The Los Azufres (México) Geothermal Reservoir: Main Processes

Transcription

The Los Azufres (México) Geothermal Reservoir: Main Processes
Proceedings World Geothermal Congress 2015
Melbourne, Australia, 19-25 April 2015
The Los Azufres (México) Geothermal Reservoir: Main Processes Related to Exploitation
(2003-2011)
Víctor Manuel Arellano1, Rosa María Barragán1, Miguel Ramírez2, Siomara López1, Adriana Paredes1, Alfonso
Aragón1, Emigdio Casimiro3, Lisette Reyes3
1
Instituto de Investigaciones Eléctricas, Gerencia de Geotermia, Cuernavaca, Morelos, México
2
Comisión Federal de Electricidad, Gerencia de Proyectos Geotermoeléctricos, Morelia, Mich., México
3
Comisión Federal de Electricidad, Residencia Los Azufres, Campamento Agua Fría, Mich., México
[email protected]
Keywords: Los Azufres geothermal field, well simulation, fluid geochemistry, well bottom conditions, production wells
ABSTRACT
In this work, the present-time thermodynamic conditions and the response to exploitation for 2003–2011 of the Los Azufres
geothermal reservoir are given. The study was based on the analysis of geochemical and production data of 39 wells, over time. In
average, due mainly to effective reservoir recharge, very moderate production declining rates were found. In the south zone, the
important effects of reinjection were observed through increases in Cl in some wells (up to 8,000 mg/kg). In contrast, in wells with
significant boiling Cl tends to decrease. In most of the North Zone wells, the variations in gas data indicated boiling and
condensation of a highly gas-depleted brine, which seems to consist of reinjection fluids. It is suggested that this process helps to
maintain the production in the zone relatively stable. In summary, the main reservoir exploitation-related processes found in Los
Azufres were: (1) production of reinjection returns; for this, it was possible to distinguish (a) wells that produce liquid and steam
from injection, and (b) wells that produce only steam from injection and sometimes this steam already condensed. (2) boiling; for
this process, two types of boiling were identified: (a) boiling with gaining steam, and (b) boiling with steam loss.
1. INTRODUCTION
The Los Azufres geothermal field is an intensely-fractured, two-phase, volcanic hydrothermal system located in the northern
portion of the Mexican Volcanic Belt, in the state of Michoacán, at an average elevation of 2,800 m.a.s.l. (Figure 1). At present, the
installed capacity of the field is 188 MWe. The field was divided into two zones (North and South) because different characteristics
in producing fluids were noticed in the beginning of the development stage of the field. Reservoir engineering and geochemical
conceptual models of the Los Azufres system were established (Iglesias et al., 1985; Nieva et al., 1987) which constitute the
reference conditions that allow characteristic changes in parameters over time to be related to exploitation. Arellano et al. (2005)
studied the reservoir’s response to exploitation for 1982–2002 based on a systematic analysis of chemical, isotopic, and production
data from 20 representative production wells. The installed capacity of the field was 88 MWe during that time. Since 2003, four
additional 25 MWe flash plants were brought on line, bringing the total installed capacity to 188 MWe. Due to this increase in
power capacity that was installed at the North Zone, concerns on reservoir performance have arisen because of higher rates in fluids
extraction.
Figure 1: Location of the Los Azufres geothermal field and locations of wells.
In order to estimate the occurrence of reservoir processes induced by exploitation (fluid extraction and reinjection) for 2003–2011,
the following objectives were stated (Arellano et al., 2012): (a) to identify the changes that have occurred in the Los Azufres
1
Arellano et al.
reservoir as a consequence of exploitation, and (b) to determine the causes of these changes through the analysis of production and
geochemical data of 39 wells (Figure 1) provided by the Comisión Federal de Electricidad. All of the injection wells are located on
the west side of the field. Injectors Az-3, 15, 52, and 61 are in the northern production area, whereas Az-7A and 8 are located in the
south (Figure 1).
