NOTICE CONCERNING COPYRIGHT RESTRICTIONS

This document may contain copyrighted materials. These materials have been made available for use in research, teaching, and private study, but may not be used for any commercial purpose. Users may not otherwise copy, reproduce, retransmit, distribute, publish, commercially exploit or otherwise transfer any material.

The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material.

Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specific conditions is that the photocopy or reproduction is not to be "used for any purpose other than private study, scholarship, or research." If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of "fair use," that user may be liable for copyright infringement.

This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law.

Geothermal Resources Council TRANSACTIONS, Vol. 19, October 1995

SEISMICITY RECORDED IN THE GEOTHERMAL FIELD DURING 1994-1995

Hubert Fabriol’, Luis Munguia Orozco’, Hector Lira2 and Gilbert0 Guillen2

‘CICESE, Divisidn Ciencias de la Tierra, Aptdo. Postal 2732,Ensenada 22830 B.C., MOxico *Cornision Federal de Electricidad, Residencia General de Cerro Prieto, Residencia de Esudios, B.C, MOxico

ABSTRACT world. The field extends over a 12 km2 area, with more than Seismicity was monitored from August 1994 to April 1995 210 wells drilled among which about 110 are producing in the Cerro Prieto Geothermal Field (), with 4 analog (Gutierrez and Rodriguez, 1994). Reinjection of waste brine seismographs. That allow the initiation of a catalog of the started in 1989 to avoid discharge of residual brines into seismicity occurring within the production zone. Although surface water bodies and streams. At present (1995), more seismic activity is scattered in time, up to 118 earthquakes than 42% of the extracted fluid is reinjected. were located during this period of time. Magnitudes range The geothermal field is situated on the sediments which from -1 to 4,O and five of these earthquakes had magnitudes fill the delta. The liquid dominated reservoir is Mdr3. Earthquake locations showed a significant change in located between 1500 and 4000m in sandstones, sandy pattern from November to December 1994. It is proposed that shales and shales (Lippmann and MaMn, 1987). The heat the October-November 1994 seismicity, located to the source is attributed to a spreading center where the crust is southeast of the field and with depths up to 12-14 km, is pulled apart between Cerro Prieto fault and Imperial fault and . related to the spreading center located between Cerro Prieto igneous rocks are being emplaced. The Cerro Prieto pull and Imperial faults. Seismicity occurring after and until April apart basin is part of the Salton Trough- 1995 is shallower and situated close to the Droduction area. system of spreading centers and transform faults which link Up to now, it is too early to state whether the latter is “normal” the to the San Andreas fault system. and linked to the spreading center and strike-slip faults or Seismological studies were carried out in the area since exploitation-related. A longer monitoring would be necessary 1969, first aimed to investigate the deep structure of to discriminate both mechanisms. Valley (Lomnitz et al., 1970), then focused on the geothermal area as an exploration tool (Reyes et at., 1979, Albores et al., INTRODUCTION 1980. Majer and McEvilly, 1981, 1982). The high level of seismic activity allowed to delimit the spreading center and Seismicity and geothermics are tightly linked. First, the extremities of Cerro Prieto and Imperial faults. Moreover, because areas of high heat flow are tectonically active, which Majer and McEvilly (1982) reported an increase of the number means high level of seismic activity, second because several of earthquakes of magnitude M~rlin the geothermal examples exist of seismicity generated by geothermal fluid production area between 1978 and 1981. They interpreted it exploitation (e.g. The Geysers, see Romero et al., 1994). A as a consequence of production increase. continuous monitoring of seismicity was undertaken beginning in August 1994 at Cerro Prieto geothermal field (Baja Continuous, local recordings indicated small seismic California, Mexico) with a small array of analog seismographs activity during 1983-1985, and increases of seismic activity in order to study the temporal and spatial distribution of local during 1987 and 1988, which concentrated in swarms (Reyes earthquakes. This is the first attempt to continuously monitor et at. 1989). The regional seismic network (RESNON) has the Cerro Prieto geothermal field seismicity since 1987. A recorded seismicity since 1977. However recording is not telemetered digital network is planned to be setup in October continuous, and earthquake and magnitudes are available 1995 and will replace the analog recorders. In this paper we only since 1987. An increase in the number of earthquakes present the data analysis of 8 months of recording, from can be observed in Cerro Prieto area in 1978, 1980, 1987 and August 1994 to April 1995. Histograms of occurrence of 1992-1993. More recently, Glowacka and Nava, (1993), earthquakes and maps of epicenters are interpreted assuming studied the correlation between seismicity and fluid extraction, that earthquakes in this particular area are very common and using data from USGS-Caltech catalogs for years 1973-1991. that the recording array does not allow location errors greater According to these authors, probabilistic analysis can not than 2 km. reject the hypothesis that strong earthquakes could be triggered by a production increase. The Cerro Prieto Geothermal Field is located in the Mexicali Valley, at the head of the Gulf of California, between From 1987 until now, no systematic monitoring of the the southeast end of the Imperial Faults and the northern geothermal area was carried out, except for a three month extremity of the Cerro Prieto Fault (Figure 1). Exploitation study with seven digital seismological stations from January started in 1973 and, at present, the nominal production is 620 to April 1993 (Fabriol and Rebollar, 1993), focused on MWe, which corresponds to the second largest field in the monitoring the effect of fluid reinjection. More than 250

