New Seismic Imaging of the Coso Geothermal Field, Eastern California

NEW SEISMIC IMAGING OF THE COSO GEOTHERMAL FIELD, EASTERN CALIFORNIA

Jeffrey Unruh William Lettis & Associates, Inc., 1777 Botelho Drive, Suite 262, Walnut Creek, CA 94598

Satish Pullammanappallil and William Honjas Optim LLC, University of Nevada, Reno 89557

Francis Monastero Geothermal Program Office, China Lake NAWC, CA 93555

ABSTRACT INTRODUCTION New multifold seismic reflection data from the This paper presents new seismic images of the Coso central Coso Range, eastern California, image brittle geothermal field in eastern California (Fig. 1). A faults and other structures in Mesozoic crystalline total of 45 line-km of 2-D reflection data were rocks that host a producing geothermal field. The acquired in the central Coso Range to test the ability reflection data were processed in two steps that of new processing methods to image structure in the incorporate new seismic imaging methods: (1) P- crystalline rocks that host the geothermal field (Fig. wave first arrivals in the seismic data were inverted 2). The data were processed using a combination of for subsurface acoustic velocities using a non-linear detailed velocity modeling and Kirchhoff pre-stack simulated annealing approach; and (2) 2-D Velocity migration to obtain accurate, depth-migrated images tomograms obtained from the inversions were 118° 117°

MOUNTAINS employed in pre-stack Kirchhoff migration to W

h WHITE Explanation

i t produce accurate, depth-migrated images of e Dextral fault M

t subsurface structure. Three-dimensional n Sinistral fault s

. f Normal fault;, ball on visualizations of the velocity structure show that a u INYO MOUNTAINS downthrown block l F tz u discrete low velocity zones are associated with the r o n Plunging anticline n a c two major producing areas of the Coso field, and that ° e e Local azimuth of Sierran- 37 O C r North American motion w e e (Dixon et al., 2000) the producing zones are separated by a “ridge” of e k n f s a V u relatively higher velocity rock. The low velocity l a t l z l e o y n zones are interpreted to be zones of pervasive e f a u H l un fracturing and/or hydrothermal alteration associated t te r M t with localized hydrothermal circulation. The . f au

lt PANAMINT RANGE Death presence of the intervening “ridge” of high velocity P a n a rock, which presumably is relatively intact and m in Valley unaltered bedrock, may indicate long-term t Valle compartmentalization of flow in the field. Kirchhoff COSO RANGE y depth-migrated seismic images reveal substantial 36° fault f a DETAILED u coherent reflectivity in the upper 6 to 8 km of the lt STUDY AREA crust. The Coso Wash fault, a Quaternary-active (Figure 2) zone fault that is a locus of surface geothermal activity, is Airport Lake well-imaged as a moderately dipping reflector that fault terminates against a reflector or reflective zone at Garlock fault approximately 4 km depth. The 4 km reflective zone N can be traced laterally beneath the geothermal field, MOJAVE 0 50 km and it may represent the brittle-ductile transition. BLOCK The Kirchhoff images also reveal a prominent high- Figure 1: Major Quaternary faults of the southern amplitude reflector at 6 km depth directly beneath the Walker Lane belt (modified from Jennings, 1994). geothermal field. The reflector exhibits slight The shaded region depicts the Sierra Nevada antiformal geometry and appears to be restricted to microplate as defined by patterns of Quaternary the northern part of the producing zone. The 6 km faulting and geodesy (Dixon et al., 1995; King et al., reflector tentatively is interpreted to be a pocket of 1999). Transfer of right-lateral shear across the right magmatic brine or partial melt. step between the Airport Lake fault and the Owens Valley fault drives localized crustal extension in the central Coso Range.

