Fluid-inclusion evidence for past temperature fluctuations in the Kilauea East Rift Zone geothermal area, Hawaii

GeothermicsVol. 24, No. 5/6, pp. 639-659, 1995

Pergamon

CNR

Elsevier ScienceLtd Printed in Great Britain

0375-6505(95)00034-8

FLUID-INCLUSION EVIDENCE FOR PAST TEMPERATURE FLUCTUATIONS IN THE KILAUEA EAST RIFT ZONE GEOTHERMAL A R E A , HAWAII KEITH E. BARGAR,* TERRY E. C. KEITHt and FRANK A. TRUSDELL* *U.S. Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025, M/S 910, U.S.A. ; t U.S. Geological Survey, Alaska Volcano Observatory, 4200 University Drive, Anchorage, AK 99508-4667, U.S.A.; and ~U.S. GeologicarSurvey, Hawaii Volcano Observatory, P.O. Box 51, Hawaii National Park, HI 96718, U.S.A. (Received February 1995; acceptedfor publication July 1995) Abstract--Heating and freezing data were obtained for fluid inclusions in hydrothermal quartz, calcite, and anhydrite from several depths in three scientific observation holes drilled along the lower East Rift Zone of Kilauea volcano, Hawaii. Compositions of the inclusion fluids range from dilute meteoric water to highly modified sea water concentrated by boiling. Comparison of measured drill-hole temperatures with fluid-inclusion homogenization-temperature (Th) data indicates that only about 15% of the fluid inclusions could have formed under the present thermal conditions. The majority of fluid inclusions studied must have formed during one or more times in the past when temperatures fluctuated in response to the emplacement of nearby dikes and their subsequent cooling. The fluid-inclusion data indicate that past temperatures in SOH-4 well were as much as 64°C hotter than present temperatures between 1000 and 1500 m depth and they were a maximum of 68°C cooler than present temperatures below 1500 m depth. Similarly, the data show that past temperatures near the bottoms of SOH-1 and SOH-2 wells were up to 45 and 59°C, respectively, cooler than the present thermal conditions; however, the remainder of fluid-inclusion T hvalues for these two drill holes suggests that the temperatures of the trapped waters were nearly the same as the present temperatures at these slightly shallower depths. Several hydrothermal minerals (erionite, mordenite, truscottite, smectite, chlorite-smectite, chalcedony, anhydrite, and hematite), occurring in the drill holes at higher temperatures than they are found in geothermal drill holes of Iceland or other geothermal areas, provide additional evidence for a recent heating trend.

Key words: Kilauea East Rift Zone, geothermal coreholes, fluid inclusions, hydrothermal alteration, past temperatures, Hawaii.

INTRODUCTION Kilauea volcano is one of the most active volcanoes on earth; the age of about 90% of the surface of this young, shield-type volcano is less than 1.1 ka (Holcomb, 1987). During much of Kilauea's recorded history, volcanic activity was confined within the caldera (Holcomb, 1987); however, since 1955, numerous eruptions have occurred along the lower part of Kilauea's East Rift Zone (ERZ, Fig. 1; Holcomb, 1987; Moore and Trusdell, 1991; Moore and Kauahikaua, 1993). Volcanic activity along the E R Z has been nearly continuous since 1983 (Moore and Kauahikaua, 1993; Smithsonian Institution, 1994). The volcanism, coupled with recently discovered gigantic submarine landslide deposits (Hilina Slump off the southeast coast of Hawaii extends over about 5200 km2; Moore etal., 1989, 1994a, b), and more than 1 km of subsidence in the past 0.5 Ma (Moore and Thomas, 1988; Ludwig et al., 1991), make the Island of Hawaii and Kilauea volcano, in particular, one of the most dynamic locations on Earth. Active volcanism along Kilauea's E R Z sparked an interest in geothermal exploration more 639

K. E. Bargar et al.

640

SOH2~...

KAPOHO

so.4 '"

Kilauea 'Summit

I.,\ ~"

,f...---(ALAPANA

156ow

lss°w

20°N

\ 0

5

10

I

I

I

15 K I L O M E T E R S

I

CONTOUR INTERVAL200m I

Fig. 1. Map showingthe locationof three SOH geothermaldrillholesand the earlierHGP-Adrillholeon the East Rift Zone of Kilaueavolcano,Hawaii.

than 30 years ago (Macdonald, 1973). The potential for a significant geothermal resource, estimated to be in the 500-700 MWe range (Olson et al., 1990), resulted in completion of 16 geothermal exploration drill holes (bottom temperatures >50°C) in this region since 1962 (Thomas, 1987; Sorey and Colvard, 1994). However, core samples, which provide much valuable information about the ERZ geothermal system, previously were obtained from only the HGP-A hole (Thomas, 1987; Fig. 1). In order to further understand the geochemistry, temperature and structural conditions of the ERZ geothermal resource and to monitor changes in the system, the Hawaii Natural Energy Institute of the University of Hawaii at Manoa drilled three additional scientific observation holes (SOH) along the ERZ, SOH-1, SOH-2, and SOH4, between 1989 and 1991, at respective surface elevations of 189, 86, and 364 m above mean sea level (Fig. 1; Olson and Deymonaz, 1991; Novak and Evans, 1991 ; Thomas et al., 1991; Evans et al., 1994). These coreholes succeeded in demonstrating that high temperatures (206.1,350.5, and 306. I°C, respectively) are present at depths of 1684, 2073, and 2000 m beneath the ground surface along a substantial portion of the ERZ (Figs 2-4). In addition, hole SOH-1 defined the northern boundary of the geothermal reservoir within which the HGP-A hole was drilled, and SOH-4 located a potential geothermal resource in a previously undrilled area (Olson and Deymonaz, 1992). About 4559 m of drill core were recovered from the 5758 m penetrated by the three SOH holes for an average recovery rate of nearly 80% (Evans et al., 1994). In addition, a few rock chips were obtained from some intervals where core was not recovered because of drilling problems (Olson and Deymonaz, 1992), and some cuttings were collected from the upper part of hole SOH-2 (Evans, 1992). Stratigraphic columns for the SOH holes show an upper subaerial section and a lower submarine section (see Figs 2-4). The subaerial deposits consist mostly of a'a and pahoehoe basalt flows with basalt dikes and/or sills and a few ash beds, while the submarine

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

641

deposits contain pillow basalts and hyaloclastites that are intruded by mafic dikes (Novak and Evans, 1991; Evans, 1992; Moore and Trusdell, 1993). The break between submarine and subaerial volcanic deposits occurs abruptly at 748 m depth in drill hole SOH-1 where it is marked by carbonate strata containing marine fossils (Novak and Evans, 1991). In SOH-2, the submarine/subaerial interface was not recovered, but it must occur within the depth interval from 520 to 580 m according to data in Evans (1992). Between depths of 1623 and 1768 m in SOH-4, a thick zone of limestone (containing marine fossils, sandy limestone, and conglomerate with water-rounded pebbles) is interspersed with subaerial basalt flows, submarine hyaloclastites, and volcaniclastic deposits (Novak and Evans, 1991). More detailed stratigraphic columns have been published for drill holes SOH-2 (Evans, 1992) and SOH-4 (Trusdell et al., 1992). In addition to lithologic descriptions of core from these drill holes, the logs also contain temperature data, and show the location of several hydrothermal alteration minerals within the holes. Recent articles published in Geothermics provide a good overview of the geology and hydrology of Kilauea volcano and its geothermal system utilizing, in part, data from the SOH drill holes (Ingebritsen and Scholi, 1993; Kauahikaua, 1993; Moore and Kauahikaua, 1993; Moore and Trusdell, 1993). Results from the SOH drill holes also were included in a data set compiled to understand possible deleterious effects of the Hawaii Geothermal Project on the ground-water resources of the island of Hawaii (Sorey and Colvard, 1994). This study is part of an investigation of the morphology, paragenesis, and formation temperatures of hydrothermal alteration minerals from the SOH holes for comparison with alteration of other drill holes in the ERZ, as well as other geothermal areas. The temperature data derived from the study of fluid inclusions in the hydrothermal minerals of the SOH drill holes are presented here; additional data on the hydrothermal minerals will be given in subsequent papers.