387 Fabriol and Mungula microearthquakes were recorded in and around the in the right part of the curve, due to the length of recording production area and the duration magnitudes ranged from period respect to the Occurrence of earthquakes of magnitude Md=-1 to Md=2.6. The period of recording was definitely too Mdr2.5. Although in volcanic or geothermal areas 6 is short to point out clear correlations between fluid injection and supposed to be larger or equal to 1, a 6-value of 0.82 is usual seismic events rate. in areas of important tectonic activity and the regression line agrees with the occurrence of earthquakes of magnitude On August 20, 1994, four analog seismographs were Md=4, at least. setup to start the continuous monitoring of the geothermal area, 9 days after a magnitude M~=4.6earthquake occurred. SPATIAL DISTRIBUTION OF EARTHQUAKES This event was located beneath the south part of the As said before, 120 earthquakes were located using the Evaporation Pond (Laguna de Evaporacion), at about 5 km HYPO71 (Lee and Lahr, 1975) algorithm. Many layered depth. From August 1994 to May 1995, five more earthquakes models are available to simulate P-wave velocity in this part of magnitude MD23 were recorded at distances shorter than of Mexicali Valley (see Gonzales, 1986 for a review). We 5 km from the power plants and depths shallower than 5km. chose the model used by the Strong Motion Network of the The local array is made up of four vertical seismometers Northwest of Mexico (Table 1) and a value of VdVs=l.73 for connected to an analog drum seismograph, distributed around hypocenter determination. As a matter of fact, some of the the geothermal area (Figure 2). The array is complemented by Md23 earthquakes shown here are located using the the permanent digital station CPX of RESNON (three accelerometers arrival times of stations GEO and DEL component), located on the Cerro Prieto Volcano, the analog (Munguia, 1995). seismograph of QKP and, as for the largest earthquakes, the Velocity (km/s) Depth (km) strong motion recorders of the Strong Motion Network of the 1.70 0.00 Northwest of Mexico. Analog recorders are certainly less 2.00 0.10. sensitive and accurate than digital ones : precision in arrival 2.30 0.73 time readings is not better than 0.05 s. Nevertheless, they 2.60 1.75 have the advantage to record everything, even if noise is 3.00 2.94 present as it happens very often in the sedimentary-filled 5.00 5.62 Valley of Mexicali. 6.00 10.00 7.80 20.00 TIME AND MAGNITUDE DISTRIBUTION OF SEISMICITY From August 20, 1994 to May 10, 1995, more than 225 Table 1 P-wave velocity model used for hypocenter earthquakes and micro earthquakes were recorded in the determinations (based on a study by McMechan and Mooney, geothermal area, among which more than 50% were located. 1980) Figure 3 shows the histogram of occurrence of events. The The mean errors in hypocenter determination (Le. the seismicity level is relatively low and the mean rate of events average of the errors of the 118 located earthquakes) are as is less than one percent per day. Rate of events reported follows: either by Majer and McEvilly in 1981, or by Fabriol and -rms residual: 0.24 f 0.16 s, Rebollar in 1993, are much more larger : 7 events of ML~1 -horizontal error: 2.0 f 1.9 km, per day, and 3-4 events of MdrO per day, respectively. Even -vertical error: 2.6 f 2.8 km. if we consider that analog recorders are less sensitive and Those figures given an idea about the limitations of the that Majer and McEvilly used a downhole sensor in M6 well, monitoring array to accurately locate earthquakes. it appears that the background level of seismicity is quite low. Nevertheless, location errors are at least half of what it is Moreover, swarms are not frequent, except during isolated expected with the permanent arrays of RESNOM or Caltech, crises as on October, 23-27 1994 or January, 10-11 1995. and of course, the magnitude threshold is lowered one order Felt earthquakes, e.g. January 11, 1995, Md=4.0, are not of magnitude, i.e. from M=2 down to less than M=l. followed by aftershock activity. That is quite different compared with the important aftershock activity of the large The whole seismicity was split in two-month periods to earthquakes in the area (Victoria, 1980, June 9, M=6.1, or show how epicenters are moving with time (Figures 6a to 6d). Cerro Prieto 1987, Feb. 7, M=5.4). At the end of August and during September 1994 (Figure 6a), few events are located in the geothermal area, the big square Magnitude of events was calculated using the following under M-6 well is the epicenter of the M~=4.6earthquake of duration formula (Gonzalez, pers. Comm.): August 11. It is likely that a quiescent time followed this Md=2.24 x lOglO(d) - 0.85 important release of stress in the area. The October- The magnitudes measured on analog seismograms with November period (Figure 6b) starts with a magnitude Md=3.4 this formulae range from -1.5 up to 4.0, which lies within earthquake situated 2 km south from Cerro Prieto Volcano, magnitude estimations of Caltech and RESNON catalog, as but afterwards, all the activity is disseminated southeast from for earthquakes with magnitude Mdrl Figure 4 shows the the geothermal field. Depths range from 2 to 14 km. This swarm like event, which exhibits relatively deep seismicity, is histogram of maximum magnitude recorded per day. The 6 in agreement with what has been reported before about earthquakes of magnitude Mdr3 are highlighted and we can earthquakes in the spreading center and along the south end observe again the characteristics noted above, namely: low Imperial Fault (Reyes et 1979, Albores et 1980). level activity, few swarms and lack of an important foreshock of al., al., These authors explained that deep swarms are the result of or aftershock activity. injection at depth of magmatic fluid in dikes within the brittle The 6 value of the Gutenberg-Richter relations was crust, according to Hill (1977) model. This could be the heat calculated with the 223 recorded events. Figure 5 shows that source of the boiling plume which rises from the northeast the catalog is far from complete since there are many holes and discharges to the west, supplying hot fluid to the

