The origin of reservoir fluids in the geothermal field of Los Azufres, Mexico — isotopical and hydrological indications

Applied Geochemistry 16 (2001) 1595–1610 www.elsevier.com/locate/apgeochem

The origin of reservoir fluids in the geothermal field of Los Azufres, Mexico — isotopical and hydrological indications Peter Birkle a,*, Broder Merkel b, Enrique Portugal a, Ignacio S. Torres-Alvarado c a Instituto de Investigaciones Ele´ctricas, Unidad de Geotermia, AP 1-475, Cuernavaca, Mor., 62001 Mexico Technical University of Freiberg, Institute of Geology, Gustav-Zeuner-Str. 12, 09596 Freiberg/Saxony, Germany c Centro de Investigacio´n en Energı´a, Universidad Nacional Auto´noma de Me´xico, AP 34, Temixco, Mor., 62580 Mexico b

Received 30 November 1999; accepted 30 December 2000 Editorial handling by H. A´rmannsson

Abstract The calculation of hydrological balance resulted in a potential, average annual infiltration rate of 446
206 mm/m2 for the Los Azufres geothermal area, which corresponds to a total of 82106 m3 per a. Due to the highly fractured and faulted structure of the volcanic formations, a considerable potential for the infiltration of recent meteoric water into deeper sections of the volcanic formations can be assumed. Isotopic data indicate the minor importance of recent meteoric water for the recharge of the geothermal reservoir. Very negative
13C values can be explained by the input of organic C from the surface, but the lack of 14C in the deep fluids reflects a pre-historic age for the infiltration event of fossil meteoric water. The dilution of the meteoric water by 14C-free CO2 gas from a shallow magma chamber complicates the exact age determination of the infiltration event, which probably occurred during the Late Pleistocene or Early Holocene glacial period. Strong water–rock interaction processes, such as sericitization/chloritization, caused the primary brine composition to be camouflaged. A preliminary hydrological model of the reservoir can be postulated as follows: the fossil hydrodynamic system was characterized by the infiltration of meteoric water and mixing with andesitic and/or magmatic water. Strong water–rock interaction processes in the main part of the production zone prove the existence of former active fluid circulation systems. Due to changes in pressure and temperature, the rising fluids get separated into liquid and vapour phases at a depth of 1500 m. After cooling, the main portions of both phases remain within the convective reservoir cycle. Isotope analyses of hot spring waters indicate the direct communication of the reservoir with the surface at some local outcrops. A recent reactivation of the hydrodynamic system is caused by the geothermal production, as indicated by the detection of lateral communication between some production and reinjection wells. # 2001 Elsevier Science Ltd. All rights reserved.

1. Introduction Since the beginning of the 1980s, a mixture of geothermal fluid and vapour has been extracted from a depth between 350 and 2500 m in the Los Azufres geothermal field, located 220 km NW of Mexico City (Fig. 1). The origin of the fluids, the influence of surface recharge by meteoric water and/or regional aquifer * Corresponding author. Fax: +52-73-182526. E-mail address: [email protected] (P. Birkle).

systems, as well as the hydraulic conditions of the deep circulation systems are still unknown. These questions will be of importance for the prediction of the future potential of the geothermal production cycle. In the past, reservoir models mainly elaborated on exploratory aspects, applying thermodynamic parameters, such as P/ T-conditions (Iglesias and Arellano, 1985) or phase distributions (Nieva et al., 1983). Artificial tracer tests with KI (Iglesias, 1983) and 192Ir (Arago´n, 1985) were not able to detect communication systems between individual wells, although reinjected brines caused elevated

0883-2927/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0883-2927(01)00031-2

1596

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

Fig. 1. The Los Azufres geothermal field is located 220 km NW of the capital Mexico City. Its northern part discharges towards the hydrological basin of the Cuitzeo Lake, whereas its southern part (Tejamaniles) discharges towards the River Balsas basin.

CO2 and N2 concentrations in adjacent production wells (Horne and Puente, 1989). This paper tries to improve the understanding of the hydrodynamic processes in the Los Azufres geothermal reservoir, which comprise potential circulation systems and recharge processes, as well as details of the type and origin of the fluids. Stable isotope analysis of surface outcrops and reservoir brines give indications about the

existence of vertical communication systems, as well as about the magnitude of water–rock interaction processes. The use of radioactive isotopes and the determination of their residence time reveal the existence of recent or fossil hydrodynamic processes within the reservoir. Hydrological balance calculations for the study area indicate the potential of recent infiltration into the shallow and deep aquifer systems.

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

2. Methods Monthly and annual climatological data from 14 stations (one inside the field, 13 in the vicinity) from 1955 to 1979 (Cedillo et al., 1981) was used to calculate the hydrological balance for the study area, such as the average and total precipitation (P) and the potential evaporation rate (ETpot) (Birkle, 1998). The actual evapotranspiration (ETactual), which represents the sum of transpiration by plants and evaporated precipitation water, was determined by semi-empirical equations. The methods of Turc (Gray, 1973), Morton (Morton, 1965), Budyko (Kuzmin and Vershinin, 1974), and Coutagne (Remendieras, 1974) are based on climatological, atmospheric and vegetation data, such as temperature, relative humidity, latent vapour heat, solar radiation, albedo and potential evaporation. During the dry season from October to May, the volume of surface discharge was derived by regular hydrometric measurements with flowmeter equipment. Due to the lack of installed hydrometric stations in the study area, significant runoff fluctuations during the wet season were not measured directly in the field. Wet season discharges from the Los Azufres area were derived by the analogue comparison of the dry/wet season discharge proportions from the lower course, external stations. A methodical error of
10%,
25% and
3% is accepted for the determination of P, ETactual, and the runoff measurements, respectively. Stable isotope analysis (D, 18O) of geothermal brines, and cold and hot springs were performed at the Unidad de Geotermia at the Instituto de Investigaciones Ele´ctricas, Cuernavaca, Mexico and the Technical University of Freiberg/Saxony, Germany. Four hundred liters of geothermal brine, with an HCO 3 concentration of 143.3 mg/l, were extracted from well Az-43. The carbonates were precipitated with BaCl2 as BaCO3. CO2 from 10 gas samples from the wells Az-2, 4, 5, 6, 9, 22, 26, 28, 38, and 41 was concentrated in a vacuum vessel with a NaOH-saturated solution. The sampling preparation for the determination of 13C from CO2 was performed according to Giggenbach and Goguel (1989). Tritium, 13C and 14C-analysis on surface samples and geothermal brines were carried out at the Institute for Applied Physics at the Technical University of Freiberg in Saxony, Germany. The residence times of the isotopes were calculated with MULTIS (Richter et al., 1993).

