LOWELL E. GARRISON 4860 Hope Lane, Sacramento, California 95821

Geothermal Steam in The Geysers-

Clear Lake Region, California

ABSTRACT

Dry, superheated natural steam is producedfor the generation of electrical power from wellsin the Mayacmas Mountains, north-centralCalifornia Coast Ranges. The indicated sourceof the heat is a shallow intrusive magma bodyemplaced in Quaternary to Holocene time.Magmatic heat is transmitted through thelargely impermeable country rock by conduc-tion. Convectional heating takes place in streamreservoirs, predominantly zones of permeabilityand porosity caused by faulting and shearing.The thermal fluid is derived from meteoricwater, but some portion is probably derivedfrom the magma. It is, at least in the exploitedpart of the system, in the vapor state. Thiscontrols the temperature and pressure regime,and reservoir pressures are lower than hydro-static for similar depths in a liquid-dominatedsystem. Generation capability was 182,000 kwhin 1971, and a capability of more than 600,000kwh is projected for 1975.

INTRODUCTIONElectrical power is generated from dry,

superheated geothermal steam obtained fromwells in the Mayacmas Mountains of the ClearLake region, California Coast Ranges, about90 mi north of San Francisco (Fig. 1).

HISTORY OF GEOTHERMALSTEAM PRODUCTION

J. D. Grant of Healdsburg began drilling awell in 1921 at The Geysers fumarole tract onBig Sulphur Creek, Sonoma County, to obtainnatural steam for electrical generation.

His well blew out with steam at a shallowdepth when closed-in, and was abandoned. Thenext year the well, known as no. 1, was drilledto 203 ft, and completed as a steam well with8-in, casing to a depth of 80 ft. The initial staticpressure was 64 lb. A second well was begunlater that year within 50 ft of well no. 1. This

was drilled to 318 ft and completed with 67 lbinitial static pressure. A third well was drilledin 1924 but was abandoned at 154 ft. Duringthese early years, The Geysers DevelopmentCompany was organized to develop the en-couraging natural steam properties. In 1925,the company drilled five steam wells, whichwere completed with varying steam capacities.The steam was superheated by 15° C to 25° C.The maximum temperature recorded was190° C in well no. 5, which also had the highestshut-in pressure of 167.5 psig.

The combined capability of wells 4, 5, 6, and7 was 137,500 lb of steam per hour, which wasestimated to represent a generating capacityof 4,500 kwh. Well no. 5 was the largest pro-ducer with 52,000 Ib/hr. As there was at thetime no market for the energy, the project wassuspended. The wells were abandoned in 1969by the present operator.

In 1955, Magma Power Co. obtained a 99-yrlease on The Geysers Development Co. prop-

S A C R A M E N T O

SAN FRANCISCO

Figure 1. Northwestern California, showing generalarea of steam wells in Clear Lake region.

Geological Society of America Bulletin, v. 83, p. 1449-1468. 15 figs., May 19721449

1450 L. E. GARRISON

erties, which included other lands and hotsprings localities. Magma Power Co. formed ajoint venture with Thermal Power Co. andfrom 1957 through 1966 drilled a total of 42geothermal wells at The Geysers, SulphurBank, and the Little Geysers.

The steam-producing capabilities of the 42wells ranged from an insignificant amount(Sulphur Banks no. 1 original hole) to as muchas 312,000 Ib/hr (Happy Jack no. 1). Few dataare officially released about the wells, but byassigning a capability of 150,000 lb per hour to

the unknown steam flows, the wells have araw capability of about 4.7 million lb of steamper hour. At the conversion rate in currentpractice of 18.53 Ib/kwh, this represents253,600 kwh.

In the summer of 1966, Union Oil Co. drilledthe Ottoboni no. 1 well in an area north of theMagma-Thermal development. The well wascompleted for 167,000 Ib/hr. Magma-Thermaland Union later agreed to pool their leases andto jointly develop the prospective and produc-tive acreage of both, with Union being desig-

Figure 2. Location of steam tests and producing areas in the Clear Lake area.

GEOTHERMAL STEAM, THE GEYSERS-CLEAR LAKE REGION, CALIF. 1451

nated operator. These leased properties lie intownships 10 N., 7 W. to 10 W.; 11 N., 7 W. to10W.; and 12 N., 8 W. to 10 W.

Union and Magma-Thermal individually orjointly had drilled 70 steam wells by January1971. In addition, several older wells weredeepened with an increase in productivity, andsome wells were abandoned. Included in the 70wells is the uncontrolled Geysers no. 4 steamblowout.

In 1967, Geothermal Resources Internationalbegan drilling on its leased property west of theSulphur Bank area. Of seven wells drilledthrough 1969, three were completed with an-nounced equivalents of 10,000 kwh or higher.

In 1967 and 1968, Signal Oil and Gas Co.drilled three wildcat tests, completing one andabandoning two. In 1969, this operator dis-covered the Castle field in Lake County. As ofJanuary 1, 1971, six steam wells had been com-pleted with an estimated capability of 540,000lb of steam per hour, equivalent to 29,142 kwh.

Again, assigning 150,000 lb of steam perhour to the unknown flows, all the wells pro-ducing and those not yet hooked up through1970 have a raw capability of more than 8.5million Ib/hr, which would generate more than458,716 kwh (Fig. 2).

Data on the steam wells are summarized inTable A1, which includes information on wellsdrilled outside the presently producing areas(Sulphur Bank mercury mine and Wilbur HotSprings).

One test was drilled for oil and gas in thisregion, the McKenzie, Evelyn Awe no. 1(SW28 13N-9W). This was drilled and aban-doned at 1,679 ft in 1949.

Pacific Gas and Electric Co. purchases thesteam for electrical generation on site. Thefirst plant was installed at The Geysers in 1960with a capacity of 12,000 kwh. With theaddition of a second plant, this capacity wasincreased to 26,000 kwh. Units 3 and 4 generate56,000 kwh from Sulphur Banks. Units 5 and 6are scheduled for operation in 1971 and willproduce 100,000 kwh. Future increments ofgenerating capability are planned through 1975.Plants will normally be built when a local area

1 Tables A and B (NAPS no. 01729) may be obtainedby writing to CCM Information Corp.—NAPS, 909Third Ave., New York, New York 10023, enclosing $5for photocopies or $2 for microfiche. Make checks pay-able to NAPS.

can demonstrate a steam capability equivalentto 100,000 kwh.

CLEAR LAKE GEOTHERMAL AREA

Crustal RelationsThe Clear Lake geothermal area is defined as

including the present site of steam productionand potential areas in the Mayacmas Moun-tains, the miogeosynclinal area to the east, andthe Quaternary volcanic field.