388 Fabriol and Munguia

geothermal reservoir (Elders et al, 1984). trend. In the same way, it could be an effect of the three- Then, we observe a change during December 1994- dimensional structure of the area. Anyhow, if the observed January 1995 period (Figure 6c). Epicenters are moving effect is real, it is likely that seismicity occurs mainly in the top towards the production area and the Evaporation Pond as of the basement, which underlies the sediments of the well as towards the southeast of Cerro Prieto Volcano. Two Colorado Delta Riva and deepens towards the core of the important earthquakes occurred: one is located north from the spreading center. That explains why seismicity and the field (square No. 1, Md=3.6, 31/12/95) and the other one bottom of the reservoir follow parallel lines. With more data it halfway between CPI and CPlll power plants (square No. 2, would be also possible to delineate vertically the brittle-ductile M=4.0, 11/01/95). Some damage has been reported in the transition where earthquakes vanish and an insight into deep power plants and the administrative buildings of CFE. As for temperatures distribution could be developed. Furthermore, the other magnitude Md>3 earthquakes, none or few it would be interesting to direct the heat source of the boiling aftershocks or foreshocks were observed. Nevertheless, plume, for which Elders et at. (1984) proposed a funnel station BIL located inside the Evaporation Pond recorded shaped basalt intrusion at a depth of 5-6 km, eastward of the several of them. Hypocenters are generally shallower than 5-6 present reservoir. km, which can explain the lack of recordings of foreshocks or The perpendicular NW-SE vertical section (Figure 8) aftershocks in the other stations, since seismic energy is shows the same trend, although less prominent: earthquakes rapidly absorbed in the sediments. During the more recent focii deepen from northwest towards southeast, pointing at period (February-April 1995, Figure 6d) seismicity is still the spreading center. This trend corresponds to a deepening concentrated in and around the production area, but with a of the basement in the same direction, through normal faults trend to migrate towards south and southeast. dipping towards east and southeast, as pointed out by Projections of hypocenters onto vertical sections help to Halfman et at. Study (1984). Again the difference between visualize changes in spatial distribution of seismicity. Planes October-November and December-April seismicity is of projection are NE-SW (Figure 7) and NW-SE (Figure 8), straightforward: the second is shallow and located under the respectively parallel and perpendicular to Hidalgo Fault (see geothermal area, meanwhile the first is deeper and shifted Figure l), according to the main tectonic directions in the towards the southeast. area. The change from the October-November period to the INTERPRETATION December-April one is clear on figure 7, since the first period We have already made a clear distinction between the events, represented by triangles, are located on the east part seismicity occurring close to the geothermal area and of the section, meanwhile the following events (plain squares seismicity occurring to the southeast. The second is related and circles) are located beneath the geothermal field at to the spreading center meanwhile the first occurs beneath shallower depths. The magnitude ML>~events, recorded by the reservoir or close to it. Both are not uncommon and were the Strong Motion Network, are represented