3. Geological setting The volcanic formations of Los Azufres have been described by several authors (Dobson and Mahood, 1985; Cathelineau et al., 1987). Geologically, two principal divisions can be distinguished.

1597

1. A silicic sequence of rhyodacites, rhyolites and dacites with ages between 1.0 and 0.15 My and thickness up to 1000 m (Dobson and Mahood, 1985). This sequence serves as a caprock to the aquifer from the surface, allowing the geothermal system to pressurise. 2. A 2700 m thick interstratification of lava flows and pyroclastic rocks, of andesitic to basaltic composition with ages between 18 and 1 Ma, forming the local basement (Dobson and Mahood, 1985). This unit provides the main aquifer with fluid flow through fractures and faults, sometimes reaching the surface. The thermal fluids are NaCl rich waters with high CO2 and H2S contents, and a pH between 5.8 and 7.2. Average Cl contents are 3100 mg/kg and CO2 may represent as much as 90% of the total gas phase. Fluid temperatures reach values as high as 320 C, but 240–280 C are normal in the field. Hydrothermal alteration has affected most rocks in the geothermal field to a varying extent. Studies of hydrothermal alteration at Los Azufres have been carried out, among others, by Cathelineau et al. (1985), and Torres-Alvarado (1996). These studies have shown that partial to complete hydrothermal metamorphism has occurred. The most important alteration assemblages with increasing depth are: argillitization/silicification, zeolite/ calcite formation, sericitization/chloritization, and chloritization/epidotization.

4. Results 4.1. Hydrological balance The area of the Los Azufres field forms two hydrological basins, mainly because of its isolated, topographical high plateau position (Fig. 1). The northern part (Marı´taro) discharges towards the flat plain of the Cuitzeo Lake basin in NW-direction, whereas the southern part of the field (Tejamaniles) discharges towards the river Balsas basin in S- and SE-directions. The detailed map of the geothermal field area (Fig. 2) shows the water division between both two lateral sections of the geothermal field. The radial structure of rivers and creeks with the Los Azufres mountain peak as the central core area confirms the unusual hydrological discharge conditions. Two case studies were developed to determine the size of the hydrological impact area of the geothermal reservoir. a. Geothermal field area: Hydrological processes, which affect the deep reservoir, are restricted to the lateral surface dimension of the geothermal field (total size: 30 km2) (Fig. 2).

1598

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

b. Extended study area: the entire mountain plateau, including the surrounding lower foothill plains influences the geothermal reservoir by means of infiltration processes (total size: 183 km2). The seasonal distribution of P, ETpot and ETactual values in the study area is described according to Fig. 3: from March to May, high ETpot rates and the lack of precipitation imply the complete absence of infiltration

processes. The same is probably true for the winter months December to February, when lower temperature conditions cause a decrease in ETpot. Major infiltration processes can be expected during the rainy season from June to September, when P exceeds drastically the ETactual rate. The triangular shaped area between the P and ETactual line represents the amount of meteoric water available for surface runoff and groundwater recharge.

Fig. 2. Topographical map of the Los Azufres geothermal field with the sampled cold (I Ciudad Hidalgo, II ‘‘Camp’’, III ‘‘Az-47’’, IV ‘‘Az-17’’, V ‘‘Az-6’’) and hot springs sites (1 San Alejo, 2 ‘‘Az-17’’, 3 ‘‘Leoncito’’, 4 Gallo, 5 Azufres, 6 Laguna Verde, 7 ‘‘Agrio’’, 8 Laguna Verde, 9 Currutaco, 10 Marı´taro), geothermal wells, as well as the water division between the northern and southern part of the field.

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

1599

Fig. 3. Seasonal distribution of the precipitation (P), potential evaporation (ETpot), and actual evapotranspiration (ETactual) in the Los Azufres geothermal area.

The quantitative hydrological balance for the extended study area resulted in a total annual input of 223106 m3 of precipitation within 183 km2, which corresponds to 1330 mm/m2/a (Fig. 4). 48% (637 mm) get lost by evapotranspiration processes, whereas 11106 m3 (62 mm) and 34106 m3 (186 mm) discharge from the study area in the form of surface runoffs during dry and wet season, respectively. The considerable amount of 446 mm (82106 m3) is available for groundwater recharge. Due to methodical errors in the semi-quantitative calculations and measurement techniques, a standard deviation of
206 mm must be accepted for the infiltration value. Due to the highly fractured and faulted structure of the volcanic formations and the low abundance of shallow aquifers in the Los Azufres zone (Birkle et al., 1997a), major fractions of meteoric water are considered to infiltrate deeper sections of the bedrock. 4.2. Stable and radioactive isotopic composition 4.2.1. Influence of andesitic and magmatic water Isotopic data from cold and hot springs and from production wells of the geothermal field of Los Azufres are given in Fig. 5. The composition of the deep fluids, recalculated under reservoir conditions, is shown for the initial (1980–1981) and intermediate (1985–1987) period of geothermal production (Nieva et al., 1983, 1987, respectively) (Table 1), as well as the recent composition of the geothermal fluids in 1998 (Barraga´n et al., 1999). The global meteoric water line, as well as the

Fig. 4. Quantitative hydrological balance for the extended study area (183 km2): column 1: total amount of precipitation (column number in mm); column 2: total amount of evapotranspiration, surface runoff during dry and wet period, and the amount available for infiltration processes.

hypothetical composition of primary magmatic water from hydrous basalts and subduction-related ‘‘andesitic’’ water (Taran et al., 1989, Giggenbach, 1992) are plotted as well.

1600

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

Fig. 5.
18O vs.
D of geothermal brines, condensates, cold and hot springs of the Los Azufres geothermal field (location and names in Fig. 2, Hot springs: site number and water type in Table 2) compared to that of andesitic (Giggenbach, 1992) and magmatic water. Table 1 Recalculated reservoir values of
18O,
D and Cl for geothermal fluids from different sample periods (Nieva et al., 19831, 19872)a Well

Date (d/m/y)


18O (%)


D (%)

Cl (mg/l)

Az-51 Az-51 Az-71 Az-81 Az-131 Az-191 Az-52 Az-92 Az-132 Az-16AD2 Az-182 Az-192 Az-262 Az-282 Az-322 Az-362

29/04/80 13/01/81 17/04/80 26/04/80 14/01/81 13/01/81 24/04/85 16/05/87 24/04/85 23/04/85 11/09/85 25/06/85 26/06/85 08/05/85 08/05/85 31/03/87

4.0 4.1 4.2 4.7 2.8 5.2 3.7 2.3 2.5 3.7 3.6 3.5 2.0 2.5 2.7 3.3

59 62 65 66 60 62 63 61 63 62 66 61 64 56 61 60

1,134 1,250 1,407 1,325 1,619 284 609 1,451 946 573 939 1,367 1,801 1,648 168 878

a

Superior 1 and 2 denote Refs.