Depth (m) i

I

i

I

i

I

• • I

I

i I~

(n

•

I-

Smectlte

Calcite

Aragonlte Truscottlte

Gyrollte Phllllpsite

•

Mordenlte Heulandite Chabazlte

0 • o

subaerlal o

o

[

Pyrite Anhydrlte Quartz+Chalcedony

Crlstoballte

• i iii • • i ~ • • E Q

Talc

IIII

Analclme

submarine o o

o o

Fig. 2. Distribution of hydrothermal alteration minerals with depth below the ground surface in the SOH-1 drill hole. Measured temperature curve (OIson and Deymonaz, 1991, 1992; Novak and Evans, 1991) is shown by + symbols.

K. E. Bargar et al.

642

Depth

(m)

t'w}

, J

o°

I O oJ

Garnet Talc Epldote Actlnollte/Tremollte Hematite Sphelerlte Galena Pyrrh. o t i t e Chalcopyrlte Pyrite Anhydrlte Alblta Adularla Quartz+Chalcedony Kaolinite/Serpentine Chlorite Chlorlte/Smectlte Smectlte Calcite Truscottlte Tobermorlte Mordenlte Erlonlte Analclme

I m m m p m m ° mn __ ~.~m.

•

I

•

m

o

m

•

•

m

i

O

n

m

*

subaerial

m I m

m m m m m

m

•

"

o

m

m •

'

•

m • _a_

•

• •

n n

mmu m

•

•

submarine

i

!

!

m

O 0

O 0

O 0

O 0

?-

T--

Fig. 3. Distribution of hydrothermal alteration minerals with depth below the ground surface in the SOH-2 drill hole. Measured temperature curve (Olson and Deymonaz, 1991, 1992; Evans, 1992) is shown by + symbols.

Depth i

I

,

(m)

I

•

,

•

I

m

i

I

m~me •

•

.m

i

E

m

~ m n

"" m i n i mm i n

0

n

/

=...4

•

~ ~

submerlal

Talc Epldote Actlnollte/Tremollte Prehnlte Hematite Pyrrhotlte Chalcopyrlte Pyrite Anhydrlte Alblte Adularla Quartz+Chalcedony OPal+Crlstobante Chlorite Chlorlte/Smectlte $mectlte Calcite Xonotllte Tobermorlte Truscottlte Phllllpslte Mordenite Heulandlte Analclme÷Walraklte

i

o

o o

o o o

o o ¢~

'P-

,o"

o o Q

Fig. 4. Distribution of hydrothermal alteration minerals with depth below the ground surface in the SOH-4 drill hole. Measured temperature curve (Olson and Deymonaz, 1991, 1992; Novak and Evans, 1991; Trusdell et al., 1992) is shown by + symbols.

Temperature Fluctuations in Kitauea East Rift Zone, Hawaii

643

ANALYTICAL METHODS Relying upon the logs of the three drill holes, we collected 356 core specimens (54 from SOH-1; 141 from SOH-2; and 161 from SOH-4) that are believed to be representative of various types of alteration present in each of the holes. However, a great many of the core boxes were not inspected. The sample subset from each drill hole was examined by binocular microscope and appropriate specimens were prepared for analysis by X-ray diffraction (XRD), scanning electron microscope (SEM) equipped with an X-ray energy dispersive spectrometer (EDS), electron microprobe, and petrographic microscope. In addition, 25 doubly-polished thick sections (along with a few unpolished cleavage chips) of hydrothermal quartz, calcite, and anhydrite were prepared from the three drill holes in order to obtain data on past subsurface temperatures and fluid salinities. Measurements of fluid-inclusion homogenization (Th) and ice-melting temperatures (Tin) were made using a Linkam THM 600 heating/freezing stage and TMS 90 temperature control system.* Successive calibration runs, using synthetic fluid inclusions (Bodnar and Sterner, 1984) and chemical compounds with known melting points recommended in Roedder (1984), suggest that the accuracy of the homogenization temperature measurements is within +2.0°C and the ice melting-point temperature values are accurate to at least +0.2°C. Salinities of the inclusion fluids were calculated, in weight % NaCI equivalent, using the equation given in Potter et al. (1978). HYDROTHERMAL ALTERATION MINERALOGY

SOH-1 Nearly all SOH-1 core specimens collected for this study contain early Fe-rich smectite coating fractures, vesicles, and breccia fragments (Fig. 2). Dark green smectite also replaces basaltic glass, as well as phenocryst and groundmass mafic crystals of the pillow basalts. Tiny (<0.1 mm) pyrite crystals or clusters of crystals usually appear to have formed later than smectite, but a few core specimens contain pyrite that may be earlier than smectite. Thin, colorless or gray to white, botryoidal silica (cristobalite and chalcedony in XRD) formed later than smectite in several specimens from the lower part of the drill hole. Occasional fracture and vesicle fillings of some of these specimens contain clusters of small, colorless or frosted, subhedral to euhedral quartz crystals that also are later than dark green smectite. Colorless, bladed crystals or occasional white nodules of anhydrite formed later than smectite and possibly the silica deposits, but it appears to have precipitated before zeolite minerals in a few specimens that contain both anhydrite and mordenite or anaicime. Fractures and cavities in several specimens contain traces of a soft, white mineral identified as calcite by XRD or by reaction with HCI. In the two upper occurrences, where measured temperatures are <50°C, the XRD analyses show that aragonite is the predominant carbonate phase. Calcite and/or aragonite appear to be paragenetically late, possibly forming after analcime in one fracture filling, but earlier than mordenite in other occurrences. In addition to analcime (NaAISi20 6- H20)t and mordenite [(Ca,Na2,K2)A12SiloO24. 7H20)], other zeolite minerals identified as late deposits in fractures and vesicles in SOH-1 include chabazite (CaAlESi4012 • 6H20), heulandite [(Na,Ca)2_3A13(AI,Si)2Si13036 • 121-120], and phillipsite [(K,Na,Ca)I_2(Si,Al)sO16 • 6H20 ]. Two other associated late vug or fracture fillings are gyrolite [NaCa16(Si23AI)O60(OH)5 • 15H20 ] and truscottite [(Ca,Mn+2)14Si24058 • 2H20]. These minerals are hydrated calcium silicates that seldom are reported from geothermal *Any use of trade, product, or firm names in this paper is for descriptive purposes only and does not imply endorsementby the U.S. Government. tMineral formulasare fromFleischerand Mandarino(1991).