by stars. It is reported in several previous studies. Nevertheless, the worthwhile to note that their depths range from 3 to 5 km, present study is the first systematic one of the seismicity except for three events. This could be due to the fact that the occurring inside the production area, it is necessary to brittle-ductile transition prevents the occurrence of magnitude investigate whether it is related to the normal "background" hA~>3 earthquakes at depth. In other words, as the temperature increases beneath 6-7 km depth it limits the seismicity in the area or to the geothermal resource. Therefore, we propose three possible mechanisms to explain propagation of large ruptures at depth, as the seismic earthquakes generation in the Cerro Prieto area: moment, and consequently the magnitude, is proportional to fault area. Source of parameters of the January 11, 1995 1- Release of tectonic stresses either within the spreading Md=4 earthquake, were estimated applying the Brune model center or along the normal faults identified within the field or (I 970) to the spectra of the strong motion records: seismic the strike-slip Cerro Prieto fault. moment M, = 2.3~1023dyne-cm., source radius r= 1-1.5 km That is the origin of "background normal" activity. Since it and stress drop ~0=71bars. The seismic moment is was reported prior to geothermal exploitation, part of the equivalent to Mw=4.85, which is much more than the present seismicity must be of what kind. For example, the estimated duration magnitude. Since the calculated depth of August 11, 1994, M~=4.6earthquake is located right in the the focus is 3.8 km, and assuming the fault plane is vertical, intersection of Cerro Prieto and Hidalgo fault. The puzzling it is likely that the source extended from the bottom of the problem is to distinguish between background and reservoir down to the top of the ductile-brittle transition. The exploitation-related seismicity, since both have a similar stress drop value is definitely high, that could explain the lack location. A further insight in focal mechanisms could help in of foreshocks or aftershocks and the relatively low b-value. determining it. Focal mechanisms should be of the normal The bottom of the geothermal reservoir is schematized by faulting kind in the spreading center and of strike-slip kind a gray line : its west limit is well known, while towards the close to Cerro Prieto or Imperial faults (Albores et at., 1980). east, there source deepens below a depth of 3 km, (Halfman Up to now, we do not know what the focal mechanisms of et al., 1993) and is still unexploited. We can also observe on exploitation-relatedearthquakes are. figure 7 an increase of hypocenter depths from SW towards 2- Thermal cooling of the basaltic intrusions injected at the NE, roughly along a deeping line which starts at 2 km depth, bottom of the spreading center. southwest from Cerro Prieto Volcano, and ends at 14 km depth, beneath Ejido Nuevo Lebn. We need more data and Even if this process is real and could generate seismicity, more precise determination of depths to determine if this it is not enough to explain the occurrence of magnitude M24 pattern is real or an artifact due to localization algorithm. The earthquakes. Moreover the b-value of Gutenberg-Richter choice of the P-wave velocity model could also be important, relations should be closer to 1, as in volcanic areas, where but a trial with another velocity distribution preserved this stress are released through a larger number of micro earthquakes.

