versión impresa ISSN 0016-7169
Geofís. Intl v.45 n.1 México ene./mar. 2006
Seismicity, focal mechanisms, and stress distribution in the Tres Vírgenes volcanic and geothermal region, Baja California Sur, Mexico
Victor Wong and Luis Munguía
Centro de Investigación Científica y de Educación Superior de Ensenada, (CICESE) Departamento de Sismología, División de Ciencias de la Tierra, Km107 Carretera TijuanaEnsenada, 22860 Ensenada, Baja California, Mexico. Email: Victor Wong firstname.lastname@example.org, Luis Munguía email@example.com
Authors mailing address in USA:
Victor Wong and Luis Munguía
CICESE Earth Sciences Division,
P. O. Box 434843
San Diego, CA 921434843, U. S. A.
Received: June 14, 2004
Accepted: August 30, 2005
En octubre de 1993 se llevó a cabo un monitoreo sísmico en la región volcánica Las Tres Vírgenes con el propósito de registrar la actividad sísmica asociada a las estructuras volcánicas, al campo geotérmico y a la tectónica local. Se localizaron 257 microsismos con hipocentros en los edificios volcánicos y a lo largo de la falla de rumbo, lateral derecha conocida como falla La Virgen. La profundidad focal de los sismos varía desde los muy cercanos a la superficie de la Tierra hasta los 8 km. Las profundidades someras ocurren principalmente en los edificios volcánicos. Los sismos más profundos ocurren fuera del área volcánica. La magnitud de duración de los microsismos localizados varía entre 1 y 3. La razón Vp/Vs y los valores bajos de Q que se estimaron en la zona sugieren un material con propiedades heterogéneas bajo las estructuras volcánicas, principalmente hacia la falla el Azufre y la caldera El Aguajito, donde se ha reportado la actividad hidrotermal. La distribución de los ejes P y T de los mecanismos focales de 90 microtemblores sugieren que la región se encuentra bajo compresión en dirección NS y extensión en dirección EO, en concordancia con el régimen de esfuerzo regional impuesto por el sistema de fallas transformadas, NOSE, lateral derecho, del Golfo de California.
Palabras clave: Sismicidad, mecanismos focales, esfuerzos tectónicos, campo volcánico, geotermia, Baja California Sur.
In October 1993 we carried out a seismic monitoring in the Tres Vírgenes volcanic region in order to record the background seismicity associated with the volcanic structures, the geothermal field and the tectonic features of the area. Hypocenters for 257 microearthquakes were located in the volcanic edifices and along the northwest rightlateral, strikeslip La Virgen fault. Focal depths range from close to the Earth surface to about 8 km. Shallow depths occur mainly in the volcanic edifices. Deeper seismic events occurred outside the volcanic area. The duration magnitudes of the located microearthquakes range between 1 and 3. The Vp/Vs ratio and the lowQ values estimated suggest heterogeneous material properties in the volcanic structures mainly toward the El Azufre fault and the El Aguajito Caldera, where hydrothermal activity has been reported. The P and Taxes of focal mechanisms for 90 microearthquakes suggest that the region is under NS compression and EW extension, in agreement with the regional tectonic stress field of the NWSE rightlateral strikeslip transform fault system of the Gulf of California.
Key words: Seismicity, focal mechanisms, tectonic stress, volcanic structures, geothermal field, Baja Califonia Sur.
Tres Vírgenes is the name of an area of three stratovolcanoes aligned in a NESW direction in the central part of Baja California peninsula (Figure 1). The volcanoes in this group are known as La Virgen, El Azufre, and El Viejo (Sawlan, 1981). The most recent volcanic activity in the area occurred in 1746 and was associated with La Virgen volcano, the youngest of the complex (Ives, 1962; Sawlan, 1981).
Hot springs and fumaroles have been observed at scattered locations throughout the Tres Vírgenes volcanic complex. The region was considered as a potential geothermal resource area. Since 1983, geological, geochemical and geophysical studies have been carried out in the region by the Comisión Federal de Electricidad (CFE), the Mexican federal power commission (CamposEnríquez, 1992; SánchezVelasco, 1996; López, 1998). In 1986 an exploratory well was drilled in the thermally active zone. After this, CFE drilled up seven more wells in the zone, to depths between 1200 and 2150 m. Temperatures went from 110° to 260° C (Lira et al., 1997). In this hydrothermally altered region the higher elevations contain fumaroles and steaming fields. At lower levels small hot springs can be found. The main results of the studies are summarized in López et al. (1995). The development of the geothermal field is still in progress.
Results from seismological studies as well as from different geophysical surveys including magnetotellurics (MT), transient electromagnetics (TDEM) and direct currentresistivity vertical soundings (Schlumberger) have been recently summarized by Romo et al. (2000). Those studies found that geothermal fluids may exist in some localized zones of the granodioritic basement, at depths of about 1200 m, pressures of ~120 bars, and temperatures near 240° C. At such depths the permeability has been increased due to an intense fracturing of the rocks (López et al., 1995). Such studies suggest that highly fractured rocks with fluids and high temperatures at depth may affect the attenuation of the seismic waves, as well as the electrical conductivity.