644

K . E . Bargar et al.

drill holes, although both minerals usually are associated with zeolites in basic igneous rocks (Juan et al., 1970; Merlino, 1988). A clearly defined paragenetic sequence of smectiteanhydrite-analcime-heulandite-gyroliteplus truscottite was observed in a breccia consisting of pillow basalt fragments from a depth of 1440 m. At 1389 m a near-vertical fracture is coated by green smectite, colorless, massive analcime, colorless calcite, white to colorless, bladed gyrolite and truscottite, and late smectite (listed in order of paragenesis). Talc was identified by XRD, along with truscottite, mordenite, and anhydrite (paragenesis was not determined) in a pillow-basalt breccia containing some unaltered glass from 1583.1 m. In the Philippine Tongonan geothermal field, talc occurs at temperatures near 300°C (Leach et al., 1983). This occurrence of talc is the only indication found for possible past high temperatures (->300°C) in core from the SOH-1 drill hole. SOH-2 Nearly all of the SOH-2 core specimens collected for this study are from the lower half of the submarine section where temperatures recorded during drilling ranged between about 100 and 350°C (Fig. 3). The most abundant hydrothermal alteration minerals identified from this drill hole are clay minerals. These minerals also appear to be the earliest hydrothermal minerals. However, tiny (<0.1 mm) pyrite crystals may have formed previously in some occurrences; rare chaicopyrite, galena, sphalerite, and pyrrhotite also formed early in the sequence of hydrothermal minerals observed in this drill hole. All but pyrrhotite are found in close association with pyrite. Various shades of green clay coat or completely fill fractures, vesicles, and void spaces between breccia fragments and replace glass, mafic phenocrysts, and groundmass minerals. Smectite is the dominant clay mineral in core specimens above a depth of 1760 m (Fig. 3). Semiquantitative chemical analysis by SEM shows that the smectite is an Fe-rich mineral (nontronite) with subordinate Mg and Ca in addition to Si and AI. Most smectite appears to consist of cobweb-like masses or platy crystals having sharp - 1 5 / ~ (001) XRD reflections that expand to --17 A following exposure to ethylene glycol vapors at 60°C for 1 h. These reflections collapse to - 1 0 / ~ after heating in a furnace at 550°C for at least 0.5 h. A single core specimen from 1503.9 m depth contains a white to pale green, waxy, fracture filling that is characterized by a 7.4 ~ major XRD reflection which suggests that the clay is a serpentine-kaolinite group mineral; EDS shows it to consist of Mg, Si, and Fe along with traces of Ca and AI. Green, euhedral, hexagonal, Mg-rich (Mg>Fe in EDS) chlorite is present in the lower part of the drill hole. Below 1760 m depth, chlorite is the predominant clay mineral; however, smectite is present sporadically, and randomly interstratified chlorite-smectite was frequently identified in XRD. The mixed-layer chlorite-smectite has an (001) reflection that varies from - 1 4 to 14.5 A and expands to -15-17 A after glycolation depending on the ratio of chlorite to smectite. Tiny (as large as 0.5 mm), colorless, euhedral adularia crystals (many are twinned) are present as open-space fillings in a few drill core specimens between 1252.4 and 1838.7 m depth. The adularia formed earlier than smectite in some cavities, but it is usually a later mineral. This and other (mostly SEM) observations suggest that there may be more than one generation of smectite. At 1437.1 m depth, an XRD analysis of a thin, colorless to white, fracture coating shows the presence of smectite, quartz, and albite. Although relatively uncommon, three other specimens from this drill hole also have open-spacedeposits that contain euhedral albite crystals (Fig. 3). One fracture filling at 1918 m depth contains small, orange-colored, subhedral, garnet crystals in association with later chlorite, actinolite/tremolite(?) and epidote. Several XRD analyses of clay minerals from near the bottom of the SOH-2 drill hole also show the presence of occasional actinolite/tremolite(?) and minor talc. Talc probably results from alteration of olivine and/or pyroxene phenocrysts and groundmass crystals; short (<0.1 mm) actinolite/tremolite fibers on a

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

645

few fractures also may have formed during alteration of pyroxene. Several core specimens collected from the lower part of the drill hole contain radiating sprays of tiny, green, epidote crystals that formed earlier than chlorite at a depth of 1842.8 m. The association of talc, actinolite, epidote, and chlorite is characteristic of low-grade greenschist facies metamorphism (Phillips and Griffen, 1981). The first three minerals occur at temperatures near 300°C in the Tongonan geothermal field, Philippines (Leach et al., 1983) and also are found at or above this temperature in the SOH-2 drill hole. Chlorite occurs over a very wide temperature range (<100-350°C; Hulen and Nielson, 1986) in numerous geothermal areas. White to bluish, botryoidal or massive chalcedony is present in cavities and fractures of a few of the shallowest SOH-2 core specimens; however, colorless euhedrai quartz crystals (<2 mm long) are the more common silica phase. Quartz frequently formed later than green clay (smectite, chlorite, or chlorite-smectite) and pyrite. However, a second generation of green clay is perched on quartz crystals in a few open-space fillings. Most other associated hydrothermal minerals, including biotite(?) at 1938.8 m depth, appear to have formed later. White to colorless, bladed to tabular, clusters of anhydrite crystals are present sporadically in the suite of collected samples. These crystals formed as late minerals in open spaces of fractures and vesicles and are sometimes associated with even paragenetically later zeolite minerals. Masses of very tiny, white gypsum needles frequently occur in the same specimens and probably formed by hydration of anhydrite. In a few specimens, small, tabular, anhydrite crystals have a colorless core, but crystal edges are altered to an opaque white mineral (calcite) that reacts with dilute HCI. Late calcite, possibly introduced during drilling, also is present as coatings on cylindrical cored surfaces of several scattered core specimens. Open-space fillings of colorless or white, fibrous, clusters of mordenite crystals and subhedral to euhedral crystalline to massive analcime, along with one specimen of white, fibrous to cottony erionite [(K2,Ca,Na2)2AIaSi14036 • 15H20], were the only zeolite minerals identified in core specimens from this drill hole. Vesicles in two specimens are filled by white, densely packed, radiating, fibrous, tobermorite [Ca9Sil2030(OH)6.4H20 ] crystal clusters; the zeolites and tobermorite precipitated later than most other hydrothermal minerals identified during this study of the SOH-2 drill core.

SOH-4 Except for the lower ~100 m of drill core, smectite is present as an open-space filling, replacement of mafic phenocrysts, or groundmass alteration in nearly every core specimen collected from SOH-4 (Fig. 4). The green, well-crystallized, Fe-rich ( F e > C a > M g in EDS) clay mineral is usually an early mineral; however, SEM studies show that some smectite formed later than anaicime, prehnite, quartz, and pyrite, indicating that more than one generation of the mineral was deposited. Randomly interstratified chiorite-smectite is present in a few scattered specimens from the submarine deposits and the lower part of the subaerial section of the SOH-4 drill hole (Fig. 4). The green mixed-layer clay mineral is similar in occurrence and EDS chemistry (Fe, Ca, Mg, AI, and Si) to smectite. In SEM studies, chlorite-smectite was determined to be paragenetically later than quartz, adularia, and chalcopyrite, and it formed earlier than anhydrite. In the submarine segment of this drill core, minor chlorite appears to be present in XRD analyses of both smectite and chlorite-smectite (Fig. 4). Mafic minerals in several core samples from the subaerial part of the drill hole are altered to green and red clay. XRD analyses for two of these clay specimens indicate that the red alteration consists of hematite. A few mafic minerals (olivine?) and one cavity filling in the submarine section consist partly of talc. Tiny (<0.5 mm), cubic, pyrite crystals are present in several specimens from the lower part of the subaerial section of the SOH-4 drill hole (Fig. 4). Most pyrite occurs as open-space fillings,

646

K . E . Bargar et al.