On a regional scale, the Clear Lake geother-mal area is on the northwestern extension of theDiablo antiform described by Bailey andothers (1964), an arch in the Mesozoic Fran-ciscan rocks of western California, extendingnearly 350 mi from Parkfield at the southeastto a point northwest of Clear Lake. About 35 misoutheast of Clear Lake, there are extensivePliocene volcanic rocks in the Mt. St. Helenaand Sonoma Mountain areas. At Clear Lake,the volcanic field is no older than Pleistocene(Brice, 1953).

The eugeosynclinal, ocean-floor Franciscanrocks are, in part, contemporaneous with theshelf-and-slope Great Valley miogeosynclinalrocks. The Great Valley sequence overlies theFranciscan rocks along thrust faults, through-out the length of the Great Valley of California.This overthrust appears to represent a Benioffseismic zone (Hamilton, 1969).

Stratigraphy

General. The Mesozoic Franciscan andGreat Valley rocks and the Quaternary ClearLake volcanic series are the most widespreadrocks in the area (Fig. 3).

The oldest Tertiary strata are the "Tejon"(Eocene) and "Martinez" (Paleocene), sepa-rated by an unconformity. The combinedthickness exceeds 5,000 ft of marine sandstoneand mudstone. They crop out to the east of theMayacmas Mountains and are separated frombeds above and below by unconformities (Brice,1953).

The lacustrine beds of the Pliocene-Pleis-tocene Cache Formation unconformably over-lie Mesozoic and Tertiary beds. Up to 2,000 ftof Pliocene sediments, assigned to the MercedFormation, unconformably overlie the Fran-ciscan rocks in the southwest part of theMayacmas Mountains (McNitt, 1968).

Franciscan Assemblage. The FranciscanAssemblage is divided by McNitt (1968) in

1452 L. E. GARRISON

the Mayacmas Mountains into an upper unit,10,000 ft thick, composed of interbedded shaleand graywacke; and a lower unit, 15,000 ft,composed of diverse rock types, the most com-mon being graywacke. Basalt is the next mostabundant rock type of the lower unit, withan aggregate thickness of 7,000 ft (McNitt,1968). The basalt commonly lies at the top ofthe lower unit. Other rock types in the lowerunit include serpentinite, chert, schist, anddiabase.

The Franciscan Assemblage ranges in agefrom Late Jurassic Tithonian to Late Creta-ceous Turonian stages (Swe and Dickinson,1970). It is a eugeosynclinal suite.

Great Valley Sequence. The Great Valleysequence consists of more than 35,000 ft ofclastic marine sedimentary rocks, ranging inage from Late Jurassic Tithonian to LateCretaceous Campanian (Swe and Dickinson,1970). Broadly, the Great Valley sequence is amiogeosynclinal suite of mudstone, sandstone,and conglomerate.

Clear Lake Volcanic Series. The volcanicrocks are expressed in the large landforms,roughly north to south, of Mt. Konocti, Mt.Hannah, Seigler Mountain, Roundtop Moun-tain, Perini Hill, Boggs Mountain, and CobbMountain. They are flows and pyroclasticrocks which lie primarily upon the GreatValley sequence or the Franciscan Assemblageeast of the Mayacmas Mountains.

Rock types include dacitic flows (Mt.Konocti), dacitic and andesitic flows (Mt.Hannah and Seigler Mountain), andesite(Boggs Mountain), rhyolite and rhyolite tuff(Cobb Mountain), obsidian (tract betweenMt. Konocti and Seigler Mountain), andminor exposures of rhyolite (Caldwell Pinesand Tyler Valley).

The volume of the exposed volcanic materialsis estimated by Brice (1953) at 5.75 cu mi ofwhich dacite is the most common.

Volcanism first occurred in the early Pleis-tocene, as evidenced by tuff and basalt that areintercalated in the uppermost Cache Forma-tion (Brice, 1953). Koenig (1969) cites ap-parently unpublished radiometric ages of from3,000,000 to 50,000 yrs old. The sharp, un-weathered topography of many of these vol-canic landforms and the freshness of exposedmaterial in many areas also testify to the geo-logically recent age of the series. From therelative degree of erosion, Mt. Hannah appears

to be younger than Mt. Konocti or CobbMountain (Brice, 1953).

Structure

The Clear Lake geothermal area consists ofthe following structural elements: the Fran-ciscan terrane of the Mayacmas Mountains;the Great Valley sequence klippe; the Quater-nary volcanic pile, and its source, the postulatednear-surface magma chamber; and the alluvialBig Valley.

Franciscan Terrane. The Mayacmas Moun-tains, made up mostly of Franciscan rocks, arebroadly homoclinal in structure, complexlyfaulted, and dip northeastward. According toMcNitt (1968), stratigraphically lower andolder Franciscan is progressively exposed fromthe southwest to the northeast (Fig. 3). Faultstrend northwest, parallel or subparallel to theregional grain, and many are curved. Figure 3is based on McNitt (1968), Swe and Dickinson(1970), and the author's interpretation of aerialphotogeology of faults at the northeast belt ofthe Mayacmas Mountains.

The faults seem to be concentrically or-ganized about 5 mi west and southwest of Mt.Hannah into a system of gently imbricate orintersecting and more concentric fractures,generally concave toward the northeast. Thiscurving loses definition with increasing distanceto the southwest. The upthrown side of faultsare generally on the northeast. These faults areregarded by the writer to be predominantlyreverse, although McNitt (1968) portraysthem as predominantly normal.

The preservation of the Merced Formationin a graben in townships 11 N., 9 W., and UN.,10W. (Fig. 3) suggests that the maximum age offaulting is late Pliocene (McNitt, 1968). If theminor outcrops of Clear Lake volcanic rocks atCaldwell Pines and Tyler Valley are erosionalremnants, then the exposures, initially moreextensive, were the object of erosion that wasaggravated by uplift of the Mayacmas Moun-tains. The Quaternary alluvium at Tyler Valley(and elsewhere) is a remnant of an older drain-age system that has been disrupted by veryyoung faulting and uplift. Quaternary terracedeposits on the east side of Big Valley andsouthwest of Kelseyville contain much obsidiandebris, which is certainly derived from theobsidian deposits near Mt. Konocti. The CacheFormation near Kelseyville has been tilted and

GEOTHERMAL STEAM, THE GEYSERS-CLEAR LAKE REGION, CALIF. 1453

faulted. Lastly, seismicity testifies to continuingtectonic activity.

Topographically, there is a rough progressionin elevation across the Mayacmas Mountainsfrom southwest to northeast. This is concordantwith the progressive uplift in that direction.

