DEMONSTRATING THAT THE 1989 LOMA PRIETA EARTHQUAKE DAMAGE IN SOUTHWESTERN SANTA CLARA VALLEY IS NON-RANDOMLY ALIGNED AND IS POSITIVELY ASSOCIATED WITH MAPPED LINEAMENTS

Damage to public infrastructure (streets, curbs, water and gas lines) in southwestern Santa Clara Valley (Schmidt and others, 1995) is used to support the interpretation of photointerpretive lineaments as evidence of tectonic features reactivated by the 1989 Loma Prieta earthquake, and to show that the orientation of the damage is non-random. Much of the deformation causing the damage was compressional, and seems to be located along faults mapped in bedrock and Quaternary alluvium at the southern end of the valley, and to some extent to follow a NW-trending pattern of photointerpretive lineaments to the NW (Hitchcock and others, 1994). Further NW the damage shows a linear pattern that could indicate continued faulting.

To quantitatively test whether the photointerpretive lineaments indicate tectonic features, we examine whether the damage is located preferentially closer to the mapped lineaments than to spatially random points. The method compares the distance of the damage from the lineaments to the distance of the theoretical distribution of randomly located points from the lineaments. To assess whether the mapped damage locations by themselves are preferentially aligned along a NW trend, we compare the orientation of pairs of damage locations to the theoretical distribution of the orientation of pairs of randomly located points. The problem is complicated by the elongate shape of the study area, which causes preferential directions parallel to the long axis in a set of randomly located points. The method distinguishes randomly aligned from non-randomly aligned pairs of points. Both methods provide a way to quantitatively specify the degree of association between a set of mapped point locations and geologic features.

The tests confirm that the damage is located preferentially close to the lineaments and that the damage is preferentially aligned along a trend of 315°, suggesting a tectonic cause for the observed patterns. This analysis provides a quantitative basis for more focused research into possible faulting beneath the alluvium in Santa Clara Valley.