The cold springs of the geothermal field are characterized by a typical meteoric composition (samples I, II and III), although two samples (IV and V) with elevated
D and
18O values are affected by evaporation processes (Birkle, 1998).

The initial reservoir discharge from the geothermal field (Nieva et al., 1983) was characterized by the deviation of the
18O and
D-values from the global meteoric water line towards more positive values. According to Giggenbach (1992), very large 18O shifts, such as observed in Los Azufres, cannot be explained exclusively by water–rock interaction. Thus, the positive isotopic trend of the initial fluid discharge, especially the increase in
D in comparison to recent meteoric water composition, indicates the influence of additional processes. The intermediate position of the Los Azufres fluids between meteoric and ‘‘andesitic’’ water and/or magmatic water implies mixing between both components. Considering the latter process exclusively, a mixing ratio of about 30:70 between meteoric water and the fossil, magmatic component respectively would be postulated. Normally, mixing between two components should be reflected by a straight-line relationship between Cl and 2 H concentrations. Fig. 6 shows the Cl and
D-values of the total discharge from different wells, sampled from 1980 to 1981 (Nieva et al., 1983) and 1985 to 1987 (Nieva et al., 1987) during the initial phase of geothermal production. In contrast to the expected straight-line relationship, a very heterogeneous distribution of the values is observed. Thus, mixing of two components is not the dominating formation process of the reservoir fluids. Considering the variation of the fluid composition with time, most wells show constant isotopic composition from 1980 to 1998 (Fig. 5). Beginning with average
18O and

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

1601

Fig. 6. Cl and
D-correlation of the total discharge from different wells, sampled from 1980 to 1981 (Nieva et al., 1983) and 1985– 1987 (Nieva et al., 1987).


D values of 4.2 and 62.5% in 1980–1981 (Nieva et al., 1983), respectively, the reservoir fluids maintain a similar average
18O and
D composition of 3.7 and 60.4% in 1998 (Barraga´n et al., 1999) (Fig. 5). A slight isotopic enrichment can be observed exclusively for production wells, which are adjacent to reinjection wells and affected by the reinjection of geothermal brines (Barraga´n et al., 1999). Similarly, the rise of the Cl content with time has been related to the reinjection of geothermal fluids into the reservoir (Sua´rez et al., 1995). 4.2.2. Vertical communication between reservoir and hot springs The hot springs and fumarole fluids from the geothermal field area show a very heterogeneous composition with strong variations from meteoric composition to extreme positive
D and
18O values (Birkle et al., 1997b) (Fig. 5). Using chemical-isotopic data and results of 3H analysis, 4 different types of warm and hot springs (A-D) can be distinguished (Table 2, Fig. 5): . Type A: high mineralization, especially of typical geothermal tracer elements, such as Cl, B and F,

as well as very high
D (24 to 34%) and
18O (3.4 to 5.6%) values, indicate the direct rise of geothermal liquid and vapour to the surface (Examples: Marı´taro, Laguna Verde spring). . Type B: missing 3H (0 T.U.), quite high
D (24 to 39%) and
18O values (1.7 to 5.4%) as well as the combination of low Cl-concentrations (16– 29 mg/l) and enrichment in SO4 (640–660 mg/l) reflect the condensation close to the surface of rising vapour enriched in H2S as the principal process determining the Currutaco spring chemical composition and probably that of the upper course spring of the Agrio river (‘‘Agrio’’). . Type C: elevated 3H values (5.1–8.3 T.U.) and a low mineralization rate with SO4 as principal component indicate the heating up of a shallow aquifer (with a residence time of more than 10 a) by ascending vapour, which causes the deviation of the
D (57 to 62%) and
18O (4.5 to 5.8%) values from the meteoric water line (examples: Gallo, Azufres) (Birkle et al., 1997b). . Type D: hot springs with a
18O vs.
D composition close to the meteoric water line are fed by

1602

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

Table 2 Chemical and isotopic composition of hot springs in the Los Azufres area Tritium (T.U.)

5.6 3.4

24 34

n.m.a n.m.

<1 15.0

5.4 1.7

24 39

0.0
0.3 n.m.

48.0

3.8

44

3.1
0.5

<0.1 2.3 0.6 6.0 2.5

4.5 5.8 8.2 7.7 10.3

57 62 64 68 72

5.1
0.6 8.3
0.9 n.m. 5.8
0.7 n.m.

Site

Sample

Date (d/m/y)

pH

T ( C)

Cond. (mS/cm)

Cl (mg/l)

SO4 (mg/l)

B (mg/l)

A A

10 8

Marı´taro Laguna Verde spring

16/05/96 17/05/96

3.2 3.5

88.0 83.0

2290 1200

407 102

71.2 273

249 <1

B B

9 7

Currutaco ‘‘Agrio’’

24/11/94 08/06/95

2.4 2.8

85.0 21.3

1240 1020

640 660

B/D

6

Laguna Verde lake

23/11/94

2.6

18.2

1400

163

C C D D D

5 4 3 2 1

Azufre Gallo ‘‘Leoncito’’ ‘‘Az-17’’ San Alejo

08/06/95 25/11/94 27/01/96 24/11/94 18/05/96

2.5 2.4 2.2 6.7 3.1

24.1 87.0 64.0 80.0 40.0

1000 730 560 580 180

a

16.3 28.5 484 62.2 16.5 8.4 28.0 27.7

205 249 268 340 36.3


18O (%)


D (%)

Type

n.m., not measured.