5 5

5 d5 5 5

5 5

5 5 5 5 5

5 5 5

5 5

5 5 5 5 5 5 5 5 5

5 5 5 5

5 5 5 5 5 5

5 5 5 5

5 5 5 5 5 5 5 5

5 555 5 5 5 5 5 55 5 5 5

55 5 5 5 5

5 5 5 5 5 5 5

5 5 5 5 5

5 5 5 5 5 5 5

5 5

5 5 5 5 5

5 5 5 5

5 5 5 5 5

5 5 5

5 5

5 5 5 5

5 55 5 5 5 5

5 5

5 5 5

5 5 5 5 5 5

55 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5

55 5 5 5 5 5

5 5

5 5 5 5 5 g2y22y 5 5 5 5 5 5 5 5 5 5 5 vine2IIHGxEP 55 5 5 5 5 5 55 5 5

5 5 5 5 5

5 5 5IPH

5 5 5 5 5 5

5 5 5 5 5 5 5

5 d 5 5 5 5 5 5 5 5 5 IHI 5 5 d 5 5 5 5 d5 5 5 55 5 5 5

5 5 5 5

d 5 5 5 5 5

5 5 5 5 d d 5 5 5 5 5 5 5 IPH 5 5 5

5 5 55

vine2IHTGiEU 5 5 5

5 5 5 5 5 5

5 5 5 5 5 5 IPH 55 5 5 5 5 5 5

5 5 5

5 5 5 5 5 5 5 5 5 5

5 5 5

vine2IHWGiEW 5 5 5 5 5 5 IHH 5 5 5 5 5 5

5 5

5 5 goso2sh2ptF

5 5 IPH 5

5 5 ITH 5

5 5

5 5 5 5

5 5 5 5 IRH

5 5 5 5 5 5

5 5 5

5 5 5 5

5 5 PHH 5 5 5 5 5 5 IRH

5 d 5 5

5 5 5 d 5 5 5 5 5 55 5 5 PPH IPH 5

5 5 5 5 vine2IHWGiEW 5 5

5 ITH 5

IVH 5

5 5 5

5 5 5 5 5 5

5 5 5 5 5 5 5

5 5 5 5

5d 5 5 5 5 5 5 5 5 5 5 5 5 5 5

5 5 5 5

5 5 5 5 5 vine2IHTGxET

5 5 5 PRH

5 5 5 5 5 d 5 5 vine2IISGxES 5

5 5 5 5

5 5 5

5 5 5 5 5 5 5 5

IVH

5 5 5 5 5 5

5 5 5 5 5 ITH5

5 5 d 5 5

d 5 5

5 5 5 5 5 5 5 5IRH

5 5 5 5 5

5 5 5

5 5 5 5

5 5 5 5 5

5 5

5 5 55 55

pigure2T d 5

5 5 5 5

5 5

5 5 5

5 5 d 5

2e2x2q2i 5

5 5 5

5 5 5 5

5 5 5

55 5 5 5 5

5 5 5

5 5

5 5 d 5 5

d 5 5 5

55 5 5

5 5 5 5

5 5 vine2IIIGiEIH

vine2IIHGxEP

5 5

5 5 IVH ITH

5 5 IRH 5 5 5

5 5 5 5 5PHH 5 5

5 5

IPH 5

5 5 5 5

5 5 5 ITH 5

5 5 d IVH

5 5 5

5 5 5 5

vine2IIIGiEII5 2y22i 5

5 5 5 ITH 5 5

5 5

5 5 5d 5 5 IHI

5 5 5 PHH 5 5 5 5

5 5 5 5 5 5 5

55 d 5 5 5

5 5 5 5 5 5

5 IHH

5 5 5

5 5 5 IPH 5 5

5 5d5 5 5 5 5 5 IRH 5 5 IVH 5 5 5 5 5

5 55 55 5 5 5 5

5 5ITH 5 5 5 5 5 5 5 5 5PPH 5

5 5 5 5 555 5 5 5 5 5 555 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 55 5 5 5 5d 5 5 5 5 5 55 5 5 5 5 5 5IPH d vine2IIRGiEIP5 5 5 5 5 5 5 5 IRH 5 5 d 5 5 5 5 5 5 5 5 5 55 5 5

