Block Modeling of Taiwan Active Faults and Its Application to Earthquake Hazard Analysis (Ii)

Taiwan locates in a high seismic activity belt of the southeastern Asia. In central Taiwan, manylarge earthquakes occurred in historic time, such as the 1906 Meishan (ML= 7.1), the 1935Hsinchu-Taichung (ML=7.1), and the 1999 Chi-Chi (Mw= 7.6) earthquakes. GPS horizontal velocities, inaddition, also show a shortening rate of about 30 mm/yr across the Central Range to the DeformationFront in the central Taiwan. However, Earthquake forecasts are necessarily an underinformed process,and we have not witnessed enough earthquakes anywhere in the world to make accurate forecasts basedsolely on historical catalogs.Extensive use of geodesy in the form of Global Positioning System (GPS) observations is a newfeature brought into the earthquake probability analysis in the past few years, such as the Working Groupon California Earthquake Probabilities (WGCEP) forecasts for the Uniform California EarthquakeRupture Forecast, version 3 (UCERF3) model. Geodetic measurements are potentially more spatiallycomprehensive than geologic offset observations, however, geodetic observations require a modelingstep to translate them into estimates of fault slip rate, and they have poor resolution on closely spaced,locked faults. Therefore the challenge is to use the geodetic and geologic data in the best, mostcomplementary way. In this proposal we employ a block-modeling method, DEFNODE, which isconstructed by dividing the crust into numerous closed, fault-bounded blocks, to evaluate theinterseismic fault slip behaviors in Taiwan with constraints from updated geological evidence oflong-term fault activity. This proposal intends to incorporate different geodetic observations such asGPS, precise leveling, and InSAR to evaluate the geodetic long-term slip, interseismic slip, andslip-deficit rates. By considering the effect of contemporary surface deformation, the results willprovide a new insight into the method of probabilistic earthquake source rupture analysis, with betterconstraints than the traditional seismic hazard analysis that have used long-term average fault slip ratesonly.