordingly, labor rates used in this estimate are composite crew rates based on Anchorage, Alaska labor union agreement. The work week is based on six 10 hour days except for drilling which is conducted on a 24 hour per day basis, seven days per week. All material and equipment is priced F.0.B. Seattle, Washington. Barging costs for all materials, construction equipment and man camp are included in the estimated based on $280/ton of shipping weight. Borrow material, sand, gravel and rocks as needed will be available to Contractor on the island free of charge. Demobilization will occur 12 months after mobilization because of the barge schedule. Standby time for construction equipment while it is waiting for retrograde barge trip is included with distributed costs. Project cost of each of the alternatives except for the 9.5 MW hybrid plant assumes that the entire project will be constructed at one time, and without any interruptions. Drilling precedes all construction activities except for the road from Driftwood Bay to the power plant site. 5.5.2 Exclusions Items not included in this study estimate are: (1) Escalation (2) Financing costs (3) Permits/fees MLF7/G 5-14 (4) Environmental study costs (5) Hos abatement costs. MLF7/G 5-15 6.0 CONCLUSIONS AND RECOMMENDATIONS 6.1 TECHNICAL CONSIDERATIONS A geothermal power plant utilizing the Makushin reservoir is technically feasible for plant sizes up to 9.5 MW capacity. The preferred location for the production well pad is at the site of the test well drilled during the explora- tion program conducted by Republic Geothermal on behalf of the Alaska Power Authority. This site offers an almost certain probability of obtaining a com- mercial scale production well. The preferred location for the power plant is on the opposite (north side) of the Fox canyon. Due to the difficulty of conveying a steam-water mixture down steeply inclined pipes, a steam water separator is recommended at the well site. Although this site requires a pair of pipelines across the canyon (one for water and one for steam) the greater ease of access to the powerplant afforded by this site offsets the cost of the pipeline. Access to the powerplant during construction will be via a road from Driftwood Bay. This approach utilizes a largely usable existing road. The only New portion of the road will be a segment from Sugarloaf to the plant site. Access to the well site will be via helicopter. No bridge across Fox Canyon is anticipated, Access to the powerplant site during operations will be via a dedicated crew boat which will be used to cross Unalaska Bay from the city of Unalaska to Broad Bay. A small dock will be built at the south side of Broad Bay. A road will be built from the dock along the edge of Makushin Valley to the base of an existing switchback road. Part of the Makushin Valley road will be built using an encapsulated fill design which will float on the marshy valley floor. The least cost powerplant configuration includes two 2.5 MW single flash steam generation units. An additional 2 MW of binary capacity can be added to this flash plant utilizing the hot water which is not flashed to steam by the two 2.5 MW flash steam units. If additional capacity is needed a third 2.5 MW unit can be added. This addition requires a second production well and a second injection well. The transmission route will generally follow the road from the plant site to Broad Bay. This will facilitate transmission line construction and maintenance MLF6/C43 6-1 the line from the plant site to the lower third of the valley will consist of 34 KV conductors on wooden H-pole towers. At this point the line will continue via three 34 KV underground cables to Broad Bay. There the line will connect to a three conductor 34 KV insulated submarine cable which will terminate at Dutch Harbor. It is technically possible to design a power plant for remote operation. This can be accomplished by setting the plant up to run at full power and to bypass unused steam to the condenser. The plant can shut