Академический Документы
Профессиональный Документы
Культура Документы
00/0
Pergamon Press Ltd.
C~ 1981 CNR
Abstract--We propose here a new geothermometer for natural waters. Analyses from many explored
geothermal fields allow us to define two empirical thermometric relationships.
One is for waters of low to moderate salinity (CI-< 0.3 M)
log Na/Li = IO00/T -0.38,
(a)
and one for marine waters and brines (CI> 0-3 M)
log Na/Li = I195/T + 0.13.
(b)
These relationships, which at present are not well understood, result mainly from the increase of Li
concentrations in waters with temperature.
Equation (a) proved to be adequate for spring waters from mostly known geologic origin; this is an
important feature in geochemical surveys for geothermal prospecting.
Furthermore, when comparison between springs and drillhole chemistry of a given geothermal field is
possible, the Na/Li geothermometer gives more reliable temperature estimates from the spring
compositions than do classical geothermometers.
INTRODUCTION
N a t u r a l h y d r o t h e r m a l systems are an i m p o r t a n t m e a n s o f s t u d y i n g water - rock i n t e r a c t i o n s .
D e t e r m i n a t i o n o f the e q u i l i b r i u m t e m p e r a t u r e existing at d e p t h in a field has always been a
m a j o r g o a l a n d in the past 15 years the e v o l u t i o n o f g e o t h e r m a l p o w e r u t i l i z a t i o n has e n h a n c e d
still f u r t h e r the interest in t e m p e r a t u r e e s t i m a t i o n . Several w o r k e r s have p r o p o s e d r e l a t i o n s h i p s
between the c h e m i c a l c o m p o s i t i o n o f g e o t h e r m a l s o l u t i o n s a n d d e e p t e m p e r a t u r e s .
CHEMICAL
GEOTHERMOMETERS
56
Additional difficulties occur because of the mixing of rising hot solutions with cold
superficial waters. In this case, application of the mixing models approach proposed by
Fournier and Truesdell (1974) and Truesdell and Fournier (1976) is sometimes biased by
surprisingly high silica contents of the colder water fraction (see for example Giggenbach, 1978;
Fouillac et al., 1976b).
Compared to the silica geothermometer, the sodium/potassium ratio is less affected by
chemical re-equilibration occurring during the ascent of solutions; but, it may be slightly
disturbed by mixing with cold dilute superficial waters. The deposition of aluminium-rich
minerals, like kaolinite, has been invoked by Michard et al., (1979) to explain the freezing of
deep Na/K ratios. However, Na/K ratios can be modified by exchange reactions with clay
minerals as shown by Weissberg and Wilson (1977), or by addition of potassium from
sedimentary rocks in subsurface reservoirs (Michard et al., 1976). The application of the Na/K
geothermometer to superficial waters very often leads to high estimated temperatures, and this
is obviously troublesome for geothermal surveys.
Fournier and Truesdell (1973) have proposed an improved Na/K/Ca empirical
geothermometer. The true meaning of their formulae is not clear, but reflects an equilibrium
between alkali feldspars, calcium-bearing silicate (or calcite) and deep solutions (Shikazono,
1976; Michard and Fouillac, 1976). Fournier and Truesdell's formulae are not applicable to cold
concentrated solutions (cf. sea-water). Also, for CO,-rich springs, as pointed out by Pa~es
(1975) and Pa~es and Cermak (1976), Fouillac and Michard (1977), the Na/K/Ca relationship is
not reliable. An attempt to correct the Na/K/Ca geothermometer for CO2-rich waters (Pa~es,
1975) was not entirely satisfactory.
In summary, chemical geothermometers provide good temperature estimates for high
temperature tapped by drillholes but their use is more difficult when working on surface
hydrothermal manifestations.
SODIUM AND LITHIUM IN WATERS
The occurrence of lithium as a common element of most hot waters is well known and was
discussed by White (1957). A few years later, Ellis and Wilson (1960) in a careful chemical study
of Wairakei (N.Z.) drillhole discharges pointed out that a low Na/Li ratio seemed to be related
to the hottest zones of the field. This was also shown by Koga (1970) for Otake and Hatchobaru
(Japan) geothermal wells.
A geochemical survey of French hot springs has been undertaken since 1975 in the Massif
Central (Fouillac et al., 1976a) and in the Eastern Pyrenees and Corsica (Boul~gue et al., 1977).
The latter two regions include fields with estimated deep temperatures down to 60C. This
enlarges the temperature range of hydrothermal systems available to test geothermometers.
When looking for possible thermometric relationships, sodium and lithium appeared to be
promising elements. A wide search for lithium data in geothermal waters was thus undertaken.
Four groups of water systems were used to build up a thermometric relation.
1. Explored geothermal systems: with measured deep temperatures.
2. Natural hot springs: temperature is obtained by Na/K and silica geothermometers. For
French thermal waters only the systems showing concordant estimates are selected. For
samples from other countries, we have taken into consideration the estimates given by authors
in the literature.
3. Analyses o f surface waters: here, measured temperatures are available. For very dilute
samples, we have only considered flameless A.A.S. lithium determinations. Some studies
showed lithium values only twice the announced detection limit and accordingly these samples
were not considered further.