In summary, the Mayacmas Mountains havebeen tectonically active from Pliocene throughvery late Pleistocene time, and probably intothe Holocene. Plutonism is likely to haveabetted this uplift. The youngest pulses of up-lift occurred in the northeastern belt.

Great Valley Klippe. The Great Valleysequence is a complexly faulted mass in thrustcontact with the Franciscan terrane. The solethrust of a number of thrusts is the Soda Creekfault of Swe and Dickinson (1970).

Tectonic Relations between Franciscan andGreat Valley units. The Soda Creek thrustfault, which separates Franciscan from GreatValley units is presumed to be the upper edgeof a fossil Benioff zone. It is masked by volcanicrocks at Boggs Mountain but cuts across theQuaternary obsidian of Camel Back Ridge.From its effect on these volcanic rocks, move-ment on this fault must have persisted into theQuaternary at least.

Quaternary Volcanic Pile and the MagmaChamber. The Clear Lake volcanic series wasapparently extruded from a system of north-west-trending fissures; Cobb Mountain is alocalized vent for the more explosive phase ofrhyolite (Brice, 1953).

The Clear Lake geothermal area is the locusof a nearly circular Bouguer gravity anomaly ofnearly 200 sq mi (R. Chapman, 1965, personalcommun.; Mineral Information Service, 1966).The center of this feature is a minimum of morethan —50 mgals, precisely underlying Mt.Hannah. Studies by Moore (1962) show adirect, causal relation between Bouguer gravityresponse and Cenozoic volcanic rocks based onNiggli ^-values. These values are high for theClear Lake volcanic rocks. Where ^-values arehigh, the Bouguer gravity response is a low.Moore (1962) concludes that the connection ismore related to the geographic location of anigneous body than to its age or mode of em-placement.

The arcuate fault pattern about Mt. Hannahand the Bouguer gravity contours show cor-respondence. This is interpreted as a reflectionof the volcano-tectonic uplift. Here is a parallelto the Italian steam-producing areas of

Larderello and Monte Amiata, where there isagreement between Bouguer gravity contoursof minima; curved, concentric fractures; anduplift (Marchesini and others, 1962).

The source of the volcanic products at ClearLake is presumed to be a relatively shallowmagma body. Its agressive emplacement wouldhave arched and uplifted the sedimentary roof,allowing the extrusion of the Clear Lake vol-canic rocks along fissures and vents. The result-ant structure is volcano-tectonic in nature.

SeismicityThe Clear Lake region is seismically active.

Epicenters of minor and medium shocks havebeen in the Mayacmas Mountains, near Kelsey-ville, and in the Lakeport area on the west shoreof Clear Lake. Historically the strongest tremorin the region was the June 6, 1962 earthquakewith an epicenter approximately 10 mi north-west of Kelseyville, at a modified Mercalliintensity of VII.

HOT SPRINGS

The principal hot springs of the Clear Lakegeothermal area are summarized on Table B,and locations are shown on Figure 4.

These thermal and mineralized waters maybe compared to examples of regional groundwater at Alexander and Cloverdale valleys tothe west of the Mayacmas Mountains (Table 1).

Certain of the hot springs have geochemicalindicators or parameters by which a geothermalpotential might be identified.

1. Calcium. Calcium is the principal cationin fresh natural waters (Hem, 1970). In waterswith a suspected magmatic contribution, thesolubility of calcite decreases with increasingtemperature, to solubilities controlled largelyby salinity and/>H (White, 1970).

2. Chloride. Chlorides are typically in lowconcentrations in acid-sulfate type waters in avolcanic environment. White (1970) remarksthat no liquid water associated with a vapor-dominated stream reservoir is known to have aCl concentration in excess of about 15 ppm, andconversely, a chloride water body with morethan 50 ppm is suggestive of a hot-water sys-tem. The Geysers and Castle hot springs, bothsurface waters of a vapor-dominated system,have low Cl contents. Cl may be in relativelyhigh concentrations in acid-sulfate-chloridewaters of volcanic environments (White and

1454 L. E. GARRISON

REGIONAL BOUGUER GRAVITYFigure 4. Regional Bouguer gravity contours, after springs.

Mineral Information Service (1966), and location of hot

HOT SPRINGS

others, 1963), although sulfate is the dominantanion.

3. Lithium. Relative contents of Li and theLi/Na ratio may reflect lithium from volcanicemanations (White and others, 1963). Relative-ly high Li/Na ratios may suggest volcanicemanations in the water.

4. Magnesium. Low Mg contents are char-acteristic of a high temperature system (White,1970). It is generally present in low amounts inwaters of volcanic environments (White andothers, 1963). Low Mg/Ca ratios are similarlycharacteristic of a high temperature hydro-

thermal system (White, 1970). High concen-trations can be derived from reaction of waterwith magnesium silicate minerals, as at SeiglerSprings (Hem, 1970).

At high temperatures, Mg favors a solidphase such as montmorillonite (White, 1970).Bailey and others (1964) have described Fran-ciscan graywacke near The Geysers largely asbeing hydrothermally altered to a montmoril-lonite-rich rock.

5. Silica. SiOj is normally present in highconcentrations in thermal waters with a sus-pected magmatic contribution. Fournier and

GEOTHERMAL STEAM, THE GEYSERS-CLEAR LAKE REGION, CALIF. 1455

TABLE 1. GROUND WATER IN ALEXANDER AND CLOVERDALE VALLEYS*

Chemical constituentor property

Ca

Mg

Na

HC03

Cl

B

total solids

PH

temperature

Concentration inparts per million

or measurement of theproperty

.5 to 60

11 to 275

15 to 143

202, 300, 680

8.5 to 9.3 in 4samples , 555 in one

zero to 40

510 and 2,650

7.1 and 7.3

14°C and 18°C

Source: (Cardwell, 1965).*Si02, apparently absent.

Rowe (1966) refined a concept for estimatingsubsurface temperature from the amount ofdissolved SiO2 and the temperature of the sur-face spring or flow. White (1970) modifiedFournier and Rowe, and his temperatureestimation chart is included as Figure 5 A.

6. Sodium. Figure 5B shows experimentaland empirical curves for atomic Na/K ratios,derived by White (1970). Curve "G" is pro-posed for general use in estimating subsurfacetemperatures of a hydrothermal system. Theatomic Na/K ratio has proved about as reliableas SiC>2 contents for predicting temperaturesfrom deep-well samples above 200° C (White,1970).

7. Total Solids. Thermal and mineralizedwaters associated with volcanism or epithermalmineral deposits should be expected to havegenerally more dissolved solids than naturalground waters from the same region.

8. Temperature. Abnormally high tem-perature of surface springs in a volcanic envi-ronment is usually the first indicator of ahydrothermal system. Temperatures of watersrelated to epithermal mineral deposits may alsobe high, for example Sulphur Bank mercurymine and Wilbur Springs.

