Seismic velocity structure and microearthquake source properties at The Geysers, California, geothermal area

Abstract

The method oC progressive hypocenter-velocity inversion has been extended to incorporate S-wave arrival time data and to estimate S-wave velocities in addition to P-wave velocities.Synthetic tests demonstrate that the joint use oC P and S-wave arrival time data has the Collowing advantages over the-use-oC-P---w,av.e-data-alone;-{-l-}--P--wav-e-v·elacity-and-slowness~gradient-structure are more accurately estimated; (2) hypocenter mislocation errors are substantially reduced, especially hypocentral depth; (3) convergence oC progressive inversions to local minima is more detectable using RMS data, misfits oC P and S-wave data; (4) velocity model and hypocenter estimates are more accurately determined when station corrections are used; (5) errors in linearized resolution and error estimated are reduced; and (6) complete elastic properties are estimated providing greater constraints Cor geologic interpretation oC velocity structure. Adding S-wave data to progressive inversion does not completely eliminate hypocenter-velocity tradeoffs, but they are substantially reduced.Results oC a P and S-wave progressive hypocente~velocity inversion at The Geysers show that the top oC the steam reservoir is clearly defined by a large decrease oC Vp/V. at the condensation zone-production zone contact. The depth interval oC maximum steam production coincides with minimum observed VpfV., and Vp/V. increases below the shallow primary production zone suggesting that reservoir rock becomes more fluid saturated. 2The moment tensor inversion method wast applied to three· microearthquakes at The Geysers. Estimated principal stress orientations were comparable to those estimated using Pwave first motions as constraints. Well constrained principal stress orientations were obtained for one event for which the 17 P-first motions could not distinguish between normal-slip and strike-slip mechanisms. The moment tensor estimates of principal stress orientations were obtained using far fewer stations than required for first-motion focal mechanism solutions. The three ,focal mechanisms lobtain!!d' here support the hypothesis that' focal mechanisms are a Junction of,depth at. The Geysers.'Progressive inversion as developed here and the moment tensor inversion method provide a complete approach for determining earthquake locations, P and S-wave velocity structure, and earthquake source mechanisms.

Fault zone structure determined through the analysis of earthquake arrival times

“…As noted by O'Connell (1986), application of the orthogonal matrix V j , calculated from the SVD of the matrix of hypocenter partial derivatives from both P-and Sarrival times, results in an annulled data set of mixed P and S type though, as with P-arrivals only, completely independent of the estimated hypocentrallocation.…”

“…We follow Pavlis and Booker (1980) and O'Connell (1986) in the formulation of the linearized and iterative inversion scheme.…”

“…Pavlis and Booker (1980) also introduced, together with Rodi et al (1980) and Spencer and Gubbins (1980), the parameter-separation technique to reduce the joint problem to a tractable size and proposed the progressive inversion scheme developed by Roecker (1982). The Pavlis-Booker method was extended to the joint inversion for P-and S-velocity models and station corrections by O'Connell (1986). Thurber (1983) also adopted progressive inversion, attacking the true three-dimensionality of most inversion problems by using a three-dimensional (3-D) ray-tracer on a model parameterized in terms of linear interpolating functions in a three-dimensional grid of node points having assigned velocities.…”

“…• .. J1p and J1s should be set as J1p/ J1s = Vs/Vp because they act on the P-and S-velocity model respectively (O'Connell, 1986;EberhartPhillips, 1989 ray density diagonal matrix having elements Each ray path is discretized in jk segments.…”

“…Pavlis and Booker (1980), O'Connell (1986), andO'Connell andJohnson (1991) are among the few authors that have addressed this topic, relying on the formulation of errors given by Backus (1970). We have addressed this topic with synthetic tests for various structures.…”