pure meteoric water with 3H values close to the recent atmospheric composition (3.5–6.0 T.U. from 1985 to 1987, Veracruz, Mexico; in: IAEA, 1992) and are slightly heated by rising vapour. It can be concluded, that some of the surface springs are in direct communication with the geothermal reservoir. Low permeability characteristics of some reservoir zones favour the exclusive ascent of geothermal vapour into the local shallow aquifer system, whereas the existence of vertical fault and fracture systems allows the direct rise of geothermal brine. The irregular distribution of the individual hot spring types (Fig. 2) reflects the heterogeneous permeability distribution of the formations and their related fault structures. 4.2.3. Vertical zonation — water–rock interaction In general, water-rock interaction processes are characterized by a positive shift of the
18O values of the liquid phase, while
D values remain unchanged. The O isotope is, therefore, a valuable indicator of such processes. Fig. 7 shows the vertical distribution of the initial
18O values of geothermal liquids (Giggenbach and Quijano, 1981; Nieva et al., 1983, 1987) and of core samples (Torres-Alvarado et al., in press) from different depths. The brine samples represent the specific production depth of the respective production wells, whereas the rock samples were extracted during the drilling of well Az-26. Due to the irregular field topography, the depths of the samples are given in m.a.s.l. Four different trends may be distinguished. a. Rhyolites (2889–2569 m.a.s.l.), forming the caprock of the andesitic geothermal reservoir, are characterized by homogenous
18O values around 9%, which represent the primary (non-altered) felsic rock isotopic composition. Brine samples are not available from this shallow section.

b. Rocks below 2600 m.a.s.l. show an abrupt shift in their
18O values, reaching values as high as +17%. After this depth the O isotopic ratios decrease continuously in felsic rock samples, as well as in andesitic rocks, which host the geothermal reservoir. As the main production zone is located about 600 m below this level, and for that reason, the interaction with geothermal fluids is not yet significant, water–rock interaction processes alone cannot explain this shift. Considering plagioclase, the most important mineral phase in these rocks, and assuming isotopic equilibrium with the geothermal fluids, the isotopic shift can be explained with reference to the temperature effect during O isotope exchange (Torres Alvarado et al., in press). According to plagioclase fractionation curves (O’Neil and Taylor, 1967), temperatures lower than 90 C cause a shift to higher
18O values in rocks, while temperatures higher than 90 C produce the opposite effect. Water–rock interaction may lower
18O values, as seen for the volcanic rocks below 2400 m.a.s.l. (Fig. 7). c. From the upper part of the geothermal reservoir (2220 m.a.s.l.) towards the main production zone (2000–1600 m.a.s.l.), the
18O brine values increase from 5.1% (well Az-17: 2300 m.a.s.l.) to 2.0% (well Az-26: 1600 m.a.s.l.), whereas the rock values decrease from 9.0% to 4.7% in the interval from 2210 to 1750 m.a.s.l. This contrasting tendency can be explained by increasing W–R-interaction processes with a maximum intensity in the main production zone. On the other hand, the linear isotopic trend from the surface meteoric water composition towards the lower part of the main production zone could indicate a decreasing influence of infiltrating meteoric water in the lower part of the geothermal reservoir. d. Below the main production zone, the isotopic

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

1603

Fig. 7. Vertical distribution of
18O values for early production geothermal brines (vertical bars: range of extraction zone), compared with the isotopic composition of the rhyolitic and andesitic drill cuttings from well Az-26.

composition of the geothermal fluids is more homogeneous, with a slight tendency to lower
18O values. Although well Az-26 was only drilled to 1668 m.a.s.l., and, therefore, no deeper core samples are available, the steady trend in the fluid isotopic composition indicates a minor effect of water–rock interaction.
18O data from calcite minerals from different depths of the main production zone were compared with the isotopic composition of the reservoir fluids to determine whether the isotopic temperature of the calcite-water system is in agreement with the actual reservoir temperature, calculated by geothermometers. As initial components, the
18O composition of calcite samples from depths of 900, 1000, 1080 and 1160 m in the production well Az-26 (Torres-Alvarado et al., in press) and a value of
18O=2.0% for the mixture of reservoir liquids from the Az-26 production zone (Nieva et al., 1987) were inserted into the equation of O’Neil et al. (1969). As a result, isotopic temperatures (Tcalcite-geo therm) of 198, 128, 244 and 249 C were derived for the calcites at the respective depths in the production zone (Table 3). Assuming an actual reservoir temperature of 292 C as determined by Nieva et al. (1987) with cationic geothermometers (Tbrine-geotherm), the O’Neil equation gives a hypothetical
18O-value of 0.45% and +4.87% for

the liquid phase and calcite minerals, respectively. On the other hand, the measured values between 2.0 and 5.0% for the range of the Los Azufres reservoir liquids (Giggenbach and Quijano, 1981, Nieva et al., 1983) indicate, that none of the liquids is in complete isotopic equilibrium with the analyzed minerals. However, the calculated temperature of 247 C at a depth of 1160 m may be considered similar to the value of 292 C, as the difference between both values is within the error limits of the Na–K geothermometer proposed by Nieva and Nieva (1987). Furthermore, Verma and Santoyo (1997) calculated by statistical methods a total propagated error (t ) between 27 and 65 C for lowest and highest temperature conditions, respectively. This may indicate the existence of almost complete isotopic equilibrium for the calcite-water system in the lower part of the production zone. The wide range of the calculated calcite temperatures indicates water–rock-interaction processes in fracture and microfracture systems, which allow the achievement of an almost complete isotopic equilibrium along the contact walls of the main flow pathways, whereas calcite minerals, enclosed in their host rocks maintain their primary isotopic composition. The calcite isotopic temperatures (Tcalcite-geotherm) for the production zone are almost identical to the calculated fluid inclusion temperatures (Tcalcite-inclus) with the exception of the calcite sample from 1000 m depth (Gonza´lez et al., 2000). This proves that there is chemical

1604

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

equilibrium between the fluid and solid phases at these temperatures (Table 3). Water–rock interaction processes, shown principally as chloritization and sericitation of the host rock throughout the reservoir (TorresAlvarado et al., in press), cause significant camouflage of the primary meteoric water composition. 4.2.4. Residence time The detection of the radioactice isotope 14C in deep aquifer systems proves its communication with the atmosphere during the last 40,000 to 200 a, whereas the abundance of 3H reflects infiltration processes since the atomic bomb test period in the 1950s and early 1960s. In general, global geothermal fluids are characterized by negligible contents of 14C. Fluids from the Geysers, Steamboat Springs and Salton Sea, USA, contain less than 0.5 pMC (Craig, 1963). Concentrations of 0.2 pMC define an age of 40 ka for the deep well fluids from Wairakei, New Zealand (Fergusson and Knox, 1959). These values were defined as maximum ages due to dilution with 14Cfree, magmatic fluids. In the case of the Los Azufres fluids, measured 3H concentrations between 0 and 0.4
0.3 T.U. from liquids and gases indicate at a first glance the absence of

infiltration of meteoric water into the reservoir at least over the last 40 a (Table 4). A more probable explanation is the complete natural decay of the radioactive isotope due to the long flow distance before reaching the reservoir at a depth from 600 to 2500 m. For 6 samples, the 14C values of the geothermal liquids and vapour are below the detection limit. Maximum values of 1.7
0.6 pMC were measured for the gas sample of the well Az-41. 14C was encountered especially in wells with a high ratio between the vapour and liquid phase. Using an initial 14C activity of 85 pMC for the infiltrating meteoric water, a minimum residence time of 32 ka was derived for all reservoir samples (Table 4). Assuming a theoretical mixing model with a ‘‘young water’’ (meteoric water) and a ‘‘fossil’’ water component for the Los Azufres geothermal reservoir fluids, the contribution of the young water component is significantly below 10% (Birkle, 1998). 4.2.5. Organic component The isotopic composition of compounds at various stages in the C cycle shows large variations due to the impact of different environmental parameters.
13C values of magmatic CO2 gases range between 5 and

