余震时空丛集对概率地震危险性分析的影响

本文以东北、华北及川滇地区为例，系统研究了余震时空丛集对概率地震危险性分析的影响.采用基于传染型余震序列模型（ETAS）的蒙特卡罗模拟方法，模拟了包含余震和不包含余震的两套地震序列，然后以模拟地震目录为基础输入，采用基于空间光滑地震活动性模型的地震危险性分析方法计算了两套地震危险性结果——PGA（Peak Ground Acceleration，峰值加速度），通过分析比较这两套PGA的绝对差值和相对差值来研究余震时空丛集对概率地震危险性分析的影响.研究结果表明余震对50年超越概率10%地震危险性计算结果的影响均值为6%左右，最大可达10%，并且随着超越概率水平的提高，余震影响也越大.弱地震活动区余震对概率地震危险性分析的影响要高于强地震活动区.研究结果还进一步揭示两套PGA结果绝对差值的最大值约为15 cm·s^-2，且出现在高PGA区，这意味着余震对概率地震危险性计算结果不会产生显著影响.因此在地震区划或一般性地震危险性分析中可考虑不用删除余震.

Effect of temporal-spatial clustering of aftershocks on the analysis of probabilistic seismic hazard

We systematically study the effect of temporal-spatial clustering of aftershocks on the probabilistic seismic hazard analysis (PSHA). We generate Poissonian and clustered catalogues using Monte Carlo simulations based on the Epidemic Type Aftershocks-Sequence model and calculate two sets of peak ground-motion acceleration (PGA) values from these synthetic catalogues. Then the differences between the two sets of PGA values are compared to study the effect of aftershocks on PSHA. The method is applied to three regions, which have significant differences in seismicity: Northeast China, North China, and Sichuan-Yunnan. The results show that the average contribution of aftershocks on PGA with 10% probability level of exceedance in 50 years (10% P.E. in 50 yr) is approximately 6% with upper bound around 10%. The effect of aftershocks on PGA increases with the probability level of exceedance, and is smaller in regions of high seismic hazard than those of low seismic hazard. Furthermore, the maximum absolute difference between the two sets of PGA values with 10% P.E. in 50 yr is approximately 15 cm·s^-2 in regions of higher seismic hazard in North China, showing that the results of the seismic hazard analysis do not significantly change when aftershocks are considered. Thus, aftershocks need not be discarded in general seismic zonations or seismic hazard analyses.