Relationships between Strong Ground Motion Peak Values and Seismic Loss during the 1999 Chi-Chi, Taiwan Earthquake
A better real-time assessment of earthquake effects (i.e. seismic intensity estimation)is crucial for hazard mitigation. Especially during the aftermath of a disastrous event,significant reduction of loss can usually be realized through timely execution ofemergency response measures. These effects include strong-ground shaking, groundfailure, and their impact on man-made structures. The descriptive Modified Mercalliintensity scale, though still in common use in many poorly instrumented areas of theworld, is out of date in areas of extensive strong-motion instrumentation. It is desirableto place the earthquake intensity scale on a more quantitative basis based on the actualrecorded ground-motion shaking and carefully compiled damage records. In this paper,we investigated the relationships between earthquake loss, intensity and strong motionpeak values, mainly based on the Chi-Chi earthquake. Both the strong-motion peakvalues and the earthquake loss are related. From the results, we found that peak groundacceleration (PGA) and peak acceleration response spectra at 1 s period (1 s Sa) valuesare two parameters that give slightly higher correlation coefficients than other parametersfor earthquake loss analysis. For intensity estimations, the peak ground velocity (PGV)values and 1 s Sa values are better parameters in the high range and PGA is not stable forsmaller earthquakes. Although PGV values give a slightly lower correlation coefficientand larger standard deviation in seismic loss analysis during the Chi-Chi earthquake, itnevertheless gives more reliable instrumental intensity over a broad magnitude range.1 s Sa is a good parameter for both seismic losses and intensity evaluation. We thusconclude that PGV and 1 s Sa are relatively more stable in damage assessment and,at least in the high end, in intensity estimation. We shall incorporate these findings inour real-time earthquake rapid reporting and early warning systems.