Journal o f Volcanology and Geothermal Research, 16 (1983) 99--129 99

Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands

TRACE-ELEMENT DISTRIBUTION IN AN ACTIVE HYDROTHERMAL

SYSTEM, ROOSEVELT HOT SPRINGS THERMAL AREA, UTAH

ODIN D. CHRISTENSEN', REGINA A. CAPUANO 2 and JOSEPH N. MOORE 2

' N e w m o n t Exploration Ltd., 327 Freeport Blvd., Sparks, N V 89431 (U.S.A.)
2Earth Science Laboratory, University o f Utah Research Institute, 420 Chipeta Way,
Salt Lake City, UT 84108 (U.S.A.)
(Received June 5, 1981; revised and accepted May 27, 1982)

ABSTRACT

Christensen, O.D., Capuano, R.A. and Moore, J.N., 1983. Trace-element distribution in
an active hydrothermal system, Roosevelt Hot Springs thermal area, Utah. J. Volcanol.
Geotherm. Res., 16: 99--129.

Chemical interaction of thermal fluids with reservoir rock in the Roosevelt Hot Springs
thermal area, Utah, has resulted in the development of characteristic trace-element disper-
sion patterns. Multielement analyses of surface rock samples, soil samples and drill cuttings
from deep exploration wells provide a three-dimensional perspective of chemical redistribu-
tion within this structurally-controlled hot-water geothermal system.
Five distinctive elemental suites of chemical enrichment are recognized, each character-
istic of a particular combination of physical and chemical conditions within the geothermal
system. These are: (1) concentrations of As, Sb, Be, and Hg associated with siliceous
material at locations of liquid discharge, fluid mixing or boiling; (2) concentrations of Mn,
Ba, W, Be, Cu, Co, As, Sb and Hg in manganese and iron oxide deposits; (3) high concen-
trations of Hg in argillized rock near fumaroles and lower concentrations in a broad diffuse
halo surrounding the thermal center; (4) concentrations of As in sulfides and Li in silicate
alteration minerals immediately surrounding high-temperature fluid flow-controlling frac-
tures; (5) deposits of CaCO3 at depth where flashing of brine to steam has occurred due to
pressure release. The geochemical enrichments are not, in general, widespread, pervasive-
ly developed zones of regular form and dimension as are typical in many ore-forming
hydrothermal systems.
As the geothermal system develops, changes and eventually declines through time, the
chemical deposits are developed, remobilized or superimposed upon each other, thus
preserving within the rocks a record of the history of the geothermal system. Recognition
of trace-element distribution patterns during the exploration of a geothermal system may
aid definition of the present geometry and interpretation of the history of the system.

INTRODUCTION

T r a c e a m o u n t s o f b a s e a n d p r e c i o u s m e t a l s are c h a r a c t e r i s t i c o f h o t s p r i n g
deposits and hydrothermally altered rocks in many of the world's high-tem-
p e r a t u r e g e o t h e r m a l s y s t e m s ( W e i s s b e r g e t al., 1 9 7 9 ) . F o r e x a m p l e , E w e r s
and Keays (1977) have shown that geothermal fluids at Broadlands, New

0377-0273/82/$03.00 © 1983 Elsevier Science Publishers B.V.

i00

Zealand, are depositing volatile and precious metals in ore-grade amounts.

Their studies confirm the existence of a crude metalliferous zoning developed
in response to decreasing temperatures and to boiling within permeable
zones. White (1955, 1967) has d o c u m e n t e d a similar geochemical zoning at
Steamboat Springs, Nevada. There, Sb and Hg are being deposited to depths
of 50 m below the surface. Base metals and silver are present in greatest con-
centrations below this depth.
The Roosevelt H o t Springs geothermal system is perhaps the most im-
portant and best documented of the recently discovered geothermal resources
in the Basin and Range Province. Since intense exploration efforts began in
1972, six wells have tapped commercial quantities of thermal fluids (Forrest,
1980). An initial 20-MW power plant is expected to be in service in 1984 and
two 50-MW plants are also planned once the reservoir capacity has been as-
sessed and technical difficulties resolved (National Geothermal Service, 1980).
This thermal area provides an unusual opportunity to examine in three di-
mensions the geological and geochemical characteristics of a high-tempera-
ture, hot-water geothermal system.
Previous geochemical work at the Roosevelt Hot Springs thermal area has
confirmed the existence of multielement chemical zoning developed within
solid materials associated with the present geothermal system. Various aspects
of the trace-element geochemistry have been described by Bamford (1978);
Bamford et al. {1980); Capuano and Bamford (1978); Capuano and Moore,
(1980); Christensen (1980) and Christensen et al. (1980b). This paper de-
scribes the major geochemical features of the Roosevelt geothermal system
based upon multielement analysis of surface rock samples, soil samples and
drill cuttings from 32 shallow temperature-gradient holes and six deep ex-
ploration and production holes. These data are used to develop a three-dimen-
sional geochemical model of the geothermal system and to assess the potential
of solids geochemistry for geothermal exploration applications.

Geologic setting

The Roosevelt H o t Springs thermal area is located in southwestern Utah

on the western flank of the Mineral Mountains near the eastern margin of the
Basin and Range (Fig. 1). The thermal area, as defined by heat flow and deep
drilling, covers about 32 km 2. The rocks of this area consist largely of Ter-
tiary intrusives and highly metamorphosed gneiss and schist of Precambrian
age (Nielson et al., 1978; Sibbett and Nielson, 1980).
During the last 10 m.y., igneous activity produced several small, primarily
basaltic, volcanic centers along the margin of the Basin and Range between
the Roosevelt Hot Springs and Crater H o t Springs thermal areas (Ward et al.,
1978). In the Mineral Mountains, this activity began with the eruption of
rhyolite and basalt around 8 m.y. ago. Renewed activity spanned the interval
between 0.8 and 0.5 m.y. ago, and produced at least ten rhyolite domes
along the crest of the range (Lipman et al., 1978; Ward et al., 1978). Although
 101

EXPL AN.~ T / O N I
P/elstocene rhyolite
flows and domes

Tertlory intrussves ~')'i

rocks undivided ~ ( J ~/ / I I II \

o82-33 e12-35 \i/

ROO$EVELT HOT ,~q~:~..Z 4 - ~

5 4 - ~ ~--~2~/# ~ /
OPAL MOUND-13--~ ¢ e m 2 ~ ~ "

l/
..4:,~'~ ~ .,~. , ~ . ~.

I o 1 2km.
1't2°50 ' L--_ 1 I

Fig. 1. Generalized geologic map o f the Roosevelt Hot Springs geothermal area, from
Nielson et al. (1978). Open circle = nonproducing wells, closed circles = geothermal produc.
tion wells.

the evolution and deep-seated configuration of the igneous complex is still

poorly understood, a deep magma b o d y related to rhyolitic volcanism is a
possible heat source for the geothermal system.
Basin and Range faulting, which produced the Mineral Mountains horst
block, became active in southwestern Utah during the middle Miocene and
has continued throughout Cenozoic time (Steven et al., 1977). The geo-
thermal reservoir lies immediately west of the Mineral Mountains range front
in an area of narrow horsts and grabens. Nielson et al. (1978) have mapped
four sets of faults that they believe control fluid movement in the crystalline
rocks of the reservoir and in the overlying alluvium. These are shown in
Fig. 1 and include westward-dipping low-angle denudation faults, northwest-
trending pull-apart structures developed in the hanging wall of the denudation
faults, east-trending high-angle faults, and north to northeast-trending high-
angle faults. It has been suggested that zones of high permeability in the up-
102

per portions of the reservoir are located at the intersections of the high-
angle faults (Nielson et al., 1978; Capuano and Moore, 1980). The northeast-
trending Opal Mound Fault appears to form the western boundary of the
reservoir.

Fluid chemistry

Capuano and Cole (1982) have provided the most recent and comprehensive
review of the chemistry of Roosevelt Hot Springs fluids. They distinguished
four types of water, including: (1) deep circulating thermal fluids from
producing wells; (2) hot mixed waters from nonproducing wells; (3) present

TABLE I

Chemistry of thermal waters

Well 14-2 a 72-16 b 54-3 c 52-21 d Surface

Constituent, ppm seep e

Na 2150 2000 2320 1900 1840

K 390 400 461 216 274
Ca 9.2 12.2 8 107 122
Mg 0.6 0.29 <2 4.0 25
Fe -- -- 0.03 6.3 --
A1 -- -- <0.5 <0.1 --
Si 299 244 263 65 81
Sr -- 1.2 1.2 -- --

Ba -- -- < 0.5 -- --

M n -- -- <0.2 -- --
Cu -- -- <0.1 -- --

Pb -- -- <0.2 -- --

Zn -- -- <0.I -- --

As 3.0 -- 4.3 -- --

Li -- 16.0 25.3 -- --

Be -- -- 0.005 -- --

B 29 27.2 29.9 27.0 --

Ce -- -- 0.27 -- --

F 5.2 5.3 6.8 3.6 --

C1 3650 3260 3860 2882 3210
HCO 3 -- 181 232 615 298
SO4 78 32 72 85 120
TDS > 6614 6444 7504 5677 6063
pH, collection T 5.9 7.53 -- 6.8 6.5
T, °C b o t t o m h o l e 268 243 >260 204 --
tsO, °/oo --13.46 . . . .

a14-2 analysis by USGS; sample collected 10/1977;Thermal Power Co., 1978.

