Chemical Geology, 25 (1979) 219--226 219

© Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands

C H E M I C A L G E O T H E R M O M E T R Y IN G E O T H E R M A L S Y S T E M S

A.J. ELLIS
Chemistry Division, DSIR, Petone (New Zealand)
(Accepted for publication August 11, 1978)

ABSTRACT

Ellis, A.J., 1979. Chemical geothermometry in geothermal systems. Chem. Geol., 25:
219--226.

The application of chemical and isotopic geothermometry to geothermal systems is

reviewed, pointing out the uses and limitations of specific reactions in estimating deep
temperatures from well, hot-spring and fumarole discharges.
At present the most reliable indicators are: the silica--water equilibria; the Na/K
ratio; the isotopic distributions A 2 H(H2--H2 O), A2 H(H2-CH4 ), A 18 O(H20--HSO~ );
and the gas reactions CO2 + 4H2 # CH 4 + 2H~ O, and 2NH3 -~ N2 + 3H2. Many other
qualitative chemical indicators exist.

INTRODUCTION

A b o u t t w e n t y years ago Ellis ( 1 9 5 7 ) , M a t s u o (1960), and Heald et al.

( 1 9 6 3 ) d e m o n s t r a t e d t h a t t h e p r o p o r t i o n s o f H-, O-, S- and C-containing
gases in h i g h - t e m p e r a t u r e volcanic s t e a m varied with the s o u r c e t e m p e r a -
tures. More r e c e n t l y , and with considerable practical significance it has
been f o u n d t h a t m u c h l o w e r sub-surface t e m p e r a t u r e s ( 1 0 0 - - 3 5 0 ° C ) can
be e s t i m a t e d f r o m t h e chemical or i s o t o p i c c o m p o s i t i o n s o f the water,
s t e a m or gas e m a n a t i n g f r o m g e o t h e r m a l systems. L a b o r a t o r y studies o f
the relevant chemical species at c o n t r o l l e d t e m p e r a t u r e s and pressures
(Ellis and Mahon, 1 9 7 7 ) have o b t a i n e d the e q u i l i b r i u m c o n s t a n t s necessary
for calibration. Chemical g e o t h e r m o m e t r y is n o w widely used in the evalu-
ation of the p o w e r p o t e n t i a l o f g e o t h e r m a l areas t h r o u g h hot-spring a n d
f u m a r o l e analyses, a n d in m o n i t o r i n g t h e e f f e c t s o f e x p l o i t a t i o n on t h e
t e m p e r a t u r e and h y d r o l o g y o f g e o t h e r m a l systems t h r o u g h t h e analysis
of well discharges (Truesdell, 1 9 7 5 a ) .
Qualitative e v i d e n c e o f high t e m p e r a t u r e s within an area and o f i n t i m a t e
r e a c t i o n b e t w e e n r o c k and h o t w a t e r can be derived f r o m the presence o f
c o n s t i t u e n t s such as Li, Cs, B, N H 3 , As and Hg in waters or v a p o u r s reach-
ing t h e surface ( T o n a n i , 1 9 7 0 ; White, 1 9 7 0 ; F o u r n i e r , 1 9 7 6 ) .
220

As pointed out by Fournier e t al. (1974), accurate chemical geother-

mometers involving fluid constituents must fulfil the following require-
ments:
(a) The elements or species involved must be freely available in the rock--
water system.
(b) The reactions must equilibrate within the time that the fluid is present
at the depths of interest.
(c) The reaction rates for re-equilibration of solution species or between
minerals and solution during upflow to cooler conditions at the surface
should be slow.
In addition, several other factors need consideration:
- -Geothermometers involving the reaction of fluid with particular minerals
can be applied only to systems in which these minerals can be expected
to exist or to form as secondary minerals -- an obvious but sometimes
overlooked fact. The mineral form of the c o m p o u n d in equilibrium
with the water is required, e.g., in silica geothermometry the nature of
the silica p o l y m o r p h present.
- -High-temperature water approaching the surface loses an appreciable
fraction of its mass as steam, so that the concentrations of solutes or of
particular isotopic species in water or steam samples need to be corrected
back to the values existing at the deep source.
-- The partial pressures of gases vary with the steam/water proportions in a
fluid flow, and for geothermometry purposes must be estimated for the
original deep conditions.
- -Surface water may enter the rising hot-water column to cause dilution
and isotopic modification.
-- Surface oxidation of H2 S and the accompanying increase in acidity leads
to accelerated rock--water interaction at shallow levels and may com-
pletely alter the nature of the original deep water, e.g., creating SO~-
and destroying HCO~.
For geothermal well discharges, the simple sampling situation and the
rapid fluid transit to the surface enable the successful application of a wide
range of chemical geothermometers. However, for natural h o t springs and
fumaroles, even with high outflows, fluids are always affected to some ex-
t e n t by the retrograde factors above. In surveying natural thermal activity,
agreement between several geothermometers is required before there can
be confidence in the temperatures derived, and in general more reliance
is placed on the reactions little affected by rock interaction.