2. METHODOLOGY
In order to obtain the thermodynamic characteristics of reservoir fluids and to investigate the dominant processes occurring with
exploitation, a method based on the analysis of the following patterns of behavior was used (Arellano et al., 2005): (a) well mass
flow-rate, well-bottom pressure, enthalpy, and temperature, (b) chloride concentration in separated water and total discharge fluid,
(c) the comparison of the total discharge enthalpy and those corresponding to the reservoir temperature estimated by both the
“slow-response” cationic (Nieva and Nieva, 1987) and the “fast-response” silica (Fournier and Potter II, 1982) geothermometers
(Truesdell et al., 1995), (d) fraction of steam entering the well, (e) total discharge and reservoir chlorides, (f) total discharge and
reservoir CO2 (g) liquid saturation, and (h) δ18O and δD in total discharge. Well-bottom thermodynamic conditions were obtained
through WELFLO simulator (Goyal et al., 1980). Input data consisted of wellhead production data and well geometry. Chemical,
isotopic and production data were provided by the Residencia Los Azufres, CFE. The study included 39 wells (Figure 1): Az-4, 5,
9, 9A, 9AD, 13, 19, 28, 28A, 30, 32, 41, 43, 45, 48, 51, 56R, 65D, 66D, 67, and 69D corresponding to the North Zone of the field,
and Az-1A, 2A, 6, 16AD, 17, 18, 22, 23, 25, 26, 33, 34, 35, 36, 37, 38, 46, and 62 corresponding to the South Zone.
3. FLUID PRODUCTION/INJECTION
Fluid production and injection rates have varied with the operation of the generating units over time. Figure 2 shows the production
and injection data as a function of time for the field. Fluid extraction started in 1978 (Figure 2) and 441,744,661 tons of fluids were
produced by December 2011. Of these, 254,376,326 tons (57.6 %) corresponded to the south, and 187,368,335 tons (42.4 %) to the
North zone. Between 1982–2011, 132,016,271 tons (30 %) of produced fluids had been re-injected. Production increased in the
South Zone in 1987, when the 50 MW power plant came on line (Arellano et al, 2005). In 2003, fluid production in the North Zone
increased due to installation of an additional capacity of 100 MW.
Figure 2: Production and reinjection fluids in Los Azufres geothermal field.
4. RESULTS
The results obtained allowed identification of the main processes that have occurred and are currently occurring in different zones
of the Los Azufres reservoir. Two main processes were identified: (1) production of reinjection returns. In this process two subprocesses were identified: (1.1) wells that receive liquid and steam from reinjection, and (1.2) wells that receive steam originated
from the boiling of reinjection fluids and/or wells that receive condensed steam originated from boiling of reinjection fluids; and (2)
boiling. In this process two sub-processes were identified: (2.1) boiling with steam gain and (2.2) boiling with steam loss.
4.1 Reservoir Processes
In order to study the evolution of the well bottom pressures and enthalpies, production data for each well was simulated using the
WELFLO (Goyal et al., 1980) simulator. WELFLO is a well simulator that considers multiphase, one-dimension, and steady state
conditions and is suitable for vertical wells with variable diameter. Input data consists of geometry of the well, mass flow rates, and
wellhead pressures and enthalpies. Reservoir temperatures were estimated by WELFLO simulations and also by a cationic
geothermometer (Nieva and Nieva, 1987) for two-phase wells and by the FT-HSH2 gas equilibrium method (Siega et al., 1999;
Barragán et al, 2013) for the steam wells.
4.1.1 Production of Reinjection Returns
In order to illustrate the process of production of injection returns, the case of the well Az-33 that receives injection returns in liquid
phase is shown. In Figure 3 time series of mass flow rates, well bottom pressure, well bottom enthalpy, chlorides, wellhead
2
Arellano et al.
AZ-33
120
(A)
80
30
20
0
0
3000
AZ-33
(B)
40
30
20
3200
2000
1600
in
(F)
AZ-33
2000
1500
ENT CCG
ENT SiO2
1000
CO2 (mol ‰)
2400
10
5
0
2500
ENT DT
500
22.50
(C)
AZ-33
2800
(E)
AZ-33
40
10
10
Well bottom
enthalpy (kJ/kg)
50
40
50
Well bottom
pressure (bar)
Total
Liquid
Steam
160
Enthalpy (kJ/kg)
Flowrate (ton/h)
200
Wellhead
pressure (bar)
pressure, comparison of total discharge enthalpy and enthalpy estimations by geothermometers, CO2, and reservoir liquid saturation
for the well Az-33 are given. This well is 683 m deep, with 81 m of slotted liner (602–683 m). As seen in Figure 3, the mass flow
rate and pressure decrease, whereas enthalpy total chloride discharge, and CO2 increase. The pattern for the enthalpy comparison
shows mixing of equilibrated liquid with steam produced by boiling away the well (Figure 3F). Figure 4A shows that well Az-33
has received reinjection fluids in some years through the increase in chlorides and the behavior of the well shows intermittent
changes as the injection rates are not constant over time (Figures 3A–H and 4A, B). In Figure 4A, the enthalpy-chloride data of
well Az-33 are found to the right hand side of the characteristic boiling line which constitutes the reference for the field (Arellano et
al., 2005). This is due to the chloride enrichment of discharged fluids because of production of reinjection returns. Chloride and
other solutes are highly concentrated in injection fluids since they are evaporated at ambient conditions before injection. Well Az33 and other wells show important recovery in liquid saturation since 2004–2005 (Figure 3H), which can be related to the operation
of the injection well Az-7AR that replaced the original injection well Az-7R. Well Az-33 produced two-phase fluids but became a
steam producer when the injector well Az-7R was closed in 2003. In 2007, Az-33 started intermittent liquid production (Figure
3A). Results from tracer tests (Iglesias et al., 2011) indicate that part of the fluids injected in well Az-8R was recovered as both
liquid and steam in well Az-33. However, well Az-33 develops a moderate and sometimes important boiling process depending on
reinjection rates, which can be seen through the enthalpy variations in Figures 3C and F.
