Loading...
HomeMy WebLinkAboutGeothermal Development in AK- An Engineering & Geologic Analysis 1984037 Géo 027 ol. 13, No. 3, pp. 241 — 264, 1984. 0375 — 6505/84 $3.00 + 0.00 t Britain. Pergamon Press Ltd. © 1984 CNR. GEOTHERMAL DEVELOPMENT IN ALASKA: AN ENGINEERING AND GEOLOGIC ANALYSIS M. J. ECONOMIDES and G. N. ARCE University of Alaska, Fairbanks, AK 99701, U.S.A. (Received May 1983, accepted for publication September 1983) Abstract—In spite of the vast geothermal potential within the state of Alaska, the economic feasibility is tenuous. Of the five sites examined in this paper, only Tenakee and Summer Bay have even marginal attractiveness for direct utilization. The geothermal reservoirs located at Copper Valley and Makushin Volcano may be feasible for power generation in the near and intermediate future. To make geothermal development feasible, an increase in the population/industrial base would be required, or a consolidation of the present power users. In general, the economic prospects of geothermal power development in Alaska are not attractive at this time, with the exception of Unalaska Island. 7 TNOARAS SN e a a Ksteet x BO CHAN POS DFS pir nrnrnrnrnrnrnrnrnrrneee edn oonrnrnnnn nono NOMENCLATURE annual distribution costs ($) annual fixed costs ($) annual labor costs ($) annual maintenance costs ($) costs per thousand BTU ($) heat capacity (BTU/Ib °F) annual costs of installed insulated piping system ($/mile) annual pump costs ($/mile) cash flow after taxes ($) cash flow before taxes ($) total outside diameter (including insulation) (in.) outside diameter of steel pipe (in.) inside pipe diameter (in.) depreciation ratio of total costs for fittings and installation to purchase cost for new pipe enthalpy rate (BTU/h) ambient (outside) heat transfer coefficient (BTU/h ft? °F) inside heat transfer coefficient (BTU/h ft? °F) total heat load for a town (BTU/h) interest rate investment ($) number of years of a viable project conductivity of insulation (BTU/h ft? °F/ft) conductivity of steel (BTU/h ft? °F/ft) annual fixed charges including maintenance expressed as a fraction of initial,cost for completely installed pipe length (ft) mass flow rate (Ib/h) a constant with value dependent on type of pipe. For steel pipes, n = 1.5 pump power requirement (kW) town population pump flow rate, gallons per min (GPM) rate of heat transfer (BTU/h) annual revenues ($) ambient temperature (°F) entering temperature (°F) exiting temperature (°F) overall heat transfer coefficient (BTU/h ft? °F) purchase cost of new pipe per foot of pipe length if pipe diameter is 1 in. ($/ft) 241 242 M. J. Economides and G. N. Arce INTRODUCTION Alaska potentially contains some of the largest geothermal resources in the United States. However, geothermal development is hindered by the remoteness of the reservoirs from population centers and the small size of Alaskan markets. These factors may render economic exploitation difficult. Five sites in Alaska are examined (Fig. 1). They are: Tenakee, Copper Valley, Pilgrim Springs (Nome), Makushin Volcano (Unalaska) and Summer Bay (Unalaska). PRUDHOE BAY @ PILGRIM SPRINGS CANADA FAIRBANKS @ a MT. McKINLEY a COPPER VALLEY ANCHORAGE JUNEAU oF eS TENNAKEE SPRINGSDY MT. MAKUSHIN fp o SITKA S ae SUMMER BAY Fig. 1. Selected geothermal sites in Alaska. In order to evaluate the geothermal potential of each site, an analysis of those variables pertinent to development are examined. A general discussion of pumping and piping costs in Alaska are presented, along with analyses of transportability factors and heat losses. Economic models for direct heating systems and power generation are also described. In addition, the geological and geochemical characteristics of each site are discussed. These engineering, economic and geologic data are then synthesized to form a comprehensive evaluation of the geothermal potential of each site. THE ECONOMIC AND ENGINEERING MODELS Two economic models are used in the evaluation of the potential of geothermal development in Alaska. For direct utilization the required cost per thousand BTU (C\,g7y) is calculated under a 20% rate of return to the prospective utility company. The possible selling of geothermal fluids would be unique to Alaska since there is no past history of town-wide heating systems. Geothermal Development in Alaska 243 Almost all of the home and business heating is done either by the combustion of fuel oil or wood. No alternatives to geothermal heating are envisioned on town-wide scales. For power generation, though, the present trends are for the creation of unitized grid systems. For example, on Unalaska Island, while there have been several power generators in operation, the city is installing a grid system that would encompass all users. Hence, the model used in this paper is to compare the rates of return of geothermal power generation versus the rates of return of diesel generation, assuming single utility, single grid system for each town. This comparison is useful since several towns are contemplating new power generation schemes that would support the new unitized electric distribution systems. The geothermal potential thus becomes important as a competing alternative to diesel generation. In the economic models presented in this paper there is no consideration for tax credits and depletion allowances that have been available in the U.S.A. in the past. The present policy of the federal administration is not clear. Further, in Alaska, the land status of many of the contemplated sites is not yet resolved. Much of the land could either be conveyed to native corporations under the Native Settlement Act or could be removed from further consideration for development, if Federal Fish and Game Agencies were to reclaim certain sites as Game Preserves. Hence, the uncertainty of government regulations and land status makes the following economic evaluation necessarily conservative. Multiple uses of the geothermal fluids are not considered in this analysis. The geothermal sites in Alaska are, normally, at a long distance from small communities. Hence, if geothermal power generation were to be utilized then an effort for ‘‘all-electric’’? homes may have to be undertaken in order to justify the project. This would necessarily undercut the possibility for multiple uses of the geothermal fluids. Major industry development is not contemplated. As a result, geothermal power plants in conjunction with the transportation of hot water for space heating cannot be envisioned in Alaska at the present time. ECONOMIC EVALUATION OF DIRECT HEATING SYSTEMS The economic scenaria envisioned are: (1) a state appropriation to initiate the project and (2) a private consortium to raise appropriate financing. The two options differ significantly. In the first case, a publicly run facility does not pay taxes, nor does it (necessarily) require a rate of return. Private ventures are subject to taxes and they have to earn an appropriate rate of return. Two quantities are required to prove or disprove the attractiveness of a project. These are the capital investment and annual operating costs. The calculated price will then be compared with present non-geothermal costs. The capital investment consists of the pumping units and the drilling costs of the wells. The in-town distribution system will be treated as an annual cost. The piping costs will also be presented as an annual cost. The purchase costs of the pumps depend on the flow rate and the pump head. Since the piping and pumping costs have identified an optimum pipe diameter, the selection of pump size (head) is also determined. The distance of the geothermal site to the market will indicate the number of pumps (or pumping stations). The operating costs consist then of pumping costs, piping costs, in-town distribution costs, maintenance costs, labor costs and fixed costs. Two engineering issues need evaluation next: (1) the pumping and piping and their associated costs and (2) the pipeline heat losses and flow rate requirements under Alaskan climatological conditions. 