5 5 5 5 5

5 5 5 5PHH 5

5 5

d 5 5 5

5 5 5 goso 5 5IVH 5 5 vine2IIQGxEQ

5 5 5 5 5 5 5 55

5 5

5 5 5 555 5 55 555 5 5 5 5 5ITH 5 d 5 55 5 55 555 5 5

55 5 5 5

5 5 5 5 5 5 5 5 5 5 5 5 5 5

5 5 5 5 5 5 5 5 5

5 5 5

5 5 5 55 5 5 5 5 5 5 5 5 5 5 5 5 5 5

5 5 5 5 5 5 2e2v2v2i2 d 5 5 5 5 5 5

5 5 5555 5 5 5

d 5 5 5 55

5PHH

5 5 5 5 5 5 5

5 5

5 5 5 5

5 5

5 5 5 5 5 5

5 5

5 5 5 5 sh

5 5 5

5 5

5 5 5

5 5 5 5 5 5 5 5 5 d 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 555 5 5 5 5 5 5 5 5 5 5 5 5

5 5 5 5 5 5 5 5 5 5 5 5 5 5

5 5

55 5 5 5 5 5

5 5 5 5

5 5 5

5 5 5 5

5 5 5 5 5 5

5 5 5 5 5 d 5 5 5 goso2rojet2ere qeology eismiity2@wgFA elluvium qrnodiorite 5 H2E2P 5 P2E2Q x endesite wetsedimentry 5 Q2E2R eismi2rofile2votions fslt wetvolni 5 R2E2T gtlsite ylder2elluvium nd2qeology hite umie hiorite yrolsti QH2meter2ilevtion2ht2ourtesy2of2illim2eevedoD2 q2xwhF httpXGGgisdtFusgsFgovGnedG pnglomerte urtz2monzonite pelsi hyodite H I P uilometers pult2mpping2from2ihrd2hitmrshD qro hyolite httpXGGgeompsFgeoFuknsFeduGhtmlGosoFhtml qrnite uff H I P wiles

Figure 2: Geologic map of the central Coso Range showing locations of 2-D seismic reflection profiles acquired and processed for this study. Map data from R. Whitmarsh (Ph.D. dissertation, in progress). of the subsurface structure (Pullammanappallil et al., also were acquired to image deeper structures and 2001, this volume). A specific goal of this study was assess their relationship to shallow faults that to image moderately to steeply dipping brittle faults accommodate active strike-slip faulting and extension and fractures that may control permeability and in the central Coso Range. localize production in the field. The reflection data TECTONIC SETTING OF THE CENTRAL A more detailed description of the data acquisition COSO RANGE parameters and processing approach is presented in The Coso Range is a tectonically and volcanically Pullammanappallil et al. (this volume). active region along the eastern margin of the Sierra Nevada microplate, which moves about 13 mm/yr RESULTS northwest with respect to stable North America (Argus and Gordon 1991; Dixon et al. 2000). Correlation of Shallow Velocity Structure to Northwest motion of the Sierran microplate primarily Producing Zones in the Coso Field is accommodated by distributed strike-slip and We interpolated the shallow velocity structure normal faulting in the Walker Lane belt, a 100-km- between individual 2-D seismic lines to develop a 3- wide zone of active deformation bordering the D model of lateral velocity variations beneath the eastern Sierra Nevada (Dixon et al., 1995; 2000). At geothermal field. A commercial software a regional scale, crustal extension in the Coso Range (SlicerDicer, ver. 3.0.3, (c) copyright 1988-2000 is driven by a releasing stepover between the Airport Pixotec LLC) was used for 3-D visualization of the 2- Lake fault and the Owens Valley fault, two major D data. right-lateral strike-slip faults (Fig. 1). Extension in the stepover region is accommodated in part by Individual 2-D velocity tomograms show that a low opening of Coso Wash, an asymmetric half graben velocity zone is present in the 1.0 to 1.5 km depth that is bounded on the west by the Coso Wash fault. range beneath the main production area of the Coso Hot springs, fumaroles and localized hydrothermal field, and beneath a producing area known as the alteration are associated with the Coso Wash fault “East Flank” (Fig. 3). The average velocity of these along the northeastern margin of the Coso geothermal zones is about 4.6 km/s (15,000 ft/s), compared to field (Roquemore, 1980). N