AZ-33
(G)
17.50
12.50
7.50
Reservoir
Total discharge
2.50
Chlorides (mg/kg)
AZ-33
(D)
Total discharge
Separated water
8000
Liquid saturation
(fraction)
1200
12000
4000
0.70
AZ-33
(H)
0.50
0.30
0.10
0
1975
1980
1985
1990
1995
2000
2005
2010
2015
1975
1980
1985
1990
1995
Year
2000
2005
2010
2015
Year
Figure 3: Time series of (A) mass flow rates, (B) well bottom pressure, (C) well bottom enthalpy, (D) chlorides, (E) wellhead
pressure, (F) comparison of total discharge enthalpy and enthalpy estimations by geothermometers, (G) CO2, and
(H) reservoir liquid saturation of well Az-33, taken as representative of production of reinjection returns.
3000
AZ-33
9
7
88
Enthalpy (kJ/kg)
(B)
11
83
Chlorides increase
reinjection effect
95
2000
93
96
92
Well bottom pressure (bar)
Vapor
2500
Az 33
(A)
0 8
97 1 98
91
94
90
1500
89
Boiling line
1000
Steam loss
100
300°C
90
94
89
92
3 98
9396
97
95
4
11
8
500
7 6
9
5
10
200°C
10
0
0
1000
2000
3000
4000
5000
500
1000
1500
2000
2500
3000
Well bottom enthalpy (kJ/kg)
Total discharge chlorides (mg/kg)
Figure 4: (A) Enthalpy versus chlorides and (B) well bottom pressure versus well bottom enthalpy for well Az-33.
3
3500
Arellano et al.
The production of injection returns is shown only by some wells from the South Zone, where injection plays an important role, such
as in Az-1A, 2, 2A, 16D, 36, and 46. Among these, wells Az-36 and 46 became steam producers since well Az-7R was closed. In
wells that receive reinjection returns in liquid phase, chloride concentrations in separated water range between 5,500–8,000 mg/kg,
whereas concentrations range between 2,500 and 4,000 mg/kg in wells with no reinjection interference.
The production of steam from the boiling of reinjection returns and sometimes condensed steam from the boiling of injection fluids,
was also identified to occur in some wells of the field. When injection fluids consist of recycled produced fluids, the effects of
reinjection in two-phase production wells is routinely depend on the salt increase in the discharged liquid (Arnórsson, 2000;
Arnórsson and D’Amore, 2000; Arellano et al., 2005; Barragán et al., 2005 and 2010) as was illustrated for well Az-33. However,
injected fluids in contact with reservoir rocks were heated and eventually evaporated by boiling. Then reinjection fluids either as
steam or as condensed steam flew to the production zones of wells. These types of reinjection returns could not be recognized in
production wells by the increase in the salinity of the liquid discharged, since condensed steam is depleted in terms of salt and
isotopes. Thus, in two-phase wells that produce such types of reinjection returns, the effect in the discharged liquid sometimes
include the decrease in salinity. In this case, identification of the effects of injection returns consisting of steam or condensed steam
from boiling of reinjection fluids in production wells, gas, and isotope data analysis is suggested. Following characteristics are
particularly observed in the produced fluids, when production of steam from reinjection occurs: (a) N2/Ar molar ratios lower than
that of air-saturated water (38), (b) low gas and CO2 (<5 ‰) concentrations, and (c) relative depletions in δD and δ18O.