9. Other Qualities. Cold CO2 seeps arelocated in sections 11 and 12, T. 11 N., R. 10W. This is the principal noncondensable gas at

50 100 150 200 250 300

TEMPERATURESILICA/TEMPERATURE

Figure 5. Silica/temperature and atomic Na/Kcharts for estimation of underground temperature afterWhite (1970). White's curve "G" is shown in theatomic Na/K chart, which is recommended for generaluse.

7060

50

40

SO-

Ii 20cr

10-

9-

5-

4-

TEMPERATURE °C

100 150 200 250 3003504005001 'i ' ' 'i ' ' ' 'i ' ' ' ' I i" h ' 1 '—

W-PROJECTED

\

y (WEIGHT! X 1.70= ^(ATOMIC)

•-V- J-r-1

2.8 2.6 2.4 2.2 2.0 L8 1.6 1.4I/T X I03 (KELVIN)Na/K RATIO

1456 L. E. GARRISON

The Geysers steam production. Seeps of CH4are known from the Clear Lake geothermalarea (Thurston Lake and Kelseyville). TheSulphur Bank no. 4 steam well was originallydrilled to 1,680 ft and suspended as a cool non-commercial well; it was, however, capable offlowing measurable amounts of CH4 at 125 lbshut-in pressure. Radioactivity was first notedat The Geysers steam well flows in the 1920s(Bradley, 1922).

The Geysers and Castle are the two groups ofhot springs known to be related to a vapor-dominated geothermal system in the ClearLake geothermal area. They are characterizedby (1) high temperatures, (2) high concentra-tions of silica, (3) low concentrations ofchloride, and (4) high concentrations of sulfate.Other geochemical and physical parametersmay be quite variable. A detailed, comparativestudy of all geochemical and physical para-meters is an essential tool in the exploration anddevelopment of geothermal resources.

STEAM PRODUCTION

NatureAs far as investigated, the Mayacmas Moun-

tains steam production is of the vapor-dom-inated type system, recently described byWhite and others (1971). It is thereby char-acterized by the produced steam being dry andsuperheated, and existing at rather uniformtemperatures and pressures lower than hydro-static for similar depths in a liquid-dominatedsystem.

The vapor within the system is derived fromboil-off from a deep water table. This deepwater table must lie at depths greater than9,000 ft, that is, beyond the penetration of thedeepest wells to date. Within the vapor range,temperature and pressure behavior is basicallythat of a continuous steam phase. The mostprobable temperature and pressure range for avapor-dominated system is 236° C to 240° Cand 32 kg/cm2 to 34 kg/cm2 (455 to 483 psi).Temperatures and pressures may exceed theselimits because of the effect of dissolved solidsand the pressures of other gases (Fig. 6).

The produced steam is derived from localpore-space water in the vapor-dominatedreservior above the deep water table. Liquidwater and vapor coexist in these rocks. Therocks are gravity drained as steam and othergases rise in the largest channels and condenseat the reservoir borders. This condensate

percolates down into the reservoir, attracted tothe narrow channels and pore spaces because ofsurface tension and the lower specific resistanceof the liquid flow in relation to that of vapor(White and others, 1971).

The reservoir water body and its boiled offvapor are primarily meteoric in origin. Thehydrothermal system is replenished by therecharge of meteoric waters. With a sufficientlypotent heat source, or a decreasing rechargerate, the boil-off will exceed recharge. At thispoint, an initially hot-water system becomesvapor dominated. Heat from the presumedmagmatic source is transferred by conductionthrough the Franciscan country rock whichhas slight to nil permeability. The steam reser-voirs are primarily shear zones, related to fault-ing. Within these zones, with amplified per-meability and porosity, convectional heating,with its lower thermal gradient and equalizingeffect, prevails2.

Superheat is due to the fact that the deep-water body is a brine and is superheated inrespect to pure water at the same pressure.Steam boiled from a 25 percent NaCl brine at35 kg/cm2 is superheated by 12° C (White andothers, 1971). Superheat also occurs withdecompression. Most of the heat content of thereservoir is stored in the solid state (White,1970). The shear zones within the hydro-thermal system are presumedly the avenues forrecharge, as well.

The steam reservoirs are isolated laterally andfrom above and below by cooler regions whichare effectively impermeable. This can be due toinnate impermeability of the rock, or deposi-tion of minerals which seal the reservoir. Silicamay be deposited by hot waters at the reservoirboundary as they flow into cooler regions.Conversely, cool waters may deposit CaCOsand CaSO4 as they flow into heated regions.

At the borders of the reservoir, wells mayencounter liquid water that may be accom-panied by steam. This liquid is derived from

2 The Franciscan "sandstones" (graywackes) areestimated (Decius, 1961) to have permeability less than 1millidarcy and porosity less than 10 percent. Thesefigures are probably high, except for shear zones orlocalized stringers. Grindley (1965) states that parts ofthe Wairakei reservoir with permeabilities of 1 millidarcyare nonproductive. Wells drawing on Wairakei acquiferstorage (that is, innate permeability) have permeabilitiesof from 5 to 30 millidarcies; wells drawing on faultreservoirs have permeabilities on the order of 1 darcyand higher.

GEOTHERMAL STEAM, THE GEYSERS-CLEAR LAKE REGION, CALIF. 1457

10120 140 160

TEMPERATURE °C180 200 220 240 260 280

oCJin

50-

100-

(O 200-trUJ

UJ500-

Q. :UJ 1000-a

4000-

REFERENCE BOILING-POINTCURVE FOR PURE WATER

X Gl MAXIMUM TEMPERATURE AT THESURFACE IN FLOWING WELLS, PLOTTEDAT DRILLED DEPTH (ALLEN 8 DAY, 1927)

+ TI MAXIMUM TEMPERATURE MEASUREDIN HOLE BY THERMOCOUPLE AT PLOTTEDDEPTH (MC NITT, 1963)

A Ml MAXIMUM TEMPERATURE MEASUREDIN HOLE BY 6EOTHERM06RAPH ATPLOTTED DEPTH (WHITE, I957o)

V SB3 SULPHUR BANK AREA OF THE GEYSERS iSATURATION TEMPERATURES CALCULATEDFROM SHUT-IN WELL HEAD PRESSURES CORRECTEDFOR WEIGHT OF STEAM. PLOTTED AT DRILLEDDEPTHS (NUMBER WELLS, 0-D.OTTE 8 OONDAN-VILLE, 1968 FROM GENERATED PRESSURES OFFIELD ASSUMED IN-HOLE. ABSOLUTE, AND NOEFFECT FROM OTHER GASES)

PREDICTEDTEMPERATURESIN 2 PHASE VAPOR-DOMINATED RESERVOIR •(PURE WATER 8 STEAM;SEE TEXT)

eo-oWATER-TABLE ASSUMED"

APPROXIMATE TEMPER-ATURES BELOW BRINEWATER TABLE

I I I3 5 10 15 20 25 30 40 50 60 70

SATURATION PRESSURE KG/CM2

Figure 6. Boiling point/depth curve for The Geysers from White and others (1971).

