EVALUATION OF EARTHQUAKE POTENTIAL FOR CHINA: A GEOLOGICAL APPROACH

We present a long-term earthquake potential model for China based on locations, slip rates, and lengths of Quaternary faults. To compare our model with the Global Seismic Hazard Assessment Program (GSHAP) model, we use the GSHAP seismic zones with some minor adjustments. For each seismic zone, we assume a tapered Gutenberg-Richter magnitude distribution with b-value of 1. The corner magnitudes are estimated using the Wells and Coppersmith [1994] relationship, applied to the longest fault in each zone. The a-values are estimated using moment rate conservation. We take the thickness of brittle crust to be 14.4 km to make the total predicted earthquake rate over the whole area consistent with observed earthquakes from 1950 to 1999.

We test the geological model and the GSHAP model retrospectively using the earthquakes from 1950 to 1999, and "prospectively" using earthquakes after 2000. Retrospectively the total number of earthquakes is consistent with both hypotheses, but the distribution of those earthquakes by zone is consistent with neither. In the prospective tests, both models pass all tests, although the geologic model fits the data slightly better.