but a few core samples contain disseminated pyrite. One fracture surface at 1281.5 m depth is lined by several bronze, hexagonal, pyrrhotite crystals; the fracture also has paragenetically early smectite and late calcite, as well as quartz and gypsum. Five open-space fillings near the bottom of the drill hole contain tiny ( - 0 . 1 - 0 . 2 mm), subhedral, reddish, tarnished, chalcopyrite (Cu, Fe, S in EDS) crystals. The chalcopyrite appears to be an early hydrothermal mineral; it was observed to be earlier than actinolite/tremolite and chlorite-smectite in SEM. Traces of amphibole (referred to here as actinolite/tremolite) were identified in two core specimens from near the bottom of the drill hole (Fig. 4). EDS semiquantitative analyses of two different specimens show variations in the chemistry of the fibrous mineral (Ca, Mg, Si in one analysis, and Mg, Fe, Ca, AI, and Si in the other) suggesting the presence of more than one phase of this solid-solution series. One brecciated core specimen from 1971.3 m depth contains tiny, green, euhedral, epidote crystals lining open spaces along with later quartz, and chlorite-smectite; adularia and chalcopyrite are also present in this core sample. Four silica minerals (opal, cristobalite, chalcedony, and quartz) were identified in open spaces of many SOH-4 drill hole specimens (Fig. 4). The first three minerals are found mostly above 1250 m depth and are similar in appearance (occurring as white to frosted, botryoidal or horizontal siliceous deposits). However, they are readily distinguished by XRD. Quartz is present primarily as colorless, euhedral or subhedral (<0.5 cm long) crystals that formed earlier (except for some green clay) than the associated hydrothermal minerals. Closely associated with quartz in the lower half of the drill hole are colorless to white, subhedrai to euhedral, adular]a and albite crystals that are present in fracture and vesicle fillings of a few scattered specimens (Fig. 4). Albite crystals at 1734.6 m depth have a light to moderate dusting of debris particles (seen in SEM) consisting mostly of clay or albite fragments, although some particles are similar in size and morphology to bacteria (Fig. 5). EDS analysis indicates that CI is present in the dimpled, bacteria-like particles. This suggests that if the particles are bacteria they may have lived in situ in pore fluids containing sea water, and did not result from contamination by the fresh-water drilling fluids. SEM and fluid-inclusion studies of hydrothermal minerals in drill core from the Medicine Lake volcano geothermal area in northern California have revealed very similar bacteria-like particles (plus rod-shaped and branching filamentous forms) trapped within (Bargar, 1992) or adhering to the crystal faces of hydrothermal minerals. Unfortunately, controls to minimize bacterial contamination were not employed during drilling or subsequent handling of the drill core from either location. While it cannot be determined with absolute certainty at present that the bacteria-like particles did not result from contamination and actually lived on the hydrothermal minerals, it would be very significant if the particles from the SOH-4 drill hole were not contaminants, because the present temperature at the depth (1734.6 m) where these particles occur is about 265°C. At present, the highest temperature at which bacteria are known to survive is about 110°C (Huber et al., 1989), although some studies (Baross and Deming, 1983) have suggested that microorganisms can withstand temperatures above 250°C.* Vesicles, fractures, and open spaces between breccia fragments in the core specimens commonly contain small amounts of calcite (Fig. 4). This colorless, white, or frosted mineral is usually found as blocky (up to 0.7 cm length) or bladed crystals that appear to have formed earlier than some of the zeolites, but it is later than smectite, chlorite-smectite, and quartz. *The implicationthat thermophilicbacteria could have lived at high temperatures in the pore fluidspenetrated by the SOH-4 drill hole is mentioned here in order to alert other researchers to that possibility. Also, we would like to recommend that consideration be given to implementation of controls during future geothermal drilling in order to avoid or at least minimize possible microbialcontamination by drillingfluids and their introductionduring handling of recovered drill core specimens.

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

647

Fig. 5. Scanningelectron micrograph of dimpled, bacteria-like particles found on the surface of plagioclase crystals filling a cavity in core from 1734.6 m depth beneath the ground surface in SOH-4. The nearly transparent connecting "trail" between two of the particles (arrow) probably consists of a viscous slime that permits adsorption and shows movementof the extreme left particle across the plagioclasecrystal surface. Colorless to white, blocky or tabular, anhydrite crystals were identified in several open-space fillings, mostly from the submarine section of SOH-4 (Fig. 4). The anhydrite formed later than most other associated hydrothermal minerals; however, a few specimens contain still later, needle-like or fibrous, gypsum which may originate owing to hydration of anhydrite. The studied samples suggest that there are two zeolite zones (Fig. 4). Colorless to white analcime; white, fibrous tufts of mordenite; colorless to frosted, tabular, heulandite; and radiating, colorless phillipsite crystals are present as open-space deposits in the upper zone between approximate depths of 750 and 1250 m. Phillipsite at 744.5 m depth is associated with a mineral that tentatively was identified as apophyllite in an SEM analysis. Zeolite minerals identified in a lower zone (about 1500-1750 m) are colorless to white, intergrown, subhedral to euhedral, trapezohedral analcime crystals and wairakite, which was identified from a single X R D analysis of a white fracture filling at 1735.5 m depth. Electron microprobe analyses of analcime from 1564 m depth show about 1 wt% CaO. The presence of Ca indicates that the mineral is intermediate in composition between analcime (Na-rich) and wairakite (Ca-rich); a complete isomorphous series exists between the two minerals (Gottardi and Galli, 1985). In SOH-4, three calcium-rich minerals [prehnite--Ca2A12Si3010(OH)2, truscottite-(Ca,Mn+2)14Si24058(OH)8. H 2 0 , and xonotlite--Ca6Si6017(OH)2 ] occur as open-space fillings over much the same depth and temperature range as the anaicime in the lower zone (Fig. 4). Tiny, euhedral, blocky prehnite crystal clusters formed later than honeycomb smectite and analcime and are earlier than a thin, late, smectite covering. Colorless to frosted, radiating, needle-like, xonotlite crystals were observed in vesicle fillings or were identified from X R D analyses of a few specimens from the upper part of the submarine section of the drill hole. Five nearby collected specimens contain fracture and vesicle fillings of white, bladed (sometimes

K. E. Bargar et al.

648

1000

'

'

SO'H-1

150C

0

TEMPERATURE,IN °C

Fig. 6. Plot of depth below ground surface vs fluid-inclusion homogenizationtemperatures (Th) for anhydrite and quartz fluid inclusions in core from SOH-1 (Table 1). Solid curve shows measured downhole temperatures. Th measurements are shown by histograms (N = number of Th measurements) with sample depth beneath the ground surface as the baseline. radiating), truscottite crystal clusters. Xonotlite and truscottite have been reported from only a few studies of hydrothermal minerals in geothermal drill holes (Bargar et al., 1981). FLUID-INCLUSION DATA

SOH-1 One quartz crystal from 1654.9 m depth has very tiny (most are between 5 and 10/~m in size), secondary, fluid inclusions. Th values for nine of these liquid-rich fluid inclusions range from 155 to 171°C (Fig. 6 and Table 1). The fluid inclusions have a eutectic temperature (To) of about -22°C indicating that the fluid is an NaCI solution (Shepherd et al., 1985); freezing studies show that the liquid trapped in the fluid inclusions has a salinity of about 3.5 wt% NaCI equivalent, which is nearly same as the salinity of sea water (Williams, 1962). Fluid inclusions were located only in one anhydrite specimen from 1526.8 m depth. These liquid-rich, secondary inclusions have T h values between 157 and 161°C, which is within the range reported above for one quartz crystal from deeper in this same drill hole (Fig. 6). The only Tm measurement obtained for anhydrite indicates a lower salinity (2.2 wt% NaC1 equivalent) than for quartz (Table 1). Thus, the anhydrite may haveprecipitated from a mixture of sea water and meteoric water that percolated down through the volcano. SOH-2 Fluid-inclusion heating and freezing data were obtained on quartz from five depths in the SOH-2 drill hole (Table 1). Minimum Th values for the upper four quartz specimens plot about 5-14°C below the measured temperatures at the respective sample depths, whereas maximum

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

cqe~l

0

~"q¢'q

0

0

e'qe~l

e~lcqe'qe~4

o~

m

~.~'~

~

0

.=.

E "N

L

o

~ o

II

0

I

I I I I I

I

I

I

I,~

I

I

i

I

.~

_=

"~. "1~ .~. ~ . , ' ~

~.~ e~

~,o_

'T. 0

cq

"T,

0ra~

0c~

"r~ ~.,.--.~ ~

~'~

I

I

649

K. E. Bargar et al.

650

SOH-2

/

1500 e

'

20(}0

____1 100 4-

__1 L__ 20o TEMPERATURE, IN ° C

0 3 0

t

Fig. 7. Plot of depth below ground surface vs fluid-inclusion homogenization temperatures (Th) for quartz fluid inclusions in core from SOH-2 (Table 1). Dashed curve is a theoretical reference boiling-pointcurve for pure water originatingat the ground surface(after Elder, 1981,Table A5). Solidcurve showsdownholemeasuredtemperatures. Th symbols as in Fig. 6.