244 M. J. Economides and G. N. Arce Annual costs of pumping and piping The annual power costs required to pump the geothermal fluids per mile of pipeline (and assuming 130% load factor and 25¢g/kWh) may be given by: Cyump = Pm X 1.3 x 24 (h/day) x 365 (day/year) x 25¢/kWh a) Then: Cyump = 2-85 X 10° P,, (2) = 34.6 hg {GPM} (3) The annual costs for the piping may be obtained from a correlation supplied by Peters and Timmerhaus (1980): Coipe = (1 + F)XG/"K;, (4) where: Cyipe = (in the equation above) cost for installed insulated piping system as dollars per year per foot of pipe length. The value is in uninflated dollars (i.e. they are based on their real value at the year of installation). In Alaska, X = $0.65/ft of 1 in. diameter pipe, K, = 0.20 and F = 1.4. Then, eq. (4) may be rewritten for a 1 mile length of pipe as: Cyine($/year) = 1.64 x 10° d'* (5) pipe The total annual transportation costs for horizontal pipes would then be the summation of the pumping costs and the piping costs. Hence, G. total = 2.85 x 10° P,, + 1.64 x 10°dj! (6) These do not include the initial costs of the pumps at the geothermal sites or the distribution system in-town. Such costs vary for each location and will be dealt with separately. Figures 2 and 3 represent optimization graphs for the total costs. Small diameter pipelines @————41 2250 GPM 4—_—-~4 1124 GPM ————-* 562 GPM @———-©_ 225 GPM ANNUAL COST / MILE ° 5 10 15 20 25 PIPE DIAMETER (in) Fig. 2. Optimization of annual total costs per mile of geothermal fluid transportation (low flow rates). Geothermal Development in Alaska 245 -—— 16857 GPM o——*% | 1240 GPM *—————* 5620 GPM ANNUAL COST / MILE ° @ PIPE DIAMETER (in) Fig. 3. Optimization of annual total costs per mile of geothermal fluid transportation (high flow rates). have low construction costs, yet they have much higher pumping requirements due to higher friction losses. Hence, for flow rates of 225 GPM, 562 GPM, 1124 GPM and 2250 GPM (sufficient for towns up to 2000 people), there is a clearcut minimum cost (Fig. 2). For larger flow rates, there is an obvious need for larger diameter pipelines (above 24 in.), in order to achieve minimum costs (Fig. 3). The correlation by Peters and Timmerhaus (1980) is not considered valid for pipelines with very large diameters. Hence Fig. 3, dealing with cities of more than 5000 people, is left as it is without a further attempt towards optimization, assuming that the 24 in. diameter pipeline would suffice. The in-town distribution costs will be fairly standardized, i.e. a set footage per user of insulated, main line plus small diameter, branching, insulated pipe. Assuming 50 ft of 4 in. line, plus 100 ft of 1 in. line per user, the annual pipe distribution costs are: ‘< P. 7 Cus = 1.52 (soy + F)X4'K, + 1.5-2 (1001 + F)X1"K, (7) using F = 1.4, X = $0.65/ftoflin., n= 1.5, K, = 0.20 Then Cui, = 105.3P, (8) The maintenance costs are treated as a fixed percentage of the pumping unit costs (1%/year is used). Cc, ‘maint = 0.017 (9) The labor costs are the most difficult to assess. Large units would require full-time employees, while smaller units could not justify such expenditures. Hence the labor costs can only be approximate depending upon whether a municipality may undertake the operation as part of ongoing activities, whether it would be contracted out, or whether a private company would operate the facility. The labor costs are-expected to be 5% of the initial investments. Cup = 0.051 (10) 246 M. J. Economides and G. N. Arce Finally, the fixed costs are normally assessed at 7% of the investment. Hence, Cy, = 0.077 ql) The amount necessary to charge for 1000 BTU of delivered heat in order to defray investment and annual operating costs would serve as the gauge of whether direct heating geothermal systems are attractive. The revenues, R, are then obtained by multiplying the annual heat load by the cost per 1000 BTU (Cypru)- R= 1.5 “2 30,000 X 24 x 365 X Cypry/1000 (12) R = 9.86 X 10* P,Cusry (13) Equation (13) assumes a load of 30,000 BTU/h for a household (assuming a four-member household). The factor 1.5 accounts for non-domestic uses. This factor has been obtained through spot-check surveys in a number of communities. It varies from 1.2 to 1.8 depending upon local activities. Then, the net cash flow before taxes is: C.F.B.T. = R — Cyige — Noump — Cais ~ Cmain ~ Cray ~ Chix (14) pump If the project is funded by the government, and assuming a 20-year life of the project, then the cost of 1000 BTU may be calculated from the relationship: 20(C.F.B.T.) = 7 (15) ain If the project is private, then the cash flow before taxes must be further reduced by the depreciation, D, and then assuming a corporate tax of 48%, the cash flow after taxes will be: C.F.A.T. = 0.52(C.F.B.T. — D) + D (16) There are several ways to assess depreciation. For simplicity, a straight line, 20-year depreciation rate will be used here. Hence, D = 1/20 (17) Finally, assuming a 20-year, 20% rate of return, the cash flow after taxes must be multiplied by the uniform series present worth factor and equated to the capital investment. Hence, (i+ if -1 = ————— (C.F.A.T. ia+iy CPAT) (8) where i is the rate of return (0.20) and j the number of years (20). The uniform series present worth factor is then equal to 4.87 and eq. (18) reduces to: I = 4.87(C.F.A.T.) (9) From this relationship and eqs. (17), (16), (14) and (13), one may easily calculate the Cugry for each site. Pipeline heat losses and flow rate requirements under Alaskan weather conditions Certain assumptions are necessary for this set of calculations. (1) Complete mixing within the pipeline is assumed. (This is assured by the high turbulence of the flow.) (2) The heat loss to the surroundings is controlled by the resistance through the pipeline Geothermal Development in Alaska 247 insulation. Hence, the reciprocal of this ‘‘controlling resistance’? can be set as equal to the overall heat transfer coefficient. The heat loss equation, through a horizontal pipe, exposed to the atmosphere is Rt (ee nmdL mE? 20 Oe a , din(d/d) din(d/dy 1 (20) hd, 2k 2k h, Steel ins where h, is taken as equal to 200 BTU/h ft? °F | from Kern (1950)| and h, is equal to 3 BTU/h ft? °F. (k,,. = 0.033 BTU/h ft? °F/ft and k,,.., = 26 BTU/h ft? °F/ft). Table | contains typical dimensions of insulated pipelines in use in Alaska. Table 1. Dimensions of insulated steel pipe (in.) Nominal diameter d, (inside) d. (+ steel) d (+ insulation) 4 4.026 4.500 12.500 8 7.981 8.625 16.625 12 12.090 12.750 20.750 16 15.250 16.000 24.000 24 23.250 24.000 32.000 For a typical case (8 in.), the values of the individual resistances may be calculated. fo ees hd, 200 x 7.981 din (d./d) _ 16.625 In (8.625/7.981) = = ="0102 ® Kee 2 x 26 . din(d/d.) 16.625 In (16.625/8.625) Pps ee 1653 2k,,. 