b72-16 analysis of USGS; sample collected 10/1977; Thermal Power Co., 1978.
c 5 4 - 3 a n a l y s i s b y E S L ; s a m p l e c o l l e c t e d 1 1 ] 1 9 7 9 ; B a m f o r d e t al., 1 9 8 0 .
d 5 2 - 2 1 a n a l y s i s f r o m G e t t y Oil C o m p a n y ; c o l l e c t e d 1 1 1 1 9 7 8 ; G e t t y Oil C o . , 1 9 7 8 .
e R o o s e v e l t s e e p ; s a m p l e c o l l e c t e d 6 / 1 9 7 5 ; W a r d e t ai., 1 9 7 8 .
-- not determined.
 103

natural surface discharges; and (4) past surface discharges. Selected water
chemical data are presented in Table I. Brines discharged from producing
wells 72-16, 14-2 and 54-3 are all similar and most closely represent the deep-
reservoir fluid. These fluids are near-neutral, dilute NaC1 brines with approxi-
mately 6000--7500 ppm total dissolved solids. Calculated cation geothermo-
meter temperatures agree quite well with measured maximum bottom-hole
temperatures of 269°C. Unusually large concentrations of F, As, Li and B
are typical of fluids both from surface discharges and from the deep wells.
Fluids from nonproducing well 52-21 and from the Roosevelt seep have
chemistries clearly different from the production wells. Greater concentra-
tions of Ca, Mg, Fe, SO~ and HCO~ and lower average Na-K-Ca geothermo-
meter temperatures are interpreted as resulting in part from mixing with
nonthermal ground water. Mixing calculations demonstrate an increasing
ground-water component with distance from well 54-3. Comparison of
recent analyses with older spring analyses indicated that the reservoir could
have undergone a moderate decrease in total salinity in the past 20--30
years although relative ionic concentrations have remained similar.

Hydrothermal alteration and deposits

Hot springs in the thermal area consist of opaline and chalcedonic sinter
and hematite, silica, calcite, or manganese oxide-cemented alluvium. Most
of the sinter and calcite-cemented alluvium deposits are located along the
trace of the Opal Mound Fault, whereas hematite and manganese oxide-
cemented alluvium occur at distances up to 0.5 km away from the main
sinter mounds. Present surface activity consists of one saline seep and several
active fumaroles which evolve hydrogen sulfide and deposit sulfur.
Two distinct types of hydrothermal alteration have been recognized within
the thermal area. A near-surface assemblage, which formed above the water
table by downward-percolating waters, is represented by acid-leached allu-
vium and crystalline rock composed of variable amounts of quartz, kaolinite,
alunite, montmorillonite and muscovite (Parry et al., 1978). Deeper altera-
tion assemblages produced by upward-convecting hydrothermal fluids are
concentrated mainly along faults and fractures that mark past and present
fluid channels.
Two assemblages of hydrothermal minerals have been recognized at depth
in the production wells: a lower-temperature mineral assemblage consisting
of montmorillonite, mixed-layer clays, abundant hematite and sparse epidote
and pyrite, and a deeper higher-temperature assemblage of chlorite after
plagioclase, abundant calcite, sparse anhydrite and increased abundance of
pyrite. Potassium-feldspar is common as a primary constituent in rocks and
appears petrographically to be a stable phase throughout the deep parts of
the system.
Capuano and Cole (1982) have modeled the chemistry of the present
reservoir fluid and compared predicted alteration products with observed
104

assemblages. In general, they concluded that the observed sequence of mineral

assemblages could be produced by the present reservoir fluid cooling from
approximately 300°C to 150°C during its migration to the surface. Although
the predicted temperatures are slightly higher than measured temperatures,
the good agreement between observed and predicted alteration mineralogy
indicates that the geothermal reservoir fluid at Roosevelt Hot Springs has not
changed character significantly in the recent past.

A N A L Y T I C A L PROCEDURES

Sampling

A variety of materials were sampled during this investigation including sur-

face rock samples, soil samples and drill-cutting samples. Rock samples,
composite samples of 20--30 chips collected from an outcrop, document the
geochemical signature of recognizable geothermal surface deposits.
Drill cuttings from six deep geothermal exploration wells and 32 shallow
temperature-gradient drill holes provide an array of samples from within and
around the geothermal system. Individual 1.5-m (5-ft) to 15-m (50-ft) washed
samples were crushed to --4 mesh, cleaned of magnetic contaminant material
derived from drilling hardware with a hand magnet, and pulverized to --80
mesh. In most instances, composite samples representing 30-m (100-ft)
drilling intervals were prepared. Pulverized samples were stored in glass vials
to minimize Hg loss and sample cross-contamination.
Soil samples were collected at approximately 152-m (500-ft) grid intervals
from a depth of 15--18 cm (Capuano and Bamford, 1978). Samples were col-
lected in polyethylene bags, dried overnight at room temperature, sieved to
--80 mesh and sealed in glass vials.
Early geochemical investigations on drill cuttings included the separation
and analysis of a greater than 3.3 specific gravity fraction (Bamford, 1978).
The intent was to concentrate hydrothermal sulfides and oxides which could
be enriched in some trace elements. This practice was discontinued when
petrographic and microprobe examination revealed that a significant amount
of exotic material from drilling hardware remained despite careful sample
cleaning. Further, the amount of sulfide present in this geothermal system is
generally minor compared to the content of other nonmagnetic heavy
minerals including zircon, sphene, ilmenite, epidote, allanite and hematite.

Analysis and data treatment

Samples were analyzed for 37 elements by inductively coupled plasma

(ICP) emission spectroscopy. Specifics of the analytical procedure and in-
strumentation as well as an evaluation of the quality of the analyses are sum-
marized in Christensen et al. (1980a). In addition, As was determined by a
colorimetric procedure (Fisher Scientific Co., 1960), and Hg was determined
 105

by gold-film mercury detector (Capuano and Bamford, 1978).

Selected mineral separates were analyzed by electron microprobe and
whole-rock samples separated into components by selective dissolution in
order to ascertain the mineralogic sites of trace elements. In the selective
dissolution experiments, six fractions were sequentially isolated and analyzed:
exchangeable cations, manganese oxides, "amorphous iron oxides", "crystal-
line iron oxides", sulfides, and remaining silicate material, following the
methods of Fuchs and Rose (1974), Chao and Sanzolone (1973), Chao and
Theobald (1976), Coffin (1963) and Olade and Fletcher (1974).
In order to distinguish chemical variations resulting from geothermal pro-
cesses from those related to the dispersion of analytical values expected in
natural background geochemical populations, cumulative probability plots
of analytical values were constructed and partitioned graphically (Lepeltier,
1969; Sinclair, 1974, 1976). This method permits estimation of population
parameters for single lognormal populations or for mixtures of two or more
statistical populations. Plots for three important elements -- Hg, As and Li --
presented in Fig. 2 illustrate the ranges and distributions of observed concen-
trations. Pearson product-moment and Spearman rank-order correlation coef-
ficients (Nie et al., 1975) for selected elements provide quantitatively assess-
ment of interelement associations.
1000 i i 1ooo
8OO 800
6O0 60c
5O O 5oc
400 400[
300 300
200

%~ ~oo ~1oot lO0 p p b

~ Bo ~ 70ppm eo~
6c p p /:,
~ 4c
~ 4o
~ :30pp~m
2O ~ 2c
Q ]C

Preclslon
L/mJt~--SPPb ~-2s--4pprn~"-
lJ L h i l L i i I I I I ~ I i
2 10 3 0 50 70 910 9J8 2 10 2 0 50 70 90 98

CUMULA TIVE PERCENTRGE

Fig. 2. Cumulative frequency distribution on log-probability coordinates of As, Hg and Li
contents o f all 30-m (100-ft) interval drill-hole cuttings samples.

Log probability plots (Fig. 2) indicate that the analytical values for As and
Hg belong to more than one lognormal population suggesting that they have
distributions related to more than one geochemical process or event. In both
cases, the ranges of concentrations that define each of the statistical popula-
tions overlap and, although threshold values may be defined, individual sam-
106

0 0 0 0 0 0 0 0 0 0 0 , - ~
z I I I I/

0 0 0 0 0 0 0 0 0 0 , - ~ 0
1 1 1 / 1

0 0 0 0 0 0 0 0 0 , - ~ 0 0
[ I J I / 1

I I / [ I

d o d d ~ d M ~ d d d o
I I / Ill
O,] 0 -~ ,-~ ,~ LO 0 ["- O~ Oa m ~"-

0 0 0 0 0 0 , - ~ 0 0 0 0 0
1 / 1 1 1

e~ 0 0 0 0 0 , ' 4 0 0 0 0 0 0
[',1 I / I I 1
0

a,
m
I/ ] I I|

C
ol I
r~ / I I } ]
~2
0 C d d ~ o d o d d d o o o

C ~ , - ~ O 0 0 0 0 0 0 0 0 0
I I i I I I
0

" e - ~ O 0 0 0 0 0 0 0 0 0 0
I
 107

ples cannot everywhere be unambiguously assigned to one or the other

population.
Pearson and Spearman correlation coefficients for selected elements in drill
hole samples are presented in Table II. Pearson coefficients are product-
m o m e n t correlations for pairs of analytical values. Spearman coefficients are
nonparametric statistics requiring value rankings rather than absolute values
for their computation. Both measure the strength of the relationship between
two variables, varying from --1 through 0 to +1. No group of elements in-
vestigated in this study is strongly correlated by these measures. It is evident
from the geochemical data presented in Figs. 6--10, however, that enrichments
of some elements, especially As and Li, are frequently coincident. This ap-
parent discrepancy is a consequence of two effects. Firstly, the correlation
coefficients, particularly the Spearman, are relatively insensitive to the few
correlative highly anomalous values when these are accumulated with the
greater number of lower, noncorrelatively, background concentration values.
Secondly, even though particular enrichments of elements are coincident
with those of others, this may be a consequence of either spatial or temporal
overlapping of noncoincident depositional events which are in fact distinct.
In this case, the low correlation coefficients accurately reflect the distinctly
separate nature of elemental distributions.