Digital and analog data from local microearthquakes recorded in October 1993 with a portable seismic network are used to study the seismic activity, the focal mechanisms and to investigate their tectonic significance. During its 23day period of operation, the portable network recorded data for a large number of microseismic events. Hypocenters for 257 events could be obtained with data from at least five stations. The microearthquakes used in this study were classified as volcanotectonic (VT) earthquakes, according to the classification method of Miller et al. (1998) or Lahr et al. (1994), among others. The seismograms produced by these events were characterized by high frequencies and clear P and Swave arrivals. Other type of volcanic earthquakes was also observed in our recording period. However, these other events were characterized by the absence of secondary phases and they were not located. A better knowledge of the seismicity pattern, the attenuation of the seismic waves, the Vp/Vs ratio, and the stress distribution in the study area will complement the results reported in previous geophysical studies.
GEOLOGIC AND STRUCTURAL SETTINGS
The first detailed studies of the volcanism and its relation to tectonic structural features in the Tres Vírgenes area were made by Colleta and Angelier (1981), Demant (1981), Ortlieb (1981), and Sawlan (1981). Those studies suggest that the Tres Vírgenes volcanic complex (El Viejo, El Azufre and La Virgen volcanoes) is the northwestern portion of the Santa Rosalía basin, which is controlled by the Gulf of California deformation system. The geological setting of the Tres Vírgenes region consists of Mesozoic granitic rocks overlain by late Cenozoic to Quaternary volcanic rocks (Gastil et al., 1975). This volcanism is considered as part of the magmatism related with the evolution of the Gulf of California (Gastil et al. 1975, 1979).
Figure 1 shows the most prominent geomorphic features recognized in the Tres Vírgenes volcanic area. These features are, from oldest to youngest, (1) the La Reforma Caldera (1.61.4 Ma), (2) the Aguajito Caldera (0.90.5 Ma), and (3) the Tres Vírgenes volcanic complex (0.44 Ma to present) (López et al. 1994). This figure also shows the El Azufre Canyon, which is the natural boundary between the Tres Vírgenes complex and the Aguajito Caldera. La Reforma and Aguajito Calderas have been considered as resurgent domes (GarduñoMonroy et al. 1993; Demant, 1981).
The oldest rock found in the region is a granitic rock of Cretaceous age (91.2 ± 2.0 Ma), which probably forms the regional plutonic basement in the Tres Vírgenes area (Schmidt, 1975). Overlaying this intrusive rock is a late Cenozoic volcanic cover that includes the Sierra Santa Lucía andesite, the northern limit of the Comondú Formation, and the Esperanza basalt (Sawlan, 1981). Beneath this volcanic cover is the Santa Rosalía Formation, which consists of shallowwater marine deposits characterized by fossiliferous sandstone. At the top of the above sequence there is a variety of pyroclastic products related to different stages of recent volcanism of the Tres Vírgenes complex (Sawlan, 1981). Chemically, the volcanic rocks emitted at the Tres Vírgenes volcanic region correspond to the calcalcaline series, excepting an alcaline rich pyroclastic flow at La Reforma Caldera, which is marginally peralkaline (López et al., 1995).
The eruptive history of the Tres Vírgenes volcanic complex is predominantly characterized by effusive activity but an exceptionally Plinian eruption occurred about 6500 years ago at La Virgen volcano (Capra et al., 1998). The absence of younger volcanic deposits in the area suggests that the historic volcanic activity reported at La Virgen volcano in 1746 by a Spanish Jesuit was not probably associated to magmatic activity, although such historic observation was probably due to a landslide on the edifices of the volcanic complex, fumarolic activity, or small phreatic eruptions (Sawlan, 1981; Capra et al., 1998).
At the study region, the highest fault density is observed near the El Azufre Canyon and around the hydrothermal zone that is outlined by a rectangle in the Figure 1 (López et al., 1995). Preferential directions of faulting in the zone are NWSE (La Virgen, Las Víboras, and El Azufre faults), NS (Las Cuevas and El Cimarrón faults), and NESW (La Puerta and El Mosquito faults). López et al. (1995) suggested that the main fault system that controls the active hydrothermalism in the zone is the NWSE Las VíborasEl Azufre system.
PREVIOUS SEISMOLOGICAL STUDIES
An exploratory experiment, consisting of a single seismic station operated in the Tres Vírgenes area from May to October of 1992, recorded more than 2000 events with SP times of less than 3 sec and magnitudes between 1 and 3 (Wong and Munguía, 1992). After that initial seismic monitoring, CFE deployed six digital stations around the Tres Vírgenes volcanic complex. Results from analyses of data produced in the first four months of network operation show that the Tres Vírgenes seismicity does not occur at random. Rather, the seismic events occur spatially clustered along the volcanic structures and to the southeast of the volcanoes (Munguía and Wong, 1993, 1995). A large proportion of the recorded seismicity was located within the area covered by the seismic network. The duration magnitudes calculated for all the located events ranged from 1 to 3 and the focal depths between 1 and 12 km.