In the South Zone, due to reinjection rates, this process seems to occur constantly in wells Az-1A, 23 and 25, and intermittently in
Az-22, 35, and 62. In contrast, many of the wells in the North Zone (e.g., Az-4, 9, 19, 28, 28A, 45, 51, 57, 66D, 67 and 69D and
Az-5, 13, 32, 42 and 48) that constantly produce steam or condensed steam from the boiling of reinjection fluids began to produce
in an intermittent way after 2005. In Figure 5, the N2-He-Ar ternary diagrams for wells Az-9 (two-phase) and Az-35 (steam) are
given. Both wells receive steam or condensed steam from the boiling of injection fluids in the years when data plot below the He
corner-meteoric line. As seen in Figure 5A, well Az-9 has received condensed steam from the boiling of reinjection fluids almost
constantly, which explains the production of slightly lighter fluids (δ 18O: -3.5‰, δD: -63.5‰) in 2010 (Figure 6A; Barragán et al.,
2010) relative to the reference isotopic composition (δ18O: -2.4‰, δD: -61‰; Nieva et al., 1987). Well Az-28 has also become
isotopically lighter compared to its reference composition (δ18O: -2.51‰, δD: -61.8‰) due to production of condensed steam from
boiling of reinjection fluids. This can be seen in Figure 6A, where 2010 data plots closely to the lighter end of the fitting line
(Barragán et al., 2011). In contrast, steam well Az-35 (South Zone) intermittently received steam produced by boiling of reinjection
fluids in 2010 and 2011 (Figure 5B). This probably causes deuterium enrichment in well Az-35 relative to the fitting line in Figure
6B, considering that deuterium partitions preferably in steam at temperatures above 220°C. As a result, isotopic compositions of
two-phase wells Az-1A, 23 and 25, which constantly produce condensed steam from reinjection fluids, are more depleted in the
South Zone (Figure 6B).
N2/100
(A)
N 2/100
(B)
1000
1000
magmatic
magmatic
Az-35
Az-9
6
100
100
aire
aire
89
89
meteoric
90
0
92
meteoric
93
91
4
6
10He
96
9
5
10
10
11
4
99
2
5
94
88
8
0
97
96 1
97
Ar10He
crust
3
8
88
7
11
95
10
crust
Figure 5: He-N2-Ar relative compositions of (A) well Az-9 (North Zone) and (B) well Az-35 (South Zone).
4
99
10
98
9
Ar
Arellano et al.
-30
3
North zone
South zone
B
A
Total discharge D (‰)
-40
7A
15
61
-50
8
2A
52
16
42
-60
16AD
65D 9A 4
56 9AD
69D
5 68D
32
43
66D
9 45 13
67
19
28
51 28A
48
30
37 3446 18
6
17 38
36
35
26
1A 33
25 62
23
22
Reinjection well
Reinjection well
Well
Well
-70
-6
-4
-2
0
2
4 -6
-4
Total discharge O (‰)
-2
0
2
Figure 6: δD versus δ18O compositions of (A) North Zone and (B) South Zone wells according to 2010 data (Barragán et al.,
2010).
4.1.2 Boiling
Boiling occurs in wells due to pressure drops induced by fluids extraction during exploitation. Boiling may occur in a well either by
steam gain or by steam loss.
Boiling by steam gain was identified in some wells of Los Azufres such as wells Az-5 and 13 from the North Zone and Az-18 and
26 from the South Zone (Arellano et al., 2005). From these wells, Az-5, 13, and 18 dried after being first two-phase producers.
Well Az-26 is taken as a representative to illustrate the process through the analysis of chemical and production data. Time series of
mass flow rates, well bottom pressure, well bottom enthalpy, chlorides, wellhead pressure, comparison of total discharge enthalpy
and enthalpy estimations by geothermometers, CO2, and reservoir liquid saturation of well Az-26 are given in Figure 7.
Well Az-26 produced two-phase fluids with high liquid/steam ratio until 1995 (Figure 7). During 1996–2003, it produced lower but
constant amounts of liquid until production decreased drastically between 2003 and 2005. From 2008 to date, a decreasing trend in
liquid production is observed due to the boiling with limited recharge. Following a relatively stable period, the well bottom pressure
started to decline 2009 (Figure 7B) and enthalpy increased to values that reach the vapor enthalpy (Figure 7C). Well shows an
decreasing enthalpy trend during 2007–2008, corresponding to a small recovery in liquid production, which could be due to low
recharge coincident with a peak in injection rates in wells Az-7AR and Az-8R. Figure 7D shows that total chloride discharge
decreases due to very small liquid fraction that is calculated for very high total discharge enthalpies. The pattern of enthalpy
comparisons in Figure 7F indicates boiling in the well, but high-temperature condensation is inferred in the well and in the reservoir
since 2010. There is an increasing trend in CO2 (Figure 7G) due to boiling, while the reservoir liquid saturation has significantly
decreased, with the exception of 2007 data by which time some recharge might have reached the well (Figure 7H). The boiling
process can also be identified through the pressure-enthalpy plot in Figure 8A, in which the data trend over time has a direction
toward higher enthalpies. In this case, the well has reached to a steam fraction of 0.9 in 2011 because of high enthalpy. In an
enthalpy-chloride diagram (Figure 8B), data from well Az-26 plot on the boiling line (enthalpy increases as chloride decreases),
trending towards the enthalpy of vapor. Az-18 and 46 in the south and Az-5, 13, 19, 28, 32, 43 and 48 in the north are the other
wells with significant boiling. Moderate boiling was identified in wells Az-22 and 62.