GEOCHEMICAL CHARACTERISTICS OF THE GEOTHERMAL SYSTEM

Chemical interaction between thermal fluids and reservoir rocks at

Roosevelt H o t Springs has resulted in distinctive trace-element enrichments
and depletions within the rocks. The chemistry of the solid materials docu-
ments the time-integrated effects of chemical redistribution within the geo-
thermal system. Within the Roosevelt system, five characteristic deposits or
suites of trace-element enrichments are recognized, each representative of
different combinations of physical and chemical conditions within the
thermal system. These are: (1) concentrations of As, Sb, Be, Mn and Hg
associated with siliceous sinter or cemented alluvium; (2) concentrations of
Ba, W, Be, Cu, As, Sb and Hg associated with manganese and iron-oxide
deposits; (3) high concentrations of Hg in argillized deposits at fumaroles and
lower concentrations in a diffuse halo surrounding the thermal center; (4)
concentrations of As and Li immediately surrounding fluid-flow-controlling
fractures or permeable horizons; and (5) deposits of CaCO3 at depth where
flashing of brine to steam has occurred.

Chemistry of surficial deposits

The chalcedonic sinter of the Opal Mound, deposited at the point of past
brine discharges, consists predominantly of thin, variably colored layers of
SiO2 with trace impurities. The variegated color of the sinter suggests that
the trace-element chemistry is not uniform and that the fluids from which
108

TABLE III

Chemistry of surface samples

Chalcedonic sinter, Soil, near Mn-cemented Altered alluvium

Opal Mound Opal Mound alluvium over fumarole

Na (%) 0.15 1.20 1.79 0.07

K (%) 0.14 1.95 3.12 0.26
Ca (%) 0.14 0.88 0.39 0.06
Mg (%) 0.01 0.59 0.10 0.03
Fe (%) 0.02 2.29 0.74 0.16
Al (%) 0.09 5.80 5.18 4.01
Ti (ppm) 19 3870 560 2040
P -- 407 651 336
Sr 33 212 386 266
Ba -- 539 4.9 (%) 326
Cr -- 34 9 24
Mn 388 504 18.8 (%) 173
Co -- 7 28 --
Ni -- 12 -- --
Cu -- 19 231 3
Mo -- -- 5 --
Pb -- 18 68 15
Zn 1 52 23 7
Cd -- -- 4 1
As 145 17 858 5
Sb 243 11 291 21
W -- 19 2940 29
Li 11 28 17 8
Be 99.8 2.5 18.6 3.4
Zr -- 107 17 42
La -- 39 37 34
Ce -- 59 42 56
Hg (ppb) 352 650 2210 49300

As determined colorimetrically; Sb, Au and Ag by AAS; Hg by gold film detector; all others by
ICP. Elements analyzed but present below detection levels include V, Te, Sn, Ag, Au and Th.
- - indicates element present below detection level.

the sinter formed may have been compositionally variable through time.
Small crystals of pyrite and pseudomorphs of hematite after pyrite occur
l o c a l l y w i t h i n t h e s i n t e r . S i g n i f i c a n t c h e m i c a l i m p u r i t i e s i n c l u d e As, Sb, Be,
Mn a n d Hg (Table III).
M a n g a n e s e - c e m e n t e d a l l u v i u m is e x p o s e d a p p r o x i m a t e l y 0.5 k m f r o m t h e
Opal Mound. Although the chemistry of the cemented alluvium reflects in
part the composite chemistry of the included lithic material, the hydro-
t h e r m a l m o d i f i c a t i o n is e v i d e n t p a r t i c u l a r l y b y c o m p a r i s o n w i t h t h e c h e m -
i s t r y o f a l l u v i a l soil ( T a b l e I I I ) . S i g n i f i c a n t e n r i c h m e n t s i n B a a n d M n a n d
lesser c o n c e n t r a t i o n s o f Co, C u , As, S b , W, Be a n d H g are c h a r a c t e r i s t i c .
 109

The altered alluvium collected from the area around an active fumarole
represents material that has been affected by interaction with geothermal
vapor rather than brine. Overall, the chemistry reflects the composition of
the parent alluvium which has been leached by sulfuric acid formed from
the oxidation of H2S-charged steam condensate. X-ray diffraction shows the
sample to consist primarily of quartz, kaolinite, sulfur, and minor gypsum.
Alkali and alkaline-earth metals are depleted while less mobile major ele-
ments -- A1 and Ti -- are relatively enriched. The concentration of Hg within
the sample at 49 ppm is more than 3 orders of magnitude greater than the
local background concentration.
Soil geochemistry
Soil samples collected from an area of approximately 11 km 2 over the
thermal system reveal evidence both of hypogene elemental enrichment due
to deposition from the cooling ascending thermal fluids and of supergene
redistribution of elements by the cooled oxidized fluids. The distribution
of As and Hg in soils are shown in Fig. 3.
Display of all results on lognormal cumulative frequency plots shows at
least two populations of Hg and As in the Roosevelt soil data. Anomalous
concentrations of Hg and As in soils over the system occur in a series of
closely spaced northeast- and northwest-trending zones that parallel major
fault directions in the geothermal field (Capuano and Moore, 1980). Zones
of high permeability, characterized by extreme enrichments of As and Hg,
occur at the intersections of these structural trends with the ground surface
and are frequently coincident with hot-spring deposits. Extension of the
anomalous Hg concentrations to the northwest, beyond the known limits
of the reservoir, reflects the emanation of Hg vapor from thermal fluids
flowing within shallow alluvial aquifers down the local hydrologic gradient.
This flow is d o c u m e n t e d as well by an extension of the shallow thermal
anomaly (Ward et al., 1978), by a characteristic pluming evident in regional
ground-water chemistry (D. Cole, pers. commun., 1980), and by anomalous
Hg concentrations immediately above the alluvium-bedrock contact in well
9-1.
A number of other elements are clearly remobilized within the near-sur-
face environment of the geothermal system. The concentrations of W and Sb
in soil are slightly greater in the vicinity of the hot spring deposits, whereas
the concentrations of Mn, Cu and Zn are coincidently lower. Capuano and
Moore (1980) propose that hydrothermal fluid convectively rises along major
structures, cools by conduction and boiling, becomes increasingly acidic due
to oxidation of hydrogen sulfide, and percolates downward, reacting with
and leaching soil and rock material. Elements which were stable in the thermal
brine m a y deposit in response to the changing conditions while other ele-
ments are leached from the soil into solution. Parry et al. (1978) recognize
such a supergene process as responsible for the formation of the alunite,
kaolinite and montmorillonite near-surface alteration assemblage near the
Opal Mound Fault.
110

• .U • . . . . ~

. . . . . ~,~ ~ ~ " M ' ~ ' ~ ~ k / l . . . . . . .

• • i ~ ;~ ~ X ~/~1 ; > ~ ~ • • .

• ' ~ h " " ~-//--~;; " / ( ' ~ ~"- . . . . . : " " ) ~'E

y" • - 7" ,'<J "~,-,. " y l ~ , ~ A ' ~ . . . . . . . . . . . . , ..... ~ . ,

• ~ ....... ,.~..~ .72,,. ~.~. ~'x: ........ ~-J,_~C~.~
..~.~.~. . . . . . r..~.~... ~'. u ~ , ~ . . ~ . ~ .-s . . . . . +.J~ • • . ~..-. • ~~.~7_~)-
.... : ~ ~
. . . . . , , ° • ° • .

: : : : : : :~., • .~. • .~_~...-. • • ~ -'--

i i i~ i~i i i i i~i !~:~-:--~: :::i ......... i...

~°
 111

Shallow temperature-gradient hole geochemistry

Analyses of drill cuttings from 32 90-m-deep temperature-gradient holes

drilled over approximately 310 km 2 exhibit pronounced enrichments in Hg
and As over the geothermal system. Anomalous As and Hg concentrations
in composite samples from the 30--60-m depth interval are grouped in an
area nearly coincident with the area of producing wells (Fig. 4). As in the
soil samples, anomalous Hg is more broadly dispersed than As, reflecting the
shallow flow of thermal fluids to the northwest. Although the number of
data points is small and the reliability of contouring low, the utility of this
observation for application in exploration is evident.

•131 J8 *4
e<r

elfll I ~ 1,18 o4
o-I
•3
•2
f ~3" 200 I
.... '-'~o) l •t2
e4
/
[ o4 *r
!
o82 •45

/357 f*
OPAL i~O°~
~
MOUND "1
FA UL'r.~./

"/
@<1

.B

• 18 / / 20f
k.j
• 17 eIB i<p

• 20 • I~ • 12 o<l , I i<I

o I 2, J ,~ MERCURY (ppb) ARSENIC (ppm)

Fig. 4. As and Hg c o n t e n t s o f temperature-gradient hole cuttings from 3 0 - - 6 0 m d e p t h

over the R o o s e v e l t H o t Springs s y s t e m . T h e l o c a t i o n s o f the Opal M o u n d Fault and o f
wells 14-2, 7 2 - 1 6 and 5 2 - 2 1 are indicated.