SPECIFIC INDICATORS

Information on individual geothermometry reactions is summarized

under the following headings: Isotope exchange; Mineral solubility; Ion
exchange; and Solute reactions.
 221

Iso tope exchange

Laboratory experiments (Hulston, 1976) have indicated that the follow-

ing reaction attains equilibrium at 250°C in the order of days or weeks:
HD + H20 ~ H2 + HDO (1)
Agreement between temperatures for isotopic equilibrium (calculations by
Bottinga, 1969) and measured well source temperatures (200--260°C) has
been obtained for some New Zealand, U.S.A. and Iceland well discharges
(Arnason, 1976; Giggenbach and Lyon, 1977).
CH3D + H2 -~ CH4 + HD (2)
Reaction (2) has also been tested as a geothermometer, the equilibrium
fractionations having been calibrated by Craig (1975). Field experience
suggests that the rate of equilibration is similar to that of the H2 --H2 O ex-
change reaction, reasonable temperature estimates being obtained for
220--260°C waters in New Zealand (Hulston, 1976; Giggenbach and Lyon,
1977).
C-isotope exchange reactions (CO2--CH4 and HCO~--CO 2 ) have been
proposed as geothermometers. With the CO2--CH4 exchange reaction,
equilibrium attainment is so slow as to be beyond simple laboratory demon-
stration at 200--300 ° C. With the ready availability of extraneous carbon
gases from the rocks in up-flow channels, the relevance of the temperature
calculations is still in doubt (Ellis and Mahon, 1977). The temperatures
estimated are frequently 50 to 200°C higher than those measured at the
base of geothermal wells (Panichi et al., 1976). Conceivably, the estimates
may refer to equilibrium conditions much deeper in the geothermal system,
but more proof is required.
HCO3 ions tend to decompose by reaction with weak acids as geothermal
waters rise and cool. In many situations this creates difficulties in applying
the HCO~ --CO2 C-isotope exchange reaction. However, for alkaline waters,
such as those occurring in part of Yellowstone Park or Iceland, useful tem-
perature estimates may be obtained (O'Neil et al., 1975).
Many O-isotope exchange reactions are eliminated as geothermometers
because of the rapid rate of equilibration, or because of the instability of
a species (e.g., HCO~ ). Through the direct collection of geothermal steam
and CO2 in liquid N2, Panichi et al. (1976) demonstrated that AlSO(H20--
CO2 ) was a measure of the steam supply temperatures at Larderello.
The following reaction with a half-life reaction time of a few months at
neutral pH's and 250°C shows a good potential for geothermometry:
H2160 + HsISo1603 --~ H21So + HS1604 (3)
Hulston (1976) summarized the calibrations of the system. Agreement
with, or sli~h~iy higher than the temperatures measured, was obtained for
200--~00°C well discharges (Sakai, 1976; Ellis and Mahon, 1977). McKenzie
222

and Truesdell (1976) pointed o u t that surface hot-spring waters may have
their O-isotope compositions altered by steam loss or by dilution, and that
sulphate may be derived from the oxidation of H2 S. Means of correction
were suggested.
S-isotope exchange between sulphate and sulphide is slower than that of
oxygen between water and sulphate (Robinson, 1973; Hulston, 1976),
and temperature estimates available based on the former reaction are far
in excess of those measured in wells (e.g., at Wairakei, 400°C instead of
250°C).