389 Fabriol and Munguia 3- Fluid flow inside or around the reservoir, which means that changes in injection flowrates. pressure gradients due to production or reinjection could In other respects, earthquake spatial distribution could induce seismicity. provide information about the deep structure of this particular First, if real, it is difficult to establish a clear distinction area and the location of the heat source. Vertical distribution between seismicity induced either by production or by of hypocenters show a deepening of earthquakes towards the reinjection. Withdrawal of fluid is supposed to increase friction core of the spreading center. This needs to be confirmed by along fractures, and therefore to impede earthquakes, but, additional data. throughout time, contraction of the reservoir could induce ACKNOWLEDGMENTS earthquakes (see Grasso, 1990, for a review of reservoir induced seismicity). At the contrary, injection of fluid in a This study was made possible owing tu the invaluable fractured medium decreases effective stress along planes of help of Comision Federal de Electricidad, Residencia de weakness and enables stress release by earthquakes. In the Estudios in Cerro Prieto, through the contract CLSO-O01/95. Cerro Prieto case, after 20 years of exploitation we would We thank in particular Ing. Hector Gutierrez Puente from expect both mechanisms to occur. Then, to discriminate CICESE, we received help from Javier Gonzalez, who applied background from exploitation-relatedseismicity we need to the QKP station recordings, and from lgacio Mendez and establish close correlations in space and time, since Francisco Farfan, who supplied the CPX arrival times. background seismicity is a constant fact in the area. As for the first point, it is obvious that horizontal and vertical location errors of 2 km and 3 km, respectively, are serious limitations REFERENCES to correlate precisely hypocenters with the reservoir and injecting wells. Consequently, we can state that seismicity Albores, A., Reyes, A., Brune, J.N., Gonzalez, J., Garcilazo, has shifted after November 1994 from the southeast towards L. And Suarez, F., 1980, Seismicity studies in the region the production area, but without any precise correlation with of the Cerro Prieto Geothermal Field,: Geothermics, Vol. one well in particular. 9, 65-77. In respect to time correlations, available data from CFE do Brune, J.N., 1970, Tectonic stress and the spectra of seismic not show large variations of the total reinjected mass shear waves from earthquakes: J. Geophys. Res., Vol. throughout years 1993 and 1994. The percentage of 75,4997-5009. reinjected fluid in respect to produced mass volume is stable since 1993 and ranges from 42 to 45% (Gutierrez et al., Elders, W.A., Bird, D.K., Williams, A.E. and Schiffman, P., 1994). Nevertheless, there is an annual change of total 1984, Hydrothermal flow regime and magmatic heat reinjected volume in winter when evaporation in the source of the Cerro Prieto Geothermal system, Baja Evaporation Pond is less and must be compensated by an California, Mexico: Geothermics, Vol. 13, No. ?4,27-47. increase in reinjection flowrates. The observed shift of seismicity around the production area at the beginning of the Fabriol, H. And Rebollar, C.J., 1993, Monitoreo de la winter of 1994 could be explained by this seasonal change in microsismicidad en el Campo Geot4rmico de Cerro Prieto, reinjection pattern. But to confirm this, we need to observe a (): lnforme T&nico, CICESE, 22 pp. decrease in seismicity in the production area during the summer 1995 and at least to observe the same effects during Glowacka, E. And Nava, F.A., 1993, On a possible relation winter 1995 and summer 1996. between fluid extraction at Cerro Prieto geothermal field CONCLUSION and major earthquakes in Mexicali Valley: Seism. Res. Monitoring seismicity during eight months with 4 analog Lett., Vol. 64, 46. seismographs in the Cerro Geothermal Field allowed to develop a catalog of earthquakes occurring close to and Gonzalez, J., 1987, Sismoiectbnica del Valle de Mexicali, within the production zone. Although seismic activity is MScthesis, CICESE. scattered in time, up to 118 earthquakes were located with 2km horizontal error and 3 km vertical error. Duration Grasso, J.R., 1992, Mechanics of seismic instabilities induced by the recovery of hydrocarbons: Pageoph., Vol. 139, pp. magnitudes range from -1 to 4.0, and 5 earthquakes had 507-534. magnitudes Md23. The January 11, 1995 earthquake had a magnitude Md=4.0 (Mw=4.85, estimated from source parameters) and was located midway between CPI and CPlll Gutierrez Puente, H. and Rib6 MuAoz M., 1994, Reinjection power plants. experiences in the Cerro Prieto Geothermal Field: Geothermal Resources Council Transactions, Vol. 18, pp. Earthquake locations showed a significant change in 261-267. . . pattern from November to December 1994. We propose that October-November seismicity, located to the southeast of the Gutierrez Puente, H. And Rodriguez, M., 1994, Cerro Prieto field and at depths up to 12-14 km, is related to the core of steam production forecasting: Geothermal Resources the spreading center, where magma-filled dykes from the heat Council Transactions, Vol. 18, pp. 599-603. source of the geothermal system. The seismicity occurring after and until April 1995 is shallower and situated close to the Halfman, S.E., Lippmann, M.J., Zelwer, R. and Howard, J.H., production area. Up to now, it is difficult to state whether the 1984. Geologic interpretation of Geothermal Fluid latter is normal and related to normal or strike-slip faulting or Movement in Cerro Prieto Field, Baja California, Mexico: it is exploitation-related. At least one more year of continuous Am. Assoc. of Petroleum Geol. Bull., Vol. 68, No. 1, pp. monitoring is necessary to investigate correlations between 18-30. seismicity occurring close to the reservoir and seasonal