2 x 0.033 ot .=— =~ = 0.33 ees Obviously, r,, r; and r, may be readily ignored. The overall heat transfer coefficient, U, can then be calculated: 1 2k, =— = ——s_ 21 U r; din (d/d,) 2) and T, — T, nLk, On 2 ee (22) In T- Tr. In (d/d,) This heat loss must be equal to the change in enthalpy of the fluid. Ah = mC,(T, — T,) (23) 248 M. J. Economides and G. N. Arce Equations (22) and (23) yield: tit = — is ore (24) In Toa In (d/d,) Finally, 2nLk, T =f. T, — T, = 25 oF AT sexo ( Gna) (25) Equation (25) is significant since it provides the exiting water temperature, 7,, following a path of length L (in ft) with a flow rate 7m (Ib/h) and subjected to an ambient temperature, 7,. The initial wellhead temperature is 7,. From eq. (25), the in-town temperature, 7,, may be calculated. The same equation allows the evaluation of the limiting distance that water may be pumped in order to deliver a certain temperature. Example 1. Calculation of limiting distance (L) in the transportation of geothermal heating fluids Assume: wellhead temperature, 7, = 200°F, population (such as Nome), P, = 2000, ambient temperature, 7, = 0°F, pipeline diameter, d, = 24 in. k,,, = 0.033 BTU/h ft? °F and temperature delivered, 7, = 170°F. The flow rate requirement for various sizes of population centers may be approximated using an average heat consumption per household and business. For the case of four-member households and a 30,000 BTU/h heating load, the total load for a town is: h, = 1.5 s x 30,000 (BTU/h) (26) as explained in eq. (13). The load implied by eq. (26) must be equal to the useful enthalpy output of the heating fluid: Ah = mC,AT (27) where AT is generally taken as equal to 20°F. The latter is considered as a fairly efficient temperature drop in heating systems. Equating 4, and Ad and solving for sh results in: m (Ib/h) = 562.5P,, (28) or, since flow rate is normally measured in GPM, then: 562.5 x 7.48 (gal/ft’) x P, m(GPM) = 65 4 Ub/te’) x 60 (s/min) = 1.124P, (29) Equation (29) allows the calculation of the heating flow rate requirements for various sizes of population. From eq. (28): mm = 5$62.5P, = 562.5 x 2000 = 1.125 x 10° lb/h Geothermal Development in Alaska 249 Using 4 in. insulation, d = 32 in. Then from eq. (25): E.= 253-210? ft = 48 miles If the ambient temperature were — 40°F, then: L = 2.09 x 10° ft = 39 miles Therefore, the transportability of geothermal fluids is significantly reduced in the harsh Alaskan climate. ECONOMIC EVALUATION OF GEOTHERMAL POWER GENERATION Geothermal power generation has a priori a limiting constraint. The power plant must be in close proximity to the reservoir. Hence, the construction costs are often large, since they must be assessed on remote and harsh sites. . In Alaska, construction and logistical costs are significantly higher than elsewhere in the United States. Drilling costs, because of the cumbersome logistics, are expected to be at least twice the level of established sites such as in The Geysers or in the Imperial Valley. As an example of the logic used in the economic evaluation, a case applicable to Unalaska (Makushin) is presented here. Table 2 presents a best estimate scenario for a 30 MW, geothermal power plant on Unalaska Island revised from the original, presented by Economides et a/. (1981). Table 2. Capital investment for a 30 MW, geothermal power plant, Unalaska Island Item Number Description Cost Well 6 8000 ft, 7% in. diameter (assumed 50% dry wells) $18,000,000 Piping _— 3000 ft, 8 in. diameter pipe, installed 250,000 Road — 3000 ft of service road, 18 ft wide gravel, at $200,000/mile 115,000 Generator 55 MW, maximum capacity generator, installed 25,000,000 Transformer station 1 55 MW, at $40/kW, installed 1,833,000 Transmission line 1 11 miles of transmission line overland (helicopter installed), 5 miles underwater, $100,000/mile 1,600,000 Road _ 16 miles of 18 ft gravel road at $200,000/mile 3,200,000 Subtotal $49,998,000 Contingency 10% of capital 4,988,000 Total to be depreciated $54,986,000 A conservative estimate of 50% dry holes is assumed. A transmission line, 16 miles long, and a connecting 16 mile gravel road are assessed to the cost of a power plant. An average geothermal well, producing 200,000 lb/h of steam (either superheated or separated) can support a 10 MW, maximum continuous capacity power plant. The operating costs for the same plant were calculated and presented in Table 3. A similar calculation was done for diesel generated power plants, assuming that consolidation of all present units would be a desirable event. A new diesel generating system, supplying 30 MW., would cost $20.6 million installed (source: General Electric, personal communication). 250 M. J. Economides and G. N. Arce Table 3. Estimated annual operating costs for a 30 MW, geothermal power plant on Unalaska Island Item Description Cost ($1000) Employee compensation Three professionals x $50,000, 25 hourly x $40,000 plus 50% benefits 1725 Wells Maintenance 100 Plant facilities 0.1% of generator cost 250 Piping 20% of pipe cost 50 Transmission line 2% of cost 37 Road 2% of cost 64 Fixed costs 7% of investment 3841 Total annual costs 6067 Assuming: labor costs = $40/kW, fixed annual costs = 7% of investment, fuel costs = $1.40/gal (363,000 gal/MW./year) and the total operating costs = $17.9 million. The two options are then compared. The revenues are calculated using 15g/kWh as the selling price and operating at 75% capacity. Assuming a 30-year straight-line depreciation and a tax rate of 48%, the rate of return is calculated. A summary of the calculation is presented in Table 4. Table 4. Comparative economics of geothermal and diesel power plants on Unalaska (30 MW,) Geothermal ($1000) Diesel ($1000) Revenues 29565 29565 Operating costs 6067 17888 Depreciation 1829 667 Cash flow before taxes 21669 11010 Minus 48% taxes 10401 5285 Cash flow after taxes (+ depreciation) 13097 6392 Rate of return (30 years) 24.0% 31.2% The ‘‘bottom line’’ comparison is the calculated rate of return. As can be seen in Table 4, the diesel option has a better rate of return than the geothermal option (31.2% vs 24%). Higher capacity plants tend to push the geothermal option closer to the diesel option and eventually they may surpass it, depending on the economic variables applicable to each site. A generalized model has been written and applied to selected geothermal sites in the state. All variables that were presented in the previous example may be changed, including the prospect of a government funded and operated project in which the tax rate is zero. Figures 4—8 include a number of runs using a base case (Fig. 4) and changing certain pertinent variables. Figure 4 assumes a private