基于四川西昌中强震数据的区域地震动预测模型

为了探讨基于有限的本地数据建立区域性地震动预测模型的可行性,面向建设西部地震科学实验场的需求,基于中强震数据较丰富的四川西昌附近21次地震的1250条地震动,分析了区域性预测模型中距离项、震源项以及场地项的特征,标定了模型系数,建立了符合西昌地区地震动特征的预测模型。结果表明:西昌地区的地震动加速度反应谱在近场短周期的衰减速度快,而在中远场加速度反应谱衰减速率差别不大;随着震级增大,峰值加速度以及加速度反应谱随距离增加衰减速率变慢;加速度反应谱在短周期衰减速率明显快于长周期的衰减速率。与NGA模型比较表明,西昌地区的加速度反应谱在短周期与之相当,而在长周期NGA模型的预测值偏高,表明NGA模型在中长周期对四川西昌地区并不适用。与国内相关模型比较表明,西昌地区中强震在中近场会产生较大的峰值加速度以及短周期反应谱。该研究表明地震动具有显著的区域性差异,宜采用当地有限的地震动数据建立地震动预测模型。

Regional ground motion prediction model for moderate magnitude earthquakes in Xichang,Sichuan,China

To explore the feasibility of establishing ground motion predicted models based on local earthquake data,meet the need of building western earthquake science experiment field,the characteristics of the distance,source and site effect of the prediction model were analysed based on 625 sets of ground motions of 21 moderate earthquakes in Xichang region in Sichuan.A prediction model based on characteristics of the local ground motions was established.Results show that the rate of distance attenuation in the acceleration response spectrum of Xichang area in the near field and short period is very large.When the fault distance is over 30 km,the rate of attenuation in each period tends to be equal.With magnitude increases,the value of the PGA and Sa decreases slower with distance,the acceleration response spectrum in short period decreases obviously faster than the long period.In short period,the predicted model in this paper is not much different from the NGA model,while in the long period,the estimated values of the model in this paper is relatively smaller,compared with the NGA model’s values,and the NGA model overestimates the values obviously in the long period in Xichang’s area.Compared with the domestic models,it is possible that a medium-strong earthquake in Xichang region will produce a larger PGA and a larger acceleration response spectrum in the short period.This study shows that there are significant regional differences in ground motion,so it is advisable to use local ground motion data to establish a ground motion predicted model.