100

UJUJu.

-1000

UJo

-10,000

condensing steam or recharging water. Withproduction, parts of a hot-water geothermalsystem may become vapor-dominated, as is thecase at Wairakei (Grindley, 1965).

The Mayacmas Mountains vapor-dominatedsystem is different geologically from the otherproved examples of the type, Larderello andMonte Amiata. At the Italian region, thereservoir is a thick (often exceeding 50 m) bodyof anhydrite-bearing limestone. The limestoneis overlain by an impermeable allochthonousseries that provides the lid for the geothermalsystem. The limestone has a high innate per-meability.

Configuration of the Reservoir at TheGeysers and Sulphur Banks

Figures 7 through 12 are constructed fromthe interpretation of certain parameters ob-tained during the drilling of the steam wells.First steam shows depth, relative to sea level,where steam was first encountered. First steamis not necessarily commercial steam, and themain reservoir lies deeper. The first, uppermost,steam zone is likely to be relatively permeablebeds in the upper unit of the Franciscan.Maximum steam flow contours isopleths ofthousands of pounds of steam per hour. The

FIRST STEAM

Figure 7. First steam as encountered in wells at The Geysers and Sulphur Bank.

MAXIMUM STEAM FLOW: OOO'S #/HR

Figure 8. Maximum flow of steam from wells at steam per hour.The Geysers and Sulphur Bank, thousands of pounds

FLOW MEASUREMENT: CROSS SECTIONAL AREA, ORIFICE X WHFPFigure 9. Flow measurement in wells at The Geysers and Sulphur Bank (derived as described in text).

ABSOLUTE SHUT IN PRESSURE AS A FUNCTION OF DEPTH (X/O)Figure 10. Absolute shut-in pressure as function of Geysers and Sulphur Bank,

depth (derived as described in text) of wells at The

PRODUCTIVE INDEX: STEAM ooo's #/HR ACCORDING TO VOLUME OF HOLE, FIRST STEAM-T.O.Figure 11. Productive index (derived as described in text) of wells at The Geysers and Sulphur Bank.

CONTOURS ON 'S' CORRELATIVE HORIZON

Figure 12. Contours "S" correlative horizons, Sulphur Bank,determined from constructed logs at The Geysers and

FranciscanAssemblageUpper Unit

FranciscanAssemblageLower Unit

SEA LEVEL-

'AIR COMPRESSOR

Figure 13. Constructed log section across Sulphur Bank.

1462 L. E. GARRISON

depth and hole volume are disregarded. Flowmeasurement is a function of the maximumsteam flow and the cross-sectional area of theorifice through which the flow passes. Absoluteshut-in pressure as a function of depth is empirical,attempting to remove the effect of depth inequating reservoir pressures. The formula usedis total depth divided by shut-in pressure timesten. Productive index is a measure of the wells'producibility according to volume of holebetween first steam and total depth. Thisattempts to equate productivity in relation tothe variables of depth and hole size. Contourson "S" correlative horizon show an interpretivestructural configuration. This is a correlativemarker on mechanical logs constructed of thesteam wells.

The cross-section (Fig. 13), made up of someof these constructed logs, shows the "S"horizon as a rudely correlative marker at ornear the base of an interval of fast drillingbelow first steam, and overlying a somewhathomogeneous interval of slower drilling, in-creasing air compressor demands, and moreefficacious steam zones. For convenience, it istaken to be the contact between the upper andlower Franciscan units of McNitt (1968).

First steam, absolute shut-in pressure, produc-tive index, and contours show an elevated trend.An elevated trend mapped by maximum steam

flow and flow measurement is offset to the north,in large part because of deeper drilling here.More steam will be won from a more exposedreservoir.

The elevated trend gains credibility due tothe agreement of the parameters and theirlocation on the broken, discontinuous, andcommonly masked anticline which is mappedthrough the area by McNitt (1968). Such anassociation between structural highs andreservoir potency is seen at Larderello andWairakei.

Wells encountering what would be classed aslarge steam flows, above 150,000 Ib/hr, en-counter such flows in a relatively thick intervalnear total depth, where pressures are high, as arule above 30 lb on an 8.75-in. surface flow line.This, the principal reservoir, is a system of shearzones. Working with limited data, it appears asif this interval becomes progressively shallower(relative to sea level) in a southwesterly direc-tion. A reverse fault with a shear zone, up-thrown to the east, would also become progres-sively shallower in this direction.

Behavior of the Reservoir

At the Sulphur Banks and The Geysers, flowand shut-in pressures of the individual wells areaffected by periods of common flow or shut-in.The Sulphur Banks well no. 3 recorded aninitial steam flow of 85,000 Ib/hr with 74 psigflowing pressures. Later, when all the then-completed wells had been continuously pro-ducing for three weeks, the flow from this wellhad declined to 57,000 Ib/hr, but the pressurehad increased to 85 psig. This well's initial shut-in pressure was 183 psig; after the flow period,it had increased to 195 psig.

Well head flowing pressures tend to increasewith decreasing steam flows. There are, how-ever, exceptions, notably the smaller steamflows (Fig. 14).

Shut-in pressures, measured at the surface,range from about 100 to 500 lb (Fig. 15). Aslittle increase in pressure with increasing depthis expected in a vapor, the maximum shut-inpressure is close to the upper range suggestedfor a vapor-dominated, dry-steam system. The

STEAM FLOW VS WELL HEAD FLOWING PRESSURES

Figure 14. Steam flow v flowing pressures forrepresentative wells at The Geysers and Sulphur Bank.

SHUT-IN PRESSURES (PSIG) VS. TOTAL DEPTH

LEGEND

• 215 REFERS TO STEAM FLOW PER IN OOO'S POUNDS

BC BARELY COMMERCIAL

TSTM TOO SMALL TO MEASURE

6000

5000

4000

LdLUU-

z

XI-o.LU

t-o

3000

2000

1000

0 100 200 300 400 500 600

Figure 15. Shut in pressure v depth for representative wells at The Geysers and Sulphur Bank.

1464 L. E. GARRISON

slight apparent excess can probably be attrib-uted to the presence of other gases. At TheGeysers, less than 2 percent of the total vaporconsists of gas other than t^O, the most abun-dant being CO2 (White and others, 1971).