Distributions of elements in drill holes

Of the elements investigated, Hg, As, Li, Sb and Ca provide the clearest
expression of hydrothermal redistribution at depth within this system.
Analyses of fluids, altered rock, and scale indicate that a number of other
elements (Na, K, Ba, Sr, Fe, Si, Bi, Be, Zn, Mn) are leached, transported,
and deposited by thermal fluids moving within the system. The magnitudes
of the redistributions, however, are not sufficient to render them distinctive
from the normal background chemical variability of the reservoir rock.
Figures 6--10 illustrate the distributions of As, Hg, Li and Sb along with

Fig. 3. As and Hg in soils over the R o o s e v e l t H o t Springs g e o t h e r m a l system. The l o c a t i o n s

o f the Opal M o u n d F a u l t and several g e o t h e r m a l wells are indicated. After Bamford et al.
(19S0).
112

simplified geologic columns and generalized t e m p e r a t u r e profiles for two

deep p r o d u c t i o n wells (14-2 and 72-16), two deep n o n p r o d u c i n g wells (9-1
and 52-21) and one t e m p e r a t u r e gradient hole (1A). Locations of the holes
are shown on Fig. 1. Other wells studied but not included are GPCR-15 and
24-36, b o th o f which showed essentially no recognizable h y d r o t h e r m a l ef-
fects, and 1B and 76-1 which are very similar to drill hole 1A.

Fig. 5. Schematic cross section through the Roosevelt Hot Springs geothermal system show-
ing the positions of wells relative to major structural and hydrologic features. Solid arrows =
groundwater flow; lined arrows = thermal water flow; lined patches = zones of thermal
water discharge into alluvium. The locations of all features are schematic only.

The positions of the drill holes relative to the interpreted geom et ry of the
geothermal system are illustrated schematically on Fig. 5. Each well occupies
a unique position relative to fluid-flow-controlling fractures and the overall
g e o m e t r y o f the reservoir. Trace-element enrichments, like h y d r o t h e r m a l
alteration, result from the chemical interaction between thermal fluids and
rock. H y d r o t h e r m a l element enrichments are not pervasively developed
t h r o u g h o u t the reservoir rocks, but rather are localized immediately adjacent
to both active fluid conduits as well as older fluid-conductive fractures which
have been sealed. Analyses of 3-m interval samples from well 72-16 (Fig. 6)
indicate th at the dimensions of the h y d r o t h e r m a l trace-element signatures
m a y be quite limited. The data summarized on Fig. 6 indicate that, while
analyses o f 30-m composite samples may s m o o t h out some chemical variabil-
ity associated with localized enrichments evident in 3-m samples, broad en-
richments associated with major chemical redistributions remain recognizable.
 113

DH 72-16
LITHOLOGY TEMP. °F PPM AS PPM Sg

-'-~o
PPB HG PPM LI

o .... iooi ~o r~s

PPM AS PPM SB PPB HG PPM LI

~o , s,o

Al
SilluicviifuieAlmdluvium~~ Ouartz
BiotitMonzoni
Grani
e teGnei
te s~s BiotiteGneiss
Fig. 6. Lithology, temperature, and As, Sb, Hg and Li concentration profiles for well
72-16; 3-m (10-ft) and 30-m (100-ft) interval samples.

Distinctive geophysical well log responses and temperature profiles in the

upper portions of producing wells 72-16 and 14-2 suggest that broad intervals
of anomalous chemistry and more intense alteration are associated with
steeply-dipping fracture zones (Glenn and Hulen, 1979). In well 14-2, the
drill appears to have followed along such a zone from 275--550 m (900--
1800 ft). Well 72-16 encountered hot-water at 95 m (312 ft) and 191 m
(628 ft) in silicified altered alluvium and at the b o t t o m of the hole in little
altered gneiss.
Three temperature-gradient holes, 1A, 1B, and 76-1, drilled near the trace
of the Opal Mound Fault, did not intersect fluid-producing fractures but con-
tain rock t h a t has been intensely altered by thermal fluids (Fig. 8).

Arsenic
In the drill cuttings, the concentration of As is not uniform throughout
the reservoir rock but rather is discontinuous, reflecting the geometry of
fluid-flow-controlling structures. This is particularly evident at 95 m (312 ft)
in well 72-16 (Fig. 6), at 275--500 m (900--1800 ft) and 878 m (2880 ft) in
well 14-2 (Fig. 7) and in drill hole 1A (Fig. 8). Wells 52-21 and 9-1 both con-
tain somewhat higher concentrations within their upper, more altered por-
tions (Figs. 9, 10}.
114

LITHOLOGY TEMR* F PPM AS PPB HG PPM LI

o o
O° o 0
,?'q o o~,
I0'

\
_\ / \ /\

/
%
~3"

u,, /

/
I-- \ ~ /
I

\ /\/
--/ \

I/ 7/\
I/ \\
I I I
ix/ ~ \ /

Alluvium [] Quartz Monzonlte

~ Microdiorite Hornblende- Biotite Gneiss
[TT~ Biotite Microgrnnlte

Fig. 7. Lithology, temperature and As, Hg and Li concentration profiles for well 14-2;
30-m (100-ft) interval samples.

Lithium
Like As, the distribution of Li is closely associated with fluid-flow-con-
trolling fractures and permeable horizons, although the relative enrichment
of Li is lower. A general correspondence is observed between the intensity of
argillic alteration (Ballantyne and Parry, 1978; Rohrs and Parry, 1978) and

~. LITHOLOGY PPM AS PPB HG PPM LI

'=" ° ,

2oo-pf f'//ll l

Altored AlluvluI [iT/;] Altered Biotite Gneiss

Fig. 8. Lithology and As, Hg and Li concentration profiles for drill hole 1A; 30-m (100-ft)
interval samples. The lithology and geochemistry of drill holes 1B and 76-1, not illustrated,
are very similar to this.
 115

~E
I I ~ i [ _ _ -- 1 i -- J -- --

•- - = = _ ~ ~ ~

"~-'-~ . . . .
..L ~ , - - = ~ = - -

col xl:l:J:l NI H.Ld:IG ~ 0

~ o

. . . . . . =.~ ~'~
~" o- 6

eOI Xl:l:It NI Hid:l(] ~

116

Li concentration. Significant concentrations of Li occur at 95 m (312 ft) in

well 72-16 (Fig. 6), at 878 m (2880 ft) in well 14-2 (Fig. 7) and in the shal-
lower levels of wells 52-21 and 9-1 (Figs. 9, 10).

Mercury
The distribution of Hg, in contrast to those of As and Li, is strongly related
to both the temperature and fluid-flow geometry of the system. Significant
Hg concentrations are closely associated with fluid pathways only in the
outer portions of the reservoir.
Comparison of present temperatures and Hg distribution suggests that
significant Hg deposition does not occur at temperatures greater than a b o u t
200°C. The 878-m (2880-ft) entry in well 14-2 is not characterized by anom-
alous Hg concentration, despite alteration and deposition of As and Li (Bam-
ford et al., 1980). Hg is enriched along through-going fractures in the cooler
fringes of the system and is somewhat concentrated as well in the little
altered, relatively cooler gneiss of well 52-51.

Antimony
The distribution of antimony within the Roosevelt geothermal system has
been difficult to d o c u m e n t because of the low concentrations present in
whole-rock samples relative to analytical detection levels. In well 72-16, Sb
concentrations up to 9 p p m in whole rock samples occur only in the upper
120 m {400 ft) of the drill hole (Fig. 6). Similar concentrations were not
measured in drill cuttings from the other deep drill holes. These limited data
suggest that Sb is concentrated within a narrow zone only at the upper
margin of the geothermal system, largely coincident with a zone of silicifica°
tion.

Other elements
The down-hole distributions of several other elements, including Be, Sr
and Ca, while probably affected by hydrothermal activity, are less clearly
related to the present geothermal system. For these elements the magnitude
of the redistribution at depth within the undisturbed system is insufficient
to be distinguished from normal lithologic variability. Be is concentrated in
sinter (Table III) and appears to be slightly enriched about the fluid entry
zones in well 14-2 (Bamford et al., 1980). Sr has been depleted near fluid
entries where hydrothermal alteration has resulted in the destruction of
plagioclase (Bamford et al., 1980).
Ca, although a minor c o m p o n e n t of thermal brines (Table I), is concen-
trated in carbonate scale. The occurrence of this deposition is well known in
producing wells and is presumed also to occur naturally in response to boiling
caused by pressure decreases.
 117

TRACE-ELEMENT DEPOSITION

Within the Roosevelt H o t Springs geothermal system, five different types

of deposits or trace element enrichments have been identified. Elements have
been concentrated in response to changing physical and chemical conditions
encountered and induced by migrating thermal liquid and vapor. While some
elemental distributions appear to be readily explained by observed active
processes, not all of these deposits formed at the same time and certainly
n o t all are forming at present. Deposition or concentration of these elements
within the solid materials m a y occur by several processes including: (1)
deposition of the element as an essential mineral c o m p o n e n t (Hg in cinnabar);
(2) deposition of a mineral containing the element as a trace, nonessential,
c o m p o n e n t (As in pyrite); (3) formation of alteration minerals containing
the element as a trace, nonessential, c o m p o n e n t (Li in clay); and (4) adsorp-
tion of trace elements upon mineral surfaces (Co in Mn-oxide).
The deposition of trace elements within the system is thus controlled
chemically not only by reactions involving each element specifically but by
the chemical and physical state of the entire system. The chemical environ-
ments identified in this discussion are not unique. Physical conditions vary
continuously and consequently the deposition of trace elements is also some-
what continuous.