down automatically in event of malfunction, Start-up will require manual operation. Power plants cannot operate for long periods of time in an unattended mode without forced outages. Such forced outages will result in lower plant availability factors than for attended power plants of the same type. Remote monitoring of plant operations can detect problems, but corrective measures must be taken by opera- tors or maintenance personnel on location. Because the geothermal power plant will operate unattended most of the time, the Unalaska power system must main- tain sufficient back up diesel electric power generation to satisfy all elec- trical loads which cannot be curtailed for several days' duration without hardship to power consumers. Under the proposed design it is anticipated that the forced outage rate for the geothermal plant will not exceed 8 percent of the time. Planned outages for maintenance will occur about 17 percent of the time. Disposing of spent geothermal fluids in the Makushin River or its tribu- taries is precluded by the presence of arsenic in concentrations too high to rely on dilution to provide acceptable in-stream concentrations. Disposal via pipeline to Broad Bay or Driftwood Bay is prohibitively expensive. An injec- tion zone located on the same plateau as the plant site offers the preferred brine disposal option. 6.2 ENVIRONMENTAL AND PERMITTING CONSIDERATIONS No environmental constraints are envisioned within the context of the proj- ect concept proposed in this report. The most significant potential environ- mental impact concerns road and transmission line construction within the Makushin Valley and possible effects on fish resources within the Makushin River and its tributaries. Close coordination with the ADF&G is essential during MLF6/C43 6-2 design and construction. Assuming that such coordination is conducted, and mitigation of potential impacts are incorporated into the construction plans, no serious adverse impacts are anticipated. Permits for the project can probably be obtained in six months to a year assuming that an EIS is not required. An EIS should not be required assuming that the above mitigation measures are undertaken and that the geothermal brine is injected rather than disposed of in surface streams or the ocean. A pre- application agency coordination meeting should be one of the first steps in project development. Air quality modeling indicates that the worst case one hour emissions from the plant will exceed Alaska's strict odor-based standard for reduced sulfur compounds, There is a very good rationale for the ADEC to grant a variance, however, because of the remoteness of the site, ambient conditions natural to the area, and because there would not be a health risk from the anticipated con- centrations of hydrogen sulfide. Waste heat utilization is problematic due to environmental concerns and to cost. Use of the waste heat at the plant site is impractical due to its remote location. Transportation of the brines from either Broad Bay or Driftwood Bay via pipeline would be prohibitively expensive and would considerably increase the permitting complexity of the project. At Broad Bay the marshy ground con- ditions and sensitive environment make aquiculture or horticulture dubious. The best option for waste heat use rather is to utilize the off-peak electrical capacity of the geothermal plant for interruptible uses such as space heating or dispatchable hot water heaters. 6.3 ECONOMIC CONSIDERATIONS A rigorous economic feasibility test indicates that a geothermal development of 5, 7 or 9.5 MW would be feasible utilizing the Makushin Reservoir. Generation systems which include geothermal capacity in the configuration pro- posed under Section 5 of this report appear to offer cost saving in excess of $3 million over the period from 1988 to 2016 except under the assumption of APA's low diesel price trend. However, a decision to proceed with geothermal develop- ment must be preceded by a financial feasibility analysis which includes the MLF6/C43 6-3 effects of debt financing on the proposed project. This financial analysis will help to determine the optimal plant capacity. Based on the economic feasibility analysis a 7 MW plant offers the greatest cost savings. 