The estimated composite focal mechanism solutions indicate strikeslip and normal faulting. Most of the obtained focal solutions showed oblique faults, combining strikeslip and dipslip motion. The variability of the focal mechanisms, as well as the occurrence of microearthquake swarmlike sequences reflects the structural complexity of the zone.
Results from recent seismic studies carried out in the Tres Vírgenes volcanic region suggest the presence of a low Pwave velocity and lowQ body around the El Azufre Canyon (Macías, 1997; Wong, 2000; Wong et al., 2001). Those studies found low coda Q values and low Pwave velocity anomalies along the El Azufre Canyon, where dense fracturing, hydrothermal activity (hot springs and fumaroles), and high temperatures have been reported (López et al., 1995). In this particular region, multiple geophysical surveys including magnetotellurics (MT), transient electromagnetics (TDEM) and direct currentresistivity vertical soundings (Schlumberger), permitted to outline the presence of a highly conductive zone along El Azufre Canyon, the limit between the Tres Vírgenes volcanic complex and the Aguajito Caldera (Romo et al., 2000).
DATA AND ANALYSIS
On October 1993, 13 portable analog and digital seismic stations were deployed in the Tres Vírgenes volcanic region for a 23day period of recording (Figure 2). The purpose of this effort was to record the local seismicity and get precise hypocenter locations. The installed instrumentation consisted of six analog seismographs (Sprengnether, MEQ800), coupled with 1Hz verticalcomponent seismometers (RangerSS 1, Kinemetrics), and seven threecomponent digital recorders (SSR1, Kinemetrics), coupled with 1Hz seismometers (Mark L4C). The digital stations operated in an eventtriggered mode and recorded at a nominal rate of 100 samples/sec on each channel. Timing for the analog seismic records was provided through internal WWVB radio time receivers. For the digital recorders, a common and absolute time base was achieved using the Omega radio time signal. Figure 3 shows examples of seismograms from a local earthquake recorded by the digital temporary seismic stations.
Stations deployment and event locations
All the recording stations were installed on volcanic rocks with an average distance among adjacent stations of less than 7 km (see Figure 2). During its 23day period of operation, this portable network recorded more than 800 microearthquakes, including a microearthquake swarmlike sequence of at least 110 events and a threeday duration.
This microearthquake swarmlike sequence was mainly observed at the analog seismograms. From the recorded seismic data, hypocenters for 257 seismic events could be estimated with the P and Swave phases from at least five stations. Seismic events that were too small to be located with five stations are not reported in this study. Most of the located events occurred within the area covered by the seismic network. A large proportion of these microseismic events were located in the volcanic edifices of the Tres Vírgenes volcanic complex, the Aguajito Caldera, and along the northwest rightlateral, stikeslip La Vírgen fault. Some of these microearthquakes occurred as clusters of events and as swarmlike sequences. These events occurred at focal depths between the Earth surface and 8 km, and had duration magnitudes between 1 and 3. Shallow depth events occurred mainly in the volcanic edifices.
The onedimensional Pwave velocity model used in the hypocenter locations was taken from Munguía and Wong (1993, 1995). This velocity model was determined from the results of Thatcher and Brune (1973), Phillips (1964) and from logdensity data of the LV2 borehole (Viggiano, 1992). Figure 4 shows our 1D Pwave velocity model and the velocity models determined by Thatcher and Brune (1973) and Phillips (1964). In order to estimate the Swave velocities, we selected 60 microearthquakes with clear P and Sphases to construct Wadati plots. Figure 5a shows the average Vp/ Vs ratio of 1.72 ± 0.08 calculated for the whole region. This value was used in the hypocentral determination of all the located seismic events in this study.
The computer program HYPO71 (Lee and Lahr, 1975) was used to locate the seismic events. The accuracy of the arrival time picks generally depended on the nature of the phase onset and the background noise. Errors of the P and Swaves arrival times picks are between ± 0.01 and ± 0.05 seconds in the digital records. On the analog seismograms errors in the Pwave time readings are of the order of 0.05 seconds. The rms residuals of epicenters are less than 0.3 seconds. Events within the network have vertical and horizontal location errors of less than ± 3 km. Events located outside the network progressively deteriorated with increasing distance from the stations to about ± 5 km in depth and horizontally. The Figures 6a,b show cross sections of hypocenters along the lines AB and CD shown in Figure 2. The hypocenters of events lying within 4 km from those lines were projected onto the vertical planes.