Deposition of arsenic and antimony

The distributions of As and Sb within the Roosevelt system indicate that

they are transported by thermal brines and deposited predominantly within
sulfide minerals (Fig. 11). Electron microprobe studies show that the host
minerals for As are pyrite and hematite after pyrite. The As content is highly
variable within and between individual pyrite grains with a maximum value
of 3.7% determined. Rims of hematite about arsenical pyrite, or hematite
pseudomorphs after pyrite, contain as much as 0.6% As. Parry et al. (1978)
and Hulen (in Nielson et al., 1978) report the possible presence of realgar
associated with opal and native sulfur in altered alluvium.
As and Sb are transportable in aqueous solutions lacking excess sulfur or
chloride ions (Barnes and Czamanske, 1967; Dickson and Tunell, 1968;
Weissberg et al., 1979). Although the complexes responsible for the ob-
served solubilities are not clearly defined, sulfur complexes are considered to
be of major importance. For both As and Sb, the experimental work has
demonstrated a direct temperature dependence of their solubility. As fluids
rise within the geothermal system, decreasing temperature and H2S loss ac-
companying decreasing pressure results in the deposition of As and Sb. This
process is expected to be particularly effective if solutions are discharged to
the surface or reach shallow zones of boiling. Concentrations of As and Sb
at shallow levels, particularly at boiling interfaces, are observed in other
geothermal systems including Broadlands, New Zealand (Ewers and Keays,
118

ARSENIC
#f #2 ~3 #4 ~5 #6 ,~-7
I0 20 4 e 4 B <,080 4 B 2 4
EXCHANGE - - ppm
Mn. OXIDE I
Am. FeOx.
XL FeOx,
SULFIDE,
SILICATE,
I RON
1 20 I 2 I 2 I Z I 2 I 2 I 2 %

.. p..:.o ZINC
. oo.o.o .L.o
COPPE R
4 IO 2 0 4 8 4 8 I0 2 0 20 40 10 2 0

LITHIUM

E×OHA
"XL"O,,
oO'L'OA'E
X
U'F,OE
O'"0"
EGEL L . . . . . . . . . .

Fig. 11. As, Fe, Z n , Cu a n d Li c o n c e n t r a t i o n s in various r o c k c o m p o n e n t s d e t e r m i n e d b y

selective a n a l y t i c a l e x t r a c t i o n s . I d e n t i f i e d c o m p o n e n t s are c a t i o n - e x c h a n g e a b l e m e t a l ,
m a n g a n e s e oxides, a m o r p h o u s iron oxide, crystalline iron oxide, sulfide, a n d r e m a i n i n g ,
p r i m a r i l y silicate, m a t e r i a l . S a m p l e # 1 = 72-16, 9 0 - - 1 2 0 m ( 3 0 0 - - 4 0 0 ft); # 2 = 72-16,
1 8 0 - - 2 1 0 m ( 6 0 0 - - 7 0 0 ft); # 3 = 14-2, 4 8 0 - - 5 1 0 m ( 1 6 0 0 - - 1 7 0 0 ft); # 4 = 76-1, 3 4 - - 4 0 m
( 1 0 2 - - 1 2 0 ft); # 5 = 1B, 6 8 - - 7 4 m ( 2 0 4 - - 2 3 0 ft); # 6 = 9 - 1 , 5 7 0 - - 6 0 0 m ( 1 9 0 0 - - 2 0 0 0 ft);
# 7 = 5 2 - 2 1 , 4 5 0 - - 4 8 0 m ( 1 5 0 0 - - 1 6 0 0 ft).

1977), S t e a m b o a t Springs, Nevada {White, 1967) and Uzon Caldera, USSR

(Ozerova et al., 1971).
The depositional mechanism for As and Sb in this system is suggested to
be coprecipitation with pyrite, as discussed by Ewers and Keays {1977) for
the Broadlands system. During deposition of pyrite, As and Sb are incor-
porated either b y entrapment of surface-absorbed ions or by ionic substitu-
tion into the growing pyrite crystals. The highly variable content of these
elements within pyrite grains suggests that the kinetic or chemical controls
on this process were both spatially and temporally variable. It is suggested
that the concentration of Sb at 60--120 m (200--400 ft) in well 72-16 repre-
sents a zone of past boiling where high-temperature solutions discharged into
the alluvium and cooled rapidly by boiling, depositing silica, pyrite and as-
sociated trace elements. Concentrations of As and Sb in the pyrite-bearing
opaline sinter of the Opal Mound are also probably related to such a process.

Deposition of lithium

The hydrogeochemical cycle of Li has been discussed by Brondi et al.

(1973). The partitioning of Li between thermal fluids and rock is extremely
temperature-dependent; small increases in the temperature at which water--
 119

rock interaction occurs greatly increase the content of Li in the waters. Li

does not form insoluble secondary minerals and is in general relatively mobile.
Clay minerals, however, can selectively remove Li from natural waters by ca-
tion exchange and can attain high Li contents. Within this process the behav-
ior of Li is affected by that of Mg since these elements o c c u p y the same sites
in the clay structure (Brondi et al., 1973). Thus we observe a general coinci-
dence of thermal fluid entries, hydrothermal argillic alteration, and Li anom-
alies. Selective analytical extraction experiments (Fig. 11) confirm the oc-
currence of Li as a constituent in silicate minerals, although the specific
mineral has n o t been identified.

Deposition o f mercury

Hg, like As and Sb, is most likely transported in solution as a mercury

sulfide complex (Barnes, 1979). It is not surprising then that Hg, As and Sb
enrichments are frequently coincident. Changes in conditions that can cause
Hg deposition as a sulfide include increasing activity of 02 or H ÷, decreasing
activity of H2S and HS- (as by boiling), dilution or cooling (Barnes, 1979).
The mobility of Hg in the temperature range of the geothermal reservoir
was investigated by measuring the temperature dependence of Hg liberation
from Roosevelt soils and drill cuttings. Pulverized samples were heated in air
at one atmosphere for approximately one hour and splits removed for analysis
by gold-film Hg detector. The temperatures of Hg release varied significantly
among the samples although the overall forms of the Hg loss vs. temperature
curves are similar (Fig. 12).
The temperatures of Hg release have been shown to be characteristic for
the various Hg c o m p o u n d s and occurrences. Hg adsorbed as an organomercury
I00

$0
Z

60

~-- 40
Z

~= 20

t
' I~o 2~o 3o0 4o0 ,oo
TEMPEItATURE,eC
Fig. 1 2 . L o s s o f m e r c u r y as a f u n c t i o n o f t e m p e r a t u r e . • = s i l i c e o u s sinter; o = a l t e r e d al-
l u v i u m near f u m a r o l e ; D = drill c u t t i n g s , w e l l 7 2 - 1 6 , 9 0 - - 1 2 0 m ( 3 0 0 - - 4 0 0 ft); ~ = drill
c u t t i n g s , w e l l 7 2 - 1 6 , 1 8 0 - - 2 1 0 m ( 6 0 0 - - 7 0 0 ft); • = drill c u t t i n g s , w e l l 1 4 - 2 , 4 8 0 - - 5 1 0 m
(1600--1700 ft); • = drill c u t t i n g s , w e l l 9 - 1 , 4 8 0 - - 5 1 0 m (1600--1700 ft).
120

species on soil is rapidly volatilized at 100--200°C (Landa, 1978), while ele-

mental Hg adsorbed on mineral surfaces is released below 110°C (Reed et al.,
1972). Mercury chlorides liberate detectable Hg at less than 100°C with com-
plete loss below 250°C (Koksoy et al., 1967). The temperatures of Hg release
from silicates have been shown to vary with metamorphic grade but to ex-
hibit in general a maximum release at a b o u t 150°C (Jovanovic and Reed,
1968, 1980). Liberation of Hg from pyrite is similarly dependent upon
metamorphic grade; pyrite from greenschist-facies rocks occurs at tempera-
tures above 450°C (Reed et al., 1972). Koksoy et al. (1967) found that Hg
in cinnabar was mobilized above 250°C with complete extraction by 350°C.
Moiseyev (1968) summarized from the work of others three types of Hg
association in rocks: (1) in an adsorbed state it is mobilized by any tempera-
ture rise and is completely removed by 300°C; (2) as mercuric sulfide it is
mobilized mainly above 250°C and completely removed at 500°C; and (3) as
a trace in pyrite it is mobilized in substantial amounts only above 700°C. The
critical temperatures proposed by the cited studies are not in exact agree-
m e n t due to differences in experimental conditions. The relative sequence of
release temperatures is, however, well established.
The release curves observed for the Roosevelt samples indicate that Hg oc-
curs in more than one form. Indeed the stepped release curve for the sample
from well 14-2 suggests that at least 3 forms of Hg may be present within
this one sample. Three temperatures, 150°C, 250°C and 400°C, appear to be
critical thresholds defining four temperature regimes.
Hg lost below 150°C from two samples from well 72-16 is interpreted to
have been Hg adsorbed upon rock material. The occurrence of adsorbed Hg
on altered material within these shallow samples is consistent with the pro-
posed transport of Hg as a vapor and its condensation within a cooler halo
a b o u t the thermal center.
Major breaks occurring in the Hg release curves for several samples at tem-
peratures above 150°C are interpreted to represent the release of Hg from
silicate rocks and minerals. All of these materials have been affected by hydro-
thermal fluids, and chalcedony or quartz veins are a c o m m o n feature.
A second, higher-temperature break occurring above 250°C for the altered
alluvium and the sample from well 14-2 may represent the thermal decomposi-
tion of cinnabar. The presence of cinnabar associated with opal was recognized
by Parry et al. (1978). The occurrence of mercury as cinnabar within altered
alluvium indicates that Hg is deposited as cinnabar from Hg and H2S-charged
vapor within the fumarolic environment.
The highest temperature release of Hg from the well 14-2 sample occurs at
temperatures in excess of 400°C, a temperature consistent with the occur-
rence of Hg in pyrite or other sulfides. This particular sample interval is highly
altered and contains pyrite as a hydrothermal mineral (Ballantyne and Parry,
1978).
R o b e r t s o n et al. (1977, 1978) have shown that during the flashing of geo-
thermal fluids, Hg is concentrated in the steam fraction whereas As remains
 121

primarily in the brine. This partitioning of Hg and As between vapor and

liquid phases provides a mechanism for the separation of these elements. Our
heating experiments have further demonstrated that a significant portion of
the Hg in the reservoir rock is heat labile at temperatures above a b o u t 200°C.
Hg transport is probably by aqueous mercury sulfide complexes in brine with-
in the reservoir, but vapor transport becomes dominant about the margin of
the reservoir.
Deposition of Hg does n o t become significant until temperatures decrease
to below a b o u t 200°C. Hg then may be deposited from solution by adsorp-
tion onto other mineral surfaces, by incorporation within silicate or sulfide
phases as an impurity, or by precipitation of cinnabar. A b o u t the margin of
the hot-water system, vapor dispersion and deposition by adsorption or cin-
nabar formation appear dominant. Deposition of vapor-transported Hg may
occur either directly from the vapor phase or from vapor condensate.