I/T
- log Li
- log Na
log N a / L i
0.84
0"82
0.93
0'82
0.78
1'63
1'56
I"56
l'71
1'55
} '53
1"52
1"51
1'52
I "51
References
CI ppm
Chile
E1 Tat(o
Well T1
Well "1-2
Well T3
Well T4
Well T5
',~ ell T6
Well TI0
Well T7
Well TI I
Well T l 2
m
m
m
m
m
m
m
m
m
m
2'07
2-00
1.9(I
2'03
2-06
2'20
1.96
1'87
2"00
1"92
2'63
2"39
2"51
2"54
2"33
El Talio Brine
2"16
3"14
-0"6.7
3'80
e
e
e
e
e
e
e
e
e
e
e
2"44
2.41
2"36
2-41
2-0'7
2"02
2" I I
2'66
2"40
2''73
2"45
3" 18
2"71
3"07
3" 14
2'69
3'00
2"89
4"03
3'67
4"3
3"10
1"23
0"95
I "21
I "21
1"07
1.41
I 19
2"09
2 10
2"03
1"09
I "95
1-77
1"85
1'92
1"61
1"59
1"70
1.94
I "57
2'26
2.00
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
2"74
2'61
2"49
2.60
2-64
2"82
2"68
2.77
3"06
2"72
2'81
2'77
2"58
2.77
2"71
4"88
4"56
4'88
4'88
4"86
5" 15
4-85
5'40
5"52
4"88
5.00
4.92
4'92
5"01
5.00
2'54
2'36
2"67
2"55
2.64
2' 73
2'63
2.87
2'77
2"50
2"50
2"57
2"55
2"72
2.60
2'34
2"20
2'21
2"33
2"22
2 "42
2"21
2"53
2'75
2'38
2"50
2"34
2'37
2"28
2"39
e
e
e
e
e
e
e
e
2'83
3" 12
2-48
2"61
2"59
2"9,1
2'45
3"02
5"OI
5"~
4"51
4.49
4"79
5"15
4.75
5"22
2.40
2"60
2'27
2" 16
2"48
2"63
2'52
2" 55
2'61
3"04
2"24
2"32
2'31
252
2"23
2' 67
5840
6020
4490
5650 ,
6690 J
2219
6544 |
6402 ?~
6348
6154 :
185,000
Giggenbach (19781
Giggenbach (19781
France
Massif Central
La Bourboule
"Croizal
Fe'lix
Gabriel
St. Nectaire
Royal
Peliljean
St. Laurent
Bagnols
La Chaldetle
Vichy
P.vrenees
Les Aigues
Amelie
Ax les Thermes
Canaveilles
Carcanieres
Dorres
Escouloubre
Fontchaude
La Prest
Lto
Naussa
St, Thomas
Thues
Usson
Vernet
Corswa
Baracci
Caldane
Caldanella
Caldaniccia
Guagno
Guitcra
Pietrapola
U rbalaeone
z757
31o6 !
~5~ /
15o5 /
17~o k,,.
873
(FouiUac
(unpub,)
1349 /
16.3[
237 1
15.6..)
326
Michard e / a l . {}976)
11.2 ~
22
12
7'1
4'5
53
4"4
3"6 "Labora o re de Geochimie des Eaux (19.78a1
1"8
5'0
14'2
7"8
71
4"5
7"B
33
49
l~0~.7
14-21
28
35 _ )
58
Locality
Wesl Indies
Bouilllmte
Well BO2
Well BO3
Paris Basin
Melun
Well P M 2 A
Well PM2B
Well P M 2 C
Challly
Well C H I
Well C H 2 6
Well CH28
Well C H I01
Iceland
Geysir area
A Geyser
Reykjenes
Hver~rerdi
Well G3
india
Puga
Well G W 2
Well G W 5
Well G W 7
Well G W 8
Well G W I0
Well G W I I
Indonesia
K a w ~ Kamojan8
Well 6
Italy
Cesano
Wel| I
Phlegraean Fields
Spring 5
T a b l e 1.
(continued)
I/ T
- log Li
- log Na
log N a / L i
CI- p p m
m
m
1-94
1'94
3"20
3'20
0'63
0"61
2"58
2'59
m
m
m
2"83
2"83
2"83
3"69
3"69
3"69
0"74
0'76
0'74
2'94
2"92
2'94
m
m
m
m
2"89
2"87
2"87
2"89
3"74
3"74
3"74
3"74
0"87
0"83
0"83
0"87
2'87
2"91
2'91
2-87
1'89
1'87
4'53
2"97
2-04
0'20
2-49
2"77
129
t9.500
1"90
4'36
2'03
2-32
260
e
e
e
e
1'84
1"84
1"84
I "96
1-96
1"96
3"02
3-09
3"09
3"06
3'02
3"03
1"51
1'65
1'60
I "60
I '60
1-61
1"50
1,44
1-49
1'46
I "42
1"42
475
411 ~
2"01
3"97
2'36
1'60
2'0
13,350
12,530
7170 "~
7171 l
7170 L
4790 ( s54o /
References
Boulc~ue (1978)
5570 ]
4790 J
Bjornsson ez aL (1972)
Ellis (1970)
K a r l o k u s u m o el al. (1976)
1"91
1"25
-0"53
l'79
42.850
Calamai et aL (1976)
2'26
3"36
0'40
2"90
14,620
e
e
e
2"14
2'07
2'05
2'11
3"18
3'20
3'13
3"08
1"43
l'45
I'39
1-32
1"75
1"75
1"74
1"76
1219~
HalChObaru
Well H I
Well H2
e
e
I '84
1"89
2'79
2'99
l'21
1'28
t "58
I "71
2327
2300
Hakone
Well II
Well Ill
Well IVa
m
m
m
3"02
2-74
295
5"00
3"45
4"40
2'41
1'18
1 "71
2"59
2'27
2-70
2570 ~"
617 P
Naruyo area
Toscnro
Farmer's h.
Higashitaya
m
m
m
2'68
2"85
2"68
4-13
4"49
4"07
1"55
l'65
1"60
2'57
2"84
2'46
~o~
Mexico
Cerro Prieto
Well'5"
Well M 31
m
m
I "63
l "71
2"56
1"95
0"56
-O'12
I "96
2"07
10.420].