DATA ACQUISITION AND PROCESSING L1 10 Mesozoic intrusive rocks of the Sierra Nevada L106a L109 L1 15 (feet) batholith underlie the Coso Range and host the Coso L1 11 L1 11a geothermal field (Duffield and Bacon, 1981). L1 14 L106 13 Layered reflective structure is absent or poorly L1 expressed in such rocks, which traditionally have Main East flank producing been regarded as “acoustically transparent” for the area purposes of seismic reflection imaging. This study uses a new processing approach to improve imaging (feet) of moderately to steeply dipping faults and fractures associated with lateral velocity contrasts in N Main crystalline rocks. The processing consists of the producing following steps: E area

(feet) L110 (feet) 1) P-wave first arrivals in the seismic data for L106a L109 individual lines are inverted to obtain the L115 L111 L111a shallow 2-D velocity structure (i.e., a L114 velocity tomogram) along the line. The L106 East flank L113 inversion is performed using a proprietary software called SeisOpt @2D (© Optim LLC 2000). (feet)

2) Kirchhoff pre-stack seismic images are developed for each line by using the velocity tomograms as a basis for migrating the reflection data. Pre- Figure 3: Selected views through the 3-D velocity processing of the data (muting, filtering, model of the central Coso Range showing low etc) was performed prior to the migration. velocity zones associated with the main producing area and East Flank production zones. crystalline bedrock velocities of about 6.1 km/s Reflection Imaging (20,000 ft/s) outside the zones. The low velocity zones also are associated with an aeromagnetic low Imaging Brittle Faults intensity anomaly that is centered on the producing A major objective of this study is to test the ability of area (Plouff and Isherwood, 1980). The 3-D velocity the data acquisition and processing approach to model helps delineate the extent and geometry of the image brittle faults and other potentially permeable two low velocity zones, and it reveals a north-south- features in crystalline rocks. To this end, multiple trending “ridge” of high velocity rocks separating seismic lines were acquired across northern Coso them (Fig. 3). Wash graben (Fig. 2). Of particular interest is the down-dip geometry of the Coso Wash fault, which is We tentatively attribute the reduced acoustic velocity associated with hot springs, zones of alteration and of the producing zones to pervasive fracturing and/or other surface manifestations of hydrothermal activity. hydrothermal alteration of the crystalline rocks. Extensive alteration could destroy primary magnetite The Coso Wash fault is clearly imaged as a in the bedrock and thus account for the magnetic low moderately east-dipping reflector in the processed over the geothermal field. If this interpretation is data (Fig. 4). West-dipping reflectors in the hanging correct, then the high velocity “ridge” between the wall of the Coso Wash fault are correlated with two low velocity zones may be relatively unaltered antithetic faults exposed along the eastern margin of bedrock, indicating long-term compartmentalization the graben (Fig. 2 and 4). The Coso Wash fault of hydrothermal circulation in the two main flattens abruptly or soles into a faint but discernable producing areas. reflecting horizon at about 4 km depth that can be traced laterally beneath Coso Wash, and westward beneath the geothermal field.

Line 106a Line 106 Coso Coso Wash Coso Wash Station number fault Station number Wash 103 108 113 118 123 133 121 126 131 136 141 146 151 156 161 166 171 176 181 186 191 196 128 103 108 113 118 123 133 121 126 131 136 141 146 151 156 161 166 171 176 181 186 191 196 128 0 3822 0 3822

5000 -1178 5000 -1178 Elevation (Feet) 10000 -6178 10000 -6178 Depth (Feet) 15000 -11178 15000 -11178