Deposition of iron and manganese

Depletions and enrichments of Mn and Fe in drill cuttings are in general

minor relative to normal rock variability. Concentration of Fe and Mn oxides
and Fe sulfides a b o u t active fluid pathways and the presence of Mn and Fe
oxide-cemented alluvium, however, are clear evidence of their mobility with-
in the system. In soil samples, the concentrations of Fe and Mn are located
at some distance, up to 0.5 km, from identified points of fluid discharge.
While some Mn is evidently leached and remobilized in the near-surface
environment, the presence of Mn in well-scale and siliceous sinter indicates
that the thermal brines deliver Mn to the surface as well. Fe and Mn discharged
from h o t springs remain in solution, are transported in the cooling, oxidizing,
probably acidic solutions for a short distance, and are precipitated as insoluble
oxides. Once formed, Fe and Mn-oxides enhance the concentration of other
elements, particularly of transition metals, because of their strong adsorption
potential (Jenne, 1968; Murray, 1975; Chao and Theobald, 1976; Means et
al., 1978; Vuceta and Morgan, 1978). The Mn oxide-cemented alluvium at
Roosevelt H o t Springs contains extraordinary amounts of W, Ba, Co, Cu,
Be, As, Sb and Hg. The association of these metals with Mn has been used by
H e w e t t et al. (1963) and H e w e t t (1964) to distinguish Mn oxide deposits of
hydrothermal origin from deposits of supergene origin.

DISCUSSION

Geothermal system model

A schematic model for the Roosevelt H o t Springs geothermal system,

based upon the known geology and geochemistry, is presented in Fig. 5. The
holes discussed in this paper are positioned according to our perception of
their location within this schematic model. Each hole occupies a unique posi-
122

tion relative to the system configuration, and consequently, each has a unique
rock chemistry. It is emphasized that this is not intended to be an accurate
section through the system.
A n u m b e r of different geothermal deposits are indicated. As is evident
from the previous discussion, the geothermal trace-element signatures are not
spatially continuous nor of regular dimension or form. The nonfracture
permeability of the various reservoir rock units is insignificant. Hence, the
reservoir and associated deposits are restricted to discrete, in places widely
separated, fractures and fracture zones. These control the migration of thermal
fluids and significantly affect the temperature configuration of the system.
Structures may act as barriers to the flow of fluids as well. The Opal
Mound Fault apparently forms a western limit to the fluid reservoir and, by
limiting the movement of thermal fluids, acts as an effective barrier between
different temperature regimes. The existence of a similar flow-restricting east-
trending structure between holes 72-16 and 52-21 has been proposed by Ward
and Sill (1976} and Nielson et al. (1978) based upon geophysical evidence.
It is expected that the specific reservoir configuration has not been static
through time, b u t that it has been somewhat transient in response to fracture
sealing by precipitation and alteration and to tectonic refracturing. At various
times during the life of the system, thermal brines have discharged at high
temperature. At present, vapors reaching the surface apparently overlie a
boiling interface at depth.
Producing wells 14-2 and 72-16 directly overlie shallow portions of the
reservoir. Anomalous concentrations of As and Li within the rock are closely
associated with specific fluid pathways. These pathways may be presently
active or may be sealed. Hg also is concentrated within the cooler outer por-
tion of the system. The enrichments of these three elements are coincident
directly over the system due to a telescoping of their distributions in response
to the high temperature gradient. Sb is concentrated in a shallow interval
in the upper level of 72-16 where coincident silicification suggests that boiling
may have occurred.
Drill holes l-A, 1-B and 76-1 were drilled in very close proximity to the
Opal M o u n d Fault. Fluids rising along this structure, cooling and possibly
boiling, have deposited elements in response to the pronounced physical and
chemical changes. Enrichments of As, Li and Hg are coincident here also.
Nonproducing drill holes 52-21, 9-1, 24-36 and GPCR-15 are peripheral to
the known shallow fluid reservoir. As and Li concentrations are in general
low, since interaction of the rocks with thermal fluid has been minimal. Only
in the upper portions of 9-1 and 52-21, where limited outflow of thermal
brine occurs, are greater than background concentrations found. Hg concen-
trations, on the other hand, are relatively high in gneiss of 52-21, well-removed
from k n o w n fractures and clearly on the fringe of the system. The absence of
broad Hg enrichments in GPCR-15 and 24-36 may be due to their distances
from the center of the high-temperature system and the absence of intercom-
municating structures.
 123

We may usefully examine the system by following a volume of thermal

fluid as it moves within the system. Figure 13 presents a very generalized model
on which several possible fluid pathways are indicated. At depth within the
system, two zones of trace-element enrichment are distinguished: a volume
of rock in the center of the system in which As and Li are deposited about
thermal fluid pathways, and a cooler surrounding envelope in which Hg is
concentrated with As and Li about fluid entries as well as being more general-
ly dispersed t h r o u g h o u t the reservoir rock. The two zones are separated ap-
proximately by 200°C isotherm. As our volume of fluid interacts with reser-
voir rock at depth, As is concentrated where pyrite precipitation occurs, and
Li is enriched by incorporation within alteration minerals. As the fluid passes
through cooler rocks, Hg may either be deposited from the thermal brines or,
if separation of a vapor phase occurs, may diffuse outward for a considerable
distance to form a Hg halo about the system.

Mn-Oxide ~ Fumarole
cw, Da,L.o. I m r ^ CL.D U ', ~:~":'!::"mg.n2~ ~"~'%

~ ~ / ~ C 0 .~. 3 ~.

./Hg ./ '~'Li ~,..i .z~ "\ .~ .

// -\2Oooc

Fig. 13. Generalized schematic model of a geothermal system indicating relative posi-
tions of physical features and geochemical deposition.

As the water rises toward the surface hydrostatic pressure decreases suffi-
ciently to permit boiling. Depending upon the structural and temperature-
depth configuration of the system, boiling can occur anywhere from great
depths to the surface. In this model (Fig. 13), three courses of fluid discharge
are indicated. In the first, thermal water is cooled as it moves upward, reach-
ing the surface as a liquid, or boiling at shallow depths. Siliceous sinters of
the Opal Mound and zones of silica-cemented alluvium, inferred to be zones
of past shallow boiling or fluid mixing, are characterized by concentrations
of As, Sb, Be and Hg. This same suite of trace elements is enriched within
soils that have been invaded by thermal fluids.
Once having reached the surface, the brines cool and are oxidized by ex-
posure to the atmosphere or mixing with ground water. Near the point of
124

discharge, oxidation may result in the formation of acidic fluids which leach
near-surface materials. As the fluids cool, deposits of Mn and Fe oxides form,
including within them significant concentrations of Ba, W, Be, Co, Cu, As,
Sb and Hg.
In the second possible course, boiling at depth results in the separation of
a vapor phase enriched in Hg, CO2, H2S and other volatile constituents. Con-
densation and/or interaction of this vapor with rock materials or other gas
constituents at fumaroles results in the deposition of significant concentra-
tions of Hg. Lower concentrations of Hg form in a similar manner by restricted
leakage of vapors into soils over the geothermal system.
In the third illustrated fluid course, mineral precipitation within structures
leads to a nearly complete sealing of the fluid conduits, effectively interrupt-
ing fluid flow and permitting a rise in temperature. If the cap is breached by
tectonic or thermally induced fracturing or by drilling, sudden flashing of
the fluid column m a y occur beneath it, leading to the precipitation of calcium
carbonate and included trace elements. Thus, boiling can occur at different
levels at different times in the history of a geothermal system. Such recurrent
boiling has apparently been particularly significant in the formation of some
epithermal ore deposits (Buchanan, 1980).
Because of the different fluid-flow paths and different sequences of physical
and chemical conditions encountered, one thermal brine may leave behind
a variety of different chemical signatures and deposits. Each is characteristic
of a particular combination of physical and chemical conditions and provides
a lasting record of the state of the system at some point in its history. Thus,
the trace-element geochemistry of a geothermal system is in fact a time-in-
tegrated record of the development of the thermal system.