14,170J
New Zealand
Beoadlcnds
Well 1
Well 2
Well 3
Well 4
Well 5
Well 6
Well 7
Well 8
Well 9
Well 10
Well 11
Well 12
Well 13
Well 14
Well 15
Well 16
Well 17
Well 18
Well 19
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
1.81
1-78
1 '80
1"83
1"93
2"40
1"81
1 "83
1"76
1-81
1 "82
1 "82
1 "82
1"75
1"75
I"82
1.77
1.78
1'83
2'80
2"77
2"75
2" 76
3'14
3"75
2'66
2'78
2'73
2-86
2"75
2"83
2'81
2"81
2"78
3"04
2"82
2.79
2.77
1'33
1"33
1"34
I" 33
1"21
17 2
1"25
1-37
1"39
1"40
I '35
I" 18
1"37
t'42
1-33
I '60
1.36
['32
l'38
1"48
1.44
I "42
1.43
1"93
2'04
1"42
I '41
1"34
1"46
I "40
I "65
1-44
1"40
1"44
I "44
1.47
1.47
1"39
1700 -,~
1740
1800
1850
1140
28
1823
1858
1709
1245
1795
1546
1670
1480
1750
528
t780
1985
1720 j
Ka~'rau
Wdl 7A
Well 8
Well 4
Well 10
Well 12
Well 14
m
m
m
m
m
m
1-85
t'85
2"11
1"91
1'84
1 "82
3" I I
3" 10
3'29
3"17
3-09
3,04
1"49
1 '49
1"73
1"55
1"56
1,48
1"62
l '61
1"54
1-62
1'52
1"56
1400"~
| 310 /
585 ~.
1098~
1003]
1136..J
Japan
Otake
Well 7
Well 8
Well 9
Well I0
1243 t~
1474|
1753J
20 l
125 ~
630 )
Koga (1970)
Ellis (1970)
Mercado (1976)
Ngawha
Well I
Orakeikorako
Well 2
Rotokaua
Well I
It~tiotapu
Well 6
21irakei
Well 471
Well 20
Well 44
Well 24
Well 48
Well 5
U.S.A.
A rizona
Casagrande
Expl. Well
Cal(fornia
Sahon Sea
Well 2D
Well I State
I T
- 1o8 Li
- Io 8 Na
lt~g N a / L i
1"98
I "81]
2"82
1'43
1.40
1658
3'34
1"62
1"72
546
2'02
2"63
1"18
1'65
1"85
3'02
1'42
1"59
m
m
m
m
e
e
2"02
1.95
I "89
1-91
1'88
2"26
2' 74
2-70
2"69
2"71
2"72
3,77
1 25
1'25
124
1'26
1"28
2-00
I "49
1"45
I -~4
1"45
1"43
1'76
2'68
3"68
I "49
2" 19
1'63
t'70
1"51
1'57
-036
-0"31
1"88
1"89
('1 ppm
2140/"
2260J
2230
2156
546
Ellis 119701
Koga (1970)
Shanklr et al. (19761
1250
Dellechaie 119761
150,000 I.
127,000 /
3'40
3.40
3"68
3"62
1'74
1.77
1"87
1"76
1'63
1"61
I.'79
I "85
3S 29E34KSI
2" 14
3.64
I'85
1"92
3S 2E25AS,l
2'23
3.48
1.76
I "69
225_J
2'68
3'68
3'44
19,500 -.,.~
18,2oo t
Tb III(81
m
m
2"68
2'68
3.57
3.67
2'68
3'73
0"47
3"26
Tb V(1)
2'68
3"79
0'52
3'28
UM (5)
LM (41
m
m
2'68
2'68
3.60
3.38
I'01
0'80
2"59
2-57
Montana
Alhambra
Boulder
3"24
3"22
221.~/
2' 36
2'36
2"33
2' 11
T b 11(8)
Ellis (1966)
1450-~
e
e
~,
e
Tb IV( t )
Ellis (19701
Ellis { 1970)
Long Valley
3S 28E 13ES3
3S 28E32E9
3S 28E21PSI
3S 29E28HSI
0'24
0'31
0"45
Reterences
150t.
1701
Helgeson 11967}
1501
16,ooo!
12,600 ~" Merino(1975bl
16,6oot
3380)
4720~
e
e
2"58
2 '43
4'26
4 "42
2"03
2"27
2"23
2" 15
Broadwater
2-45
4.09
2"03
2-06
Grcgson
Norris
Silver star
e
e
e
2" 50
2"49
2"42
4"00
4"85
4'26
2" I 0
2"02
2" 12
1-89
2"79
2' 14
2"48
2"42
0"45
- 2"82
165,000
2"83
2"83
2'85
2'95
-0"46
- 0"49
3"31
3'44
143,500"[
158.000I
Lebedev 0 9 7 3 )
m
m
m
m
m
3"38
3" 14
3"29
3"39
3"30
4"43
3'83
3.95
4.5
4-02
0"24
- 0"08
- 0"05
0"33
- 0'08
4'19
3"91
4'00
4" 17
4. I0
155,000 "~
155,000
155,000
42,000
42,000
726 9
727 9
727 23
726 23
m
m
m
m
3-37
3"38
3'03
3"03
4.42
4-48
3-21
3.17
0"24
0"26
- 0"69
- 0'68
4"18
4"22
3-90
3"85
Station 722 I 9
3'32
4.27
0-18
4-09
722 2110 - 5) m
722 2 23
m
715 10
m
715 13
m
715 23
m
718 14
m
718 I
m
3"38
3"15
3"38
3'29
3"03
3"39
315
4"45
3'65
4"52
3"58
3"16
4'49
3"58
0"23
-0"33
0"15
- 0'36
- 0"71
0"24
- 0"34
4"22
3'98
4"37
3.94
3"87
4"25
3"92
80,000
80,000
80,000
80,000
80,000
80,000
80,000
80,000
80,000
80,000
80,000
80,000 J
Mississipi
Homer Currie
U.S.S.R.