20000 -16178 20000 -16178

25000 -21178 25000 -21178

Figure 4: Composite north-south depth-migrated seismic line through Coso Wash formed by joining lines 106 and 106a (see Fig. 2 for locations of seismic lines). The crossing geometry of lines 106 and 106a provides three dimensional views of the antithetic faults and their termination against the Coso Wash fault. Note discontinuous deep reflectors in the 6-7 km depth range (20,000-23,000 ft). Producing area Producing area Coso Coso Sugarloaf Mt. Wash Sugarloaf Mt. Wash West Station Number fault East West Station Number fault East 228 233 240 238 173 113 118 168 138 153 188 193 203 208 178 143 148 158 163 133 198 213 218 183 223 108 228 233 240 238 103 128 123 173 113 118 168 138 153 178 188 193 203 208 133 143 148 158 163 213 218 183 198 223 108 103 128 0 4240 0 123 4240

5000 -760 5000 -760 Elevation (Feet) 10000 10000 -5760 B -5760

15000 -10760 15000 -10760 Depth (Feet) A

20000 -15760 20000 -15760

C 25000 -20760 25000 -20760

Figure 5: East-west depth-migrated line 109 passes through the main producing areas of the Coso field (see Fig. 2 for location of line). The surface trace of the Coso Wash fault is correlated with an east-dipping reflector that can be traced down-dip to a depth of about 4.3 km (14,000 ft), where it appears to terminate against or merge with the subhorizontal reflective zone (A) imaged on lines 106 and 106a (Fig. 4). Moderately east-dipping reflectors west of the Coso Wash fault (B) also appear to terminate against the 4 km reflecting horizon. West-dipping reflectors appear to terminate against the B reflectors. A high amplitude reflector (C) present at 6 km depth (20,000 ft) beneath the geothermal field is associated with a zone of reflectivity or discontinuous reflectors that can be traced laterally beneath most of the seismic reflection array.

the upper 4 km of the crust is accommodated by Deep Reflectors brittle faulting within the field and opening of the The depth-migrated data consistently image a 1- to 2- Coso Wash graben to the east (Fig. 6). The brittle km-wide zone of discontinuous reflectors in the 6-9 faults appear to sole into or terminate against a km depth range beneath the seismic array (Fig. 4). subhorizontal reflecting horizon at 4-5 km depth; we The reflectors can be traced laterally and clearly tentatively interpret this horizon to be the brittle- correlated on crossing lines, and thus do not appear to ductile transition, consistent with the observation that be processing artifacts. most seismicity in the vicinity of the geothermal field and western Coso Wash graben is confined to the upper 4 to 5 km (Fig. 6). Ductile stretching of the A striking high-amplitude reflector is present within crust and emplacement of shallow igneous bodies this zone at 6 km depth beneath the main production may accommodate extension at depth, and area (Fig. 5). This feature is located beneath and particularly beneath the geothermal field. The high adjacent to some of the youngest volcanic features in amplitude reflector at 6 km depth beneath the the central Coso Range (Duffield and Bacon, 1981). geothermal field may be a magmatic sill or a zone of The reflector is coincident with a low-velocity fluid derived from a magma body below 6 km. The teleseismic S-wave converter at 6 km depth observed intrusions provide a localized source of heat that via receiver function studies (C.H. Jones, personal confines brittle deformation to the upper 4-5 km of communication, 2000). We interpret the high the crust and drives hydrothermal circulation in the amplitude reflector to be a localized zone of fluids field. and/or partial melt. DISCUSSION TECTONIC INTERPRETATION The structure beneath the Coso field in Fig. 6 is The Coso geothermal field is associated with a zone similar to a generalized model proposed by Fournier of localized crustal extension in a releasing stepover (1999) for hydrothermal processes in magmatic- between two major strike-slip faults. Extension of epithermal environments. In Fournier’s model, Approximate East-West Extent of Reflection Data Coverage Coso Wildhorse Geothermal Rose East Mesa Coso Wash Field Valley West Sugarloaf Mt. 2

t l

u B a 0 f B h s a -2 W o s Co Depth (km) on -4 ctile Transiti A Brittle-Du C -6

0 2 4 6 8 10121416182022242628 Horizontal Distance (km)