Significance of results

Geochemical studies at Roosevelt H o t Springs, Utah, have d o c u m e n t e d a

number of element enrichments which are directly associated with physical,
chemical and hydrologic conditions of the geothermal system. While con-
siderable caution must be exercised in the extrapolation of the model devel-
oped for this system to other geothermal systems, we suggest that the recogni-
tion of trace-element geochemical distributions a b o u t geothermal systems
can be of significant importance in the exploration for geothermal systems or
for ore deposits that m a y have formed in a geothermal environment. The ele-
ments which have been definitively redistributed by thermal fluids include
Fe, Mn, As, Sb, Li, Hg, Be and W, elements that have been shown to be
mobile in many other geothermal systems (Ellis, 1979; Weisberg et al., 1979).
Despite the c o m m o n element association, however, the specific associations,
enrichment patterns, and concentration magnitudes for each system will be
different. The physical and chemical conditions under which each element is
leached, transported and deposited are unique, as are the physical and chemical
configurations of each geothermal system. For example the felsic gneisses
 125

and granites of the Roosevelt system have lower background geochemical con-
centrations of As, Sb and Hg than the Franciscan reservoir rocks of the
Geysers or the mixed volcanic suite of the Broadlands thermal areas. The fluid
chemical differences between the relatively dilute brines of the Roosevelt
system and the highly concentrated chloride brines of the Salton Trough
(Skinner et al., 1967) have profound effects on the solution properties of the
fluids. Finally, the configuration of the fractures through these relatively
impermeable rocks has influenced the development both of the thermal con-
figuration and of the chemical distributions. Thus, although there are similar-
ities between systems, we should not be surprised to discover significant dif-
ferences.
The recognition and documentation of geochemical zoning about geo-
thermal systems can be of significant importance to exploration. The redistribu-
tion of elements in response to the temperature configuration and fluid flow
within a system imprints upon the rock a time-integrated history of the devel-
opment of the geothermal system. Points of discharge or fluid passage may be
recognized, and limits to past activity defined. Relative to the costs of ob-
taining drill chip samples, geochemical analysis is very inexpensive. The cost-
effectiveness of a soil survey is particularly high (Bamford et al., 1980).
The similarities between geothermal systems and epithermal ore-forming
hydrothermal systems have long been recognized (White, 1955; Joralemon,
1951) and are currently being carefully reexamined for applications to prospect
ing for previous metal deposits (Worthington et al., 1980; Buchanan, 1980).
Recognition of the trace-element enrichments described from geothermal
systems could provide valuable insight into the geometry and development of
ore-forming systems.

CONCLUSIONS

The determinations of 38 major-, minor- and trace-element concentrations

in drill cuttings, surface rock samples and soils from the Roosevelt Hot
Springs geothermal system provide a three-dimensional picture of chemical
redistribution within this hot-water, structurally controlled geothermal sys-
tem. Five distinct suites of trace element enrichment have been recognized,
each characteristic of a distinct combination of physical and chemical condi-
tions within the geothermal system. The geothermal deposits are not general-
ly continuous, widespread, or of regular form or dimension; that is, they do
not form continuous geochemical "zones" such as those found about many
metal-depositing hydrothermal systems.
Three of the recognized trace-element suites are characteristic of the sur-
face and near-surface environment. These are: (1) concentrations of As, Sb,
Be and Hg associated with siliceous material at the location of liquid dis-
charge, fluid mixing, or at boiling interfaces; (2) deposits of Mn and Fe
oxides containing concentrations of Ba, W, Be, Co, Cu, As, Sb and Hg formed
by the oxidation of cooled brines; and (3) high concentrations of Hg within
surficial materials, especially in the vicinity of fumaroles.
126

Continuing to greater depths within the system are concentrations of Hg

along with As and Li near fluid channels and of Hg alone in a more diffuse
halo within rocks outside of the 200°C isotherm. Two other suites at depth
within the system are: (4) concentrations of As in sulfides and Li in silicate
alteration minerals in the vicinity of high-temperature fluid conduits; and (5)
deposits of calcium carbonate where flashing of brine to steam due to pres-
sure release has occurred. The depths of these geothermal deposits are not
fixed but depend upon the physical conditions in the system.
Deposition of the elements occurs as a result of decreasing temperature,
boiling, mixing of ground water with thermal brines, and vapor loss as fluids
convect within the geothermal system. As and Sb, probably carried as sulfide
complexes, are precipitated within a sulfide phase. Li is likely to be incor-
porated w i t h ~ the structure of layer-silicate alteration minerals. Hg, carried
either within thermal brine probably as a sulfide complex or transported as
a vapor, apparently is not deposited in significant amounts at temperatures
greater than a b o u t 200°C. If a vapor phase separates, Hg is strongly partitioned
into the vapor and is effectively dispersed about the periphery of the system
where it may be deposited either as a sulfide or adsorbed as metallic Hg. Mn
and Fe oxides form upon the cooling and oxidation of thermal fluids, in-
corporating within them concentrations of Ba, W, Be, Co, Cu, As, Sb and Hg.
The model of the Roosevelt H o t Springs geothermal system developed
from the geochemistry is consistent with geological and geophysical models
developed by Nielson et al. (1978) and Ward et al. (1978). The application of
geochemistry to geothermal exploration seems to be an extremely cost-effec-
tive and useful technique. As an aid to interpreting the present configuration
and past development of a geothermal system, geochemistry provides insights
not obtainable by other techniques. Like other techniques, geochemistry is
most useful when used in conjunction with other geological and geophysical
methods.

ACKNOWLEDGEMENTS

The authors wish to express thanks to the staff of the Earth Science Labora-
tory of the University of Utah Research Institute for their cooperation. Dis-
cussions with David Cole and Dennis Nielson have been particularly helpful.
This work was initiated by Robert Bamford, formerly at ESL. Analytical
work was performed by Ruth Kroneman. This study was funded by U.S.
Department of Energy, Division of Geothermal Energy contract number
DE-AC03-79ET-27002 to the Earth Science Laboratory, UURI.

REFERENCES

Ballantyne, J.M. and Parry, W.T., 1978. Hydrothermal alteration at the Roosevelt Hot
Springs thermal area, Utah: petrographic characterization of the alteration to 2 km
depth. Univ. Utah, Dep. Geol. Geophys. Rep., 26 pp.
 127

Bamford, R.W., 1978. Geochemistry of solid materials from two U.S. geothermal systems
and its application to exploration. Univ. Utah Research Inst., Earth Science Lab. Rep. 6,
196 pp.
Bamford, R.W., Christensen, O.D. and Capuano, R.M., 1980. Multielement geochemistry
of solid materials in geothermal systems and its applications, Part I: The hot-water
system at the Roosevelt Hot Springs KGRA, Utah. Univ. Utah Research Inst., Earth
Science Lab. Rep. 30, 168 pp.
Barnes, H.L., 1979. Solubilities of ore minerals. In: H.L. Barnes (Editor), Geochemistry
of Hydrothermal Ore Deposits. John Wiley and Sons, New York, N.Y., pp. 404--460.
Barnes, H.L. and Czamanske, G.K., 1967. Solubilities and transport of ore minerals. In:
H.L. Barnes (Editor), Geochemistry of Hydrothermal Ore Deposits. Holt, Rinehart
and Winston, Inc., New York,N.Y., pp. 334--381.
Brondi, M., Dall'Aglio, M. and Vitrani, F., 1973. Lithium as a pathfinder element in the
large scale hydrogeochemical exploration for hydrothermal systems. Geothermics, 2 :
142--153.
Buchanan, L.J., 1980. Ore controls of vertically stacked deposits, Guanajuato, Mexico.
Soc. Min. Eng., AIME, Preprint, 80-2, 26 pp.
Capuano, R.M. and Bamford, R.W., 1978. Initial investigation of soil mercury geochem-
istry as an aid to drill site selection in geothermal systems. Univ. Utah Research Inst.,
Earth Science Lab. Rep. 13, 32 pp.
Capuano, R.M. and Cole, D.R., 1982. Fluid--mineral equilibria in a high temperature
geothermal system, Roosevelt Hot Springs, Utah. Geochim. Cosmochim. Acta, 46 :
1353--1364.
Capuano, R.M. and Moore, J.N., 1980. Hg and As geochemistry as a technique for mapping
permeable structures over a hot water geothermal system. Geol. Soc. Am., Abstr. with
Programs, 12(6): 269.
Christensen, O.D., 1980. Trace element geochemical zoning in the Roosevelt Hot Springs
thermal area, Utah. 3rd Int. Symposium on Water--Rock Interaction, Proceedings.
Alberta Research Council, Edmonton, pp. 121--122.
Christensen, O.D., Kroneman, R.L. and Capuano, R.M., 1980a. Multielement analysis of
geologic materials by inductively coupled plasma-atomic emission spectroscopy. Univ.
Utah Research Inst., Earth Science Lab. Rep. 32, 33 pp.
Christensen, O.D., Moore, J.N. and Capuano, R.M., 1980b. Trace element geochemical
zoning in the Roosevelt Hot Springs thermal area, Utah. Geotherm. Resour. Council,
Trans., 4: 149--152.
Chao, T.T. and Sanzolone, R.F., 1973. Atomic absorption spectrophotometric deter-
mination of microgram levels of Co, Ni, Cu, Pb, and Zn in soil and sediment extracts
containing large amounts of Mn and Fe. J. Res. U.S. Geol. Surv., 1 : 681--685.
Chao, T.T. and Theobald, P.K., 1976. The significance of secondary iron and manganese
oxides in geochemical exploration. Econ. Geol., 71: 1560--1569.
Coffin, D.E., 1963. A method for the determination of free iron in soil and clays. Can. J.
Soil Sci., 43: 1--17.
Dickson, F.W. and Tunell, G., 1968. Mercury and antimony deposits associated with ac-
tive hot springs in the western United States. In: J.D. Ridge (Editor), Ore Deposits of
the United States, 1933--1967. Am. Inst. Mining, Metall. Pet. Eng., New York, N.Y.,
pp. 1673--1701.
Ellis, A.J., 1979. Explored geothermal systems. In: H.L. Barnes (Editor), Geochemistry
of Hydrothermal Ore Deposits. John Wiley and Sons, New York, N.Y., pp. 632--683.
Ellis, A.J. and Mahon, W.A.J., 1964. Natural hydrothermal systems and experimental hot-
water/rock interactions. Geochim. Cosmochim. Acta, 28: 1323--1357.
Ewers, G.R. and Keays, R.R., 1977. Volatile and precious metal zoning in the Broadlands
geothermal field, New Zealand. Econ. Geol., 72 : 1337--1354.
Fisher Scientific Company, 1960. "Reagent of choice" for arsenic in parts per billion.
Fisher Scientific Company Technical Data TD-142, 5 pp.
128