6.4 RECOMMENDATIONS 1. MLF6/C43 A financial feasibility analysis should be undertaken as soon as is practical. Load growth in Unalaska is expected to increase at a rapid rate over the next five years. If a geothermal development is to be economically feasible it must utilize the resource before investment in diesel capacity is committed to serve the burgeoning demand. The project concept recommended in Section 5.0 of this report should be the basis for design and construction bids. Other concepts were rejected as either more costly or more environmentally damaging. Final siting and alignment for the Makushin Valley road, transmission line and dock facilities should be determined in the field with input from ADF&G and other interested agencies. As soon as a decision to develop is reached, or possibly before if development is considered likely, it would be prudent to begin the environmental approval process. This will avoid a situation in which the narrow development season is missed due to permitting delays. An additional incentive to prompt action in geothermal development is the presently depressed construction and drilling costs. If a decision to develop is reached during the current depressed drilling market it may be possible to obtain a drill rig at a fraction of its book value. This rig could be committed to the project offering considerable saving over day-rate rental. AH IZ? HOT WATE FROM PRODUCTION WELLS TUBE REMOVAL ffACE a\¢ HoT WATERL To PAE HEATER ZPACE HEATER. (TYR oF 4) HER Ree _ |4 DPEN TANE _ FEED PUMPS CLieNT ALASKA POWER AUTHORITY eed PROECT WO DRAWING NO. ae APPvO. bare OS Ecion Pn A Bee ee ( OCALE | DO NOT SCALE DRAWING | a/c". | O" |oueeT or 7 v I ao Lalas a INSTRUMENT AIR CAMPREOSOR. W/ RECEIVER. 4 AIR DRYER TRANSPOMER Db). ABSAR. ¢ Moc UNIT FLOOR. TRENCH FoR ELECT. CABLES pr AENERATIOR SW. SEAR ¢ Moco FoR BINARY CYCLE it 20? STEAM FROM PRODUCTION WELL STEAM | | SPACE HEATER (TYP OF AD RECEIVER TANK, TEAM JET EIECTORS To CONDESATE RECEIVER — eee et eee OLieat ALASKA POKER AUTHORITY 2'? BIECTORS 212 CONDENSATE INJECTION WELLS 3030 Patrick Henry Dr., Santa Clara, Ca., 95050 ZS0OKW FLASH STEAM PLANT UNALACEKA CAEOTHERMAL PROJECT CAENERAL ARKAN@ZEMENT ee beiag. |” Flew S| ae a DO NOT SCALE DRAWING ae) ce ee H i i i | j } TO ATMS. TO FLASH STZAM UNIT No.2 STEAM JET BIECTORS INTER CONDENSER STEAM WATER SEK Toe HIS Mbt _PRocucyion N@eL CONDENSATE PUMPS T PI - x i v : ial A xX TO BINARY FROM BINARY CYCcLle UNITS fi | CYCLE UNITS ><] $-+4><t--J re) ~TRITIEF,| TORS VSS WE P= Ft CJ _SPARE PRODUCTION | | gS WELL | FEINJIECTION WELL terion sntnntnenperninme tert onhmpamnestanteih snide rte TOTAL ~ 550,000 = eee erennonni PRESSURE( IA) BO PSIA | sekde lpceiiaai eehets AAA a at eras emacecc ale eu TEMPSRAT RECE BNTHALPY ATU/LB 1 335.7 al. LINS SIZE NRE | IG K ASOUMS SATURATZO STBAM, Adi: SAI ENGINEERS, INC. 3030 Patrick Henry Or. Santa Clara Ca. 95050 UNALASKA GEOTHERMAL JOSGmD by oaTe en? Pp RO J G He . fea = SINGLE FLAGH 2500 KW GENERATOR i 5 ae —e Fig. 5-3 fo _| CHK'D GY} APPROVED NO. r DESCRIPTION DESCRIPTION —__ : : _PREP'O ey i i oo z= REVISIONS REVISIONS PREP'O BY CHK'D BY | APPROVED ak PI; ee / FROM FLASH pec ( STEAM PLANT —1 \ g J AIR COOLED prmnerey — Ce CONDENSER | cb Lie eee vst) Se TURBINE VENT TRIP TURBINE TRIP | ISOPENTANE CONDENSATE RECEIVER. TURBINE GENERATOR SET TURBINE | TRIP = poy —{ TT | os Lo wn V7 TSH) (PSL (1) (Pr ) a | E Ne | on eee ee ee uae | 2 a | ISO PENTANE = | STORAGE ISOPENTANE Gl L L L FEED PUMPS A A A ISO PENTANE 2 REMOVAL PUMP b Foe tet fev < ~ ae STS RENIECTION @ . 7 WELL ISOPENTANE a fee UNLOADING AND TRANSFER PUMP FROM BINARY UNIT No. 2 To BINARY _ UNIT No. 2 Bak SAI ENGINEERS, INC. 3030 Patrick Henry Or., Santa Clara Ca. 95050 UNALASKA GEOTHERMAL PRoJdec 7 BINARY CYCLE P ¢ ID PREP'D BY DESCRIPTION REVISIONS