Th measurements are nearly coincident with the measured temperatures for three of the four sample depths (Fig. 7). These data suggest that the fluids trapped within the fluid inclusions at these four sample depths very likely are representative of the present-day geothermal waters. Assuming this conclusion to be valid would justify application of a pressure correction (Potter, 1977) to the Th data, effectively raising the minimum Th value even closer to the measured temperature curve than is shown in Fig. 7. Salinities, calculated from Tm data for these fluid inclusions (Table 1), indicate that the water within the fluid inclusions probably consists either of sea water or sea water diluted to varying degrees by meteoric water. Fluid-inclusion heating and freezing data for the deepest quartz specimens, where measured temperatures are close to the theoretical reference pure water boiling-point curve (Fig. 7), suggest that the quartz crystals were not precipitated from the present-day geothermal fluids. The Th values range from about 17 to 59°C cooler than the measured temperature at the sample depth. Most fluid inclusions are liquid-rich, secondary or pseudosecondary(?) and are estimated to contain about 10-20 volume % vapor (Fig. 8a); however, rare vapor or vapor-rich (Fig. 8b) fluid inclusions were also observed. The presence of this type of inclusion may indicate that boiling took place at some time during or following crystallization of the quartz, although the vapor-rich inclusions do not appear to be coeval with the liquid-rich inclusions. Th measurements were not obtained for the vapor-rich fluid inclusions because of their optical characteristics, i.e. the temperature at which the vapor phase completely filled the inclusion cavity was difficult to determine. Boiling also is indicated by freezing studies, where some of the liquid-rich fluid inclusions were found to have salinities as high as 13.7 wt% NaCI equivalent (Fig. 9) or nearly four times

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

651

greater than the salinity of sea water. The chemical composition of water from one other deep well in the E R Z (near well HGP-A in Fig. 1) shows that the chloride content is two to three times greater than that of sea water; Janik et al. (1994) indicate that this well probably tapped fluids from a boiling reservoir. Such concentrated salinities could exist in the SOH fluid inclusions if the trapped fluid is composed of sea water (or sea water mixed with meteoric water) that has been heated in a closed system to temperatures above the boiling point, as described by Bischoff and Pitzer (1985) and Fournier (1987). Brine also can form through condensation of highly saline gas at a temperature above the critical temperature (Fournier, 1987). According to Fournier (1987), brine forms more readily by these processes if there are repeated magmatic intrusions or intrusion of magma at shallow depth. A wide range in the salinities of fluid inclusions in the deepest quartz specimens from this drill hole (Table 1 and Fig. 10) indicates that the fluid inclusions trapped geothermal waters consisting of different proportions of meteoric water and sea water. Furthermore, as previously stated, the fluid inclusions could not have been produced from the present geothermal waters. Instead, these inclusions must have formed from water that boiled at a substantially lower temperature than the present measured temperature at some time during the past history of the Kilauea volcano.

Fig. 8. Photomicrographsof (a) liquid-rich,secondary(?)fluidinclusions,and (b) single-phasevapor and vapor-rich fluidinclusionsin hydrothermalquartzcrystalsfrom2019.8 m depth belowthe groundsurfacein SOH-2.

K. E. Bargar et al.

652 oo Z

/

/

l

z

_o

"

i

÷+

•

,

-I.4-

+

•

2oo

-Z

O

O SOH-1 + SOH-2

uJ i

• i

I0°0

i

SOH-4

i

i

3 6 9 12 SALINITY (~n'. % NaCI EQUIVALENT)

Fig. 9. Plot of fluid-inclusion homogenization temperatures vs salinity (calculated from ice-melting temperatures) for fluid inclusions in hydrothermal anhydrite, calcite, and quartz crystals from the S O H drill holes.

SOH-4 Fluid-inclusion Th measurements were made on hydrothermal quartz crystals from six depths in the SOH-4 drill hole (Table 1; Th data for depths of 1959.7, 1969.9, and 1973.9 m were combined in Fig. 11). The quartz T h values range from 7 to 51°C below the measured temperature curve (Fig. 11). The majority of these fluid inclusions are liquid-rich, and appear to be secondary or pseudosecondary in origin. However, at 1959.7 m depth several vapor-rich fluid inclusions are present. In the specimen studied, these inclusions are not coeval with the liquidrich fluid inclusions because the vapor-rich fluid inclusions homogenized at several tens of degrees Celsius higher (the exact Th could not be determined) than the liquid-rich fluid inclusions. These vapor-rich fluid inclusions indicate that previous boiling conditions occurred

500

-

-

,

•

I

O

SOH-1

+

SOH-2

•

SOH-4

O m

1000

z O m

1500

O

ta~

÷ 4.+ 2000 0

-fl~- ~ J 3

SALINITY

' 6 (WT.

'

%

'

' 9

NBCI

4-1-~ . . . . . 12

15

EQUIVALENT)

Fig. 10. Plot of depth below m e a n sea level vs salinity (from ice-melting temperatures) for fluid inclusions from the S O H drill holes.

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

"'1

653

N=ll

SOH-4

/

• Calcite I'1 Quartz [] Anhydrite

Measured Temperature Curve

/ /

"

/ "

/ :

uJ

I~ N = 15

~m=Is

~ 1500

Reference Boiling-Point with Depth Curve

I/

z w a

)i

,, : \ N=56

2ooo 100

/ I

I

200

I

30O

40

I

TEMPERATURE, IN DEGREES CELSIUS

Fig. 11. Plot of depth below ground surface vs fluid-inclusion homogenization temperatures (Th) for anhydrite, calcite, and quartz fluid inclusions from 10 locations in SOH-4 (Table 1). Dashed curve is a theoretical reference boiling-point curve for pure water originating at the ground surface; dotted curve is a theoretical reference boiling-point curve for pure water originating at 1 km beneath the present ground surface (after data in Elder, 1981, Table AS). Solid curve is a profile of the measured temperatures for the SOH-4 drill hole. T h symbols as in Fig. 6.

at this depth. While most of the Tm measurements on the quartz-hosted fluid inclusions from SOH-4 suggest mixing of meteoric water and sea water, the salinities of the liquid trapped within some of the deeper fluid inclusions in this core hole (Table 1; Figs 9 and 10) are substantially greater than that of sea water. This suggests that the fluid inclusions formed as a result of boiling (Shepherd et al., 1985). If so, boiling occurred sometime in the past when the fluid temperature, as indicated by the Th measurements, exceeded a theoretical reference boiling point curve (dotted curve in Fig. 11) originating at a ground surface, or, alternatively, at a water-table as much as 1 km below the present-day ground surface. Fluid-inclusion heating and freezing analyses were obtained for two calcite specimens from SOH-4 (Fig. 11). In the shallower sample, liquid-rich, secondary, fluid inclusions trapped meteoric water (Tin = 0). Th values range between 9 and 22°C higher than the measured

654

K.E. Bargar et al.