390 Fabriol and Munguia Hill, D.P., 1977, A Model for Earthquake Swarms: J. Majer, E.L. and McEvilly, T.V., 1981, Detailed micro Geophys. Res. Vol. 82, No. 8, pp. 1347-1352. earthquake studies at the Cerro Prieto Geothermal Field: Proceedings, Third Symposium on the Cerro Prieto Lee, W.H.K. and Lahr, J.C., 1975, HYPO71 (revised): A Geothermal Field, U.S. Department of Energy, CONF- computer program for determining hypocenter magnitude 8410179, El Centro, California, October 16-19, 1984, pp. and first motion pattern of local earthquakes. U.S.G.S. 347-352. Open-file report, 75-3N, pp. 113. Munguia, L., 1995, Aceleraciones del suelo producidas por Lippmann, M.J. and MaAdn A., 1987, The Cerro Prieto temblores del Valle de Maxicali, Baja California, en el Geothermal Field: Geotherm. Sci. & Tech., Vol. 1, pp. 1- periodo 1993-1995, lnforme Tknico, CICESE, Abril 1995. 38. Reyes, C.A., Mendoza, L., Acosts, J., Alvarez, S.,Soares, J. Lomintz, C., Mooser, F., Allen, C., Brune, J., N. and Thatcher, And Avila, L., 1989, Estudio de sismica pasiva en el W., 1970, Seismicity and tectonics of the Northern Gulf of campo geotermico de Cerro Prieto, lnforme Tecinco, California region, Mexico, preliminary results: Geofisica CICESE, pp. 28. lnternacional, Vol. 10 (2), pp. 37-48. Reyes Zamora, L.A., 1979, Estudio de microsismicidad del McMechan, G.A. and Mooney, W.D., 1980, Asymptotic ray sistema de fallas transformadas Imperial-Cerro-Prieto. theory and synthetic seismograms for laterally varying lnforme Tecnico, CICESE, GEO-79-01. structures: theory and application to the Imperial Valley, California: Bull. Seism. SOC.Am., Vol. 70, pp. 2021-2035. Romero, A.E., Kirkpatrick, A., Majer, E.L. and Peterson, J.E., 1994: Geothermal Resources Council Transactions Vol. Majer, E.L. and McEvilly, T.V., 1982, Seismological studies in 18, pp. 331-338. the Cerro Prieto Geothermal Field: 1978-1982: Proceedings, Fourth Symposium on the Cerro Prieto geothermal field, pp. 145-151. Com. Fed. de Electr., Guadalajara, Mexico, August 10-12. 1982.