Table 3 Measured
18O values of calcite minerals and geothermal fluids, both from the production well Az-26. Isotopic temperatures of the brine Tbrine-geotherm were calculated with cationic geothermometers (Nieva et al., 1983), calcite isotopic temperatures Tcalcite-geotherm using the equation from O’Neil et al. (1969), and the fluid inclusion temperatures Tcalcite-inclus from fluid inclusions from calcite minerals Depth (m)


18Ocalcite (%)


18Obrine (%)

103Ln

Tbrine-geotherm ( C)

Tcalcite-geotherm ( C)

Tcalcite-inclus ( C)

900 1000 1080 1160

7.1 11.9 5.0 4.9

2.0 2.0 2.0 2.0

9.11 13.91 7.01 6.83

292 292 292 292

198 127 243 247

196 206 245 250

Table 4 Results of 3H, 13C and 14C-analysis, and calculated residence time (n.d. not detected; n.m. not measured). The gas content of production wells and the average CO2-percentage of the gas phase is derived from Sua´rez et al. (1995) and Nieva et al. (1987), respectively Well

Tritium (T.U.)


13C (%)

Sample type (14C-method)

Gas (%w)

CO2 (% of gas phase)

14 C-activity (pMC)

Residence time (y)

Az-4 Az-5 Az-9 Az-28 Az-41 Az-43 Az-2 Az-6 Az-22 Az-26 Az-38

n.m.a 0.0
0.3 n.m. 0.1
0.3 n.m. n.m. 0.0
0.3 0.4
0.3 0.0
0.3 n.m. n.m.

12.7 5.4 8.4 11.1 8.2 19.5 12.6 11.4 6.9 14.5 16.1

Gas (CO2) Gas (CO2) Gas (CO2) Gas (CO2) Gas (CO2) Liquid (HCO 3) Gas (CO2) Gas (CO2) Gas (CO2) Gas (CO2) Gas (CO2)

1.4 1.4 0.5 0.3 n.m. 1.0 0.7 4.2 0.8 4.5 2.8

n.m. 97.7 93.6 93.3 97.4 n.m. n.m. 98.2 97.0 98.2 98.3

n.d.b 41.4 n.d. n.d. 1.7
0.6 n.d. n.d. 40.9 n.d. <0.1 <0.1

– 534,000 – – 32,000 – – 537,000 – 537,000 537,000

a b

n.m., not measured. n.d., not determined.

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

8% (Taylor, 1986), whereas Gerlach and Thomas (1986) describe a
13C-value of 3.2% for CO2 gas from Kilauea, Hawaii, derived directly from the mantle. Giggenbach et al. (1983) define a general composition of 10 to 5% for geothermal and volcanic systems. Most of the geothermal fluids range between 5 and 1% (Craig, 1953; Hulston and McCabe, 1962; Panichi et al., 1977). Carbonates with very negative
13C-values are interpreted as the result of organic material input (Presley and Kaplan, 1968). Biogenic processes, such as respiration and the cultivation of compost from plants in the unsaturated soil zone cause the formation of soil CO2 with typical soil values of 20 to 30% in humid areas and 25 to 7% in semiarid to arid regions (Pearson and Hanshaw, 1970; Mann, 1983) (Fig. 8). The generation of CH4 by bacteria results in a variation of
13C between 50 and 93% for biogenic CH4 (Schoell, 1980; Grossman et al., 1989; Aravena et al., 1995), whereas magmatic CH4 varies between 52 and 12% (Fritz et al., 1987, 1992; Sherwood Lollar et al., 1993) with values generally above 40% (Clark and Fritz, 1997). Thermocatalytic CH4 varies between 25 and 38% (Barker and Fritz, 1981) (Fig. 8). Carbonate sediments seem to affect the 13C composition of geothermal fluids such as in the case of metasedimentary graywacke basement rocks of the Taupo

1605

Volcanic Zone in New Zealand (Hulston, 1998). Crude oil, similar to coal, ranges from 34 to 18% (Stahl, 1979). Torres-Alvarado et al. (in press) found very negative
13C values (around 25%) for two hydrothermal calcite samples from the deepest part of well Az-26 (1080 and 1160 m depth). Based on fractionation factors between CO2 and calcite (Bottinga, 1969), these calcites appear not to be in equilibrium with the geothermal CO2 in the field, contrasting to more shallow samples. Although the regional basement at Los Azufres is built up by metamorphosed sediments, these rocks have not been intercepted by any of the geothermal wells. Furthermore, no widespread distribution of paleosoils has been observed in the underground of the Los Azufres geothermal field. Thus, there is no evidence of organic matter in the form of paleosoils or carbonate sediments, which may lead to the observed negative
13C values. Also, the exclusive input of magmatic CO2 cannot explain the very negative
13C values of the Los Azufres fluids (down to 20% in well Az-43). The low values for the Los Azufres reservoir may be due to the influence of surface water, which is enriched in atmospheric and organic CO2 and depleted in 13C. The wide range of the
13C-values for the Los Azufres fluids (5 to 20%) indicates a heterogeneous structure

Fig. 8. Comparison of
13C-value ranges from different natural environments with measured
13C-values of Los Azufres reservoir fluids.

1606

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

of the reservoir with zoned flow regimes (Fig. 9). In detail, the northern geothermal zone — Marı´taro — shows a decrease in the organic component content with depth, which is reflected in increasing
13C-values. For both zones, Marı´taro and Tejamaniles, the most positive values are observed in the lower part of the main production zone. 4.3. Flow velocity and recharge area Contreras et al. (1984) obtained permeability values (K) of 0.041, 0.181, and 2.224 mD for core samples from

the wells Az-19, Az-5, and Az-9, respectively. Applying the conversion factor of 9.66106 between the permeability (K) and hydraulic conductivity (kf) parameter (Freeze and Cherry, 1979), a kf value of 2.1108 m/s corresponds to the maximum rock permeability of 2.224 mD (core sample Az-9) (Table 5). On the other hand, the permeability of a fractured formation on a regional scale is considered to be 1–3 orders of magnitude higher than the local rock permeability. Therefore, K and kf values of 222.4 mD and 2.1106 m/s, respectively, seem to be realistic estimations of the formation parameters on a regional scale.