temperature (Fig. 11, Table 1), which suggests that at this depth past temperatures have been somewhat higher than the present measured one. The lower temperature may be due to continued influx of cool meteoric water. Conversely, at 1886.4 m depth the liquid-rich, secondary, fluid inclusions in calcite have Th values that are 25-44°C lower than the measured temperature for this depth (Fig. 11). These Th measurements are nearly coincident with fluid-inclusion T h values for quartz from the same sample depth. The colorless, euhedral, quartz crystals were observed to have formed earlier than the colorless, blocky, calcite crystals. The distribution of fluid-inclusion Th values in calcite indicates that the fluid inclusions could have formed either during a cooling trend, or during subsequent heating, which may have led to the current temperature at this depth. T m values for the calcite fluid inclusions (Table 1) suggest that the trapped liquids largely represent sea water, whereas higher T m measurements indicate that fluid inclusions in quartz trapped a mixture of sea and meteoric waters. This suggests that, even though the quartz fluid inclusions and some of the calcite fluid inclusions have the same Th values, fluid inclusions in the two minerals could not have formed at the same time. The above interpretations do not consider possible effects of stretching or leakage within the calcite fluid inclusions. Studies have shown that shape and size of fluid inclusions in soft, easily cleavable minerals such as calcite can be altered by increases in temperature or pressure, and Th values of the stretched fluid inclusions would increase accordingly (Prezbindowski and Larese, 1987). For this study, no attempt was made to determine whether or not the fluid inclusions in calcite had been deformed, although it was noted in several specimens that the smaller inclusions have higher T h values than the larger ones, which suggests that stretching did not occur (Prezbindowski and Larese, 1987). Even if some deformation did occur owing to heating following formation of the fluid inclusions, the interpretations given above, that the Th values of the calcite fluid inclusions reflect past (higher or lower) temperature conditions, would not change. Anhydrite is another soft, cleavable mineral in which fluid inclusions might be affected by leakage or stretching (Moore and Adams, 1988). Again, other than recognition of the potential problem, no attempt was made to determine whether deformation owing to overheating of the fluid inclusions in anhydrite utilized in this study had occurred; however, it was observed that larger-size fluid inclusions did not have higher Th values. As with calcite fluid-inclusion data, interpretation of anhydrite data would not be substantially different even if some inaccuracies in the reported T h values were found to occur because of leakage or stretching. Temperature and salinity data were obtained for liquid-rich, secondary fluid inclusions in anhydrite from depths of 1452.4 and 1472.5 m. The anhydrite fluid inclusion Th measurements range from ll°C below to 64°C above the present-day temperatures (Fig. 11, Table 1). The Th values from both sample depths suggest that past temperatures have been substantially higher than the measured ones. Applying a pressure correction (Potter, 1977) would have the effect of increasing the minimum Th value at 1472.5 m depth to nearly coincide with the measured temperature curve. This indicates that, of all the fluid inclusion Th measurements made for SOH-4 minerals (Fig. 11), only these few values for anhydrite could have occurred during the present thermal conditions. The few Tm values obtained for the fluid inclusions indicate that the trapped liquid consists mostly of meteoric water. (Table 1). Th values for liquid-rich, secondary fluid inclusions in anhydrite from 1771.7 m depth are 58-68°C lower than the measured temperature at this depth (Fig. 11, Table 1). These and data for calcite fluid inclusions from 1886.4 m depth, discussed above, both suggest that temperatures at these depths are substantially hotter than when the fluid inclusions formed. Only one Tm value was obtained from these anhydrite fluid inclusions. The trapped liquid in this inclusion appears to be slightly more saline than sea water, although no evidence for boiling (vapor-rich fluid inclusions) is present.

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

655

Table 2. Approximate measured temperatures (in °C; from Figs 2-4) at which hydrothermal minerals occur in the Kilauea E R Z SOH drill holes compared with drill holes in Iceland and other geothermal areas Mineral Analcime Chabazite Erionite Heulandite Mordenite Phillipsite Wairakite Apophyllite Gyrolite Truscottite Tobermorite Xonotlite Aragonite Calcite Smectite Chlorite-smectite Chlorite Kaolinite-serpentine Talc Opal Cristobalite Chalcedony Quartz Adularia Albite Prehnite Epidote Actinolite-Tremolite Biotite(?) Garnet Anhydrite Gypsum Pyrite Chalcopyrite Pyrrhotite Galena Sphalerite Hematite

SOH-1 <50-205 <50--60

SOH-2 134--300

SOH-4

Iceland*

< 100--265

< 100-300 <75

<100-152 70-152 70 265 70

60-170 80-230 60--85 180-300

138 133-165 120-206 <50--60 <50-133 110-185 <40--42 <40-206 40-206

175 165-200 156-205 156-206

93-185 <40-93 <40-205

160-270

240 138-165 170-347 <60-340 173-349 270-349 214 295-350 167-349 175-349 134-302 176-265 285-338 324-349 326 322 138-348 168-177 <60-349 302-326 334 302 302 290-338

Other areast

<110

226--265 230 230-260 <100-305 <100-306 158-298 260-306 290-302 < 100 126-150 150-290 154-306 148--252 228-304 230-232 304 300-302 222-306 <100-222 100-255 298-306 167 < 100-246

9 <200 <200 24-149 <270 <200 200--240 <100-240+

? ? <100--350 <150 100--200 <100-350 ? 290-320

<90 <100 100-300+ 220-300 210-240 >230-300+ 280+

< 10(O210 <40-240 150-300+ 150-300+ 100-350 220-350 220-350 260--400 270-340+ 250-300+

60-300 <100-350+

<70 100-300+ 250-350 150-350 280-320 < 100-250

*From: T6masson and Kristmannsd6ttir (1972); Kristmannsd6ttir (1975); Kristmannsd6ttir and T6masson (1978); Kristmannsd6ttir (1979); Jakobsson and Moore (1986); Fridleifsson (1991). tFrom: Honda and Muffler (1970); Keith et al. (1978); Holland and Malinin (1979); Elders et al. (1979); Cavarretta et al. (1982); Leach et al. (1983); Aumento and Liguori (1986); Hulen and Nielson (1986); White et al. (1988).

FINAL REMARKS Table 2 lists hydrothermal minerals identified in core specimens from the three Kilauea ERZ SOH drill holes. Many of these minerals previously were reported from earlier drill holes at Kilauea (Stone and Fan, 1978; Waibel, 1983; Thomas, 1987), as well as in prior reports on the SOH drill holes (Novak and Evans, 1991; Evans, 1992; Trusdell et al., 1992; Evans et aI., 1994). However, several of the hydrothermal minerals were identified for the first time during this detailed investigation of selected drill core samples. A considerable volume of data has been compiled on the hydrothermal alteration of drill hole specimens from geothermal areas in Iceland (T6masson and Kristmannsd6ttir, 1972; Krist-

656

K.E. Bargar et al.

