Evidence for dyke intrusion earthquake mechanisms near Long Valley caldera, California

Abstract

A re-analysis of the magnitude 6 earthquakes that occurred near Long Valley caldera in eastern California on 25 and 27 May 1980, suggests that at least two of them, including the largest, were probably caused by fluid injection along nearly vertical surfaces and not by slip on faults. Several investigators1,2 have reported difficulty in explaining both the long-period surface-wave amplitudes and phases and the locally recorded short-period body-wave first motions from these events, using conventional double-couple (shear fault) source models. They attributed this difficulty to: (1) complex sources, not representable by single-fault models; (2) artefacts of the analysis methods used; or (3) effects of wave propagation through hypothetical structures beneath the caldera. We show here that the data agree well with the predictions for a compensated linear-vector dipole (CLVD) equivalent-force system3 with its principal extensional axis horizontal and trending N 55–65° E. Such a mechanism is what would be expected for fluid injection into dykes striking N 25–35° W, which is the approximate strike of numerous normal faults in the area.