Cheleken Brine
Well PI52
Well U I
Red Sea
Discovery
Station 717 1 - 2
Station 728 I
Station 728 3 - 13
Station 729 20
Station 729 23
Atlantis
Station
Slation
Station
Station
Station
Station
Station
Station
Station
Station
Station
8"5~
24
--~'5
27
15 _9
Brooks
et al.
(19691
Data for surface waters are given in Table 2. Values derived from the experimental work of
Ellis and Mahon (]967) are in Table 3.
From Tables ] - 3, log mNa/mLi and log m/L~ are plotted vs l / T (Figs. ] and 2). The N a / L i
plot exhibits two linear relationships. We can see that the first line (curve a) corresponds to
60
C. Fouillac a n d G. M i c h a r d
Table 2. Lithium concentrations and sodium/lithium ratios in surface waters of various origins.
Number of
sample
I / T x 10 '
Average
Sandstones
Lauraguals
S . W . France
i9
3"50
- 6"23
Limestones
S. A I p s
France
13
Granite
Margeride
Massif Central
France
15
D e s c r i p t i o n of s a m p l e s
Log Li
Range
Average
Log N a / L i
Range
Relerencc~
2.52
3"48
- 7
3'50
3'13
- 6
3"58
- 7
Volcanic Rocks
Guadeloupe
West Indies
2-99
- 6"85
3'40
12
- 6"7
3"02
- 7
2.77
- 6" 17
Quartzite
S a n g r e de C r i s t o
Range New Mexico
U.S.A.
Sandstones
S~mgre d e C r i s t o
Range New C r i s t o
U.S.A.
3'86
2"3
- 6"79
3-08
- 6"36
2"38
- 7.20
2"63
- 7.13
M i l l e r ( 196l )
2.77
- 6"79
2'29
-6"78
3'50
M i l l e r (196D
3"78.
- 7.23
3'58
F o u i l l a c and
Ouzounian
1977.
3" 3 I
- 7"36
3"55
Beaucaire, lundt
1979.
3 "0
- 5-7
Granite
S a n g r e de C r i s t o
Range New Mexico
U.S.A.
G r i m a u d , D., et el.,
unpublished
values
2"59
-6"33
3 "60
Boulesue, J ,
unpublished
values
3" ! 8
-- 5"8
2"55
- 6"53
Miller (1961)
2"62
3'55
- 6"29
Susquehana River
U,S.A.
3"55
- 6.49
3'50
- 5-96
Japanese
Rivers
14
Major North
American Water
2"82
2"82
2'83
- 6"63
- 5"38
3"01
Livingstone
(1963)
3"47
2"34
2.66
3"55
Sea W a t e r
- 5"48
3-62
4'29
Table 3. Sodium/lithium behaviour in experimental water- rock interactions. After Ellis and Mahon (1967)
Rock type
I / T x ~0"+
Pumice
Ol~idian
Ignimbtite
Rhyolite
Dacite
Andesite
Basalt
Greywackes
1-48
1"91
1'48
1.60
1"91
I "91
1,91
I "91
L o s Li
-
3'58
4' 18
3"79
3"76
4.02
4-14
4.36
4.14
L o g Na
-
2.22
2.40
2-49
2.49
2"55
2'55
2" I 8
2'52
Los Na/Li
CI- ( p p m )
I. 35
1"59
I ' 29
1"27
1-47
I" 59
2-17
I "62
157
9
32
I l0
40
52
250
3
waters where chlorinity is less than 0.2 M) (about 7000 ppm); the second line corresponds to
high chlorinity ( > 10,000 ppm or 0"3 M) waters.
Thus, the relationships between N a / L i and temperature can be summarized by
(i) log N a / L i = l O 0 0 / T - 0"38 for C l - < 0"2 M,
the mean square correlation coefficient is r = 0.965. This line will be called the general line in
the following sections.
(ii) log N a / L i = 1195/T + 0.19 for C l - > 0-3 M (r = 0"982).
The correlation between log
log mLi
--
mLi
-2
-7
'
ii{
:-::
I/T
I
,t
300
~i::~..
x I0 -3
200
I00
25
t*C
- - \I,
4"
300 200
I00
25
t C
Fig. l.
Fig. 2.
Fig. 1. Log N a / L i in waters vs the inverse of absolute temperature. Key: (a) general line, closed symbols; (b) brine line,
open symbols. O , Measured temperatures. Estimated temperatures. + Surface waters. Experimental
water - rock interactions.
Fig. 2. Log Li in waters vs the inverse of absolute temperature. Same symbols as in Fig. I. Darker shading is for saline
waters, lighter shading for low salinity waters.
If we separate the data into the two chlorinity groups, we obtain two well-defined linear
relationships:
log mL, = - 2258/T + 1.44 for C l - < 0 . 2 M (r = 0"936),
log mL, = - 1436/T + 0.61 for C l - > 0 . 3 M (r = 0-91).
The selection of data needs some comments:
For Red Sea thermal brines, we have considered temperatures and chemical analyses quoted
in Brooks et al. (1969). More recently, several workers (Brewer et al., 1971 ; Ross, 1972, Schoell,
1976) have reported higher brine temperatures, but detailed chemical analyses are not yet
available. Neither do we consider the high temperature estimates for the input brine derived
from geophysical models (Scholl, 1976). It is likely, after Shanks and Bischoff (1977) that the
chemical composition of brine must reflect equilibrium at a measured temperature, due to reequilibration during the time of water residence in low thermality layers.