Mineral solubility

Kharaka and Barnes (1973) developed a c o m p u t e r programme to examine

the ratio between the concentrations of individual constituents in natural
solutions and concentrations expected for chemical equilibrium at the
temperature.
High-temperature geothermal waters have been found to be saturated
with silica in equilibrium with quartz. Since this fact was established
(Fournier and Rowe, 1966, Mahon, 1966), silica concentrations in geother-
mal waters have been widely used as a means of measuring water-supply
temperatues for wells. Pressure and salinity have little effect, b u t for
alkaline waters the silicate ion concentration must be subtracted from the
total silica concentration found by analysis. Above about 200°C, equilibra-
tion between water and rock quartz occurs in the order of a few hours or
less, yet it is only for initial water temperatures considerably in excess of
300°C that any silica precipitates in upflow pipes. With careful collection
and analysis, and making corrections for steam loss, silica temperatures are
reproducible and accurate to within -+ 2--3°C in the range 200--300°C
(Ellis and Mahon, 1977).
At temperatures below a b o u t 16U~C, amorphous silica and chalcedony
often dominate silica chemistry, and for Iceland wells Arnorsson (1975)
found that geothermal waters above 180°C were saturated with quartz,
and below 110°C were in equilibrium with respect to chalcedony. Between
these temperatures there was little consistency.
Silica--quartz temperatures estimated directly from hot-spring water
analyses are rarely much in excess of 200°C, the temperature where, on
cooling, equilibration becomes slow. In addition, it may be uncertain
whether or n o t the formation of amorphous silica at spring temperatures
controls the concentrations of silica. Nevertheless, silica concentrations will
indicate in a semi-quantitative way the minimum underground temperatures
in an area.
In applying concentration-based geothermometers to springs the effects
of steam loss and mixing with near-surface waters must be considered.
Truesdell and Fournier (1975) and Fournier (1976) showed h o w graphical
"mixing models" containing the silica and chloride concentrations for the
 223

major springs in an area, could indicate the most probable solute concen-
trations and temperatures of the deep supply water.
The solubility of calcium salts such as CaSO4, CaF2 and CaCO3 decreases
rapidly with increasing temperature and many deep thermal waters approach
saturation with these c o m p o u n d s (Sakai and Matsubaya, 1974; Nordstrom
and Jenne, 1975; Ellis and Mahon, 1977). Concentrations of fluoride and
sulphate vary systematically with water temperature and Ca concentrations,
and may be used as semi-quantitative geothermometers.
The dominance of calcium chemistry in h o t waters by CO2 concentra-
tions and the pH--salinity relationship prevent the direct use of calcite
solubility as a geothermometer in hot-spring surveys, b u t deductions can be
made from well discharges (Ellis, 1970).
The concentrations of metal ions such as Mg, Fe and Mn in geothermal
waters are usually low (Goguel, 1977) b u t vary with temperature, and in
particular with water salinity. Through established relationships the con-
centrations of these metals can be used for geothermometry with well dis-
charges, but application to spring waters is dubious in view of the likely
contamination (Ellis and Mahon, 1977).

Ion exchange

The ratio Na/K in geothermal waters is widely used as a geothermometer.

The partitioning of Na and K between aluminosilicates and solutions is
usually attributed to ion exchange between coexisting Na- and K-feldspars,
and reasonable agreement exists between feldspar--water reaction experi-
ments and the natural-water results (White, 1965; Ellis and Mahon, 1967;
Mercado, 1970; Fournier and Truesdell, 1973). The simple ratio Na/K
gives a good measure of most geothermal well-water temperatures in the
range 180--350°C (Ellis, 1970). Although some back reaction occurs in
natural upflow channels, for boiling springs of good flow the Na/K ratio
has been of considerable value also in indicating minimum deep-water
temperatures in undrilled areas. The rate of re-equilibration is slower than
for silica and as a result the Na/K ratios frequently indicate higher tem-
peratures from deeper levels.
Fournier and Truesdell (1973) demonstrated that the simple Na/K ratio
gave low temperatures or was inadequate for waters with high Ca concen-
trations, and for waters below 180°C. Through the correlation of data from
a wide variety of sources and water types they proposed an empirical
Na--K--Ca geothermometer to be applied to high-Ca and cooler waters
(molal concentrations):

1647 -- 273.15 (4)

T(°C) = log(Na/K) + ~log(CaV'/Na) + 2.24

/J - 4/-~ ~_: Ca w/Na ~ 1, and water T is under 100°C

= 1/3 for Ca~2/Na ~ 1, or if T4/3 is over 100°C
224

Further, PaSes (1975) showed that the application to low-temperature

(under 75 ° C) waters was complicated by a dependency of derived tempera-
tures on CO2 concentrations. Due to the wide variations possible in rock
mineralogy, results from the uncritical application of the Na--K geother-
mometers to warm springs should be treated with reserve.
The N a / R b ratio in waters, potentially has a parallel application. Studies
of rock alteration in drill cores show that the uptake of Rb by K host
minerals (e.g., feldspar, illite) is less favoured than the competing host ion.
Discrimination against Rb by the solid phase gives the N a / R b ratio of a
geothermal fluid a longer high-temperature m e m o r y than the Na/K ratio.
As useful as this feature could be, so far there is no consistent temperature
calibration for different geothermal areas. To fill this need further experi-
mental work on rock--water reactions of the type described by Ellis and
Mahon (1967) is needed with longer equilibration times (R.L. Goguel, pers.
commun., 1978).