Large steam flows tend to occur at high shut-in pressures, although the association is notlinear.

In 1968, after an 11-yr production history,a decline of 50 lb was noted at the shallow zoneof The Geysers, at depths from 400 to 2,000 ftand probably the upper unit Franciscanlithologic reservoir. Reservoir studies of thiszone predict an ultimate recovery of 110 billionlb of steam, from which 47 billion lb have beenproduced as of January 1968 (Reich and Reich,1968).

Steam temperatures cluster about a referenceboiling-point/depth curve for pure water to amaximum of about 240° C (White and others,1971; Fig. 6 of this paper). This temperaturemaximum conforms to the vapor-dominatedmodel. There does not appear to be a directrelation between temperature and rate of flow.Temperature, however, does have a dependenceupon steam pressure (McNitt, 1963): tem-perature increases with increasing pressure.McNitt (1963, p. 15) shows an increase of tem-perature from 169° C to 188° C accompany-ing an increase of flowing steam pressures from59.7 psia to 132.1 psia for three wells at TheGeysers.

The temperature/pressure relation is alsoapparent in static pressures, although thisrelation again is not linear. In Figure 6, the SB10 well has a temperature of approximately233° C; the SB 7, 228° C; the SB 6,214° C; theSB 3, 197° C; T 10, 186° C; and T 1, 182° C.The maximum initial shut-in pressures for thesewells were, in the order above, 460, 450, 320,195, 140, and 80 lb (psig).

Enthalpies of 1206 and higher Btu/lb arefound in the steam reservoirs of the MayacmasMountains. As Whiting and Ramey (1969)point out, when liquid and gas coexist in asteam system, the enthalpy of the steam cannotexceed the maximum two-phase envelopeenthalpy of 1204.6 Btu/lb. This concept mustbe modified by regarding the produced steam,only, as being in the vapor state.

While the data are scanty, they suggest thattemperatures can in places actually decline withdepth, as revealed in drilling experience. TheGeysers no. 12 had a bottom hole temperatureat 1,450 ft of 109° C, but at total depth, 1,950

ft, the temperature was 102° C. The SulphurBanks no. 2 had bottom hole temperatures of185° C at 1,825 ft. These decreased in less than100 ft more penetration to 149° C. Both thesewells were drilled with drilling mud which musthave insulated the temperature, but the spreadin the Sulphur Banks example particularly isdifficult to explain away by simple mud ordrilling conditions. A recent example of airdrilling, the Signal, McKinley no. 4, measuredflow-line temperatures of 126° C when drillingat 5,945 ft. Temperatures decreased to 82° Cwhen at 7,020 ft. Some 400 ft deeper, theywere 137° C, declining again to 112° C at totaldepth of 8,047 ft.

In the Mayacmas Mountains, it is felt thatoccasions of lower temperatures in the sub-surface can be partly due to influxes of cooler,recharging waters. Also, the shear zone reser-voirs might be hotter than other sections of thecountry rock.

A zone nearly saturated with liquid waterderived from the condensing steam overlies thereservoir. This zone (McNitt, 1963) has beenobserved to rise when reservoir pressures werebeing reduced through discharge and to fallwhen reservoir pressures were being increasedthrough shut-in. James (1968) suggested thatthis indicates that the underlying steam reser-voir seeks hydrostatic balance.

Current Production and Exploration inRegional Framework

All production to date is directly related toprominent, regionally northwest-trendingfaulting and its junction with intersectingfaults, or as seen in its arcuate, imbricate aspect.Steam production moreover is localized toMayacmas Mountains blocks that have under-gone relatively the most uplift. Except possiblyfor the shallow zone at The Geysers, productionis restricted to the in-bulk impermeable lowerunit of the Franciscan. Reservoirs are shearzones associated with the faulting. More shear-ing exposes more reservoir. Spatially, produc-tion to date is located from 7 to 8}/2 rni southand southwest from Mt. Hannah. This is theyoungest volcanic landform, and is marked bya gravity anomaly that conceivably reflects thenearest-to-surface apophysis of the magma heatsource.

Certain unsuccessful tests of Signal andUnion are located off fault intersections, or elseare relatively downdip on blocks betweenprincipal regional faulting. The Signal, Wild-

GEOTHERMAL STEAM, THE GEYSERS-CLEAR LAKE REGION, CALIF. 1465

horse no. 1, recorded flow-line temperatures of155° C at 4,830 ft and 209° C from 6,500 ft tototal depth. Insignificant puffs of steam issuedfrom some of the numerous zones of fast drill-ing, a quantity of which must represent fis-sures. These flow-line temperatures are withinthe range recorded at Castle steam field, butare generally lower than those of The Geysers.

The G.R.I. Rorobaugh area is on the west ofthe faulting of the Sulphur Bank and TheGeysers fields. This development is not assuccessful as Sulphur Bank, one-half mileaway. Some of these wells produce hot waterwith steam. Considering the Sulphur Bank-Geysers faulting as reverse in nature leads to theconclusion that it does not cut the G.R.I, wellsand they therefore do not benefit from it. Theyare located on the cooler boundary of a vapor-dominated system.

DRILLING PRACTICES

The Grant and Geysers Development wellswere drilled with rotary equipment, the steamin the bore hole being controlled by injectedstreams of cold water. Surface pipe (usually10-in. diam.) was set as far as possible into hardrocks. Drilling progressed until the steam flowwas sufficiently large to warrant the setting ofan intermediate string (usually 8 in. diam.).The well was then drilled beyond to total depthas an uncased, open hole.

The first Magma-Thermal wells were drilledconventionally with rotary equipment anddrilling mud. The string usually included per-forated casing opposite the steam zones, as inearly oil-field practice. The mud, baking ondown-hole hot surfaces, was found to inhibitsteam flow, and starting with the Geysers no. 7,drilling was carried on with air below the sur-face pipe. Hole for conductor and surface pipe,however, was still drilled with mud.

For shallow wells, conductor and as much as500 ft of surface pipe (usually 10% or 13% in.)are set, and the well is uncased and open tototal depth (usually 9%-in. hole). In deeperwells, hole size is commonly reduced near thefinal third of penetration. Later deep wellsnormally set conductor and surface pipe as wellas an intermediate string at depths from 2,000to 4,000 ft.

Air, the circulating medium for drilling, isproduced by compressor banks. The recentSignal wells use four compressors, of whichthree are used and one is kept on standby. Eachis capable of putting out some 900 cu ft/min.

This is sufficient to carry cuttings up theannulus to be blown out the flow line. Drillingbeyond the producing zone is carried on undera condition of "controlled blowout" (Hun-nicutt, 1970). The mass flow of even a 100,000Ib/hr steam well will carry abrasive cuttingsfrom the bit that can destroy a drill string ina few hours (Hunnicutt, 1970).