Boiling with steam loss consists of decrease in the boiling rate and it was identified in very few wells including Az-23 and 28A.
Well Az-28A was chosen to illustrate this process. Enthalpy versus chloride and well bottom pressure versus well bottom enthalpy
for well Az-28A are given in Figures 9A and B, respectively. On an enthalpy versus chloride plot, boiling with steam loss can be
recognized by a data trend towards lower enthalpy and higher chloride values (Figure 9 A). The trend of decreasing enthalpy over
time can also be seen in Figure 9B.
Time series of mass flow rates produced, total discharge enthalpy and estimated silica and Na/K enthalpies, CO2, and liquid
saturation at reservoir for well Az-28A are given in Figure 10. Steam production starts to decrease almost in the beginning (Figure
10A) and continues to decreases over time while liquid production varies intermittently possibly due to the entry of returns from
reinjection. Enthalpy comparison pattern indicates that boiling decreases over time as confirmed by the data trend in Figure 10B.
As boiling rate decreases, CO2 decreases with decreasing enthalpy (Figure 10C). Although dilution is not clear, it is possible that
well Az-28A is producing a condensed fluid from the boiling of degasified fluid (reinjection) as indicated by its highly depleted
isotope composition (Figure 8A). This well maintains a very high liquid saturation (> 0.8; Figure 10D). It is possible for this well
that the pressure is controlled by a constant pressure boundary; the well bottom pressure is gradually stabilized and the boiling front
5
Arellano et al.
does not expand anymore. The temperatures tend to equilibrate within the stabilized boiling zone; ceasing the heat transfer from the
rock to the fluid and causing a decrease in the excess enthalpy to negligible levels.
50
300
Flowrate
(ton/h)
200
Wellhead
pressure (bar)
Total (A)
Liquid
Steam
AZ-26
100
30
20
10
0
0
120
80
40
1600
CO2 (mol ‰)
Well bottom
enthalpy (kJ/kg)
1500
1000
AZ-26
25.00
15.00
Reservoir
Total discharge
5.00
(D)
3000
2000
Total discharge
Separated water
1000
0.90
Liquid saturation
(fraction)
AZ-26
4000
(G)
35.00
500
Chlorides (mg/kg)
ENT CCG
ENT SiO2
ENT DT
1200
45.00
(C)
AZ-26
(F)
2000
800
2000
5000
in
AZ-26
2400
0
3000
2500
10
5
0
2800
(B)
AZ-26
Enthalpy (kJ/kg)
Well bottom
pressure (bar)
160
(E)
AZ-26
40
(H)
AZ-26
0.70
0.50
0.30
0.10
0
1975
1980
1985
1990
1995
2000
2005
2010
1975
2015
1980
1985
1990
1995
2000
2005
2010
2015
Year
Year
Figure 7: Time series of (A) mass flow rates, (B) well bottom pressure, (C) well bottom enthalpy, (D) chlorides, (E) wellhead
pressure, (F) comparison of total discharge enthalpy and enthalpy estimations by geothermometers, (G) CO2, and
(H) reservoir liquid saturation of well Az-26, which is used to illustrate boiling with gaining steam.
3000
AZ-26
(A)
(B)
AZ-26
Vapor
11
10
6
9
100
Enthalpy (kJ/kg)
Well bottom pressure (bar)
2500
300°C
93
95
94 97
99
96
2
0
98 3
4
7 8
6
5
9
10
5
8
2000
7
4 86
3
1
2
98 97
1500
94
93
95
90
96
85
Boiling line
1000
11
Steam loss
500
200°C
10
0
500
1000
1500
2000
2500
3000
3500
Well bottom enthalpy (kJ//kg)
0
1000
2000
3000
4000
Total discharge chlorides (mg/kg)
Figure 8: (A) Well bottom pressure versus well bottom enthalpy and (B) enthalpy versus chloride for well Az-26.
6
5000
Arellano et al.