mannsd6ttir, 1975, 1979; Kristmannsd6ttir and T6masson, 1978; Fridleifsson, 1991). The Icelandic studies are applicable to studies of the Hawaiian geothermal systems because: (1) both oceanic islands include basaltic lavas with similar permeabilities, (2) both systems encountered the same degree of heating, (3) both are intruded by sea water mixed with varying amounts of meteoric water, (4) both systems produced similar suites of hydrothermai alteration minerals, and (5) both have experienced retrograde and prograde metamorphism owing to periods of volcanic activity with intervening periods of cooler temperatures (T6masson and Kristmannsd6ttir, 1972). During studies of Icelandic geothermal systems, apparent formation temperatures for many hydrothermal minerals were recorded (Table 2). With a few notable exceptions, most hydrothermal alteration minerals identified in cores from the three SOH drill holes are present at temperatures similar to those at which the minerals precipitated in Icelandic and other geothermal areas (Table 2). Gypsum occurs in at least two of the SOH holes at depths where the rock temperatures are far hotter than the temperature range (<70°C) at which gypsum can form (Holland and Malinin, 1979). Most likely, the gypsum precipitated during or following retrieval of the drill core at the surface due to hydration of anhydrite, with which it commonly is found. Erionite, tobermorite, and truscottite are present at significantly higher temperatures than in drill-hole specimens from Iceland or other areas; this discrepancy may be only apparent because these minerals are rarely reported. Chalcedony, smectite, and mixed-layer chlorite-smectite all occur in the SOH drill holes at temperatures as high as 350°C (Table 2). In the drill holes from Iceland and other area~, these minerals are believed to have precipitated at temperatures below 240°C. Mordenite, talc, anhydrite, and hematite also are present at somewhat higher temperatures in SOH-2 than elsewhere. It seems likely, therefore, that temperatures of the rocks penetrated by the SOH drill holes have increased substantially subsequent to the formation of several hydrothermal minerals. Waibel (1983) observed similar discrepancies between the temperatures at which a few hydrothermal minerals from the HGP-A drill hole occur and the temperatures at which the same minerals formed in Icelandic and other geothermal areas. Similarly, a comparison of temperatures at which hydrothermal minerals (particularly anhydrite) were found in HGP-A and several other nearby drill holes in the ERZ suggested to Thomas (1987) that one possible explanation might be that "the part of the rift penetrated by these wells has undergone a complex history of heating and cooling associated with successive episodes of magmatic intrusions and subsequent cooling by hydrothermal circulation". The fluid-inclusion studies conducted for this investigation confirm that there have been past fluctuations in temperatures of fluids contained in the rocks penetrated by the SOH drill holes. Homogenization temperature measurements for quartz crystals from four depths in SOH-2 straddle the measured temperature curve for this drill hole (Fig. 5). It seems probable that quartz formed under the present temperature conditions within this drill hole. A few anhydrite and calcite specimens from the SOH-1 and SOH-4 drill holes have Th values that are substantially higher than the measured temperatures. The data suggest that past temperatures at these depths have been a few degrees to several tens of degrees hotter and subsequently cooled to the present temperatures. Additional anhydrite and calcite samples from SOH-4, along with the remaining quartz from all three drill holes, have fluid-inclusion Th measurements that plot below the present measured temperature curve indicating that past temperatures also have been cooler. Tm measurements show that the waters trapped within fluid inclusions from the three SOH drill holes range from meteoric water at shallow depths to modified sea water in the deep parts of the drill holes (Fig. 10). Sea water trapped within fluid inclusions from near the bottom of SOH-2 and SOH-4 (Fig. 10) must have boiled. Tm values for fluid inclusions in quartz crystals

Temperature Fluctuations in Kilauea East Rift Zone, Hawaii

657

from these depths (Table 1) indicate that the amount of dissolved salts within some fluid inclusions exceeds that of sea water; such high salinities undoubtedly were produced by boiling. In addition, a few fluid inclusions are single-phase vapor or vapor plus small amounts of liquid, which also characterizes boiling conditions. If boiling did occur at these depths, conditions that would be used to construct a reference boiling-point with depth curve must have been very different from the present-day thermal regime. The simplest explanation would be for these fluid inclusions to have formed at an earlier time when the elevation of the ground surface in the ERZ was as much as one kilometer lower than at present. However, Kilauea is a very dynamic volcano, and the precise explanation probably is much more complex. In addition to extensive volcanism, recent studies have shown that subsidence and giant submarine landslides have substantially influenced the size and shape of the Kilauea volcano. Acknowledgments--The authors thank their USGS colleagues R. O. Oscarson, for his assistance with the SEM work, and R. O. Fournier and T. G. Theodore for their helpful reviews of the manuscript. They are also grateful for comments by J. N. Moore and G. O. Fridleifsson that improved the manuscript. Conversations with S. R. Evans, M. L. Sykes, and D. M. Thomas were of considerable value in this investigation.

REFERENCES Aumento, F. and Liguori, P. E. (1986) Conceptual reservoir models through geoscientific investigations. Geothermics 15,799-806. Bargar, K. E. (1992) Video-tape of bacteria-like moving particles in fluid inclusions from Medicine Lake volcano, northern California. EOS, Trans. Am. geophys. Union 73, no. 43,640. Bargar, K. E., Beeson, M. H. and Keith, T. E. C. (1981) Zeolites in Yellowstone National Park. Mineral. Record 12, 29-38. Baross, J. A. and Deming, J. W. (1983) Growth of 'black smoker' bacteria at temperatures of at least 250°C. Nature 303,. 423--426. Bischoff, J. L. and Pitzer, K. S. (1985) Phase relations and adiabats in boiling seafloor geothermal systems. Earth Planet. Sci. Lett. 75,327-338. Bodnar, R. J. and Sterner, S. M. (1984) Synthetic fluid inclusions in natural quartz I. Compositional types synthesized and applications to experimental geochemistry. Geochim. cosmochim. Acta 48, 2659-2668. Cavarretta, G., Gianelli, G. and Puxeddu, M. (1982) Formation of authigenic minerals and their use as indicators of the physicochemical parameters of the fluid in the Larderello--Travale geothermal field. Econ. Geol. 77, 1071-1084. Elder, J. (1981) Geothermal Systems, 508 pp. Academic Press, London. Elders, W. A., Hoagland, J. R., McDowell, S. D. and Cobo, J. M. (1979) Hydrothermal mineral zones in the geothermal reservoir of Cerro Prieto. Geothermics 8,201-209. Evans, S.R. (1992) Summary geological report and index log of the scientific observation hole #2 on the Kilauea East Rift Zone, Hawaii. Trans. Geother. Resour. Counc. 16, 157-166. Evans, S. R., Keith, T. E. C., Bargar, K. E., Trusdell, F. A., Sykes, M. L., Olson, H. J., Thomas, D. M. and Novak, E. (1994) Preliminary geologic results of three scientific observation holes (SOH), Kilauea East Rift Zone, Hawaii, USA. Proc. VII Int. Syrup. on the Observation of the Continental Crust Through Drilling, pp. 138--141, Santa Fe, NM, 25-30 April 1994. Fleischer, M. and Mandarino, J. A. (1991) Glossary of Mineral Species, 256 pp. The Mineralogical Record, Inc., Tucson, AZ. Fournier, R. O.(1987) Conceptual models of brine evolution in magmatic-hydrothermal systems. In Volcanism in Hawaii (Edited by Decker, R. W., Wright, T. L. and Staufer, P. H.), U.S. Geol. Surv. Prof. Pap. 1350, pp. 14871506. Fridleifsson, G. O. (1991) Hydrothermal systems and associated alteration in Iceland. Geol. Surv. Japan Rept 277, 8390. Gottardi, G. and Galli, E. (1985) Natural Zeolites, 409 pp. Springer-Verlag, Berlin. Holcomb, R. T. (1987) Eruptive history and long-term behavior of Kilauea volcano. In Volcanism in Hawaii (Edited by Decker, R. W., Wright, T. L. and Staufer, P. H.), U.S. Geol. Surv. Prof. Pap. 1350, pp. 261-350. Holland, H. D. and Malinin, S. D. (1979) The solubility and occurrence of non-ore minerals. In Geochemistry of Hydrothermal Ore Deposits (Edited by Barnes, H. L.), pp. 461-508. John Wiley, New York. Honda, S. and Muffler, L. J. P. (1970) Hydrothermal alteration in core from research drill hole Y-l, Upper Geyser Basin, Yellowstone National Park, Wyoming. Am. Mineral. 5S, 1714--1737. Huber, R., Kurr, M., Jannasch, H. W. and Stetter, K. O. (1989) A novel group of abyssal methanogenic archaebaeteria (Methanopyrus) growing at ll0°C. Nature 342,833-834. Hulen, J. B. and Nielson, D. L. (1986) Hydrothermai alteration in the Baca geothermal system, Redondo Dome, Valles Caldera, New Mexico. J. geophys. Res. 91, B2, 1867-1886.

658

K.E.