Solute reactions

Temperature-dependent reactions between solutes also present possibili-

ties for geothermometry. For example, Ellis and Mahon (1977) showed
that the ratio mHCO~/mco2 in deep h o t waters decreased rapidly with
temperature for a given water salinity.
The gases associated with deep high-temperature geothermal waters are
generally in chemical equilibrium for the temperature and pressure (Hulston
and McCabe, 1962; Giggenbach, 1978), and gas-reaction geothermometers
have been applied in many surveys of natural thermal activity.
The principal gases used in this respect are H2 O, CO2, CH4, H2, N2 and
NH 3 , which are inter-related chemically by the following equations:
CO2 + 4H2 ~ CH4 + 2 H 2 0 (5)

2NH a -m N2 + 3H 2 (6)
Estimation of temperature of equilibrium involves the partial pressures
of the reactant gases and it is not sufficient to just consider the fraction
ci of individual gases found by analysis in a steam sample (except in the
case of a dry-steam reservoir). In a liquid-dominated system the solubilities
of the gases in the deep h o t solution must be taken into consideration.
Giggenbach (1978) successfully utilized a form of equilibrium constant
K2 which incorporates the known distribution coefficient A for each gas
1 v
between the liquid and vapour phase (Ag = Xg/Xg), and F the fraction of
steam formed from an original single liquid phase by the time it reaches
the point of sampling. For example, for reaction (5) above:
K2 = K1 PH20 QA = (CCH4/CCO 2 c h 2 ) F 4 (7)
where K 1 is the conventional partial-pressure equilibrium constant, and
 225

QA = ACI-I4/A~, ACO 2" The analytical gas fractions are then related to
the value of K2 at various temperatures, using appropriate values of F,
until a temperature fit is obtained.

REFERENCES

Arnason, B., 1976. The hydrogen--water isotope geothermometer applied to geothermal

areas in Iceland. Geothermics, 5: 75--80.
Arnorsson, S. 1975. Application of the silica geothermometer in. low-temperature hydro-
thermal areas in Iceland. Am. J. Sci., 275: 763--784.
Bottinga, Y., 1969. Calculated fractionation factors for carbon and hydrogen isotope
exchange in the system calcite--carbon dioxide--graphite--methane--hydrogen--
water vapour. Geochim. Cosmochim. Acta, 33: 49--64.
Craig, H., 1975. Isotopic temperatures in geothermal systems. I. At. E. A. Adv. Group
Meet. on Application of Nuclear Techniques to Geothermal Studies, Pisa, Sept.,
1975.
Ellis, A.J., 1957. Chemical equilibrium in magmatic gases. Am. J. Sci., 255: 416--431.
Ellis, A.J., 1970. Quantitative interpretation of chemical characteristics of hydrothermal
systems. Geothermics (Spec. Issue 2), 2 (Part 1): 516--528.
Ellis, A.J. and Mahon, W.A.J., 1967. Natural hydrothermal systems and experimental hot
water/rock interaction (Part 2). Geochim. Cosmochim. Acta, 31: 519--538.
Ellis, A.J. and Mahon, W.A.J., 1977. Chemistry and Geothermal Systems. Academic
Press, New York, N.Y., 392 pp.
Fournier, R.O., 1976. Chemical geothermometers and mixing models for geothermal
systems. Geothermics, 5: 41--50.
Fournier, R.O. and Rowe, J.J., 1966. Estimation of underground temperatures from
the silica content of water from hot springs and wet steam wells. Am. J. Sci., 264:
685--697.
Fournier, R.O. and TruesdeU, A.H., 1973. An empirical Na--K--Ca geothermometer
for natural waters. Geochim. Cosmochim. Acta, 3 7 : 1 2 5 5 - - 1 2 7 5 .
Fournier, R.O., White, D.F. and Truesdell, A.H., 1974. Geochemical indicators of sub-
surface temperatures--I, Basic assumptions U.S. Geol. Surv., J. Res., 2: 259--262.
Giggenbach, W.F., 1978. Chemical gas geothermometers. Geochim. Cosmochim. Acta.
(submitted).
Giggenbach, W.F. and Lyon, G.L., 1977. The chemical and isotopic composition of
water and gas from the Ngawha geothermal field. Open File Rep. 30/555/7 Chem.
Div., Petone.
Goguel, R.L., 1977. Improved analytical values for aluminium, iron, manganese and
magnesium in Wairakei waters. In: Geochemistry 1977. Bull. 218, N.Z. Dep. Sci.
Ind. Res. Wellington, 144 pp.
Heald, E.F., Naughton, J.J. and Barnes, I.L., 1963. The chemistry of volcanic gases.
J. Geophys. Res., 68: 539--557.
Hulston, J.R., 1976. Isotope work applied to geothermal systems at the Institute of
Nuclear Sciences, New Zealand. Geothermics, 5: 89--96.
Hulston, J.R. and McCabe, W.J., 1962. Mass spectrometer measurements in the thermal
areas of New Zealand, 2. Carbon isotope ratios. Geochim. Cosmochim. Acta, 26:
299--410.
Kharaka, Y.K. and Barnes, I., 1973. SOLMNEQ: Solution--mineral equilibrium calcula-
tions. U.S. Geol. Surv., Rep. WRD-73-002, Menlo Park, Calif., 81 pp.
Mahon, W.A.J., 1966. Silica in h o t water discharged from drillholes at Wairakei, New
Zealand. N . Z . J . Sci., 9: 135--144.
226

