Possible biases in the estimation of earthquake recurrence and seismic hazard from geologic data

Steven G. Wesnousky

Under Review 2009, SCEC Contribution #1407

Aseismic deformation, an integral part of the earthquake process, may be leading to systematic biases in the estimation of earthquake size, recurrence, and attendant strong ground motions in seismic hazard analyses founded on the geologic description of the locations, lengths and slip rates of active faults. Observations are reviewed and presented here to suggest that large earthquakes systematically rupture to increasingly greater depths below the seismogenic layer and that the portion of slip accommodated by aseismic processes during earthquakes is inversely proportional to rupture length. When taken together, the expected seismic moment of earthquakes on mapped faults may be systematically overestimated as a function of rupture length when derived from regressions of seismic moment and aftershock area, and estimates of the seismic moment rate on mapped faults derived from geologic measures of fault offset may be systematically overestimated for smaller faults. When predicting seismicity over a regional population of mapped faults and slip rates using the expression T= Uexp / geo, where Uexp and geo are the expected coseismic slip and geologic slip rate of the respective faults, the biases may be leading to an overprediction of the relative number of small to large earthquakes.

Citation
Wesnousky, S. G. (2009). Possible biases in the estimation of earthquake recurrence and seismic hazard from geologic data . Bulletin of the Seismological Society of America, (under review).