project, 15g/kWh the price for electric power and $1.40/gal the price for diesel. There are two wells per 10 MW, (i.e. assuming 200,000 Ib/h wells, 50% dry). In a repeat of the above run, with the price of diesel set at $1.20/gal, the point of intercept shifted to the right, approaching 55 MW.. When the price of diesel was increased to $1.60/gal, the reverse effect was observed, reducing the intercept to 22 MW.. Geothermal Development in Alaska 251 60 40 CIRCLES ARE FOR GEOTHERMAL 7 TRIANGLES ARE FOR DIESEL RATE OF RETURN % POWER GENERATION CAPACITY (MWe) Fig. 4. Comparison between diesel and geothermal power plants (base case). Tax rate 48%, fuel $1.40/gal. 80 S ° CIRCLES ARE FOR GEOTHERMAL TRIANGLES ARE FOR DIESEL 20 RATE OF RETURN % Papers wae ee ae ey ee 10 20 30 40 50 60 POWER GENERATION CAPACITY (MWe) Fig. 5. Comparison between diesel and geothermal power plants. Base case, fuel $1.40/gal, power $0.20/kWh. The impact of higher electric prices that the market may bear is shown in Fig. 5. Higher revenues, offsetting the high operating costs of the diesel option, render geothermal unfeasible within the 55 MW, market capacity. Figure 6 demonstrates the feasibility of a government sponsored project (0% tax). Finally, Figs 7 and 8 point out the effects of highly successful versus lackluster drilling programs. The first shows the impact of 100% wet (and prolific holes), while the second shows a case in which three wells are required per 10 MW.. In general, the intercept (if any) between the geothermal plant rate of return and that of the diesel powered plant, as shown in Figs 4—8, provides the break-point in plant capacity above which geothermal is economically attractive. 252 M. J. Economides and G. N. Arce 100 80 60 CIRCLES ARE FOR GEOTHERMAL TRIANGLES ARE FOR DIESEL 40 RATE OF RETURN % 20 oe ee te ee te tor eg ee ee | 10 20 30 40 50 60 POWER GENERATION CAPACITY (MWe) Fig. 6. Comparison between diesel and geothermal power plants. Base case, tax rate 0%, fuel $1.40/gal. 80 60 %o CIRCLES ARE FOR GEOTHERMAL TRIANGLES ARE FOR DIESEL 20 RATE OF RETURN POWER GENERATION CAPACITY (MWe) Fig. 7. Comparison between diesel and geothermal power plants (1 well/10 MW,). Wells down 50%, fuel $1.40/gal, power $0.15/kWh. Figure 9 presents a picture of the comparison between the two options for Unalaska (31 MW,) and Copper Valley (21 MW.). These results are approximate. The rates of return are for comparison only; they do not include in-town distribution costs (already there), maintenance and labor. The conclusion is that if these sites could support a market (both industrial and domestic) that would demand those levels of power, then geothermal could be an attractive option if a new diesel generator (for the entire location) were to be installed instead. RATE OF RETURN % Geothermal Development in Alaska 253 30 CIRCLES ARE FOR GEOTHERMAL TRIANGLES ARE FOR DIESEL 20 10 10 20 30 40 50 60 POWER GENERATION CAPACITY (MWe) Fig. 8. Comparison between diesel and geothermal power plants (three wells/10 MW,). Wells up 50%, fuel $1.40/gal, % RATE OF RETURN power $0.15/kWh. 805 UN ALASKA GEOTHERMAL UNALASKA DIESEL — COPPER RIVER GEOTHERMAL 704 — — —coPPeR RIVER DIESEL 6074 504 4074 304 2054 / 10 EL T T T T T 7 ° 10 20 30 40 50 60 70 PLANT CAPACITY (MWe) Fig. 9. The economic feasibility of geothermal power plants replacing diesel in selected sites in Alaska. Tenakee SITE DESCRIPTIONS AND EVALUATIONS The thermal springs at Tenakee occur in Cretaceous granitic rocks on the eastern shore of Chicagof Island, roughly 75 km southwest of Juneau (Brew and Morrell, 1980). High-angle joint sets are common at orientations of NSOE and N40W. Classified as part of the Alexander 254 M. J. Economides and G. N. Arce tectonostratigraphic terrane (Jones ef a/., 1981), the intrusive bodies are bordered on the north by Paleozoic hornfels and amphibolite, along with Silurian graywacke and sandstone of the Pt. Augusta Formation. To the south are large plutonic masses of Silurian and Cretaceous age, along with extensive exposures of a Juro— Cretaceous melange consisting of limestone, chert, layered gabbro and serpentinite (Beikman, 1980). Preliminary paleomagnetic data in the Craig region of Prince of Wales Island indicate that the rocks have undergone about 35° of counterclgckwise rotation and over 15° of northward movement since late Ordovician to Pennsylvanian time (Berg et a/., 1978). The village of Tenakee Hot Springs, Alaska (photo by M. J. Economides). The Tenakee springs were originally reported to have a total discharge of 22 GPM and a maximum temperature of 41°C (Waring, 1917). Subsurface reservoir temperatures recently calculated by Ivan Barnes, using silica and alkali geothermometers, indicate temperatures between 101 and 110°C (Reeder, 1982, written communication). This liquid-dominated system may derive its fluids from connate waters associated with expansive sedimentary units located at depth. The prominent joint sets in the granitic rocks provide avenues for the transportation of the heated fluids to the surface. A fault has been postulated to exist south of Tenakee Village beneath the waters of Tenakee Inlet (Loney ef al., 1975; Reeder, unpublished map). If this feature does exist, it may be acting as a no-flow boundary and thus impede southward escape of thermal fluids. Seven shallow test wells drilled during the summer of 1981 encountered cold to warm artesian waters. The warmest waters (38°C) were obtained from the deepest well (55 m), which also had a discharge of 0.5 GPM (Miller, 1981). The most obvious consumer of the geothermal resource would be the community of Tenakee. Since the springs are located within the village confines, direct utilization for purposes other than power generation could be accomplished easily. Geothermal Development in Alaska 255 The annual heat load demand for Tenakee is [eq. (26)]: 1.5 h, = 4g (200)(30,000)(24)(365) = 1.97 x 10'° BTU The required flow rate is [eqs. (28) and (29)]: m = 112500 Ib/h = 225 GPM This flow rate requires a modest well (~ $2,000,000) plus a 225 GPM pump. The required pressure head, assuming 5 miles of 4 in. pipeline network, is: A, (ft) = 6.94P,'75d.-475 = 15 ft From Peters and Timmerhaus (1980, Fig. 13—40), a 3 x 3, 3 HP pump would be adequate. Cost is $46,000 (for stainless steel). Then, J = 2,000,000 + 46,000 = $2,046,000. The revenues and operating costs may then be calculated. R = 1.972 x 10’ Cupru Coup [€4- (1)] = $5.84 x 10° pump C,;, leq. (8)] = $2.1 x 10* Crain 1G: (9)] = $2.05 x 10° Cay [eq (10)] = $1.02 x 10° Cy, leq. (11)] = $1.43 x 10° C.F.B.T. [eq. (14)] = 1.972 x 10’ Cry — 8-52 x 10° If government project. 