INVERTING FOR HETEROGENEOUS SLIP ON THREE-DIMENSIONAL FAULT SYSTEMS: A FIRST STEP TOWARD UNDERSTANDING FAULT MECHANICS

Many studies of earthquake triggering and fault interaction have relied on highly-idealized fault geometries and slip distributions. Geological and geophysical observations, however, reveal that faults typically are not single planar surfaces with uniform slip bounded by rectangular tiplines, but are composed of multiple curved surfaces with curved tiplines and heterogeneous slip distributions. The segments typically are organized into echelon, conjugate, and intersecting patterns. The discontinuities, bends, intersections, and slip heterogeneities generate stress concentrations that may promote or inhibit slip on nearby faults and thus play an important role in the mechanics of fault systems. It is therefore important to incorporate both realistic fault geometry and slip distributions when evaluating models of fault mechanics.

We have developed a new three-dimensional slip-inversion method based on the analytical solution for an angular dislocation in a linear-elastic, homogeneous, isotropic, half-space. The approach uses the boundary element code Poly3D that employs a set of planar triangular elements of constant displacement discontinuity to model fault surfaces. The use of triangulated surfaces as discontinuities permits construction of fault models that better approximate curved three-dimensional surfaces with no overlaps or gaps, bounded by curved tiplines. Slip inversion on three-dimensional surfaces therefore allows investigations of fault models that incorporate more realistic geometry and heterogeneous slip.

We have applied the method to invert for coseismic slip associated with the 1999 Hector Mine and 1995 Kozani-Grevena earthquakes, using InSAR and GPS observations of surface displacements. Three dimensional fault models were constructed by integrating available data sets including mapped surface ruptures, relocated aftershocks, and previous inversions for subsurface geometry. The resulting models improve the fit to the near-field geodetic data and more faithfully honor observations of fault rupture geometry. Models such as these form the starting point for more complete evaluations of fault mechanics and failure criteria.