Fig. 9. Correlation between the
13C-values from Marı´taro and Tejamaniles and their extraction depth. (dashed vertical lines: range of extraction zone for each individual well). Table 5 Measured permeability values and calculated hydraulic conductivity, linear velocity and flow distance values on a local (drilling core) and regional (geological formation) scale

Local scale: Average Local scale: Maximum Regional scale

Permeability K (mD)

Hydraulic conductivity kf (m/s)

Linear velocity va (m/s)

Flow distance (m/y)

0.815 2.224 222.4

7109 2.1108 2.1106

5.31010 1109 1.6107

0.016 0.05 5.2

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

The velocity va is defined as the average linear velocity of the groundwater flow, whereby dh=
l represents the hydraulic gradient, kf the hydraulic conductivity, and ne the effective porosity [Eq. (1)]. a ¼ 

kf @h ne @l

ð1Þ

Sua´rez (1991) defined an effective total porosity of 6.3% for the Los Azufres reservoir with a range of 11.9% in the upper and 2.9% in the lower section. The application of the parameters cited in Eq. (1), such as a hypothetical hydraulic conductivity of 2.1106 m/s, an average effective porosity of 6.5%, and a hydraulic gradient of 0.5% resulted in a calculated flow velocity of 1.6107 m/s or 5.2 m/a (Table 5). Therefore, the hypothetical infiltration of meteoric water into the underground about 25 ka ago with an average flow velocity of 5.2 m/a would result in a total flow distance of 130 km. The relatively fast travel time of the meteoric water could imply a relatively small size of the former recharge area. 4.4. Flow circulation model Based on results of several stable and radioactive isotope determinations, a flow model of the geothermal

1607

circulation system with its individual components was developed (Fig. 10). Meteoric water with a typical atmospheric stable and radioactive isotopic composition, such as negative 13C values (12 to 14%) and enriched 14C activity, infiltrated from the surface into the reservoir. It is possible, that this fossil recharge event was related to major periods of increased humidity, such as observed for Northern Mexico at the end of Late Pleistocene (18–11 ka BP) and Early Holocene (11–8.9 ka BP) (Ortega-Ramı´rez et al., 1998). Minor portions of geothermal fluids may be derived from a deep lying magma chamber, which conducts heat and 14C-free CO2 gas with very negative
D values (60%) towards more shallow zones. Additionally, a subduction related sedimentary slab, the result of a compressional tectonic regime from the Cretaceous to Recent time period, could have provided andesitic water to the regional aquifer system. Due to changes in pressure conditions, the single fluid phase gets separated into liquid and vapour phases at a depth of 1500 m. Both phases are characterized by differences in their isotopic composition. After cooling in the upper section of the reservoir, major parts of the phases remain within the convective geothermal cycle, whereas a minor part rises directly towards shallow aquifer systems and towards surface springs.

Fig. 10. Flow circulation model of the Los Azufres geothermal reservoir, based on stable and radioactive isotopic data.

1608

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

5. Discussion The trend of both isotopes,
18O and
D, towards an andesitic water composition seems to indicate possible mixing between a meteoric and a magmatic/andesitic component in the geothermal reservoir. The andesitic component could be related to the subduction of oceanic crust with its sedimentary slab below the North American Plate from the Cretaceous to Recent time period. In contrast, the lack of a positive correlation between the Cl and 2H compositions of the deep fluids reflects minor effects of mixing processes. The effects of secondary alteration processes, mainly water/rock interaction, which are reflected in the approximation of the isotopic brine and rock composition in the main production zone, seem to be more important. The influence of an atmospheric component is reflected by the abundance of very negative 13C-values, which indicate organic input into the reservoir by meteoric water. The combination of negative 13C values and the lack of 14C in the geothermal reservoir indicates the infiltration of meteoric water during pre-historic time, probably as part of the Holocene or Late Pleistocene glacial period, as well as dilution of paleo-meteoric water by ascending, magmatic CO2 without any radioactive C. The latter effect makes it difficult to determine the exact schedule of the infiltration events. The former existence of a hydrodynamic flow system and active circulation paths is reflected by strong waterrock interaction within the main production zone. Recently, the fossil hydrodynamic system has been reactivated by geothermal production, which is indicated by the occurrence of reinjected, natural tracers in production wells. Despite low permeability characteristics of the upper and lower part of the reservoir, communication to the surface seems to take place along fractures and fault systems (Sua´rez et al., 1995).

6. Conclusions Highly fractured volcanic rock in the Los Azufres area permits fast and deep infiltration of meteoric water into the deeper aquifer systems. The results of hydrological balance calculations prove the presence of a considerable amount of meteoric water available for recharge. Subtracting the fraction of meteoric water lost by evapotranspiration and discharging surface water, the amount of 446
206 mm per m2/a (total annually: 82106 m3) is nowadays available for feeding of deeper aquifers. On the other hand, radioactive isotopes indicate that natural recharge of the reservoir by ‘‘fossil’’ meteoric water occurred during pre-historic time, probably during the Holocene and/or Pleistocene glacial period. As a second component, the tendency in
D and
18O

towards an andesitic water composition could reflect the additional influence of oceanic crust and its sedimentary slab, subducted during the Cretaceous to Recent time period below the North American Plate. The advanced state of water–rock interaction processes, especially in the main production zone of the reservoir, reflects on the former existence of a fossil, hydrodynamic system with active fluid circulation. The recent reactivation of the hydrodynamic system by geothermal production is indicated by chemical and isotopic changes of the brine composition in production wells adjacent to reinjection wells.

Acknowledgements The authors are grateful to Professor Dr. D. Hebert and Dr. O. Nitzsche, Institute for Applied Physics at the Technical University of Freiberg/Saxony, Germany, for the tritium, 13C, and 14C analysis. We thank H. A´rmannsson, J. Hulston and B. Steingrı´msson for their critical reviews and helpful comments.