20(1.972 x 10’ Cugry — 8.52 x 10°) = $2,046,000 or Cygry = $0.048/1000 BTU This is a good price, assuming an average 30,000 BTU/h consumption per household, spanning the entire year. If private project. C.F.B.T.. = +1.972: x 10’ Cyr 1 8:52. x 10° — 2;046,000/20 = 1,972 x 10? Cyn; — 9.54 10° C.F.A.T. = 0.52(1.972 x 10’ Cygne — 9-54 x 10°) + 2,046,000/20 = 1,025 x 10 Con - 3M x and [eq. (19)]: 2,046,000 = 4.87(1.025 x 10’ Cyr. — 3-94 x 10°) Then Cire = $0.079/1000 BTU This is equivalent to $1.10/gal fuel oil (~ 140,000 BTU/gal). The conclusion for Tenakee is that geothermal heating may be equally attractive to fuel oil if _ one good well could be drilled on the first try. There is no margin for a dry hole. Copper Valley The Klawasi hydrothermal springs occur on the western flank of Mt Drum, 30 km east of the 256 M. J. Economides and G. N. Arce Copper River Basin. Quaternary alluvium and glacial deposits blanket much of the basin, and zones of discontinuous permafrost are located at a depth of 3 m. Isolated windows of Eocene continental sediments are exposed locally. Schist, greenstone, graywacke and shale of Paelozoic to Cretaceous age form the upland areas which border Copper Valley on the north and south. The eastern portion of the basin is underlain by Cenozoic lavas emanating from the Wrangell Mountains (Beikman, 1980). Extrusion rates from these volcanoes are about an order of magnitude greater than anything reported from the Cenozoic circum-Pacific, and individual andesitic edifices are among the largest in the world. Isotopic data do not indicate the involvement of abundant crustal material, since the values for Nd (0.5129), Sr (0.7033) and Pb are consistent with mantle numbers. The voluminous production rates have, therefore, been interpreted as indicating a higher rate of magma production or an increase in extrusive versus intrusive magmatism immediately following accretion of the Yakutat Block in southeast Alaska (Nye, 1982). The structural framework of the Copper River Basin is dominated by east — west trending orogens which are concave to the south (Alaska Geological Consultants and Geonomics, 1975). Two regions of low gravity, the Glennallen and Gakona lows, were found by Andreasen ef al. (1964) to trend from Mt Drum into the Copper River Basin. Of the three thermal springs in the Klawasi area (Upper Klawasi, Shrub and Lower Klawasi), two occur near the axis of the Glennallen low. This low may be due to the presence of a thick succession of sedimentary rocks (Reeder et a/., 1980). The regional aeromagnetic map of the Copper River Basin suggests that the basaltic and andesitic lavas of the Wrangell massif underlie the mud volcanoes of the Klawasi hydrothermal springs at a relatively shallow depth (Andreasen et al., 1964). A rapid decrease in the magnetic gradient westward probably indicates that the lavas are thinner and buried at increasingly deeper depths under the alluvium of the basin. The thermal waters emerging at the Mt Drum mud volcanoes were found to have flow rates of 0.3—10 GPM and temperatures of 12—30°C. The silica, potassium, sodium and bicarbonate contents were relatively higher than the contents measured from other mud volcanoes in the Copper Valley region (Nichols and Yehle, 1961). Subsurface reservoir temperature have been calculated at 104— 157°C and 187°C, based on silica and Na— K—Ca geothermometers (I. Barnes, U.S.G.S.). An exploration well was recently drilled at Moose Creek by the Pan American Petroleum Corporation. Located just west of Glennallen on the axis of the Glennallen gravity low, the wellbore penetrated olivine basalt of the Talkeetna formation after passing through over 2200 m of Cretaceous sedimentary units. High pressure water was encountered at a depth of 1646 m in a bentonitic shale horizon (Reeder et a/., 1980). The hydrothermal reservoirs in the Klawasi area are probably artesian aquifers lying at depths of up to 2 km within the sedimentary sequence. A significant amount of cooling and mixing likely occurs during the upward migration of these fluids (Reeder et a/., 1980), especially in their passage through glacial till and permafrost. Possible recipients for geothermal energy include Copper Center (population 213), Glennallen (488) and adjoining areas (750). Summing the populations and assuming an average distance of 16 miles from the Klawasi Springs yields: h, = 1.43 x 10'' BTU m = 815,000 Ib/h = 1630 GPM Using a 16 in. pipeline: h, = 300 ft (elevation + friction) Geothermal Development in Alaska 257 There is a need for four to six wells at a cost of $10 million ($2 million/well). Pump costs (two 8 x 6, 125 HP units) = $2.8 x 10°. Then / = $10.3 million. R = 1.43 X 10° Cugry Cyipe = $1.68 x 10° Cyump = $2.70 x 10° Cy, = $1.53 x 10° (this is conservative, the population is scattered) Crain = $1.04 x 10° Cy» = $5.15 x 10° Cy, = $7.21 x 10° C.F.B.T. = 1.43 x 10® Cygry — 2.11 x 10° Then 20(1.43 x 10* Cygry — 2-73 x 10%) = 10.3 x 10° Custy = $1.91/1000 BTU significantly above present prices (one order of magnitude). Pilgrim Springs The thermal activity at Pilgrim Springs occurs in the Pilgrim River Valley on the Seward Peninsula. This valley is a fault-bounded tectonic depression located 75 km north of Nome. Precambrian amphibolites and Mesozoic plutons are the common lithologies in the area, with local exposures of conformable and unconformable (overthrust) Paleozoic carbonates (Hudson, 1977). Potassium — argon dating indicates a cooling age for the plutons of 84 m.y. (Turner and Swanson, 1981), which suggests intrusive igneous activity in the Upper Cretaceous. A permafrost horizon has also been identified which encloses an area of 1 — 1.5 km’, and is over 100 m thick. Gravity surveys conducted in the region (Turner and Forbes, 1980) indicate that Pilgrim Springs is located near the intersection of two possible fault zones which form the corner of a downdropped basement block. Seismic data imply that normal faulting is presently occurring and this subsidence is further substantiated by surficial geologic mapping. One or more of these faults could provide deep conduits for the geothermal anomaly. In order to explain these geological and geophysical observations, Wescott and Turner (1982) postulate that the Pilgrim River Valley graben represents an incipient rift extending 250 km across the central Seward Peninsula and offshore into the Bering Sea. Based on this hypothesis, the anomalous heat flow in the Pilgrim Springs area is due to tensional tectonics and active rifting. The possible existence of a major rift system is significant for the regional geothermal potential, since extensional tectonics allow for the shallow emplacement of high-temperature magma. A helium soil survey was conducted to test this rift model, and nine out of 11 helium anomalies (which indicate abnormally high heat flow) occur near the proposed rift segments. Furthermore, extensive basaltic fields north of the Pilgrim Springs region have been interpreted as resulting from eruption in a zone of crustal weakness produced by north — south extension (Turner and Swanson, 1981). The amount of separation along this proposed rift is less than the widths of the observed Quaternary depressions, since the depressions have been affected by the interaction of normal faulting, subsidence and rifting. The thermal waters emerging at Pilgrim Springs are alkali— chloride fluids with a flow rate of 67 GPM and a temperature of 81°C. Preliminary Na— K — Ca geothermometry of this liquid- 258 M. J. Economides and G. N. Arce dominated system suggests deep reservoir temperatures approaching 150°C (Wescott and Turner, 1982). Two 50 m test wells were drilled in 1979, and artesian aquifers were encountered