Cordilleran Section - 101st Annual Meeting (April 29–May 1, 2005)

Paper No. 10
Presentation Time: 11:20 AM

A STRONG LOMA PRIETA FAULT, AS INFERRED FROM THE 1989-1990 LOMA PRIETA AFTERSHOCK SEQUENCE


TWISS, Robert J., Univ California - Davis, 1 Shields Ave, Davis, CA 95616-8605 and UNRUH, Jeffrey R., William Lettis & Associates, Inc, 1777 Botelho Drive, Suite 262, Walnut Creek, CA 94596, twiss@geology.ucdavis.edu

We inverted aftershock focal mechanisms of the 1989 Loma Prieta earthquake for the orientations of the principal instantaneous strain axes, their relative magnitude, and the relative vorticity parameter for each of 17 spatially clustered sets subdivided into 33 subsets each of which accommodates a homogeneous deformation. The aftershocks illuminate three main segments (South, Central and North) of a restraining bend in a blind fault beneath the Santa Cruz Mountains having an upper tip line between 4 and 6 km depth. Above the tip line is a set of faults striking counterclockwise from, and en echelon along, the main fault trend. Planes of hypocenter alignments in each set define a more complex substructure. Solutions show a common orientation for the principal shortening axes d3 of approximately [06°, 019°]. The principal lengthening axes d1 occupy a girdle with sub-horizontal and sub-vertical maxima. Bulk triaxial strains for individual hypocenter sets are accommodated by a partitioning into a pair of plane strains with one principal axis in common and the other two axes approximately exchanging orientations. Faults above the tip line dominantly accommodate crustal thickening. Slip lines on the main fault segments for each solution are dominantly strike-slip on the Southern Segment, a mixture of strike-slip and reverse on the Central Segment, and strike-slip with a large component of reverse slip on the Northern Segment. Previous studies of local strong motion inversions and broad band teleseismic inversions for the main shock show similar slip distributions. This similarity implies an incomplete stress drop during the main shock, and thus a relatively strong fault. Angles between the normals to the local hypocenter alignment planes and the solution for d3 for each set average between 40° and 55°. Assuming that instantaneous strain axes parallel the stress axes, these angles again imply a strong fault. The angle between the average d3 for all solutions, and the normal to the plane fit to all the hypocenters, is 29°. Previous studies inferred a weak fault from comparable small angles that were derived from similar averaging. This angle, however, is the result of inappropriately combining into a single solution, a set of heterogeneous instantaneous strains for a heterogeneous fault structure, thereby masking the actual mechanical relations.