Vp/Vs ratio and Coda Q value
According to their epicentral location, the seismic events were separated into three main groups in order to carry out an analysis of Vp/Vs variations in the volcanic area (see Figure 2). Events located on the El Aguajito Caldera are referred as to the northeast group. The central group is composed of events located in the Tres Vírgenes volcanic complex (El Viejo, El Azufre and La Virgen volcanoes), while the events located along or near to the La Virgen fault are included in the southeast group. The estimated Vp/Vs values of each group range from 1.68 to 1.76 compared to values of 1.73 to 1.87 which are usually obtained in laboratory and field investigations. These velocity ratios led to a Poisson's ratios from 0.22 to 0.26 compared with values 0.25 to 0.30 which are normally observed. In a series of carefully performed laboratory experiments, Nur and Simmons (1969) noted that dry rocks exhibit low values of Poisson's ratio (<0.20) while saturated rocks show normal to high values (>0.25). This means that lower than normal poisson's ratio may indicate partial saturation of reservoir rocks.
Figure 5 shows the Vp/Vs values of each epicentral group. In the northeast group (El Aguajito Caldera) the Vp/ Vs value is 1.68 (Figure 5b), while in the Central and southeast group (Tres Vírgenes volcanic complex and La Virgen fault) the mean Vp/Vs value is 1.76 (Figures 5c,d). Those variations suggest a heterogeneous structure and lateral variation in the velocities at the volcanic region, mainly toward the northeast region were intense fracturing and hydrotermal activity has been reported (López et al., 1995).
On the other hand, Wong (2000) and Wong et al. (2001) performed a study of seismic wave attenuation based on the same digital data of this study. They found that the low values of Qc (Coda Q) obtained from data of station E1 reflect the effects caused by the local geology below that particular station. Furthermore, at this site Wong et al. (2001) found an abnormally strong reduction of Qc at low frequencies (6 Hz), and an attenuation that is three to fourth times larger than for the other station sites. A strong frequencydependent attenuation is expected to play an important role in areas where the presence of highly fractured rocks, shallow high temperatures, and hydrothermal activity (hot springs and fumaroles) could influence the frequency content of the seismic signals (Gao, 1992). Since these environment conditions prevail around the E1 station, we expect low Qc values correlated with the dense fracturing, hydrothermal activity and high temperatures that are common around the LV2 production well. In addition, results from geophysical studies suggest the presence of a highly conductive body below this particular area (Romo et al., 2000). Similar results have been reported by Macías (1997) in his study of Coda Q at the Tres Vírgenes volcanic area. This study report low Qc values in the northeast region, where hydrothermal activity, dense fracturing and high temperatures have been reported by López et al. (1995).
FOCAL MECHANISM AND STRESS
Single and composite focal mechanism solutions were calculated from Pwave first motions using Reasenberg and Oppenheimer (1985) programs. Equal area projections of the lower focal hemisphere were used. Single focal mechanism solutions were constrained by at least 6 Pwave first motions. Composite fault plane solution was calculated with events closely related in time and space (Figure 7). The firstmotion patterns can be adequately modeled by pure doublecouple sources, although the data are not dense enough to exclude the presence of a significant nondoublecouple source component. The quality of the focal mechanism solutions varies considerably from moderately to well constrained, depending of the coverage of the focal sphere. As expected, poorly constrained focal solutions were obtained towards the edges of the seismic network. In most cases, the strike of one of the nodal planes matches the lineation of the epicenters. The locations of the estimated focal mechanism are shown in Figure 7. These focal mechanism diagrams are equalarea projections of the lower hemisphere; open circles and triangles denote compressional and dilatational motions, respectively. The P and T on the beach balls denote the position of the P and T axes of deformation. These axes approximate the maximum and minimum compresional stress directions in the region (MacKenzie, 1969).
The estimated composite focal mechanism solutions in this study show strikeslip, normal and oblique faulting; combining strikeslip and dipslip motion (Figure 7). Most solutions exhibit obliqueslip with normaldextral motion, so that the average direction of extension is near EW (Figure 7b). In many cases, the apparent direction of extension is not perpendicular to the average strike of the faults, suggesting a systematic strikeslip component of motion on normal faults. This permits to infer that most of the strikeslip faults moved when both compression and extension were horizontal. On the other side, the variability of the focal mechanisms may reflect spatial inhomogeneities in the stress field or slip on faults of different orientations subjected to a uniform stress field.
A plot of all P and T axes obtained shows that the tensional axes have an average trend that is nearly horizontal and in the EW direction (Figure 7b), while the average compressive stress axis is nearly horizontal and in the NS direction. Although some vertical tensional axes were observed, they are less numerous in comparison to the horizontal ones. Thus we considered that they had much less quantitative importance than the ensuing normal and strikeslip faulting. This stress axes distribution suggest that the tectonic regime at the Tres Vírgenes volcanic region is dominated by a conjugate system of NWSE dextral strikeslip faults associated with NS normal faults, maybe in response to the regional tectonic stress imposed by the NWSE rightlateral strikeslip transform rift system of the Gulf of California.