B a r g a r e t al.

lngebritsen, S. E. and Scholl, M. A. (1993) The hydrogeology of Kilauea volcano. Geothermics 22,255-270. Jakobsson, S. P. and Moore, J. G. (1986) Hydrothermal minerals and alteration rates at Surtsey Volcano, Iceland. Geol. Soc. Am. Bull. 9'/, 648--659. Janik, C. J., Nathenson, M. and Scholl, M. A. (1994) Chemistry of spring and well waters on Kilauea volcano, Hawaii, and vicinity. U.S. Geol. Sur. Open-File Rep. 94-586, 166 pp. Juan, V. C., Youh, C.-C. and Lo, H.-J. (1970) The dehydration reaction of natural truscottite. Proc. Geol. Soc. China 13, 34-40. Kauahikaua. J. P. (1993) Geophysical characteristics of the hydrothermal system of Kilauea volcano, Hawai'i. Geothermics 22,271-299. Keith, T. E. C., White, D. E. and Beeson, M. H. (1978) Hydrothermal alteration and self-sealing in Y-7 and Y-8 drill holes in northern part of Upper Geyser Basin. Yellowstone National Park, Wyoming. U.S. Geol. Surv. Prof. Pap. lf154-A, 26 pp. Kristmannsd6ttir, H. (1975) Clay minerals formed by hydrothermal alteration of basaltic rocks in Icelandic geothermal fields. Geol. Fdreningens Stockholm Forhandl. 9?, 289-292. Kristmannsd6ttir, H. (1979) Alteration of basaltic rocks by hydrothermal activity at 100-300°C. In International Clay Conference 1978 (Edited by Mortland, M. M. and Farmer, V. C.), pp. 359-367. Elsevier, Amsterdam. Kristmannsd6ttir, H. and T6masson, J. (1978) Zeolite zones in geothermal areas of Iceland. In Natural Zeolites: Occurrence, Properties, Use (Edited by Sand, L. B. and Mumpton, F. A.), pp. 277-284. Pergamon Press, New York. Leach, T. M., Wood, C. P. and Reyes, A. G. (1983) Geology and hydrothermal alteration of the Tongonan geothermal field, Leyte, Republic of the Philippines. Fourth International Symposium on Water-Rock Interaction (Abs. Vol.)~ pp. 275-278, Misasa, Japan, 29 August-3 September 1983. Ludwig, K. R., Szabo, B. J., Moore, J. G. and Simmons, K. R. (1991) Crustal subsidence rate off Hawaii determined from 234U/238Uages of drowned coral reefs. Geology 19, 171-174. Macdonald, G. A. (1973) Geological prospects for development of geothermal energy in Hawaii. Pac. Sci. 27,209-219. Merlino, S. (1988) Gyrolite: its crystal structure and crystal chemistry. Min. Mag. 52, 377-387. Moore, J. G., Clague, D. A., Holcomb, R. T., Lipman, P. W., Normark, W. R. and Torresan, M. E. (1989) Prodigious submarine landslides on the Hawaiian Ridge. J. geophys. Res. 94, B12, 17,465-17,484. Moore, J. G., Normark, W. R. and Holcomb, R. T. (1994a) Giant Hawaiian underwater landslides. Science 2,64, 46-47. Moore, J. G., Normark, W. R. and Holcomb, R. T. (1994b) Giant Hawaiian landslides. Ann. Rev. Earth Planet. Sci. 22, 119-144. Moore, J. G. and Thomas, D. M. (1988) Subsidence of Puna, Hawaii inferred from sulfur content of drilled lava flows. J. Volc. geotherm. Res. 35, 165-171. Moore, J. N. and Adams, M. C. (1988) Evolution of the thermal cap in two wells from the Salton Sea geothermal system, California. Geothermics 17,695-710. Moore, R. B. and Kauahikaua, J. P. (1993) The hydrothermal-convection systems of Kilauea: an historical perspective. Geothermics 22,233-241. Moore, R. B. and Trusdell, F. A. (1991) Geologic map of the lower East Rift Zone of Kilauea volcano, Hawaii. U.S. Geol. Survey Misc. Inv. Series Map 1-2225, scale t :24,000. Moore, R. B. and Trusdell, F. A. (1993) Geology of Kilauea volcano. Geothermics 22,243-254. Novak, E. A. and Evans, S. R. (1991) Preliminary results from two scientific observation holes on the Kilauea East Rift Zone. Trans. Geother. Resour. Counc. 15,187-192. Olson, H. J. and Deymonaz, J. E. (1991) The Hawaiian scientific observation hole program--preliminary results and status report. Proc. 13th New Zealand Geothermal Workshop 1991, pp. 115-120. Olson, H. J. and Deymonaz, J. E. (1992) The Hawaii scientific observation hole (SOH) program summary of activities. Trans. Geother. Resour. Counc. 16, 47-53. Olson, H. J., Seki, A., Deymonaz, J. E. and Thomas, D. M. (1990) The Hawaii scientific observation hole (SOH) program. Trans. Geother. Resour. Counc. 14, pt. 1,791-798. Phillips, W. R, and Griffen, D. T. (1981) Optical Mineralogy: The Nonopaque Minerals, 677 pp. W. H. Freeman, San Francisco, CA. Potter, R. W. II (1977) Pressure corrections for fluid-inclusion homogenization temperatures based on the volumetric properties of the system NaCI--H20. J. Res. U.S. Geol. Survey 5,603-607. Potter, R. W. II, Clynne, M. A. and Brown, D. L. (1978) Freezing point depression of aqueous sodium chloridc solutions. Econ. Geol. 73, 284-285. Prezbindowski, D. R. and Laresc, R. E. (1987) Experimental stretching of fluid inclusions in calcite--implications for diagenetic studies. Geology 15,333-336. Roedder, E. (1962) Studies of fluid inclusions I: Low temperature applications of a dual-purpose freezing and heating stage. Econ. Geol. 57, 1045-1061. Roedder, E. (1984) Fluid inclusions. In Reviews in Mineralogy (Edited by Ribbe, P. H.), Vol. 12,644 pp. Mineralogical Society of America, Washington, D.C. Shepherd, T. J., Rankin, A. H. and Alderton, D. H. M. (1985) A Practical Guide to Fluid Inclusion Studies, 239 pp. Blackie, London. Smithsonian Institution (1994) Kilauea. Bull. Global Vole. Network 19(6), 7. Sorey, M. L. and Colvard, E. M. (1994) Potential effects of the Hawaii Geothermal Project on ground-water resources on the island of Hawaii. U.S. Geol. Sur. Water Res. Inv. Rep. 94-4028, 35.

T e m p e r a t u r e Fluctuations in Kilauea East R i f t Z o n e , H a w a i i

659

Stone, C. and Fan, P. F. (1978) Hydrothermal alteration of basalts from Hawaii geothermal project well-A, Kilauea, Hawaii. Geology 6, 401-404. Thomas, D. M. (1987) A geochemical model of the Kilauea East Rift Zone. In Volcanism in Hawaii (Edited by Decker, R. W., Wright, T. L. and Staufer, P. H.), U.S. Geol. Survey Prof. Pap. 1350, 1507-1525. Thomas, D. M., Novak, E. A., Evans, S. R. and Olson, H. J. (1991) Scientific observation holes on Kilauea volcano. EOS, Trans. Am. geophys. Union 72,562. T6masson, J. and Kristmannsd6ttir, H. (1972) High-temperature alteration minerals and thermal brines, Reykjanes, Iceland. Contr. Min. Petrol. 36, 123-137. Trusdell, F. A., Novak, E. and Evans, S. R. (1992) Core lithology state of Hawaii scientific observation hole 4 Kilauea volcano, Hawaii, U.S. Geol. Survey Open-File Rept. 92-0586, 72 pp. Waibel, A. (1983) A review of the hydrothermal mineralogy of Hawaii geothermal project well-A, Kilauea, Hawaii. Trans, Geother. Resour. Counc. 7,205-209. White, D. E., Hutchinson, R. A. and Keith, T. E. C. (1988) The geology and remarkable thermal activity of Norris Geyser Basin, Yellowstone National Park, Wyoming. U.S. Geol. Survey Prof. Pap. 1456, 84 pp. Williams, J. (1962) Oceanography--An Introduction to the Marine Sciences, 242 pp. Little, Brown, Boston, MA.