at a depth of 20-30 m. Flow rates were estimated at 200 and 300-400 GPM, respectively, with temperatures of 90°C (Wescott and Turner, 1982). During the summer of 1982, deeper exploration wells were completed. The temperature data from these six exploration wells were then graphically plotted to yield temperature vs depth curves (Economides et a/., 1982). The curves display a trend toward a maximum temperature at depths of 40-100 ft (12-30 m), followed by a rapid temperature decrease at depths of 100 —250 ft (30—75 m), and finally by a constant geothermal gradient ranging from 1.8 to 2.1°C per 100 ft, down to a depth of 900 ft (270 m). Two deep wells (PS4 and PSS) show temperature trends which would intersect at 155°C and a depth of 4875 ft (1477 m), suggesting that all of the wells overlie a source reservoir which is located at a depth of 4875 ft. The shallow temperature anomaly observed in all the wells suggests that somewhere in the immediate vicinity, hot water is flowing upward through a fault or fissure system. This conduit is inferred to extend vertically from a depth of about 50 ft to the deep source reservoir at 4875 ft. By simple observation, the variables between Pilgrim Springs and Nome are even less favorable than in the case of Copper Valley. There is no apparent need to calculate Cygry- Summer Bay Unalaska Island is the second largest island west of the Alaska Peninsula, in the eastern Aleutian arc. The rocks on Unalaska Island may be grouped into three major categories, which have been correlated with those found throughout the central and eastern Aleutian Islands (Marsh, 1982). The oldest and most extensive unit is the lower Tertiary Unalaska Formation. Drewes et al. (1961) described this formation as a thick sequence of coarse and fine sedimentary and pyroclastic rocks interbedded with basaltic, andesitic and dacitic lavas. Graywacke, tuffaceous sandstone and argillite are the common sedimentary rocks, and the entire formation contains abundant alteration products (such as chlorite, pyrite, epidote and albite). The depositional environment of the Unalaska Formation is interpreted to be a perched interarc basin on the summit of the Aleutian ridge, which would account for the intercalated debris flow and proximal turbidite units (Lankford and Hill, 1979). Upper Tertiary calc-alkaline plutons comprise the second lithologic category. Three batholithic intrusions and 25 smaller plutons have been mapped (Drewes ef a/., 1961). The larger intrusions are commonly zoned with marginal phases as mafic and gabbro. Widely spaced parallel joint sets within the three batholiths are separated by massive unjointed rock. The most abundant intrusive rock type is hypidiomorphic-granular granodiorite, with quartz diorite, quartz monzonite and aplitic granite occurring locally. These intrusive bodies were emplaced by the complex interaction of assimilation, forceful intrusion and stoping mechanisms. Detailed studies on the Captains Bay pluton by Perfit (1977) indicate inward fractional crystallization of a parental andesite or high-alumina basalt. Fractionation took place at shallow depth (less than 20 km), with a temperature between 950 and 1210°C. The final group consists of basalt and andesite flows of the Quaternary Makushin Volcanics which unconformably overlie the older lithologies. Although the thickness of the Makushin Volcanics is not known exactly, it probably does not exceed 1500 m. The volcanics consist of approximately 80% basalt and andesite lava flows, and 20% agglomerate, tuff breccia and flow breccia (Drewes et al., 1961). Lava flows are commonly 5 — 20 m thick and may be separated by thin tephra or debris flow horizons. The flows form a radial pattern around the summit of Makushin Volcano, which is now capped by an ice-filled caldera. The first of the two major hydrothermal areas on Unalaska Island is at Summer Bay. These Geothermal Development in Alaska 259 warm springs flow into shallow pools located in a north — south trending glacial valley, roughly 2 km south of Summer Bay and 6 km east of Dutch Harbor. The largest of the thermal springs has a temperature of 35°C and a discharge of 2 GPM. Subsurface reservoir temperatures are predicted to be 60 — 86°C, based on silica geothermometry (Motyka ef a/., 1981). Several steeply-dipping faults have been identified in this region, and may provide deep- seated conduits for the circulation of meteoric waters. A large normal fault striking N45W and dipping 60—70° south is well exposed along the coast just south of Summer Bay. This fault may be projected across Summer Bay Lake and through the Summer Bay warm spring, although exposures become increasingly poor. Because of this uncertainty, joints trending N45W are highly suspected as controlling the source waters of the springs (Reeder, 1981). During 1980, two shallow exploration wells were drilled. Both wells encountered a warm water aquifer in sandy soil at a depth of approximately 13 m, with bedrock at 17 m. Water temperatures were 43 — 50°C, with flow rates of 7 and 50 GPM from the 4 in. diameter wells. It is not clear what lithologic unit is acting as a cap for this aquifer, because no significant thickness of impermeable material was detected during drilling. However, a lightly cemented horizon of clay was observed at a depth of roughly 10 m in the wellbore (Dames and Moore, 1980), and this is apparently capping the system. The waters from the two test wells and from the surficial warm springs are chemically very similar. All have chloride and sulfate as their major anions, with sodium and calcium as the major cations. Ratios of Na: Cl, K : Cl, SO.: Cl and Na: K are nearly identical. This suggests that the thermal waters have a common source which undergoes varying degrees of mixing with cold surface waters (Motyka ef a/., 1981). To further delineate the extent and characteristics of the thermal area, several types of geophysical and geochemical studies were undertaken in 1981 (Reeder, 1982, written communication). An EM-31 Geonomics survey verified that the N45W joints mentioned earlier are acting as conduits for the warm springs. Schlumberger electric soundings and dipole-dipole resistivity surveys, along with mercury and helium soil surveys, have demonstrated the limited regional extent of the thermal manifestations. The twin communities of Unalaska/Dutch Harbor support a population of 1300, and are located approximately 5.5 miles from Summer Bay. The necessary flow rate would require five wells at a cost of $10 million. Also, two pumps 8 x 6, 125 HP ata cost of $2.8 x 10°. J = $10.3 million. m = 731000 lb/h = 1462 GPM R = 1.28 x 10° Cyry Cyipe = $5.77 x 10° (16 in. pipeline) Cyump = $1.55 x 10° C,,, = $1.36 x 10° Crain = $1.03 x 10° Can = $5.2 x 10° Cy, = $7.2 x 10° C.F.B.T. = 1.28 X 10° Cupp — 1.39 x 10 Coote Then 20(1.28 x 10° Core — 1-75 x 107) = 10.3 x 10° and C\yyr, = $0.14/1000 BTU equivalent to $2.01/gal of fuel oil. A government sponsored project may be attractive, depending on the immediate future of oil prices. The figure, thus reached, is very near the present prices on the island. No private investment can be attractive at this time. 260 