On the other hand, the state of the stress in the study region is also evident by the northeastsouthwest orientation of the Tres Vírgenes volcanic complex and the Aguajito Caldera. The usefulness of aligned volcanic centers in tectonic regional studies was highlighted by Nakamura (1977), who showed that such alignments are perpendicular to the horizontal minimum stress (σhmin) which in an active volcanic area must be equal to either σ3 or σ2 in order for magmas to reach the surface. This is coincident with the results reported by Umhoefer and Teyssier (1992) at the Tres Vírgenes volcanic region. They considered that the trend defined by the volcanoes of the Tres Vírgenes complex and the Aguajito Caldera could be a consequence of a nearly NS compresional zone (EW direction of opening) which, through time, has migrated to the south. They also suggested that the extension may have resulted from dextral strikeslip movements on the Las VíborasEl Azufre and the La Virgen faults, which form a rightstepping offset centered on the Tres Vírgenes complex. In the same way, Sawlan (1981), GarduñoMonroy et al. (1993), and López et al. (1995) pointed out that the lineament of the major volcanic edifices and the fault geometry at this region indicate that σhmin is essentially EW.
On the other side, Fabriol et al. (1999) carried out a marine geophysical study in the central part of the Gulf of California, between Isla Tortuga and La Reforma Caldera region. Based on backscattering images of the sea floor, they suggested that the onshore and offshore volcanic structures that trend NS and EW can be related to more regional tectonic structures, like fracture zone extension, which resulted from the transtensive regime of the Gulf of California transform pullapart system.
DISCUSSION AND CONCLUSION
The most active area observed in this study was located at the northeast zone, in the Aguajito Caldera (Figure 7). Some events of the microearthquake swarmlike sequence that was recorded during a period of three days were located in this region. This seismic behavior is generally interpreted as indicator of stress heterogeneity. The seismic events of this swarm were concentrated at a 5km average depth (Figure 6a). Such seismic activity occurred in a zone where SWdipping normal faults and hydrothermal activity have been reported (López et al, 1994). However, a leftlateral strikeslip composite focal mechanism solution with northeastsouthwest strike seems to be in agreement with the seismic events located in this zone (Solution 2, Figure 7). Although a unique choice of the fault plane cannot be made from the first motion data, the northeastsouthwest striking plane can be chosen due to alignment of the epicenters.
Other active area seems to be the El Cimarrón fault (ECF, Figure 1), the NS trending leftlateral strikeslip fault that cuts the El Aguajito Caldera in its northwest flank (GarduñoMonroy et al., 1993). Focal depths of the epicenters in this area are concentrated between 5 and 7 km (Figure 6a). The focal mechanism solutions of this sequence show normal motion with a leftlateral strikeslip component, accordingly with the results reported by GarduñoMonroy et al. (1993)(Solution 1, Figure 7). They interpreted the general fault pattern as a system of NNWSSE and NNESSW Riedel shears, dextral and sinistral respectively, associated with NS enechelon normal faults and extension fractures. Although normal faults with NNESSW direction, dominated by sinistral strikeslip, are much less abundant than normal faults with rightlateral strikeslip motion (López et al., 1995). The orientation of this fault pattern seems to be compatible to the stress field associated with the rightlateral faulting related to the opening and evolution of the Gulf of California.
Microearthquakes also occurred in the south portion of the El Aguajito Caldera and along the El Azufre Canyon. This zone, characterized by shallow high temperatures (~200° C) and hydrothermal activity, shows a poorly defined trend of seismic activity. The microearthquake activity in this zone lies at depths between 4 and 7km (Figure 6a). The focal mechanisms show predominantly strikeslip and obliqueslip faulting (Solutions: 3, 14, and 15, Figure 7). Candidate planes of the fault plane solution 15 seem to be rotated from those shown for the nearby solutions 3 and 14. The latter solutions feature more strikeslip than dipslip faulting. The strikeslip focal mechanism solutions estimated in this area seem to be associated with the NESW trending La Puerta fault and the southoriented extension of the Las Víboras fault in the vicinity of the LV2 well, respectively. This particular region consists of many fault structures with dominantly strikeslip motion and steeplydipping planes, implying possible dextral movement on normal faults (Umhoefer and Teyssier, 1992; López et al., 1995).
From Figures 6a and 6b, we can see that some seismic events occur directly beneath the Tres Vírgenes volcanic complex. Most of the microearthquakes are concentrated between 3 and 7km depths, with some scattered events located at depths from 1 to 8 km (Figure 6a). Microearthquakes of the volcanic structures were located within the igneous basement, beneath the volcanic overlaying. Focal mechanism solutions for these events show predominantly obliquenormal faulting (Solutions: 12 and 13, Figure 7), suggesting extensional faulting beneath the volcanic alignments. The seismic events that were located in the southsouthwest flank of the Tres Vírgenes volcano show mainly strikeslip faulting with some component of normal motion (Solutions: 11 and 10, Figure 7). The candidate planes for the apparent obliquereverse mechanism (solution 10, Figure 7) seem to be rotated from those shown by solutions of nearby events (Solutions: 9 and 12, Figure 7). The apparent rotation of the fault planes may reflect local changes in the stress field, or slip on faults of different orientations under a uniform stress field. On the other hand, Angelier et al. (1981) suggested that reverse faults in the area could be related to fault motions on originally normal faults that have been subsequently rotated by block tilting, due to more recent normal faulting.