Drilling times for representative deep wellsare shown on Table 2.

Drilling is expensive, a 6,000-ft range testcosting $150,000 or more. Completion equip-ment is a minor additional expense. Electricallogging of steam wells is not practical becauseof the extreme heat down-hole, except thatoperators sometimes electrically log the upperportion of the bore through which surface orintermediate pipe is to be set.

Steam zones are often identified by fastdrilling and a surge from the air compressors inreaction to a steam flow.

The problem of disposal of borated watersfrom the condensing generators arose in 1968.In 1969 the suspended Sulphur Banks no. 1 wastested for water disposal. Injection rates of upto 10,000 barrels per day had no apparenteffect on nearby steam producers. The operatorultimately drilled a separate disposal well one-quarter mile to the west of Sulphur Banks.

STEAM RESERVES

Reserves of producible steam should bethought of as being measured by the life of thesystem. The geothermal system should endureas long as heat is being supplied by the source,withdrawal of steam does not exhaust thethermal fluid, the system is recharged, and theessential nature of the reservoir and cap doesnot change. Nevertheless, with continuingdraw-off, the steam reservoir will undergo somechange.

Material and energy balance reservoir cal-culations have been applied to Wairakei(Whiting and Ramey, 1969). The pressure drop

TABLE 2. REPRESENTATIVE DRILLING TIMES

Depth (in feet) Days drilling

4,8684,9005,8125,9496,2777,2547,532

19182938273053

1466 L. E. GARRISON

in the shallow zone at The Geysers permits apressure/cumulative estimation of reservesthere.

The operator at The Geysers and SulphurBanks has not released reservoir performancedata. It is presumed that he has conductedflowing and static pressure/cumulative tests.

Larderello has been experiencing a generaldecline in pressure and output, with scanty datasuggesting an increase in temperature (EnteNazionale per 1'Energia Elettrica, 1971). Acontinuous drilling program is necessary to sus-tain the required steam flows for its 365.1 MWgenerating capacity. Production and powergeneration have risen from first production in1913 to the present (2.587 million kwh in1969). This growth is in part due to exploitationof the entire boraciferous region, with thedevelopment of new, often separated producingareas within the hydrothermal province.

The Mayacmas Mountains province's actualcapability, measured in steam producibility, isat present higher than the projected generatingcapacity of 600 MW. The ultimate capabilitymight be in the 2,000 MW to 5,000 MW range.The life of the system can be measured in in-crements of a century. The expected lifemultiplied by the median projected capacitywould give approximate "reserves."

BRIEF ECONOMICS OFGEOTHERMAL STEAM

Purchase Price of Steam

The original purchase price for steam at units1 and 2 at The Geysers was 2.5 mills/kwh. Foradditional units, the price is calculated yearlyon the basis of a weighted average of 2.5 millsadjusted for current fuel costs and the best heatrate for fossil fuels compared with the Decem-ber 1958 fuel costs and the best heat rate andthe average nuclear fuel costs. Recently, .14mills was awarded the operator for his under-taking to dispose of condensate from the plant.

Production Expenses

The production expenses, including purchaseprice, average 5.65 mills/kwh for units 1 and2. The figure is 4.71 mills for units 3 and 4. Newunits 5 and 6 are designed to operate at 4.38mills/kwh at 90 percent plant factor, 5.02 millsat 70 percent, and 5.51 mills at 60 percent. The

average of the three plants at 90 percent is 4.91mills/kwh. Subtracting the steam purchaseprice results in a generating cost of 2.27 mills/kwh.

Using units 3 and 4 as an example, the netyearly production at 90 percent is 44,150,400kwh. At 4.71 mills per kwh, the yearly produc-tion expense would be $207,948, or annualproduction costs of $2.12 per kilowatt.

Plant CostsThe capital cost of the 26 MW units 1 and 2

is $3,800,000 or $146 per gross kilowatt in-stalled; for the 56 MW units 3 and 4, the cost is$6,900,000 or $123 per gross kilowatt.

Steam ConversionIn practice at The Geysers and Sulphur

Bank, 18,53 lb of steam will produce 1 kwh.Huge volumes of steam are utilized in geo-thermal plants in contrast to a fossil fuel plant.In this sense, a geothermal generating plantmay have a low efficiency (for example, 14.3percent for unit 3).

History of GenerationThe following (Table 3) is a 10-yr history of

electrical generation (courtesy Pacific Gas andElectric Co.).

Incremental BenefitsGeothermal power is incremental; that is,

plants can be augmented as producing capabil-ity is increased. Hydroelectric and fossil fuelplants, in contrast, must be built up at first tomaximum anticipated use.

TABLE 3. NET ANNUAL ELECTRICAL GENERATIONFOR TEN YEARS 1960 TO 1970

AT THE GEYSERS AND SULPHUR BANK

Year-Annual net generation

(MW hours)

1960

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

Total

33,750

94,504

100,461

167,953

203,790

189,210

188,001

316,310

435,827

614,710

525,177

2.869,693

Note: Units 3 and 4 at Sulphur Bank on stream 1967.

GEOTHERMAL STEAM, THE GEYSERS-CLEAR LAKE REGION, CALIF. 1467

Retail Value

The retail market value of 1,000 kwh inCalifornia is $360 per day. One kilowatt there-fore is worth 1.5fS. Therefore, 18.53 lb of steamare worth in retail value equivalent, 1.5jf; onepound of steam is worth .OSljS. Union Oil Co.equates a 100,000 Ib/hr steam well to a 200barrel of oil per day well in energy terms.

Future ProjectionsBy 1975, an incremental 600,000 kwh

capacity is projected for the area developed byUnion Oil Co. The natural producing potentialof the Mayacmas Mountains geothermal prov-ince is estimated, however, to be many timesthis figure.

Comparison to Other Forms of ElectricalGeneration3

Production Expense. The Geysers: 2.27mills/kwh; steam-electric (1968 weightedaverage): 3.43 mills/kwh; hydroelectric (1968weighted average): .53 mills/kwh. Despite thehigher production expense in geothermal plantscompared to hydroelectric, the annual produc-tion costs are lower, $2.12 per kilowatt com-pared to $2.72 per kilowatt for hydroelectricplants.

Plant Costs. The Geysers (average of units1 through 4): $130 per kilowatt installed;steam-electric (three selected plants aboutequivalent in size to pairs at Sulphur Bank andThe Geysers): 70 MW-$182, 73.4 MW-$157, and 50.5 MW-$176 per kilowatt in-stalled; hydroelectric (average of 21 U.S.systems in 1968): $270 per kilowatt installed.