All the chemical analyses of Central Mississipi brines (Carpenter et al., 1974) are not
considered in this compilation as their temperatures assume that all drillholes had the same
temperature profile as H o m e r Curie No. 4 (original sample Ref. 57).
Waters from Yellowstone Park are not used in examining the thermometric relations,
because of uncertainties in the aquifer temperature. The problem of this area is dealt with later.
62
TENTATIVE INTERPRETATION
The different qualitative interpretation for the temperature dependence of the Na/Li ratio
(and Li molality) suggested in the following pages indicate the need for further work to derive
thermometric relationships.
(1) To help this discussion, the log mNa vs l / T p l o t is given in Fig. 3. The sodium behaviour
in hydrothermal systems is generally influenced by the deposition of sodium-rich minerals
(albite, sodium-rich plagioclase or analcime). Albite is a very common secondary mineral in
geothermal systems (Steiner, 1953; Sigvaldason, 1962; Muffler and White, 1969; Browne and
Ellis, 1970; Hayashi and Yamasaki, 1974; Oki et al., 1974; Merino, 19.75a). In Iceland,
analcime has also been found in many drillholes in geothermal fields (Kristmannsdottir and
Tomasson, 1974; Kristmannsdottir, 1976; Tomasson and Kristmannsdottir, 1972). For dilute
waters, we observe an increase of Na with T, whereas the correlation between CI and Tis poor.
This can be explained at high temperature, by the increase of albite solubility; and chemical
composition, from one field to another, is similar enough to preserve the influence of
temperature on sodium concentration. At low temperature, equilibrium is not achieved and the
sodium concentration is still less than the equilibrium value.
1line
A.
tl
-2
~A
A
waters
~:
A
a
-3
3
I/T
=10 - 3
Fig. 3. Log Na in waters vs the inverse o f absolute temperature. Same symbols as in Fig. 1.
The case of marine waters is different: Na is generally at least three orders of magnitude
greater than silica; thus, any silicate deposition cannot rule the sodium concentration.
(2) The dependence of Li on temperature is stronger than the Na dependence: about 4
orders of magnitude between 0 and 300C for Li instead of about 2 orders for Na.
64
oETB
Z 3
8'
-3
.f..<
I
I/Tx I0"~'
Fig. 4. Behavior of some particular hot waters. Key: KCS--Kettelman interstitial solutions; concentrated samples.
KDS--Ketteimim interstitial solutions, diluted samples. CB.--Cesano Brine. ETB--E! Tatio Brine. WBlS--Water
Basalt in interaction resulting solution. HG--Hveragerdi and Geysir respectively.
other reasons originate in the deep aquifer; they are more critical and will be discussed for the
N a / L i thermometer (see Fig. 4).
Three dilute waters fall far outside the general line: two o f them are Icelandic waters
(Hveragerdi hole G3 and a geyser); in this case the rock-type can have an effect; the third one is
the run with basalt as starting material in Ellis and Mahon as experiments. As a matter o f fact,
the general line describes the behaviour o f waters for which marine originated constituents are
negligible compared to total solutes. In Iceland, the rainwater contribution to the chemistry of
dilute hot waters cannot be neglected as shown by Arnason and Tomasson (1970). This input
o f solutes induces a shift from the general line towards the brine line. A slight shift of this kind
has been observed in Corsica (Ouzounian et al., 1979). In Ellis and Mahon's experiments,
which are not equilibrium experiments, no development o f a secondary sodium phase has been
observed in the run with basalt: the increase o f Na is not limited and the N a / L i ratio remains
high.
Two hot brines do not follow the brine trend: Cesano and El Tatio brines. At Cesano,
Calamai et al. (1976) have noticed that quite large amounts of sodium chloride precipitated in
the well, which produces a mixture o f water and solid NaCI and Na2SO,. Furthermore, thermal
equilibrium in the well was not achieved when recording stopped and the temperature value was
given as a lower estimate. We could presume that the analysed fluid did not represent the deep
aquifer chemistry. We cannot conclude whether Cesano brine is apart from the brine line or
not. The problem o f El Tatio is more critical. It is possible to take this solution to be from a
different origin. Note that Mahon and MacDowell (1977) after Cusicanqui et al. (1976)do not
refuse the idea that acid, volcanic, Cl-rich emanation could have produced E1 Tatio brine by
reaction with ignimbrite. As Giggenbach (1978) has not reported isotopic data for E1 Tatio
brine, the problem is still unsolved.
Spring
TC
Na/K
Na/Ca
Na/Li
LI I~Y' (m/I)
SiO/(ppm)
F 11
F 12
F 32
F 3
F 28
F 26
F 25
Feerique
Jade
Mesdames
Genereuse
Hopital
Grande
] l'5
12
13"5
17
24
34
40-2
21-I
25'9
23'5
28"2
32"9
3 I" I
33"3
30"5
35"9
18"2
30"5
80"2
20-8
31"7
98-2
100*7
95"2
104'2
I01 "5
102.8
101.25
0"69
0-79
0"71
0"76
0-79
0-79
0"80
2l'0
25"2
22'8
17'4
95"0
54'6
70"2
F 33
F 34
Boussang
Dome
41
66
23"5
29.7
24.9
64"0
97'8
101-8
0"82
0.74
84-0
83"0
Saint Ange
Grille
Table 5. Comparison between Na - Li geothermometer and classical geothermometers, Taupo volcanic zone -- New
Zealand.