The seismic activity recorded to the southsoutheast of La Virgen volcano seems to be associated with the northwestsoutheast trending dextralnormal La Virgen fault described by López et al. (1995). The trace of this fault coincides with the southern Main Gulf Escarpment, which has been generally interpreted as the boundary between the Gulf Extensional Province and the topographically higher Peninsular Basin and Range (Gastil et al., 1975). Small concentrations of epicenters show an apparent lineament along the La Virgen fault, except for a few clusters and some isolated events that seem to be parallel or nearly parallel to the trace of this fault. This activity occurred at depths from near the surface to about 8km. Most of these events are concentrated at depths between 3 and 7 km (Figure 6b). Focal mechanism solutions for these seismic events show strikeslip, obliquenormal and obliquereverse fault type. In contrast, normal fault type is prominent at the La Virgen valley, about 5 km eastward from the La Virgen fault. The strikeslip and the apparent obliquereverse focal mechanisms seem to be related to local variability in fault orientations of the La Virgen fault. Similarly, the obliquenormal fault plane solutions at La Virgen valley seem to be related with the NWSE trending normal faults located in the area, although more definitive data are necessary to define the mode of faulting of the seismicity at this area.
On the other side, the variation of the Vp/Vs ratio at each epicentral group suggests a heterogeneous velocity structure and lateral variation in the volcanic region, mainly toward the northeast region, were hydrotermal activity has been found. Similarly, low values of Qc are expected around the E1 station, where dense fracturing, hydrothermal activity and high temperatures are commonly observed. Also, we noticed that the seismic waves propagating through this area were conspicuosly attenuated. This fact suggests that a hightemperature, ductile, and lowQ body may exist under the El Azufre Canyon. The tectonic regional framework, combined with the local stress field and the high thermal gradient in the Las Tres Vírgenes volcanic area, may produce a higher proportion of small magnitude local events in comparison with events of higher magnitude that are characteristic of the Gulf of California.
The directions of the principal stress axes estimated in this study are consistent with the tectonic framework found at the Santa Rosalia Basin, about 1520 km south of La Virgen volcano, where a mean EW to ENEWNW nearhorizontal tensional stress predominantes. The averaged NS and EW orientation of the principal stresses estimated here suggest that the tectonic regimen of the Tres Vírgenes volcanic region is dominated by a conjugate system of NWSE dextral strikeslip to obliqueslip faults associated with NS normal faults with dipslip motions. This occurs probably in response to the regional tectonic stress imposed by the NWSE rightlateral strikeslip transform fault system in the Gulf of California.
The authors thank Dr. Gerardo Hiriart, who was Director of Gerencia de Proyectos Geotermoeléctricos, of Comisión Federal de Electricidad, for his cooperation and logistical assistance. We are grateful to personnel at the Tres Vírgenes Geothermal Field, for their assistance in the fieldwork. We also thank Miguel Navarro, Tito Valdéz and Oscar Galvéz, for their valuable help in the fieldwork. We acknowledge the helpful suggestions of Dr. Cecilio J. Rebollar, who reviewed a draft of this paper.
