Space-time analysis of observed and modeled aftershock sequences

极端地震可以在一个尚不清楚的松弛过程中产生数万次余震。余震的速率在时间上按幂规律衰减(大森定律)，类似于在其他极端现象之后观察到的不随时间变化的衰减率。如果能够利用物理论证和数值模型来确定余震的驱动机制，并与大量的余震模式数据集进行比较，那么在理解地震及其极端极限方面将取得重大进展。目前关于余震产生的一种假说是地震后对深部被困高压震源的脱气作用。在这种情况下，随之而来的压力脉冲“侵入”上覆的地壳，应该会沿着流动路径产生复杂的树状结构，这在余震目录中应该是可以辨认出来的。这项建议的目的是寻找、量化和模拟这样的结构。我们建议详细分析不同构造环境的三维余震图像的时间演化，以评估观察到的图像是否指示了流体压力诱发的余震。这将需要对非线性时空模式进行分析，并开发新的算法来可视化和模拟这种情景。具体地说，我们将开发可视化技术来研究余震序列的时空演化，2)通过识别假设的基本网络结构来量化余震的演化，3)开发包括主要物理和化学过程作用的三维数值模式，以及4)在模型结果和余震观测之间进行详细的比较。

Extreme earthquakes can generate tens of thousands of aftershocks in a relaxation process that is not understood. The rate of aftershocks decays as a power law in time (Omori’s Law), similar to the time-invariant decay rates observed following other extreme phenomena. If the mechanisms driving aftershocks can be determined using physical arguments and numerical models to compare with vast datasets of aftershock patterns, then significant advances will be made in understanding earthquakes and their extreme limits. A current hypothesis for aftershock generation results from post-seismic degassing of trapped high pressure sources at depth. In this scenario, the ensuing pressure pulse 'invades' the overlying crust that should produce complex tree-like structures along flow paths that should be discernible in aftershock catalogs. The purpose of this proposal is to search for, quantify, and model such structures. We propose to analyze in detail the time evolution of 3-dimensional aftershock patterns from a variety of tectonic environments to evaluate whether observed patterns are indicative of fluid-pressure-induced aftershocks. This will require nonlinear space time pattern analyses and the development of new algorithms to visualize and model this scenario. Specifically, we will 1) develop visualization techniques to study the space-time evolution of aftershock sequences, 2) quantify the evolution of aftershocks by identifying the hypothesized underlying network structure, 3) develop 3- dimensional numerical models that include the dominant physical and chemical processes acting, and 4) make detailed comparisons between model results and aftershock observations.