Seismic magnitude calculation based on rate- and state-dependent friction law

基于速率状态摩擦定律的地震震级计算

• Heng-tao Yang  (杨横涛), Bing Bai  (白冰) & Hang Lin  (林杭)

Abstract

In the engineering related to seismic activities, such as shale gas extraction and EGS system, numerical simulation is a common research method. In current numerical simulations, the critical value of a variable (e.g., sliding velocity) is often set artificially to determine whether the element is destabilized, on which the seismic magnitude is calculated. This practice is highly subjective and arbitrary, which leads to inaccurate calculation results. In this paper, we propose a new method for calculating seismic magnitude based on the rate-state dependent friction law. Firstly, the sliding history of each element on the sliding surface is divided into stages, and then all elements in the coseismic stage are clustered in time and space. Finally, the seismic moment of an event is obtained by calculating the sum of the corresponding areas of the “cumulative slip-fault location” plot according to the clustering labels, which in turn gives the seismic magnitude.

摘要

在研究与地震相关活动的过程中,例如:页岩气开采,EGS系统等,数值模拟是一种常用的研 究手段。在目前的数值模拟中,往往通过人为设置某个变量(例如滑动速度)的临界值来判断单元是否 失稳,并以此为基础计算地震震级。这有很强的主观随意性,附加的人为因素可能对计算结果产生较 大影响,导致结果不准确。本文基于速率状态摩擦定律,提出了求解地震震级的新方法。首先,对滑 动面上的每个单元进行阶段划分。然后,对处于同震阶段的所有单元在时间和空间上进行聚类。最 后,根据聚类标签,计算累计滑移-断层位置图中对应区域的面积和,即可得到某事件的地震矩,进而 得到震级。本方法从摩擦定律本身出发,选择合适的判断标准对单元滑动过程进行阶段划分,并通过 聚类得到震级,减少了人为主观性对结果的影响。随后,通过一个垂直的二维光滑断层的案例分析对 该方法进行验证,结果证明该方法是合理的。另外,总结和讨论了目前几种判断地震的方法,重点分 析了临界功率密度在应用过程中存在的问题。