ANGELIER, J., B. COLLETA, J. CHOROWICZ, L. ORTLIEB and C. RANGIN, 1981. Fault tectonics of the Baja California Peninsula and the opening of the Sea of Cortés, Mexico. J. Struct. Geol. 3, 4, 347357. [ Links ]
CAMPOSENRÍQUEZ, J. O., 1992. Geophysical prospecting for geothermal resources in Mexico. Geofís. Int. 31, 4, 339340. [ Links ]
CAPRA, L., J. L. MACÍAS, J. M. ESPÍNDOLA and C. SIEBE, 1998. Holocene plinian eruption of La Virgen volcano, Baja California, Mexico. J. Volcanol. Geotherm. Res. 80, 239266. [ Links ]
COLLETA, B. and J. ANGELIER, 1981. Faulting evolution of the Santa Rosalía Basin, Baja California, Mexico. In: L. Ortlieb and O. Roldan, (Eds.), Proceedings of the Geology of Northwestern Mexico and Southern Arizona Congress. Univ. Nac. Autón. Mex., Hermosillo, Sonora, Mexico, 265274. [ Links ]
DEMANT, A., 1981. PlioQuaternary volcanism of the Santa Rosalía area, Baja California, Mexico. In: L. Ortlieb and O. Roldan (Eds.), Proceedings of the Geology of Northwestern Mexico and Southern Arizona Congress. Univ. Nac. Autón. Mex., Hermosillo, Sonora, Mexico, 295307. [ Links ]
FABRIOL, H., L. A. DELGADOARGOTE, J. J. DAÑOBEITIA, D. CÓRDOBA, A. GONZÁLEZ., J. GARCÍAABDESLEM, R. BARTOLOMÉ, B. MARTÍNATIENZA and V. FRIASCÁMACHO, 1999. Backscattering and geophysical features of volcanic ridges offshore Santa Rosalía, Baja California Sur, Gulf of California, Mexico. J. Volcanol. Geotherm. Res. 93, 7592. [ Links ]
GAO, L. S., 1992. Physical meaning of the coda envelopes. In: Gasparini, P., Scarpa, R., Aki, K. (Eds.), Volcanic Seismology, IAVCEI, Proceedings in Volcanology 3, SpringerVerlag, New York, 391403. [ Links ]
GARDUÑOMONROY, V. H., H. VARGASLEDEZMA and J. O. CAMPOSENRÍQUEZ, 1993. Preliminary geologic studies of Sierra El Aguajito (Baja California, Mexico): a resurgenttype caldera. J. Volcanol. Geotherm. Res. 59, 4758. [ Links ]
GASTIL, R. G., R. P., PHILLIPS and E.C., ALLISON, 1975. Reconnaissance geology of the state of Baja California. Geol. Soc. Amer., Mem. 140, 170 pp. [ Links ]
GASTIL, R. G., D. KRUMMENACHER and J. MINCH, 1979. The record of Cenozoic volcanism around the Gulf of California. Geol. Soc. Am. Bull. 90, 839857. [ Links ]
IVES, R. L., 1962. Dating of the 1746 eruption of the Tres Vírgenes volcano, Baja California Sur, Mexico. Geol. Soc. Am. Bull. 73, 647648. [ Links ]
LAHR, J. C., B. A. CHOUET, C. D. STEPHENS, J. A. POWER and R. A. PAGE, 1994. Earthquake classification, location and error analysis in a volcanic environment: implications for the magmatic system of the 19891990 eruptions at the Redoubt volcano, Alaska. In: Miller, T. P., Chouet B. A. (Eds.), The 19891990 eruptions of Redoubt volcano, Alaska. J. Volcanol. Geotherm. Res. 62, 137151. [ Links ]
LEE, W. H. K. and J. C. LAHR, 1975. HYPO71 (Revised): A computer program for determining hypocenter, magnitude and first motion pattern of local earthquake. U. S. Geological Survey, Open File Report 75, 311 pp. [ Links ]
LIRA, H. H., M. L. GONZÁLEZ and F. G. ARELLANO, 1997. Actualización del modelo conceptual del Campo Geotérmico de Tres Vírgenes, Baja California Sur. Technical Report, RE12/97, Gerencia de Proyectos Geotermoeléctricos, Comisión Federal de Electricidad, 26 pp. [ Links ]
LÓPEZ, H. A., 1998. Síntesis geológica de la zona geotérmica de las Tres Vírgenes, B.C.S., México. Geoterm. Rev. Mex. Geoenerg. 14, 1, 314. [ Links ]
LÓPEZ, H. A., G. H. GARCÍA and F. G. ARELLANO, 1994. Geological and geophysical studies at Las Tres Vírgenes, B.C.S., Mexico, geothermal zone. Geotherm. Res. Counc. Trans. 18, 275 280. [ Links ]
LÓPEZ, H. A., G. H. GARCÍA and F. G. ARELLANO, 1995. Geothermal exploration at Las Tres Vírgenes, B.C.S., Mexico. In: Barbier E. et al., (Eds.), Proceedings of the 1995 World Geoth. Congress, Int. Geoth. Assoc. 2, 707712. [ Links ]
MACÍAS, C. M., 1997. Análisis de la información sísmica del campo geotérmico de Tres Vírgenes, B. C. S., período eneroabril de 1994. B. S. thesis, Univ. Nac. Autón. Mex., D. F., Mexico, 39 pp. and appendix. [ Links ]
MACKENZIE, D. P., 1969. The relation between fault plane solutions for earthquakes and the directions of the principal stresess. Bull. Seismol. Soc. Am. 59, 591601. [ Links ]