M. J. Economides and G. N. Arce Makushin Volcano The second major hydrothermal area on Unalaska Island is at Makushin Volcano. In contrast to the small, low-temperature thermal activity in the Summer Bay region, the area around Makushin Volcano contains large hot springs and several fumarole fields (Reeder, 1981). These thermal fields occur in the vicinity of Glacier Valley and Makushin Valley. The summit of Makushin Volcano (photo by J. W. Reeder). Glacier Valley originates on the rugged south flank of Makushin Volcano and trends southwest for about 10 km to sea level. Located 22 km west of Unalaska/Dutch Harbor, the thermal activity (which includes warm springs, hot springs and superheated fumaroles) emanates from both the Unalaska Formation and the gabbroic plutons which intrude it. Motyka ef a/. (1981) identified the key features of the water chemistry, such as the extremely low levels of chloride, the near neutral pH, the relatively low cation content and the abundance of magnesium and calcium. The sum of the chemical characteristics and the surficial fumarolic activity indicate the presence of a shallow vapor-dominated zone, while the calcium and magnesium contents of the thermal waters suggest they are meteoric in origin. The surficial waters are thought to infiltrate to relatively shallow depths where they are heated by steam and volcanic gases rising through a vapor-dominated zone from a much deeper reservoir. Compositionally, the primary reservoir may be a hot sodium chloride brine overlying a cooling magma body. Silica geothermometry of the spring waters indicates that temperatures in the shallow perched reservoir may approach 150°C, while the large area covered by fumarolic and hot-spring activity suggests a hot geothermal system which may exceed 150°C (Motyka ef al., 1981). A steep normal fault which trends N50W has been identified southeast of the thermal sites (Drewes et a/., 1961). The strike of this fault is directly in line with the thermal site, and could, Geothermal Development in Alaska 261 therefore, be acting as a conduit for heated fluids. Nakamura et a/. (1977) have determined that this NSOW orientation is approximately the expected fracture direction, in light of the stresses generated by the subductive interaction of the Pacific and North American lithospheric plates presently occurring in the northern Pacific Ocean. Other fault and joint systems have been mapped in the area (Reeder, 1981; Reeder, 1982), and these also appear to be contributing to the hydrothermal convection process. Prominent near-vertical joints commonly trend N55W, N80W and N60E. The N60E orientation may be locally significant, for the thermal activity at Fields 2 and 3 appears to follow this direction. Furthermore, N40E-N60E is also the approximate trend of a line drawn between several igneous intrusions, and this may indicate that the contacts between the plutons and the Unalaska Formation are providing avenues for the circulation of heated fluids. Thus, the necessary conduits for the geothermal system appear to be the result of regional tectonics (subduction) and local plutonism. These processes are also important to the thermal activity in the vicinity of Makushin Valley. Makushin Valley has its source on the eastern flank of the volcano, and the valley trends west for roughly 13 km to its terminus at Broad Bay. The thermal sites are located in the upper reaches of Makushin Valley, a distance of about 20 km from Unalaska/Dutch Harbor. Hydrothermal activity in Makushin Valley is similar in many respects to that in Glacier Valley. For both areas, thermal manifestations consist of both hot springs and fumaroles. Also, the thermal waters are chemically similar; both locations are characterized by low chloride, high silica, high calcium and magnesium, and high bicarbonate and sulfate. Motyka et al. (1981) state that the comparatively high amounts of Ca and Mg, relative to the other cations, indicate that the waters are derived from the mixture of surface waters with deeper hotter fluids. Thermal springs in the area commonly occur at the base of the fumarole fields, and this suggests that at least part of the spring waters may originate as condensation of steam in surface waters. These meteoric waters then percolate into the substrata to finally discharge as springs. However, the high silica content of the thermal waters indicates that a large portion of the waters must have originated from a subsurface reservoir where temperatures exceed 150°C. But, since silica can equilibrate comparatively quickly (within several days to a few weeks), this suggests that the reservoir supplying the spring waters lies at a fairly shallow depth, and that this perched aquifer is heated by a much deeper reservoir (Motyka et a/., 1981). The inferred deep reservoir is likely a high-temperature system, for three temperature gradient wells drilled during the summer of 1982 encountered maximum bottomhole temperatures of 200°C at a depth of 550 m in the plutonic rocks (Economides er a/., 1982). In the summer of 1983 a new ‘‘slim hole’’ well, ST-1, produced extremely encouraging results. At a depth of 1950 ft (594 m) it encountered hot water (7 = 366°F, 186°C and p = 508 psi, 35.7 kg/cm’). The well produced 50,000 Ib/h (22.7 ton/h) through a slim pipe (diameter 77 mm). The wellhead conditions were: T = 281°F, 138°C, p = 51 psi, 3.6 kg/cm? and approximately 15% steam by mass. Recent workers (Kay ef a/., 1982) have stated that this portion of the Aleutian arc may be divided into four principal segments, as deduced from the geographic alignment of volcanoes and earthquake aftershock zones. Furthermore, the type of magmatic differentiation (calc- alkaline or tholeiitic) depends on the tectonic position of the volcanic centers with respect to the segment boundaries. If this hypothesis is correct, it could have important implications for the geothermal potential around Makushin Volcano. In the classification scheme of Kay er al. (1982), tholeiitic magmas occur at segment boundaries and show characteristics consistent with low-pressure, high-temperature crystallization in large, shallow magma chambers; these characteristics include no hydrous phenocrysts, an iron enrichment trend on AFM diagrams, parallel REE patterns and vitrophyric lavas. In contrast, calc-alkaline magmas are generated within segments and display characteristics consistent with higher pressure and lower 262 M. J. Economides and G. N. Arce Well ST-1 on the flanks of Makushin Volcano (photo by M. J. Economides). temperature crystallization. From the viewpoint of resource potential, tholeiitic magmas are desirable since they represent large, shallow, high-temperature energy sources. Since Makushin Volcano appears to display features of both calc-alkaline and tholeiitic sources, it is classified as ‘mixed’ in the initial report. However, chemical data from other studies (Drewes ef al., 1961; Perfit, 1977; Arce, 1983) indicate that Makushin Volcano is a tholeiitic edifice. Due to the attractiveness of Summer Bay (albeit marginal), there is no need to explore the direct heating potential of Makushin, especially since it is a prime target for power