Forrest, R.J., 1980. Historical synopsis of the Roosevelt Hot Springs geothermal field,
Utah. In: D. Nielson (Editor), Geothermal Systems in Central Utah, Guidebook to
Field Trip no. 7. Geothermal Resources Council, pp. 18--24.
Fuchs, W.A. and Rose, A.W., 1974. The geochemical behavior of platinum and palladium
in the weathering cycle in the Stillwater Complex, Montana. Econ. Geol., 69: 332--
346.
Getty Oil Co., 1978. Data for Roosevelt Hot Springs KGRA. Univ. Utah Research Inst.,
Earth Science Lab. open-file release.
Glenn, W.E. and Hulen, J.B., 1979. Interpretation of well log data from four drill holes
at Roosevelt Hot Springs KGRA. Univ. Utah Research Inst., Earth Science Lab. Rep. 28,
74 pp.
Hewett, D.F., 1964. Veins of hypogene manganese oxide minerals in the southwestern
United States. Econ. Geol., 59: 1429--1472.
Hewett, D.F., Fleischer, M. and Conklin, N., 1963. Deposits of the manganese oxides:
supplement. Econ. Geol., 58: 1--51.
Jenne, E.A., 1968. Controls on Mn, Fe, Ce, Ni, Cu and Zn concentrations in soils and
water. In: Trace Inorganics in Water. Advances in Chemistry Series, No. 73, American
Chemical Society, Washington, D.C., pp. 337--385.
Joralemon, P., 1951. The occurrence of gold at the Getchell Mine, Nevada. Econ. Geol.,
46: 267--310.
Jovanovic, S. and Reed, G.W., 1968. Hg in metamorphic rocks. Geochim. Cosmochim.
Acta, 32: 341--346.
Jovanovic, S. and Reed, G.W., 1980. Hg and Pt-metals in meteorite carbon-rich residues:
suggestions for possible host phase for Hg. Geochim. Cosmochim. Acta, 44: 1399--
1407.
Koksoy, M., Bradshaw, P.M.D. and Tooms, J.S., 1967. Notes on the determination of
mercury in geological samples. Inst. Min. Metall. Trans., 76: B121--124.
Landa, E.R., 1978. The retention of metallic mercury vapor by soils. Geochim. Cosmochim.
Acta, 42: 1407--1411.
Lepeltier, C., 1969. A simplified statistical treatment of geochemical data by graphical
representation. Econ. Geol., 64: 538--550.
Lipman, P.W., Rowley, P.D., Mehnert, H.H., Evans, S.H., Nash, W.P. and Brown, F.H.,
1978. Pleistocene rhyolite of the Mineral Mountains, Utah -- geothermal and archeo-
logical significance. J. Res. U.S. Geol. Surv., 6: 133--147.
Means, J.L., Crerar, D.A., Brocsik, M.P. and Duguid, J.O., 1978. Adsorption of Co and
selected actinides by Mn and Fe oxides in soils and sediments. Geochim. Cosmochim.
Acta, 42: 1763--1773.
Moiseyev, A.N., 1968. The Wilbur Springs district (California) example of a study of hy-
drothermal processes by combining field geology and theoretical geochemistry. Econ.
Geol., 63: 169--181.
Murray, J.W., 1975. The interaction of metal ions at the manganese dioxide-solution in-
terface. Geochim. Cosmochim. Acta, 39: 505--519.
National Geothernal Service, 1980. Steam contract signed for power plant at Utah's
Roosevelt KGRA. National Geothermal Service, 2(41), 2 pp.
Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. and Bent, D., 1975. SPSS Statistical
Package of the Social Sciences. McGraw-Hill, New York, N.Y., 675 pp.
Nielson, D.L., Sibbett, B.S., McKinney, D.B., Hulen, J.B., Moore, J.N. and Samberg, S.M.,
1978. Geology of the Roosevelt Hot Springs KGRA, Beaver County, Utah. Univ. Utah
Research Inst. Rep. 12, 120 pp.
Olade, M. and Fletcher, K., 1974. Potassium chlorate-hydrochloric acid: A sulfide selec-
tive leach for bedrock geochemistry. J. Geochem. Explor., 3: 337--344.
Ozerova, N.A., Noboko, S.I. and Vinogradov, V.I., 1971. Sulphides of mercury, antimony,
and arsenic forming from the active thermal spring of Kamchatka and Kuril Islands.
Soc. Min. Geol. Jpn., Spec. Issue, 2: 164--170.
 129

Parry, W.T., Bryant, N.L., Dedolph, R.E., Ballantyne, J.M., Ballantyne, G.H., Rohrs, D.T.
and Mason, J.L., 1978. Hydrothermal alteration at the Roosevelt Hot Springs thermal
area, Utah. Univ. Utah, Dep. Geol. Geophys. Rep., DOE contract EG-78-C-07-1701,
29 pp.
Reed, G.W., Jovanovic, S. and Tennessen, A.C., 1972. Concentrations and thermal release
of Hg from iron and zinc sulfides. Earth Planet. Sci. Lett., 14: 26--30.
Robertson, D.E., Crecelius, E.A., Fruchter, J.S. and Ludwick, J.D., 1977. Mercury emis-
sion from geothermal power plants. Science, 196: 1094--1097.
Robertson, D.E., Fruchter, J.S., Ludwick, J.D., Wilkerson, C.L., Crecelius, E.A. and
Evans, J.C., 1978. Chemical characterization of gases and volatile heavy metals in geo-
thermal effluents. Geotherm. Resour. Council, Trans., 2: 579--582.
Rohrs, D. and Parry, W.T., 1978. Hydrothermal alteration at the Roosevelt Hot Springs
thermal area, Utah: Thermal Power Co. Well Utah State 72-16. Univ. Utah, Dep. Geol.
Geophys. Rep., 23 pp.
Sibbett, B.S. and Nielson, D.L., 1980. Geology of the Central Mineral Mountains, Beaver
County, Utah. Univ. of Utah Research Inst., Earth Science Lab. Rept. 33.
Sinclair, A.J., 1974. Selection of threshold values in geochemical data using probability
graphs. J. Geochem. Explor., 3: 129--149.
Sinclair, A.J., 1976. Applications of Probability Graphs in Mineral Exploration. Assoc.
of Explor. Geochem., Special Paper no. 4, 95 pp.
Skinner, B.J., White, D.E., Rose, H.T. and Mays, R.E., 1967. Sulfides associated with the
Salton Sea geothermal brine. Econ. Geol., 62: 316--330.
Steven, T.A., Cunningham, C.G., Naeser, C.W. and Mehnert, H.H., 1977. Revised strati-
graphy and radiometric ages of volcanic rocks and mineral deposits in the Marysvale
area, west-central Utah. U.S. Geol. Surv., Open-File Rep. 77-569, 45 pp.
Thermal Power Co., 1978. Data for Roosevelt Hot Springs KGRA, Utah. Univ. Utah
Research Inst., Earth Science Lab. open-file release.
Vuceta, J. and Morgan, J.J., 1978. Chemical modeling of trace metals in fresh waters:
role of complexation and adsorption. Environ. Sci. Technol., 12: 1302--1308.
Ward, S.H. and Sill, W.R., 1976. Dipole-dipole resistivity delineation of the near-surface
zone at the Roosevelt Hot Springs Area. Univ. Utah, Dep. Geol. Geophys. Rep. 76-1,
7 pp.
Ward, S.H., Parry, W.T., Nash, W.P., Sill, W.R., Cook, K.L., Smith, R.B., Chapman, D.S.,
Brown, F.H., Whelan, J.A. and Bowman, J.R., 1978. A summary of the geology, geo-
chemistry, and geophysics of the Roosevelt Hot Springs thermal area, Utah. Geo-
physics, 43 : 1515--1542.
Weissberg, B.G., Browne, P.R.L. and Seward, T.M., 1979. Ore metals in active geothermal
systems. In: H.L. Barnes (Editor), Geochemistry of Hydrothermal Ore Deposits. John
Wiley and Sons, New York, N.Y., pp. 738--780.
White, D.E., 1955. Thermal springs and epithermal ore deposits. Econ. Geol., 46: 267--
310.
White, D.E., 1967. Mercury and base metal deposits with associated thermal and mineral
waters. In: H.L. Barnes (Editor), Geochemistry of Hydrothermal Ore deposits. Holt,
Rinehart, Winston, New York, N.Y., pp. 575--631.
White, D.E., Muffler, L.J.P. and Truesdell, A.H., 1971. Vapor-dominated hydrothermal
systems compared with hot-water systems. Econ. Geol., 66 : 75--97.
Worthington, J.E., Kiff, I.T., Jones, E.M. and Chapman, P.E., 1980. Applications of the
hot springs or fumarole model in prospecting for lode gold deposits. Min. Eng., 32:
73--79.