Figure 6: Interpreted relationship between major features imaged on lines 106 and 109, mapped geology and seismicity (events recorded by the Southern California Seismic Network). Reflector A lies at or near the base of seismicity and is interpreted to be the local brittle-ductile transition zone beneath the Coso field. Brittle faulting accommodated by the southeast-dipping Coso Wash fault and antithetic faults of Wildhorse Mesa is confined to the upper 4 km above the brittle-ductile transition. East-dipping reflectors west of the Coso Wash fault (B) underlie Sugarloaf Mt. and some of the youngest volcanic domes in the Coso Range. These reflectors may be moderate to low-angle normal faults or magma conduits. The high-amplitude reflector at 6 km depth beneath the geothermal field (C) is interpreted to be a lens of fluid or a magmatic sill. emplacement of a magma body may locally elevate brittle faulting and hydrothermal alteration. If this the brittle-ductile transition zone to shallow depths. interpretation is correct, and if it can be generalized The transition zone separates a hydrostatically to other geothermal regions, then it suggests that pressured domain above, in which meteoric water velocity models obtained in the course of processing flows convectively, from a lithostatically pressured reflection data may be a useful geothermal domain below. Depending on the orientations and exploration tool. Depth-migrated images that relative magnitudes of the principal stresses, fluids incorporate detailed velocity data reveal substantial expelled from underlying magma bodies may reflective structure in the upper 6-8 km of the crust, accumulate in horizontal lens-like bodies below the including brittle faults and deeper features that are brittle-ductile transition zone. According to Fournier possibly related to magmatic activity. These data (1999), the self-sealed transition zone may provide new insights into active tectonic and periodically be breached by brittle failure, allowing magmatic processes in the central Coso Range. trapped hypersaline brine and gas to be expelled upward into the brittle domain. REFERENCES Argus, D.F., and Gordon, R.G. (1991), “Current Sierra SUMMARY AND CONCLUSIONS Nevada-North America motion from Very Long New processing approaches used for this study baseline interferometry: implications for the demonstrate that seismic imaging can provide kinematics of the western United States”, Geology, valuable information about the structure of a 19, 1085-1088. geothermal field hosted in crystalline (i.e., Dixon, T.H., Robaudo, J.L., and Reheis, M.C. (1995), ‘transparent”) rocks. Data on the shallow velocity “Constraints on present-day Basin and Range structure from inversion of P-wave first arrival times deformation from space geodesy”, Tectonics,14, in the seismic data reveal that producing areas of the 755-772. Coso field are associated with relatively lower Dixon, T.H., Miller, M., Farina, F., Wang, H., acoustic velocities, which we attribute to localized Johnson, D. (2000), “Present-day motion of the Sierra Nevada block and some tectonic King, R.W., Hager, B.H., McCluskey, S.C., and implications for the Basin and Range province, Meade, B.J. (1999), “Reduction and utilization of North American Cordillera”, Tectonics, 19, 1-24. GPS data from the Navy geothermal crustal motion Duffield, W.A., and Bacon, C.R. (1981), “Geologic network”, Final Technical Report prepared for the map of the Coso Volcanic field and adjacent areas, Geothermal Program Office, Naval Air Warfare Inyo County, California”, U.S. Geological Survey Center, China Lake, CA, Contract No. N68936-95- Miscellaneous Investigations Series, Map I-1200, C-0371, 30 p. scale 1:50,000. Plouff, D., and Isherwood, W.F. (1980), Fournier, R. (1999), “Hydrothermal processes related “Aeromagnetic and gravity surveys in the Coso to movement of fluid from plastic into brittle rock Range, California”, Journal of Geophysical in the magmatic-epithermal environment”, Research, 85, 2491-2501. Economic Geology, 94, 1193-1212. Pullammanappallil, S., Hanjas, W., Unruh, J., and Jennings, C.W. (1994), “Fault activity map of Monastero, F. (2001), “Use of advanced data California and adjacent areas”, California processing techniques in the imaging of the Coso Department of Conservation, Division of Mines geothermal field”, this volume. and Geology, Geologic Data Map No. 6, scale Roquemore, G. (1980), “Structure, tectonics and stress 1:750,000. field of the Coso Range, Inyo County, California”, Journal of Geophysical Research, 85, 2434-2440.