Matsuo, S., 1960. On the origin of volcanic gases. J. Earth Sci. Nagoya Univ., 8: 222--
245.
McKenzie, W.F. and Truesdell, A.H., 1976. Geothermal reservoir temperatures esti-
mated from the oxygen isotope compositions of dissolved sulfate and water from
h o t springs and shallow drillholes. Geothermics, 5: 51--62.
Mercado, S., 1970. High activity hydrothermal zones detected by Na/K, Cerro Prieto,
Mexico. Geothermics (Spec. Issue 2), 2 (Part 2): 1367--1376.
Nordstrom, D.K. and Jenne, E.A., 1975. Fluorite solubility equilibrium in selected
geothermal waters. Geochim. Cosmochim. Acta, 41: 175--188.
O'Neil, J.R., Truesdell, A.H. and McKenzie, W.F., 1975. A13C(CO2--HCO-3 ) as a possible
geothermometer, 2nd U.N. Syrup. on the Development and Use of Geothermal Resources,
San Francisco, Calif., Abstr. III-71.
Pates, T., 1975. A systematic deviation from Na--K--Ca geothermometer below 75°C
and above 10 -4 atm P c o 2 . Geochim. Cosmochim. Acta, 39: 541--544.
Panichi, C., Ferrara, G.C. and Gonfiantini, R., 1976. Isotope geothermometry in the
Larderello geothermal field. Geothermics, 5: 81--88.
Robinson, B.W., 1973. Sulphur isotope equilibrium during sulphur hydrolysis at high
temperatures. Earth Planet. Sci. Lett. 18: 443--450.
Sakai, H., 1976. Sulfate--water isotope geothermometry applied to geothermal systems.
Geothermics, 5: 67--74.
Sakai, H. and Matsubaya, O., 1974. Isotope geochemistry of the thermal waters of Japan
and its bearing on the Kuroko ore solutions. Econ. Geol., 69: 974--991.
Tonani, F., 1970. Geochemical methods of exploration for geothermal energy. Geo-
thermics (Spec. Issue 2), 2 (Part 1): 492--515.
Truesdell, A.H., 1975a. Geochemical techniques in exploration. 2nd Proc. U.N. Symp.
on the Development and Use of Geothermal Resources, San Francisco, Calif., 1:
liii--lxii.
Truesdell, A.H. and Fournier, R.O., 1975. Calculation of deep temperatures in geo-
thermal systems from the chemistry of boiling spring waters of mixed origin. 2nd
Proc. U.N. Syrup. on the Development and Use of Geothermal Resources, San
Francisco, Calif., 1: 837--844.
White, D.E., 1965. Saline waters of sedimentary rocks. In: Fluids in Subsurface En-
vironments - - A Symposium. Mere., Am. Assoc. Pet. Geol., 4: 342--366.
White, D.E., 1970. Geochemistry applied to the discovery, evaluation, and exploitation
of geothermal energy resources. Geothermics (Spec. Issue 2), 1: 58--80.