References Arago´n, A.A., 1985. Uso de trazadores radioactivos en la zona norte del campo geote´rmico de Los Azufres. CFE, Depto. Yacimientos, Morelia, Mexico, Internal Report 1385-004. Aravena, R., Wassenaar, L.I., Plummer, L.N., 1995. Estimating 14C groundwater ages in a methanogenic aquifer. Water Resources Research 31, 2307–2317. Barker, J.F., Fritz, P., 1981. The occurrence and origin of methane in some groundwater flow systems. Canadian Journal of Earth Sciences 18, 1802–1816. Barraga´n, R.M., Portugal, E., Arellano, V.M., 1999. Estudio isoto´pico de fluidos de pozos productores y de reinyeccio´n del campo geote´rmico de Los Azufres, Michoaca´n. Instituto de Investigaciones Ele´ctricas, Cuernacava, Morelos, Mexico, Internal report, IIE/11/11172/1 037F. Birkle, P., 1998. Herkunft und Umweltauswirkungen der Geothermalwa¨sser von Los Azufres, Mexiko. Wissenschaftliche Mitteilungen, Technical Univ. Freiberg, Freiberg/Saxony, Germany, Vol. 6. Birkle, P., Torres, R.V., Gonza´lez, P.E., 1997a. El papel del sustrato geolo´gico en el escurrimiento superficial y en los acuı´feros del campo geote´rmico de Los Azufres. Ingenierı´a Hidrau´lica en Me´xico XII, 37–48. Birkle, P., Torres, R.V., Gonza´lez, P.E., 1997b. Estudios hidroquı´micos e isoto´picos sobre el origen de las fumarolas y manantiales calientes de Los Azufres, Michoaca´n. Proceedings, 5to. Congreso, Asociacio´n Geote´rmica Mexicana, Guanajuato, Gto., Mexico, pp. 91–96. Bottinga, Y., 1969. Calculated fractionation factors between carbon and hydrogen isotope exchange in the system calcitecarbon dioxide-graphite-methane-hydrogen-water vapour. Geochim. Cosmochim. Acta 33, 49–64.

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610 Cathelineau, M., Oliver, R., Garfias, A., Nieva, D., 1985. Mineralogy and distribution of hydrothermal mineral zones in the Los Azufres (Mexico) geothermal field. Geothermics 14, 49–57. Cathelineau, M., Oliver, R., Nieva, D., 1987. Geochemistry of volcanic series of the Los Azufres Geothermal Field. Internacional 26, 273–290. Cedillo, F.R., Silva, R.P., Vargas, H.L., 1981. Estudio geohidrolo´gico del a´rea de Los Azufres-Cuitzeo, Estado de Michoaca´n. CFE, Subgerencia de Estudios Geote´rmicos, Internal report 37/81, 4 volumes. Clark, I., Fritz, P., 1997. Environmental Isotopes in Hydrogeology. Lewis Publishers. Contreras, E., Iglesias, E., Oliver, R., 1984. Determinacio´n de propriedades fı´sicas y estudios petrogra´ficos de nucleos de perforacio´n del campo geote´rmico Los Azufres. IIE, Cuernavaca, Internal report IIE/11/3753/I 01/F. Craig, H., 1953. The geochemistry of the stable carbon isotopes. Geochim. Cosmochim. Acta 3, 53–92. Craig, H., 1963. The isotopic geochemistry of water and carbon in Geothermal Areas. In: Tongiorgi, E. (Ed.), Nuclear Geology on geothermal areas. Spoleto, Consiglio Nazionale delle Ricerche, Laboratorio di Geologia Nucleare, Pisa, pp. 17–53. Dobson, P.F., Mahood, G.A., 1985. Volcanic stratigraphy of the Los Azufres Geothermal Area, Mexico. Journal of Volcanology and Geothermal Research 25, 273–287. Fergusson, G.J., Knox, F.B., 1959. The possibilities of natural radiocarbon as a ground water tracer in thermal areas. N.Z. J. Sci. 2, 431–441. Freeze, R.A., Cherry, J.A., 1979. Groundwater. Prentice Hall, Englewood Cliffs, N.J. Fritz, P., Frape, S.K., Miles, M., 1987. Methane in the crystalline rocks of the Canadian Shield. In: Fritz, P., Frape, S.K. (Eds.), Saline Water and Gases in Crystalline Rocks, Geol. Ass. Canada Spec. Pap. 33, 211–223. Fritz, P., Clark, I.D., Fontes, J.-Ch., Whiticar, M.J., Faber, E., 1992. Deuterium and 13C evidence for low temperature production of hydrogen and methane in a highly alkaline groundwater environment in Oman. In: Kharaka, Y., Maest, A.S. (Eds.), Proc. 7th Internat. Symp. on Water–Rock Interaction, Balkema, pp. 793–796. Gerlach, T.M., Thomas, D.M., 1986. Carbon and sulphur isotopic composition of Kilauea parental magma. Nature 319, 480–483. Giggenbach, W., 1992. Isotopic shifts in water from geothermal and volcanic systems along convergent plate boundaries and their origin. Earth Planet. Sci. Lett. 113, 495–510. Giggenbach, W., Goguel, R.L., 1989. Collection and analysis of geothermal and volcanic water and gas discharges. Chemistry Division, Department of Scientific and Industrial Research, Petone, New Zealand. Report No. CD 2401. Giggenbach, W., Quijano, J.L., 1981. Estudio isoto´pico de las aguas del campo geote´rmico de Los Azufres. Comisio´n Federal de Electricidad, Mexico, Internal report. Giggenbach, W., Gonfiantini, R., Panichi, C., 1983. Geothermal systems. In: Agency Guidebook on Nuclear Techniques in Hydrology, International Atomic Energy, Technical Reports Series 91, pp. 359–380. Gonza´lez, E.P., Birkle, P., Torres-Alvarado, I.S., 2000. Evolution of the hydrothermal system at the geothermal field of Los Azufres, Mexico, based on fluid inclusion, isotopic and petrologic data. J. Volcanol. Geoth. Res. 104 (1–4), 277–296.