391 Fabriol and Munguia

p 10 'p

p0Q1 4/ 19/9! -6L 0 54 E, 22

0

I 1 16'30' 1 15'30' 114"30' Figure 3 Chronological histogram of number of events per day (August 20, 1994 to April 31, 1995) Figure 1 Location map of Cerro Prieto Geothermal Field

c 3 3.00 5 2.00 .-z 5 1.00

0.00

Figure 4 Chronological histogram of maximum magnitudes

V \< 2.50

2.M (1ScMdc3.6) 5- 1.50 m 3 1.M lW21' 115'18' 115'15' 115'12' 115" .9 Figure 2 Location of the seismological stations used in this study. Black triangles are stations. (After O="%, 0.M-1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 December 1994, OAX and D9 were shifted towards Oaxb and D5 emplacement, respectively). Open Magnitude triangles are injection wells: CPI, CPll and CPlll are power plants Figure 5 Frequency - Magnitude relation between the logarithm of cumulated number of events of magnitude MDzM,. The b-value is the slope of the linear regression of the right part of the curve

392 Fabric11ill 111 Munguia

I II \II

DCP*

A QW I CICtSE 05/95] I I I I 115021 115°18' 115V5' 1 W12' 115" 9' 115"21 ' 115"18' 1 W15' 115"12' 115" 9'

Fig. 6a :Recorded seismicity, August-September 1994. Fig. 6b :Recorded seismicity, October-November 1994. MAGNITUDE

'\

- \ \ \

A QKp I CICESE 05/951 I '21 ' 115"18' 115"15' 115"12' 115",9'

Fig. 6c :Recorded seismicity, December 1994-January 1995. Fig. 6d : Recoded seismicity, February-April 1995.

Fig. 6a to 6d : Location of epicenters from August 1994 to Apnl 1995. Size of crosses is proportional to magnitude. Open squares are the earthquakes located by the Strong Motion Network of Northwest of Mexico.

393 Fabriol and Munguia

SW 0

2

4

6

n E 38 N

10 A \ A 0 \

12 A t 3J A 14 --

A

~ ~ ~~ Fig. 7 : Vertical projection of hypocenters onto a SW-NE plane, parallel to Hidalgo fault. Abcissa are measured in km respect to a point of coordinates 32°20’N-115021’W. Diamonds correspond to August-September 94 earthquakes, open triangles to October-November 94, Black dots to December 94-January 95 and squares to February-April 95. Stars are earthquakes located by the Strong Motion Network of Northwest of Mexico. The associated numbers are magnitudes. M9, M133 and NL-1 are geothermal wells and CPI, the Cerro Prieto 1 power plant. The gray line represents the bottom of the geothermal reservoir schematically .The dip of Cerro Meto fault (CPF)is approximative.

I I SE 4 6 8 18 20 22 NW 24 I I I ~0 - I I I 0 2 a . =wA moo 4

6 a mu 3.4 A? A n A 0 E A I 58r . N

10 0 A A e 12 A x 3.7

14 -- A

.---- Fig. 8 : Vertical projection of hypocenters onto a SE-NW plane, perpendicular to Hidalgo fault. Abcissa are measured in km respect to a point of coordinates 32°18’N-115008’W. Diamonds correspond to September 94 earthquakes, open triangles to October-November 94, Black dots to December 94-January 95 and squares to February-April95 Stars are earthquakes located by the Strong Motion Network of Northwest of Mexico. The associated numbers are magnitudes. M 189 and M 1 10 are geothermal wells and CPI, CPII, the Cerro Prieto 1 and 2 power plants, respectively. The gray line represents the bottom of the geothermal reservoir schematically.

394