Molal ratios
Li
110"m/i)
Thermal areas
Na/K
Na/Ca
Temperature
Na/Li
TNa/K
TN - K - Ca
TNa/Li
TNa/L i
Obs
1_~/3L
= Hole 44
Wairakei
~i Spring 97
Spr og 190
[+ Hole 6
Waiotapu
Kawerau
Orakeikorako
Taupo
Broadlands
", Spring 64
I. Spring 20
Hole 7A
+, Spring 2
~=Spring 4
J Hole 3
[ Spring 98
J Hole I
/ Spring
Hole 10
Spring i
i Hole I
Ngawha
2.06
0,98
1"45
0.96
1'30
0"58
0-77
0"39
0-48
0-45
0'58
2"0
0.67
1"39
1.07
1"55
10-0
16"6
26
9-4
13'0
35
10'3
11'5
12"8
10"7
11.3
9.7
14.7
10-1
17-8
19'6
135
26
83
150
60
87
500
44
53
440
190
16-0
64
737
575
53
2%9
29"3
28"5
39
40'6
34
41'8
36"7
36'2
24-5
21
27.7
26"4
28"2
34
25"2
261
199
156
269
227
131
257
242
229
252
24,*
265
217
259
192
182
265
205
193
266
236
171
238
226
223
254
242
269
217
281
236
205
275
268
272
234
229
250
226
241
242
292
314
275
282
273
250
288
274
234
254
233
248
209
222
192
201
198
209
272
216
252
239
258
1"15
22"0
185
31
171
240
261
242
- Jubilee
' Spring
270
260
260
250
267*
270
220
66
Orakeikorako, Taupo, Broadlands) underestimates should have been caused by mixing with
cold waters. Na/Ca is very different in springs and in drillholes. Na/K values in springs provide
good estimates in some places (Orakeikorako, Kawerau) but not in others (Wairakei,
Broadlands). Furthermore, Na/K ratios are different among springs of a given area. This can
be seen in more detail in the spring chemistry of Broadlands area (Table 6, after Mahon and
Finlayson, 1975). Here the Li thermometer gives too low temperatures. This is related to mixing
with cold waters and the more concentrated spring gives the best value. Mixing or steam loss
can therefore be pointed out by comparison of the temperatures given by the Li and Na/Li
geothermometers.
Table 6. Temperature estimates from spring chemistry in Broadlands area. After Mahon and Finlayson (1975).
Maximum temperature measured: N 270oc
SprinB n u m b e r
TNa - K
I
3
4
5
6
7
A
B
189
192
164
165
136
117
163
170
Direct
/'Na
K -
( 1/3)
223
197
182
185
178
167
177
188
Ca
TNa - Li
TSiO,*
TLi
247
269
243
223
240
237
248
237
219
204
203
183
196
191
185
189
239
216
205
191
205
209
205
201
The situation is more complex in this area. As far as we know, there are no definite physical
temperature measurements in the deep aquifer. Only shallow drillholes (max. depth: 135 m)
have been used for physical records, and in some cases the wells have not reached thermal
equilibrium (White et al., 1975). Besides this, numerous careful analyses of natural springs and
geysers are available (Rowe et al., 1973). Several geochemical studies have provided a good
knowledge of possible deep temperatures in the area: Fournier et al. (1976), Truesdell and
Fournier (1976).
From these studies the following situation can be summarized:
(i) the existence of shallow aquifers, different from each other in temperature and chemical
composition, is certain;
(ii) a unique parent water can be expected at great depths; its temperature must be very
close to 340C;
(iii) individual basins in the area (Upper, Lower, Norris Basins, Mammoth area) could be
distinguished as aquifers feeding springs and geysers;
(iv) enthalpy chloride models developed in the two previously cited papers show that Norris
Basin waters are quite directly related to the inferred parent water. Norris Basin is the hottest
zone in Yellowstone Park and the chemistry of waters from this basin is rather homogeneous;
(v) Upper and Lower Basins are fed by several water types. These aquifers could result from
mixing of the parent water with shallower water tables. Chemical re-equilibrations can occur
after mixing. Classical thermometers yield temperatures lower than in Norris Basin;
(vi) springs of Mammoth area can be derived from Norris aquifer by mixing with dilute
waters and reactions between solution and carbonate host rock play an important role.
Using chemical analyses listed in Rowe et al. (1973), we compute the Li and Na/Li
temperatures. Acid sulfate springs are not considered. The Li thermometer always gives
underestimates of deep temperatures. This is not surprising considering the numerous mixings.
A histogram of the results given by the Na/Li geothermometer is shown on Fig. 5. This graph
reinforces the theory that:
Norris Basin waters are related to a common aquifer, the temperature of which is 300C;
-
Numbel
of
Sampte~,
Mammoth
Area
L o w e r Basin
Upper
200
220
Basra
240
260
21G
300
320
340
3116
t'C
Fig. 5. Histogram of frequencies for N a - Li computed temperature of Yellowstone Park thermal manifestations.
Let us finally note that the Na/Li derived temperatures are in good agreement with those
determined by Fournier et al. (1976) and Truesdell and Fournier (1976) using a complex model.
CONCLUSIONS
The study of lithium concentrations and sodium lithium ratios in waters has suggested a new
geothermometer. The relationships between temperature and Na/Li ratios for most of the
explored geothermal fields are correctly described by two linear functions. Chloride contents of
the waters provide a very simple way to choose the appropriate equation.
Obviously the Na/Li geothermometer should not be used alone, but comparison with other
geothermometers and also lithium concentrations will be very useful for geochemical surveys.
An interesting point is the remarkable constancy of the Na/Li ratio from aquifer to surface
manifestations.
The theoretical basis for describing the Na/Li ratios by two linear relationships of the
reciprocal of the temperature are not yet understood. Careful geochemical studies of well
known geothermal areas and experimental work are still necessary for a better understanding of
sodium and lithium behaviour under hydrothermal environments.