3000
Az-28A
Az-28A
(A)
(B)
Steam
2500
7
9
8
11
1500
Steam loss
Boiling line
1000
100
300°C
8
9
11
500
7
6
10
200°C
0
5
4
x=0.8
10
x=0.6
6
2000
x=0.4
Enthalpy (kJ/kg)
3 5
x=0.2
Well bottom pressure (bar)
4
10
0
1000
2000
3000
4000
Total discharge chlorides (mg/kg)
500
5000
1000
1500
2000
2500
Well bottom enthalpy (kJ/kg)
3000
3500
Figure 9: (A) Enthalpy versus chloride and (B) well bottom pressure versus well bottom enthalpy for well Az-28A.
Flowrate (ton/h)
50
40
(A)
AZ-28A
30
20
10
Total
0
CO2 (mol ‰)
Enthalpy (kJ/kg)
2400
Li quid
(B)
AZ-28A
2000
Steam
ENT CCG
ENT SiO2
1600
ENT DT
1200
800
1.00
(C)
Reservoir
Total discharge
AZ-28A
0.60
Liquid saturation
(fraction)
0.20
0.90
(D)
AZ-28A
0.70
0.50
0.30
0.10
2000
2005
Year
2010
2015
Figure 10: Time series of (A) mass flow rates, (B) comparison of total discharge enthalpy and enthalpy estimations by
geothermometers, (C) CO2, and (D) reservoir liquid saturation of well Az-28A, which is used to illustrate boiling
with steam loss.
4.2 2005-2011 Reservoir Evolution
The occurrence of the described reservoir exploitation-related processes has changed the distribution of geochemical indicators
through time as follows. In Figure 11 the temperature distributions of the Los Azufres reservoir for 2005 and 2011 are given.
Relatively more important changes include temperature variations in the North Zone, where the 300°C contour is located towards
east in comparison to 2005. In 2011, the temperature iso-contours are aligned in an approximately N–S direction in the center of the
North Zone with a decreasing trend towards west, where injection wells are located. In 2011, the 290°C contour lies along both
zones, whereas minimum temperatures are located in the west of the South Zone, where injection wells are located.
7
Arellano et al.
52D
Maritaro
61
15
54R
El Tule
El Chino
3
61
San Alejo
Fault
54R
El Tule
Well considered
2A
46
Drilled well
47
34
17 36 38
33
7
6
16
Tejamaniles
7A
16AD
24
Well considered
31
18
34
17 36 38
46
26 31
7
6 33
16
39
Tejamaniles
31D
11
24
7A
16AD
Los Azufres
Los Azufres
El Chinapo
50
2A
2
Drilled well
31D
11
El Chinapo
50
58D
35
55
47
37
8
Injection well
26
39
1A
22 62
1
12
18
25
12D
35
55
2
San Alejo
23
Fault
37
8
Injection well
9AD
67D
25
1A
22 62
1
9 9A
56
El Chino
3
12D
12
10
15
9AD
67D
23
Maritaro
9 9A
56
53
27A 27
42
Espinazo del Diablo 4853R
5
19
57
21
43
14D
19D 64
65D
66D 28A 32
29R
13
4
30 28
44
45
67
69D
41
49
52D
53
27A 27
42
Espinazo del Diablo 4853R
5
57 19
21
43
14D
19D 64
65D
66D 28A 32
29R
13
4
30 28
44
45
67
69D
41
49
58D
10
300 (°C)
300 (°C)
280 (°C)
280 (°C)
Reservoir temperature
(°C) 2011
Reservoir temperature
(°C) 2005
1 km
0
20
1 km
0
20
Figure 11: 2005 and 2011 distributions of reservoir temperature in the Los Azufres geothermal field.
Total chloride discharge distributions for 2005 and 2011 are given in Figure 12. In both years, maximum chloride discharge is
found to the west of both zones due to injection effects. Considering that two-phase wells receive some liquid from reinjection in
Az-2A, Az-33, Az-42, Az-4, diluted fluids were observed in 2011 to the east of the South Zone and in the center of the North Zone.
52D
Maritaro
61
15
54R
El Tule
53
27A 27
42
Espinazo del Diablo 4853R
5
57 19
21
43
14D
19D 64
65D
66D 28A 32
13 29R
4
30 28
44
45
67
69D
41
49
56
El Chino
3
Fault
54R
El Tule
2A
2
Drilled well
23
Fault
1A
22 62
1
46
24
Well considered
39
Los Azufres
El Chinapo
50
10
35
55
18
17 36 38 34
46
26 31
7
6 33
16
39
Tejamaniles
31D
11
24
7A
16AD
Los Azufres
31D
11
1A
22 62
1
47
2A
2
Drilled well
58D
El Chinapo
58D
1600 (mg/kg)
1600 (mg/kg)
400 (mg/kg)
400 (mg/kg)
Total discharge chloride
(mg/kg) 2005
20
San Alejo 25
37
8
Injection well
26 31
33
9AD
12
18
17 36 38 34
9 9A
12D
35
55
47
7
6
16
Tejamaniles
7A
16AD
50
56
67D
San Alejo 25
37
8
Injection well
53
27A 27
42
Espinazo del Diablo 4853R
5
57 19
21
43
14D
19D 64
65D
66D 28A 32
13 29R
4
30
44
28
45
67
69D
41
49
El Chino
3
12D
Well considered
15
9AD
12
10
61
9 9A
67D
23
52D
Maritaro
Total discharge Cl
(mg/kg) 2011
0
1 km
20
0
1 km
Figure 12: 2005 and 2011 distributions of total chloride discharge in the Los Azufres geothermal field.