development (as previously discussed). Should a power plant be constructed to utilize the resource around Makushin Volcano, the plant would need to be close to the wells in order to minimize the loss in quality which accompanies steam transportation. Unfortunately, this necessitates a plant location danger- ously near an active volcano. At least 17 ‘‘eruptions’’ have been reported since 1760, and seven of these appear to be actual volcanic eruptions. Based on these seven authentic events, Makushin erupts on average every 30 years. Since the last extrusive episode was in 1951, the volcano is statistically ready for another eruption. The most likely hazards to life and property in this vicinity would be from pyroclastic flows, mudflows, jokulhlaups and tephra fallout. Hazards associated with tsunamis, noxious gases, lightning and so on, would not be as serious. Therefore, any structures near the volcano should be built and maintained with these hazards in mind (Arce and Economides, 1982). CONCLUSION While the engineering feasibility of geothermal development has been demonstrated, the economic feasibility is, as expected, tenuous. Geothermal Development in Alaska 263 Tenakee and Unalaska (Summer Bay) are the only sites that have even marginal attractiveness for direct utilization. For power generation physically hindered by the required quality of the geothermal fluid, there are two sites that may be attractive in the near and intermediate future: Unalaska (Makushin) and Copper Valley. The reduction in petroleum prices has a detrimental effect on geothermal prospects. Further, the minimum capacity requirements to make geothermal feasible would need either a consolidation of the present power users or an increase in the population or industrial base. In general, the present prospects of geothermal development in Alaska are bleak, with the possible exception of Unalaska. Therefore, capital outlays should be limited to resource identification via geological/geophysical surveys and perhaps limited exploratory drilling. REFERENCES Alaska Geological Consultants and Geonomics (1975) Technical feasibility of geothermal energy development, Drum thermal springs area, Alaska. Report to Geothermal Energy Program and Ahtna Regional Native Corporation. Andreasen, G. E., Grantz, A., Zeitz, I. and Barnes, D. F. (1964) Geologic interpretation of magnetic and gravity data in the Copper River Basin, Alaska. U.S. Geol. Surv. Prof. Paper 316-H. Arce, G. N. and Economides, M. J. (1982) Analysis of volcanic hazards from Makushin Volcano, Unalaska Island. Proc. 1\V N.Z. Geother. Workshop, pp. 93-99. Arce, G. N. (1983) Volcanic hazards from Makushin Volcano, northern Unalaska Island, Alaska. M.S. thesis, University of Alaska, Fairbanks. Beikman, H. M. (1980) Geologic map of Alaska. U.S. Geol. Surv. and Alaska Div. Geol. Geophys. Surv. Berg, H. C., Jones, D. L. and Coney, P. J. (1978) Pre-Cenozoic tectonostratigraphic terranes of southeastern Alaska and adjacent areas. U.S. Geol. Surv. Open-File Report 78-1085. Brew, D. A. and Morrell, R. P. (1980) Intrusive rocks and plutonic belts of southeastern Alaska, U.S.A. U.S. Geol. Surv. Open-File Report 80-78, pp. 1 — 34. . Dames and Moore (1980) Geothermal drilling studies near Unalaska, Alaska. Report to Alaska Div. of Energy and Power Development. Drewes, H., Fraser, G. D., Snyder, G. L. and Barnett, H. F. Jr. (1961) Geology of Unalaska Island and adjacent insular shelf, Aleutian Islands, Alaska. U.S. Geol. Surv. Bull. 1028-S, 583 - 676. Economides, M. J., Ansari, J., Arce, G. N. and Reeder, J. W. (1982) Engineering and geological analyses of the geothermal energy potential of selected sites in the state of Alaska. Proc. VIII Workshop Geother. Reservoir Engng, Stanford University. Economides, M. J., Reeder, J. W. and Markle, D. (1981) Unalaska geothermal development. Proc. N.Z. Geother. Workshop, pp. 7-12. Hudson, T. (1977) Geologic map of Seward Peninsula, Alaska. U.S. Geol. Surv. Open-File Report 77-796A. Jones, D. L., Siberling, N. J., Berg, H.C. and Plafker, G. (1981) Map showing tectonostratigraphic terranes of Alaska, columnar sections, and summary description of terranes. U.S. Geol. Surv. Open-File Report 81-792. Kay, S. M., Kay, R. W. and Citron, G. P. (1982) Tectonic controls on tholeiitic and calc-alkaline magmatism in the Aleutian arc. J. geophys. Res. 87, 4051 — 4072. Kern, D. Q. (1950) Process Heat Transfer. McGraw-Hill, New York. Lankford, S. M. and Hill, J. M. (1979) Stratigraphy and depositional environment of the Dutch Harbor Member of the Unalaska Formation, Unalaska Island, Alaska. U.S. Geol. Surv. Bull. 1457-B. Marsh, B. D. (1982) The Aleutians. In Andesites (Edited by Thorpe), pp- 99-114. Wiley. Miller, D. S. (1981) Tenakee drilling project. Report to Alaska Div. of Power and Energy Development. Motyka, R. J., Moorman, M. A. and Liss, S. A. (1981) Assessment of thermal springs sites in the Aleutian arc, Atka Island to Becherof Lake—preliminary results and evaluation. Alaska Div. Geol. Geophys. Surv. Open-File Report 144, pp. 68 — 89. Nakamura, K., Jacob, K. H. and Davies, J. N. (1977) Volcanoes as possible indicators of tectonic stress orientation— Aleutians and Alaska. Geofis. pura appl. 115, 87 ~ 112. Nichols, D. R. and Yehle, L. A. (1961) Mud volcanoes in the Copper River Basin, Alaska. Proc. Ist Int. Symp. Arctic Geology, Calgary, Vol. 2, pp. 1063 — 1087. a Nye, C. J. (1982) The Wrangell Volcanoes—voluminous volcanism accompanying microplate accretion. Geol. Soc. America Cordilleran meeting, Anaheim, California, Vol. 14, p. 221.) Peters, M. S. and Timmerhaus, K. D. (1980) Plant Design and Economics for Chemical Engineers (3rd edn.). McGraw- Hill, New York. Perfit, M. R. (1977) The petrochemistry of igneous rocks from the Cayman Trench and the Captains Bay pluton, Unalaska Island. Their relations to tectonic processes at plate margins. Ph.D. dissertation, Columbia University. Reeder, J. W.. Coonrad, P. L., Braggs, N. J., Denig-Chakroff, D. and Markle, D. R. (1980) Alaska geothermal implementation plan. Draft to Alaska Dept. of Natural Resources and U.S. Dept. of Energy. Reeder, J. W. (1981) Vapor-dominated hydrothermal manifestations on Unalaska Island, and their geologic and tectonic setting. JA CEI Symp. Arc Volcanism, pp. 297 — 298. 264 M. J. Economides and G. N. Arce Reeder, J. W. (1982) Hydrothermal resources on the northern part of Unalaska Island, Alaska. Alaska Div. Geol. Geophys. Surv. Open-File Report 163. Turner, D. L. and Forbes, R. B. (1980) A geological and geophysical study of the geothermal energy potential of Pilgrim Springs, Alaska. University of Alaska Geophysical Institute Report UAG R-271. Turner, D. L. and Swanson, S. E. (1981) Continental rifting—a new tectonic model for the central Seward Peninsula. In Geothermal Reconnaissance Survey of The Central Seward Peninsula, Alaska (Edited by Wescott, E. M. and Turner, D. L.). University of Alaska Geophysical Institute Report UAG R-284. Waring, G. A. (1917) Mineral springs in Alaska. U.S. Geol. Surv. Water Supply Paper 418. Wescott, E. M. and Turner, D. L. (1982) Geothermal energy resource assessment of parts of Alaska. Report to the Div. of Geothermal Energy, D.O.E.