REFERENCES
Armand, L. (1933) Sur quelques sources minerales du plateau du Massif Central francais. Revue de I'lndustrie
Min#rale 300, 270 - 278.
Arnason, B. and Tomasson, J. (1970) Deuterium and chloride in geothermal studies in Iceland. Geothermics Special
Issue 2, 2, pt. 2, 1405- 1415.
68
Arnorsson, S. (1970) Underground temperatures in hydrothermal areas in Iceland as deduced from the silica content of
the thermal water. Geothermics Special Issue 2, 2, pt. 1, 536- 541.
Arnorsson, S. (1973) Application of the silica geothermometer in low temperature hydrothermal areas in Iceland. Am.
J. Sci. 275, 763- 784.
Baldi, P., Ferrara, G. C. and Panichi, C. (1976) Geothermal research in Western Campania (Southern Italy): chemical
and isotopic studies of thermal fluids in the Campi Flegrei. Proc. Second United Nations Symposium on
Development and Use o f Geothermal Resources, !, 687 - 697
Bargar, K. E., Besson, M. H., Fournier, R. O. and Muffler, L. J. P. (1973) Present day deposition of lepidolites from
thermal waters in Yellowstone National Park. Am. Mineralogist 58, 901 -904.
Beaucaire, C. and lundt, F. (1979) Etude des el(~ments traces dans quelques sources de terrains granitiques et
volcaniques du Massif Central Francais. In preparation.
Bjornsson, S., Arnorsson, S. and Tom'asson, S. (1972) Economic evaluation of the Reykjanes thermal brine area,
Iceland. Bull. Am. Assoc. Petrol. Geologists 56, 2380- 2391.
Boul~gue, J., Fouillac, C., Michard, G. and Ouzounian, G. (1977) Geothermal applications of the geochemical study
of thermal springs in Pyr~'n~es. Proc. Seminar on Geothermal Energy Brussels 2, 455 -468.
Boul~gue, J. (1978) Metastable sulfur species and trace metals (Mn, Fe, Cu, Zn, Cd, Pb) in hot brines from the French
Dogger. Am. J. Sci. 278, 1394-1411.
Brewer, P. G., Wilson, T. R. S., Murray, J. W., Munns, R. G. and Densmorre, C. D. (1971) Hydrographic
observations on the Red Sea brines indicate a marked increase in temperature. Nature 231, 37 - 38.
Brooks, R. R., Kaplan, I. R. and Peterson, M. N. A. (1969) Trace element composition of Red Sea geothermal brine
and interstitial water. In: Hot Brines and Recent Heavy Metal Deposits with Red Sea. (Eds. Degens, E. T.,
Ross, D. A.), pp. 180-205.
Browne, P. R. L. (1970) Hydrothermal alteration as an aid in investigating geothermal fields. Geothermics Special
Issue 2, 2, pt. 1,564- 570.
Browne, P. R. L. and Ellis, A. J. 0970) The Ohaki-Broadlands hydrothermal area, New Zealand: mineralogy and
related geochemist~. Am. J. Sci. 269, 9 7 - 131.
Calamai, A., Cataldi, R., Dail'Aglio, M. and Ferrara, G. C. (1976) Preliminary report on the Cesano hot brine
deposit (Northern Latium, Italy) Proc. Second United Nations Symposium on the Development and Use o f
Geothermal Resources 1, 305 - 313.
Carpenter, A. B., Trout, M. L. and Pickett, E. (1974) Preliminary report on the origin and chemical evolution of lead
and zinc rich oil field brines in Central Mississipi. Economic Geology 69, 1191 - 1205.
Cusicanqui, H., Mahon, W. A. J. and Ellis, A. J. (2976) The geochemistry of the E1 Tatio geothermal field, Northern
Chile'. Proc. Second United Nations Symposium on the Development and Use o f Geothermal Resources l,
703 - 71 I.
Dellechaie, F. (1976) A hydrochemicat study of the south Santa Cruz Basin near Coolidge, Arizona. Proc. Second
United Nations Symposium on the Development and Uses o f Geothermal Resources I, 339- 349.
Ellis, A. J. 0966) Volcanic hydrothermal areas and the interpretation of thermal water compositions. Bulletin
volcanologique 29, 575 - 584.
Ellis, A. J. (1970) Quantitative interpretation of chemical characteristics of hydrothermal systems. Geothermics
Special Issue 2, 2, pt. l, 516-528.
Ellis, A. J. and Mahon, W. A. J. 0964) Natural hydrothermal systems and experimental hot water - rock interactions
(Part 1). Geochim. Cosmochim. Acta 21, 1323 - 1340.
Ellis, A. J. and Mahon, W. A. J. 0967) Natural hydrothermal systems and experimental hot water - rock interactions
(Part 2). Geochim. Cosmochim. Acta 28 A, 519-538.
Ellis, A. J. and Wilson, S. H. (1960) The geochemistry of alkali metal ions in the Wairakei hydrothermal system.
New Zealand ,/. Sci. 3, 593-617.
Fouillac, C. and Michard, G. 0977) Sodium potassium calcium relationships in hot springs of Massif Central.
Proc. Second Int. Symposium on Water- Rock Interaction, Strasbourg, 3, p. 109- 116.
Fouillac, C. and Michard, G. (1979) Un geothermom~tre empirique: le rapport Na/Li des eaux. C. R. Acad. Sc. Paris
288, 123- 126.
Fouillac, C. and Michard, G. (1979) G~ochimie des eaux thermales des Pyrenees Oriemales et de la Corse. To be
published.
Fouillac, C., Cailleaux, P., Michard, G. and Merlivat, L. (1976a) Preliminary geothermic studies on mineral water
in French Massif Central. Proc. Second United Nations Symposium on the Development and Use o f Geothermal
Resources l, 726 - 729.