1609

Gray, D.M., 1973. Handbook of principles in hydrogeology. Water Information Center, Huntington, NY. Grossman, E.L., Coffman, B.K., Fritz, S.J., Wada, H., 1989. Bacterial production of methane and its influence on groundwater chemistry in east-central Texas aquifers. Geology 17, 495–499. Horne, R.N., Puente, H., 1989. Tracer testing at Los Azufres. Proc., 14th Workshop Geothermal Reservoir Engineering, Stanford, California, pp. 197–199. Hulston, J.R., 1998. Correlations betwen B/Cl ratios and other chemical and isotopic components of Taupo Volcanic Zone, NZ geothermal fluids — evidence for water–rock interaction as the major source of boron and gas. In: Arehart, G. and Hulston, J.R. (Eds.), Water–Rock Interaction. Balkema, pp. 629–634. Hulston, J.R., McCabe, W.J., 1962. Mass spectrometer measurements in the thermal areas of New Zealand, Part II. Carbon isotopic ratios. Geochim. Cosmochim. Acta 26, 398–410. IAEA, 1992. Statistical treatment of data on environmental isotopes in precipitation. Technical Reports Series No. 331, Vienna. Iglesias, E.R., 1983. Reporte de avance del primer intento de utilizacio´n de trazadores en el campo geote´rmico de Los Azufres. Memorias de la Primera Reunio´n Interdisciplinaria CFE-IIE, Los Azufres, Michoaca´n, Me´xico, pp. 161–170. Iglesias, E.R., Arellano, V.M., 1985. Apoyo en ingenierı´a de yacimientos para el campo geote´rmico de Los Azufres. Fase 1: Integracio´n de un modelo ba´sico del yacimiento. Instituto de Investigaciones Ele´ctricas, Cuernavaca, Internal report IIE/11/1864/I 04/F. Kuzmin, P.P., Vershinin, A.P., 1974. Determination of evaporation in case of absence or inadequacy of data. Proceed. of the Madrid Symposium. Studies and reports in hydrology. UNESCO-WMO-IASH 16, 217–226. Mann, W.B., 1983. New contemporary radiocarbon dating standards. In: Stuiver, M., Kra, R. (Eds.), Proceedings of the 11th International Radiocarbon Dating Conference. Radiocarbon Vol. 25, p. 2. Morton, F.I., 1965. Potential evaporation and river basin evaporation. Proceedings of the ASCE, J. Hydraulics Division 91, 67–97. Nieva, D., Quijano, J., Garfı´as, A., Barraga´n, R.M., Laredo, F., 1983. Heterogeneity of the liquid phase, and vapor separation in Los Azufres (Mexico) geothermal reservoir. 9th Workshop on Geothermal Reservoir Engineering, Stanford, California, pp. 253–260. Nieva, D., Barraga´n, R.M., Cathalineau, M., Gonza´lez, E., Oliver, R., Portugal, E., Santoyo, E., Verma, M., 1987. Estudio de la estructura hidrolo´gica y de la distribucio´n de para´metros fisico-quı´micos en el yacimiento de Los Azufres, Mich., Fase II. IIE, Internal report IIE/11/2011/1. Nieva, D., Nieva, R., 1987. Developments in geothermal energy in Mexico. Part 12 — a cationic composition geothermometer for prospecting geothermal resources. Heat Recovery Systems 7, 243–258. O’Neil, J.R., Taylor, H.P., 1967. The oxygen isotope and cation exchange chemistry of feldspars. Am. Mineral. 52, 1414–1437. O’Neil, J.R., Clayton, R., Mayeda, T., 1969. Oxygen isotope fractionation in divalent metal carbonate. J. Chem. Phys. 51 (12), 5547–5558.

1610

P. Birkle et al. / Applied Geochemistry 16 (2001) 1595–1610

Ortega-Ramı´rez, J.R., Valiente-Banuet, A., Urrutia-Fucugauchi, J., Mortera-Gutie´rrez, C.A., Alvarado-Valdez, G., 1998. Paleoclimatic changes during the Late Pleistocene — Holocene in Laguna Babı´cora, near the Chihuahuan Desert, Me´xico. Can. J. Earth Sci. 35, 1168–1179. Panichi, C., Ferrara, G.C., Gonfiantini, R., 1977. Isotope geochemistry in the Lardarello geothermal fields. Geothermics 5, 81–88. Pearson, F.J. Jr., Hanshaw, B.B., 1970. Sources of dissolved carbonate species in groundwater and their effects on carbon-14 dating. In: Isotope Hydrology, Proc. Symp., Wien, IAEA-SM-129/18, pp. 271–286. Presley, B.J., Kaplan, I.R., 1968. Changes in dissolved sulfate, calcium and carbonate from interstitial water of near-shore sediments. Geochim. Cosmochim. Acta 32, 1037–1048. Remendieras, G., 1974. Tratado de Hidrologı´a Aplicada. Barcelona, Espan˜a. Richter, J., Szymczak, P., Abraham, T., Jordan, H., 1993. Use of combinations of lumped parameter models to interpret groundwater isotopic models. J. Contam. Hydrol. 14, 1–13. Schoell, M., 1980. The hydrogen and carbon isotopic composition of methane from natural gas of various origins. Geochim. Cosmochim. Acta 44, 649–662. Sherwood Lollar, B., Frape, S.K., Weise, S.M., Fritz, P., Macko, S.A., Welhan, J.A., 1993. Abiogenic methanogenesis in crystalline rocks. Geochim. Cosmochim. Acta 57, 5087– 5097. Stahl, W.J., 1979. Carbon isotopes in petroleum geochemistry.

In: Ja¨ger, E., Hunziker, J.C. (Eds.), Lectures in Isotope Geology, pp. 274–282. Sua´rez, M.C., 1991. Capacidad energe´tica del sector Tejamaniles. Geotermia, Rev. Mex. Geoenergı´a 7, 291–324. Sua´rez, M.C., Ortega, L.P., Luis del, Rı´o, 1995. Fifteen years of development of Los Azufres geothermal field, Mexico. World Geothermal Congress Proc., Florence, Italy 3, 1791– 1796. Taran, Y.A., Pokrovsky, B.G., Dubik, Y.M., 1989. Isotopic composition and origin of water from andesitic magmas. Dokl. (Trans) Ac. Sci. USSR 304, 440–443. Taylor, B., 1986. Magmatic volatiles: Isotopic variation of C, H, O, and S. In: Valley, W.J., Taylor, P.H., O’Neil, R.J. (Eds.), Stable Isotopes in High Temperature Geological Processes. Reviews in Mineralogy, Vol. 16. Torres-Alvarado, I.S., 1996. Wasser/Gesteins-Wechselwirkung im geothermischen Feld von Los Azufres, Mexiko. Mineralogische, thermochemische und isotopengeochemische Untersuchungen, Tuebinger Geowissenschaftliche Arbeiten, Reihe E, 2. Torres-Alvarado, I.S., Satir, M., Metz, P., Birkle, P., in press. Stable isotopes (O, C, H) constraints of hydrothermally altered rocks and hydrothermal minerals at the Los Azufres geothermal field, Mexico, J. Volcanol. Geoth. Res., submitted. Verma, S.P., Santoyo, E., 1997. New improved equations for Na/K, Na/Li and SiO2 geothermometers by outlier detection and rejection. J. Volcanol. Geoth. Res. 79, 9–23.