The N2/Ar molar ratio distributions for 2005 and 2011 are given in Figure 13. Contours for 2005 indicate that the entry of
condensed steam originated by boiling of de-gasified fluids (reinjection) is localized from the north central part of the North Zone
to the center of the field, where N2/Ar < 38. N2/Ar contours for 2011 indicate the presence of fluids generated by boiling of degasified brines (injection) in a large portion of the North Zone, in the center part, and in the northeast of the South Zone. It is
evident in Figure 13 that production of reinjection returns in the form of steam or condensed steam generated by boiling of injection
fluids seem to be more important due to exploitation (Barragán et al., 2013).
8
Arellano et al.
27A27
53
42
Espinazo del Diablo 4853R
5
19
57
21
43
14D
19D
28A 32
29R
13
4
30
28
45
69D
41
49
67
Maritaro
61
15
La Cumbre
El Tule
61
15
La Cumbre
El Tule
9 9A
56
9AD
El Chino
3
9AD
67D
San Alejo 25
23
12D
Well considered
Injection well
Drilled well
N2/Ar 2005
1A
22 62
1
12
Fault
8
2A
2
7
46
16D
Tejamaniles
7A
17 36 38 34
Well considered
18
24
Injection well
31
Drilled well
39 Puentecillas 31D
11
Los Azufres
1A
1 22 62
12
Fault
26
6
San Alejo 25
12D
35
55
47
37
9 9A
56
El Chino
3
67D
23
27A27
53
42
Espinazo del Diablo 4853R
5
19
57
21
43
14D
19D
28A 32
29R
13
4
30
28
45
69D
41
49
67
Maritaro
37
8
2A
2
7
46
16D
Tejamaniles
7A
17 36 38 34
24
50 (mol ratio)
50 (mol ratio)
120 (mol ratio)
120 (mol ratio)
N2/Ar 2011
31
39 Puentecillas 31D
11
Los Azufres
El Chinapo
1 km
18
26
6
El Chinapo
0
35
55
47
0
1 km
Figure 13: 2005 and 2011 distributions of N2/Ar in the Los Azufres geothermal field.
5. CONCLUSIONS
The analysis of chemical and production data for the Los Azufres geothermal field allowed identification of the main reservoir
processes related to exploitation. According to these results, after 27 years of commercial exploitation, very moderate decline in
production parameters were estimated due to an efficient artificial and natural recharge to the reservoir. The extraction and
reinjection of fluids from and to the reservoir have mainly induced the occurrence of two physical processes: (a) the production of
returns from reinjection, and (b) boiling. Production of reinjection returns either as liquid or steam was relatively more important
than the production of steam and sometimes condensed steam from boiling of injection fluids. It is reported for the first time that
when returns from injection consist of condensed steam, salinity decreases in two-phase wells and isotopic compositions become
relatively-depleted. In order to identify the production of such reinjection returns, analysis of gas data (in particular the N2/Ar molar
ratio) is necessary. The other process induced by exploitation was boiling, which can take place either with gaining steam, in which
rates of boiling over time tend to increase, or with steam loss, in which the boiling rates over time tend to decrease.
It is recommended to continue the monitoring of geochemical and production data to trace the evolution of the reservoir in response
to exploitation and to support decisions on extending the resource lifetime.
ACKNOWLEDGMENTS
Results of this work were part of the project “The Response of the Los Azufres Geothermal Reservoir to Exploitation” conducted in
2012 and funded by the Gerencia de Proyectos Geotermoeléctricos (Comisión Federal de Electricidad, CFE). The authors want to
express their gratitude to the authorities of CFE for authorizing this publication. Recognition is given to the technical staff of the
Residencia Los Azufres (CFE) who provided support for this work.
REFERENCES
Arellano, V.M., Torres, M.A., and Barragán, R.M.: Thermodynamic Evolution of the Los Azufres, Mexico, Geothermal Reservoir
from 1982 to 2002, Geothermics, 34, (2005), 592-616.