Fouillac, C., Michard, G., Merlivat, L., Javoy, M. and Jouzel, J. (1976b) Etude gtSachimique des eaux
thermomin~rales de Chateauneuf-Les-Bains. J. Francais d'Hydrologie 7, 151 - 257.
Fouillac, C. and Ouzounian, G. (1977) G6ochimie des eaux thermales de la Guadeloupe. Institut National
d'Astronomie et de G(ophysique. Unpublished Report.
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 Truesdell, A. H. (1973) An empirical Na-K-Ca geothermometer for natural waters. Geochim.
Cosmochim. Acta 37, 1255- 1275.
Fournier, R. O. and Truesdeli, A. H. 0974) Geochemical indicators of subsurface temperature. Part 2: estimation of
temperature and fraction of hot water mixed with cold water. U. S. Geol. Survey J. Research 2, 263 - 269.
Oki, Y., Hirano, T. and Suzuki, T. (1974) Hydrothermal metamorphism and vein minerals of the Yugawara
geothermal area, Japan. Proc. Int. Symposium on Water-Rock Interaction, Prague 1974, pp. 209-222.
70
Ouzounian, G., Michard, G., Fouillac, C. and Beaucaire, C. (1979) Relations entre les concentrations des ions
alcalins dans les eaux thermales sulfur6es sodiques du Midi de la France. Chemical Geology (Submitted).
PaCes, T. (1975) A systematic deviation from N a - K - C a geothermometer below 75C and above l0 -+ atm PCO,
Geochim. Cosmochim. Acta 39, 541 - 544.
Pa~es, T. and Cermak, V. (1976)Subsurface temperature in the Bohemian Massif: geophysical measurements and
geochemical estimates. Proc. .Second United Nations Symposium on the Development and Use o f Geothermal
Resources, 1,803 - 807.
Reed, M. J. (1976) Geology and hydrothermal metamorphism in the Cerro Prieto geothermal field, Mexico. Proc+
Second United Nations Symposium on the Development and Use o f Geothermal Resources 1, 539 - 547.
Robertson, E. C., Fournier, R. O. and Strong, C. P. 0976) Hydrothermal activity in southwestern Montana. Proc.
Second United Nations Symposium on Development and Use o f Geothermal Resources 1, 553 - 561.
Ross, D. A. (1972) Red Sea hot brine area--revisited. Science 175, 1455- 1456+
Rowe, J. J., Fournier, R. O. and Morey, G. W. (1973) Chemical analysis of thermal waters in Yellowstone National
Park, Wyoming, 1960-65. U.S. Geol. Survey Bull. 1303 31 pp.
Schoell, H. (1976) Heating and convection within the Atlantis It deep geothermal system of the Red Sea. Proc.
Second United Nations Symposium on the Development and Use o f Geothermal Resources l, 583 - 590.
Shanker, R., Padhi, R. N., Arora, C. L., Prakash, G., Thussu, J. L. and Dua, K. J. S. (1976) Geothermal
exploration of the Puga and Chumathang geothermal fields, Ladakh, India. Proc. Second United Nations
Symposium on the Development and Use o f Geothermal Resources I, 245 - 258.
Shanks, Ill W. C. and Bischoff, J. L. (1977) Ore transport and deposition in the Red Sea geothermal system: a
geochemical model. Geochim. Cosmochim. Acta 41, 1507 - 1520.
Shikazono, N. (1976) Thermodynamic interpretation of N a - K - Ca geothermometer in the natural water system.
Geochem. J. 10, 47 - 50.
Sigvaidason, G. E. 0962) Epidote and related minerals in two deep geothermal drillholes, Reykjavik and Hveragerdi,
Iceland. U.S. Geol. Survey Prof. Paper E152, 258- 261.
Steiner, A. (1953) Hydrothermal rock interaction at Wairakei, New Zealand. Econ. Geology 48, 1 - 13.
Tomasson, J. and Kristmannsdottir, H. (1972) High temperature alteration minerals and thermal brines, Reykjanes,
Iceland. Contr. Miner. Petrol. 36, 123 - 134+
Tonani, F. (1970) Geochemical methods of explorations for geothermal energy. Geothermics Special Issue 2, 2, pt. 1,
492-515.
Truesdell, A. H. and Fournier, R. O. (1976) Calculation of deep temperatures in geothermal systems from the
chemistry of boiling spring waters of mixed origin. Proc. Second United .Nations Symposium on the Development
and Use o f Geothermal Resources 1 , 8 3 7 - 844.
Volfinger, H. (1976) Effet de la temperature sur les distributions de Na, Rb, et Cs entre la sanidine, la muscovite, la
phlogopite et une solution hydrothermale sous une pression de I Kbar. Geochim. Cosmochim. Acta 40, 2 6 7 - 282.
Weissberg, B. G. and Wilson, P. T+ (1977) Montmorillonite and the Na/K geothermometer in "Geochemist~ 77".
New Zealand Dept. o f Scientific and Industrial Research Bull. 218, 31 - 35.
White, D. E. (1957) Thermal waters of volcanic origin. Bull. o f the Geological Society o f Amerwa 68, 1637- 1658.
White, D. E. (1965) Saline water of sedimentary rocks in fluids in subsurfaces environment - - a symposium. Am.
Assoc. Petroleum Geologists Mere. 4, 342 - 366.
White, D. E. (1968) Environment of generation of some base metal ore deposits Bull. Soc. Econ. Geologists 63,
301 -335.
White, D. E., Fournier, R. O., Muffler, L. J. P. and Truesdell, A. H. (1975) Physical results of research dritting in
thermal areas of Yellowstone National Park, Wyoming. U.S. Geol. Survey Pro]++Paper 892, 70 pp.