基于芦山主震及余震的地震预警参数与快速震级估算研究

Research on EEW parameters and rapid magnitude estimation for Lushan earthquake and aftershocks

我国四川地区继08年汶川地震后,在2013年再次发生了7级强烈大地震,造成了重大人员伤亡并诱发了多处灾难性滑坡灾害.地震预警系统是防震减灾中最为行之有效的方法之一,其中对地震震级的快速估算是非常重要且困难的.目前世界上提出较早、应用较为广泛的震级估计的方法主要有:基于最大卓越周期(τ_p^(max))、特征周期(τ_c)和最大位移幅值(Pd)方法.本文利用2013年芦山主震及余震P波初期部分的信息,研究了地震震级快速估算中三个预警参数与震级的相关性,提出了四川地区的震级估计模型,然后对其进行验证分析,并与前人得出的震级估计模型进行对比和评价.结果表明:三种方法均能在短时间内(3 s)有效地进行震级估算,总体而言,P_d方法估算效果最优,τ_c方法次优,τ_p^(max)方法较弱.在震级较大的主震震级估计中,三种方法均出现明显的震级低估(震级饱和)现象.对于τ_p^(max)方法,基于芦山地震的估计模型与基于汶川地震的估计模型较为接近,但与南加州地区模型区别较大,可能与四川地区龙门山断裂带的复杂地质条件有关;而τ_c方法的估计模型大体均较为接近,更具有普适性和稳定性.在地震预警系统的实际应用中,由于目前我国仍处于刚起步阶段,尚未建立密集的地震监测台网系统,因此在短时间内难以得到较为准确的震源距,所以与震源距相独立的τ_p^(max)和τ_c两种算法则显得较为实用,其中τ_c方法略优于τ_p^(max)方法,同时能较好地满足地震预警系统的精度要求,因此推荐使用τ_c方法应用于四川地区地震预警系统中的快速震级估算.需要注意的是,地震预警参数具有一定的区域适用性,即在不同震源机制、地形和地质条件影响下,不同地方的震级估算回归模型存在一定的差别.

In 2013, five year after the 2008 Wenchuan earthquake,a strong earthquakes of magnitude 7. 0 shocked Sichuan region again,which caused heavy casualties and induced the several devastating landslide disaster. Earthquake Early Warning system（ EEWs） is one of the most effective ways for seismic hazard mitigation,in which predicting earthquake magnitude is one of the most important and difficult tasks. Currently,different empirical models have been used to estimate earthquake magnitude based on τ_p^（max） or τ_c or P_dmethods. We studied the correlation between magnitude and three early warning parameters obtained from the early portion of P-waves for the Lushan earthquake. And we analyzed,verified,evaluated and compared the estimation models with previous studies. Results indicate that three methods can be effectively used for magnitude estimation in a short time（ 3 s）. In general,P_dmethod works best in magnitude estimation,and thenτ_cmethod,finally τ_p^（max）method. For large scale earthquakes,all of three methods come up the obvious magnitude underestimation（ saturation） phenomenon. For τ_p^（max）method,the model built up by Lushan sequence is close to the one built up by Wenchuan sequence instead of Southern California, which might be because of the complicated geological conditions in Longmenshan Fault Zone. Forτ_cmethod, the modelsare close to each other with the higher feasibility and stability. In the practical application of EEWs,at present our country is still at the fledgling stage and the seismic networks have not yet set up. Hence,it is difficult to get a relatively accurate hypocentral distance in a short time. So τ_p^（max）and τ_cmethod which are independent with distance are more practical. What is more,τ_cmethod is slightly better than τ_p^（max）method and it can well satisfies the requirement of the EEWs. To draw a conclusion,τ_c method is recommended to apply in rapid magnitude estimation of EEW system in Sichaun region. However,it should be noted that the earthquake early warning parameters have certain regional applicability. The parameters perform differently under the influence of different focal mechanism,the terrain and geological conditions.