MILLER, A. D., R. C. STEWART, R. A. WHITE, R. LUCKETT, B. J. BAPTIE, W. P., ASPINALL, J. L. LATCHMAN, L. L. LYNCH and B. VOIGHT, 1998. Seismicity associated with dome growth and collapse at the Soufrière Hills volcano, Monserrat. Geophys. Res. Lett. 25, 34013404. [ Links ]
MUNGUÍA, L. and V. WONG, 1993. Analísis e interpretación de la información sísmica digital registrada en el campo geotérmico de Tres Vírgenes, B.C.S. en el período febreromayo de 1993. Technical Report., Contract CLSGPG301393/CFECICESE, Gerencia de Proyectos Geotermoeléctricos, Comisión Federal de Electricidad, 128 pp. [ Links ]
MUNGUÍA, L. and V. WONG, 1995. Estudio de sismicidad en la zona geotérmica las Tres Vírgenes, Baja California Sur. In: MedinaMartínez, F., DelgadoArgote, L. A., SuárezReynoso, G. (Eds.), La sismología en México: 10 años después del temblor de Michoacán del 19 de septiembre de 1985 (M=8.1). Unión Geofísica Mexicana, Monografía No. 2, 212228. [ Links ]
NAKAMURA, K., 1977. Volcanoes as possible indicators of tectonic stress orientationprinciple and proposals. J. Volcanol. Geotherm. Res. 2, 116. [ Links ]
NUR, A. and G. SIMMONS, 1969. The effect of saturation on velocity in low porosity rocks. Earth. Planet. Sci. Lett. 7, 183193. [ Links ]
ORTLIEB, L., 1981. Sequences of pleistocene marine terraces in the Santa Rosalía area, Baja California, Mexico. In: Ortlieb, L., Roldan, O. (Eds.), Proceedings of the Geology of Northwestern Mexico and Southern Arizona Congress. Univ. Nac. Autón. Méx., Hermosillo, Sonora, Mexico, 275293. [ Links ]
PHILLIPS, R. P., 1964. Seismic refraction studies in the Gulf of California. Marine Geology of the Gulf of California. Amer. Assoc. Petrol. Geol., Mem. 3., Tulsa, Oklahoma. 90 pp. [ Links ]
REASEMBERG, P. and D. OPPENHEIMER, 1985. FPFIT, FPPLOT and FPPAGE: FORTRAN computer programs for calculating and displaying earthquake faultplane solutions. U. S. Geol. Surv. Open File Rep. 85739, 109 pp. [ Links ]
ROMO, J. M., V. WONG, C. FLORES and R. VÁZQUEZ, 2000. The subsurface electrical conductivity and the attenuation of coda waves at Las Tres Vírgenes geothermal field in Baja California Sur, México. Proceedings of the 2000 World Geotherm. Congress, Int. Geotherm. Assoc., 16451650. [ Links ]
SÁNCHEZVELASCO, R., 1996. Aspectos Generales del Proyecto Geotérmico de las Tres Vírgenes, B.C.S., Mexico. Geoterm. Rev. Mex. Geoenerg. 12, 115124. [ Links ]
SAWLAN, M. G., 1981. Late Cenozoic volcanism in the Tres Vírgenes area. In: L. Ortlieb and O. Roldan (Eds.), Proceedings of the Geology of Northwestern Mexico and Southern Arizona Congress. Univ. Nac. Autón. Méx., Hermosillo, Sonora, México, 309319. [ Links ]
SCHMIDT, E. K., 1975. Plate tectonics, volcanic petrology and ore formation in the Santa Rosalía area, Baja California, México. MS thesis, Arizona University, Tucson, 196 pp. [ Links ]
THATCHER, W. and J. N. BRUNE, 1973. Surface waves and crustal structure in the Gulf of California region. Bull. Seismol. Soc. Am. 63, 16891698. [ Links ]
UMHOEFER P. J. and C. TEYSSIER, 1992. A preliminary structural geology study of the El Azufre Canyon area, north of the Tres Vírgenes volcano, northern Baja California Sur. In: CarrilloChávez A., ÁlvarezArellano, A. (Eds.), Proceedings of the First International meeting on Geology of Baja California Peninsula. Peninsular Geol. Soc., Univ. Autón. of Baja California Sur, La Paz, B.C.S., Mexico, 4554. [ Links ]
VIGGIANO, J.C., 1992. El pozo desviado LV2A, Las Tres Vírgenes, Baja California Sur: Petrología e interpretación. Geoterm. Rev. Mex. Geoenerg. 18, 4, 373394. [ Links ]
WONG, V., 2000. Estudio de sismotectónica, atenuación y tomografía sísmica en la región volcánica y geotérmica Las Tres Vírgenes, Baja California Sur, Mexico. PhD thesis, CICESE, Seism. Depart., Earth Sciences Division, Ensenada, Baja California, Mexico, 182 pp. [ Links ]
WONG, V. and L. MUNGUÍA, 1992. Monitoreo sísmico del área geotérmica de Tres Vírgenes, Baja California Sur. Technical Report, Contract CLSCPG301192/CFECICESE, Gerencia de Proyectos Geotermoeléctricos, 16 pp. [ Links ]
WONG, V., C. J. REBOLLAR and L. MUNGUÍA, 2001. Attenuation of coda waves at the Tres Vírgenes volcanic area, Baja California Sur, Mexico. Bull. Seismol. Soc. Amer. 91, 4, 683693. [ Links ]