Plant Factor. The Geysers and SulphurBank: 90 percent by design; steam-electric(average in U.S. 1968): 54 percent; hydro-electric (average of 21 U.S. systems in 1968):percent.

Geothermally generated electrical power isthus competitive with or superior to otherforms of generation.

ACKNOWLEDGMENTSThe writer wishes to acknowledge the

stimulus given by E. B. Towne of San Fran-cisco, a geologist and developer of geothermal

3 Data concerning steam-electric and hydrothermalsystems are taken from Federal Power Commissionannual supplements and are not cited in the references.

resources, in the writing of this paper. Thebenefit and encouragement of D. E. White'scritical review and interest in the manuscriptis acknowledged with special thanks. Signal Oiland Gas Co. was most co-operative in providingmaterial for use in the paper.

REFERENCES CITEDAllen, E. T., and Day, A. L., 1927, Steam wells and

other thermal activity at "The Geysers,"California: Carnegie Inst. Washington Pub. 378,p. 1-106.

Averill, C. V., 1947, Mines and mineral resources ofLake County, California: California Jour.Mines and Geology, v. 43, no. 1, p. 15-40.

Bailey, E. H., Irwin, W. P., and Jones, D. L., 1964,Franciscan and related rocks, and their signif-icance in the geology of western California:California Div. Mines and Geology Bull. 183,p. 1-177.

Berkstresser, C. F., Jr., 1968, Data for springs in thenorthern Coast Ranges and Klamath Moun-tains of California: U.S. Geol. Survey open-filerept., p. 1-49.

Bradley, W. W., 1916, Lake County, in Fourteenthreport of the State Mineralogist: CaliforniaState Mining Bur., v. 14, p. 202-240.1922, Radioactivity in thermal gases at The

Geysers, Sonoma County, California, inEighteenth report of the State Mineralogist:California State Mining Bur., v. 18, p. 545-550.

Brice, J. G., 1953, Geology of Lower Lake quad-rangle, California: California Div. Mines andGeology Bull. 166, p. 1-72.

Cardwell, G. T., 1965, Geology and ground waterin Russian River Valley areas and in Round,Laytonville and Little Lake Valleys, Sonomaand Mendocino Counties, California: U.S.Geol. Survey Water-Supply Paper 1548,p. 1-154.

Decius, L. C., 1961, Geological environment ofhyperthermal areas in continental UnitedStates and suggested methods of prospectingthem for geothermal power: United NationsConf. on New Sources of Energy, Rome, PaperG48, p. 166-178.

Ente Nazionale per 1'Energia Elettrica (ENEL),1971, Larderello and Monte Amiata: Electricpower by endogeneous steam: Rome, p. 1-41.

Fournier, R. O., and Rowe, J. J., 1966, Estimationof underground temperatures from the silicacontent of water from hot springs and wet-steam wells: Am. Jour. Sci., v. 264, p. 685-697.

Grindley, G. W., 1965, The geology, structure, andexploitation of the Wairakei geothermal field,Taupo, New Zealand: New Zealand Dept. Sci.and Indus. Research, Seismol. ObservatoryBull. 75, p. 1-131.

1468 L. E. GARRISON

Hamilton, W., 1969, Mesozoic California andunderflow of Pacific Mantle: Geol. Soc,America Bull, v. 80, no. 12, p. 2409-2430.

Hem, J. D., 1970, Study and interpretation of thechemical characteristics of natural water: U.S.Geol. Survey Water-Supply Paper 1473, p.1-363.

Hunnicutt, N. B., 1970, A professional driller'sevaluation of geothermal drilling and produc-tion problems: in Am. Petroleum Inst.,Northern California Chap., Field Trip to TheGeysers geothermal field, Lake and SonomaCounties, June 4th, 1970, p. 1-11.

James, R., 1968, Wairakei and Larderello—geo-thermal power systems compared: NewZealand Jour. Sci., v. 11, p. 706-719.

Koenig, J. G., 1969, The Geysers, in Mineral In-formation Service: California Div. Mines andGeology, v. 22, no. 8, p. 123-128.

Marchesini, E., Pistolesi, A., and Bolognini, M.,1962, Fracture patterns of the natural steamarea of Larderello, Italy, from airphotographs:Symposium Photo Interpretation, Delft, p.524-532.

McNitt, J. R., 1963, Exploration and developmentof geothermal power in California: CaliforniaDiv. Mines and Geology, Spec. Rept. 75.1968, Geology of the Kelseyville quadrangle,

California: California Div. Mines and GeologyMap Sheet 9.

Mineral Information Service, 1966, Gravity map ofGeysers area: California Div. Mines andGeology, v. 19, no. 9, p. 148-149.

Moore, J., 1962, Na/K ratio of Cenozoic igneousrocks of the western United States: Geochim.et Cosmochim. Acta, v. 26, p. 101-130.

Reich and Reich, Petitioners v. Commissioner ofInternal Revenue, 1969 Tax Court of theUnited States, 52. T. C. No. 74, 1970, inCompendium of papers, Imperial Valley-Sal ton

Sea Area, Geothermal hearing, State of Califor-nia, The Resources Agency, October 22 & 23,1970, Sacramento, tech. presentation EE, p.1-36.

Swe, W., and Dickinson, W. R., 1970, Sedimenta-tion and thrusting of late Mesozoic rocks in theCoast Ranges near Clear Lake, California:Geol. Soc. America Bull., v. 81, no. 1, p. 165-188.

Waring, G. A., 1915, Springs of California: U.S.Geol. Survey Water-Supply Paper 338, p.1-410.

White, D. E., 1957, Thermal waters of volcanicorigin: Geol. Soc. America Bull., v. 68, no. 12,p. 1637-1658.1970, Geochemistry applied to the discovery,

evaluation, and exploitation of geothermalenergy resources: United Nations Symposiumon the Development and Utilization of Geo-thermal Resources, Pisa, Sec. V, Rapporteur'srept. [preprint] p. 1-25.

White, D. E., Hem, J. D., and Waring, G. A., 1963,Chemical composition of sub-surface waters:U.S. Geol. Survey Prof. Paper 440F, p. 1-67.

White, D. E., Muffler, L.J.P., and Truesdell, A. H.,1971, Vapor-dominated hydrothermal systemscompared with hot-water systems: Econ. Geol-ogy, v. 66, no. 1, p. 75-97.

Whiting, R. L., and Ramey, H. J., 1969, Applica-tion of material and energy balances to geo-thermal steam production: four. PetroleumTechnology, July, p. 893-900.

MANUSCRIPT RECEIVED BY THE SOCIETY JUNE 10,1971

REVISED MANUSCRIPT RECEIVED SEPTEMBER 16,1971

PRINTED IN U.S.A.