Arellano, V.M., Barragán, R.M., López, S., Paredes, A., and Aragón, A.: Respuesta del Yacimiento de Los Azufres a la
Explotación 2003-2011, Final Report, IIE/11/14283/02F, Instituto de Investigaciones Eléctricas for the Comisión Federal de
Electricidad, Cuernavaca, Morelos, México (2012).
Arnórsson, S.: Injection of Waste Geothermal Fluids: Chemical Aspects, Proceedings, World Geothermal Congress, KyushuTohoku, Japan (2000).
Arnórsson, S., and D´ Amore, F.: Monitoring of Reservoir Response to Production, in: Isotopic and Chemical Techniques in
Geothermal Exploration Development and Use, Ed. International Atomic Energy Agency, Vienna, (2000).
Barragán, R.M., Arellano, V.M., Portugal, E., and Sandoval, F.: Isotopic (δ18O, δD) Patterns in Los Azufres (Mexico) Geothermal
Fluids Related to Reservoir Exploitation, Geothermics, 34, (2005), 527-547.
Barragán R.M., Arellano, V.M., Aragón, A., Martínez, J.I., Mendoza, A., and Reyes, L.: Geochemical Data Analysis (2009) of Los
Azufres Geothermal Fluids (Mexico): Proceedings Water Rock Interaction, Birkle & Torres–Alvarado (Eds.), Taylor &
Francis Group, London, (2010), 137- 140.
9
Arellano et al.
Barragán, R. M., Arellano, V. M., Mendoza, A., and Reyes, L.: Chemical and Isotopic (18O, D) Behavior of Los Azufres
(Mexico) Geothermal Fluids Related to Injection as Indicated by 2010 Data, Geothermal Resources Council Transactions, 35,
(2011), 603-608.
Barragán, R. M., Arellano, V. M., Aragón, A., Torres, R., Reyes, N., López, S., and Patiño, A.: Estudio Isotópico de Fluidos de
Pozos Productores y de Reinyección del Campo Geotérmico Los Azufres, Mich. 2013, Final Report, IIE/11/14154/I 01/F,
Instituto de Investigaciones Eléctricas for the Comisión Federal de Electricidad, Cuernavaca, Morelos, México (2013).
Fournier, R.O., and Potter II, R.W.: A Revised and Expanded Silica (quartz) Geothermometer, Geothermal Resources Council
Bulletin, (Nov. 1982), 3-12.
Goyal, K.P., Miller, C.W., and Lippmann, M.J.: Effect of Measured Wellhead Parameters and Well Scaling on the Computed
Downhole Conditions in Cerro Prieto Wells, Proceedings, 6th Workshop on Geothermal Reservoir Engineering, Stanford
University, Stanford, CA (1980).
Iglesias, E.R., Arellano, V. M., Garfias, A., Miranda, C., and Aragón, A.: A One-Dimensional Vertical Model of the Los Azufres,
México, Geothermal Reservoir in its Natural State, Geothermal Resources Council Transactions, 9 (2), (1985), 331-336.
Iglesias, E.R., Flores-Armenta, M., Torres, R.J., Ramírez-Montes, M., Reyes-Piccaso, N., and Reyes-Delgado, L.: Estudio con
Trazadores de Líquido y Vapor en el Área Tejamaniles del Campo Geotérmico de Los Azufres, Mich., Geotermia, Revista
Mexicana de Geoenergía, 24 (1), (2011), 38-41.
Nieva, D., and Nieva, R.: Developments in Geothermal Energy in Mexico-Part Twelve: A Cationic Geothermometer for
Prospecting of Geothermal Resources, Heat Recovery Systems & CHP, 7, (1987), 243-258.
Nieva, D., Verma, M., Santoyo, E., Barragán, R.M., Portugal, E., Ortíz, J., and Quijano, L.: Chemical and Isotopic Evidence of
Steam Upflow and Partial Condensation in Los Azufres Reservoir, Proceedings, 12th Workshop on Geothermal Reservoir
Engineering, Stanford University, Stanford, CA (1987).
Siega, F.L., Salonga, N.D., and D’Amore, F.: Gas Equilibria Controlling H2S in Different Philippine Geothermal Fields,
Proceedings, 20th Annual PNOC-EDC Geothermal Conference, Manila, Philippines (1999).
Truesdell, A.H., Lippmann, M., Quijano, J.L., and D’Amore, F.: Chemical and Physical Indicators of Reservoir Processes in
Exploited High-Temperature, Liquid-Dominated Geothermal Fields, Proceedings, World Geothermal Congress, Florence,
Italy (1995).
10