What controls Coulomb stress changes on thrust and normal faults? Insights from three-dimensional finite-element models including pore fluid pressure changes and postseismic viscoelastic relaxation

大陆内部断层上的地震对人口稠密地区构成了很大的地震危险，但与板块边界的地震危险相比，它们引发大地震的可能性受到的关注较少。最近的破坏性大陆内地震的例子包括2008年的汶川(中国)地震以及2009年L的阿奎拉和2016年的意大利诺西亚-阿马特里切地震。除了直接的破坏外，震源断层上的大地震还会在相邻断层上引起应力变化，这最终可能会触发或推迟其他地震。接收器断层应力状态的变化通常用库仑应力变化来表示，库仑应力变化是由同震位移和包括孔弹性效应和粘弹性松弛在内的震后瞬时过程引起的。到目前为止，许多关于库仑应力变化的研究都集中在自然界中的个别地震和断层上，对于这些地震和断层的瞬变和非瞬变过程的贡献往往存在很大的争议。这种模糊性的主要原因是，目前不可能直接测量库仑应力变化，因此库仑应力变化分析完全基于模型计算，而模型计算必须使用关于断层几何或实施瞬变过程的简化假设。例如，广泛使用的软件库仑根本不考虑孔弹性或粘弹性效应。此外，缺乏概念性模型来提供理论上预期的应力变化模式的信息，以及与自然地震引起的应力变化进行比较的基础。拟议的项目的目标是促进我们对瞬变和非瞬变机制如何控制大陆内逆冲和正常断层上的同震和震后库仑应力变化的理解。作为一种方法，我们将应用ABAQUS软件进行三维有限元建模。模型断层将被实现为摩擦接触界面，根据模型中暂时演化的应力和应变场积累滑移。在不同的实验中，我们会改变接收器断层的倾角和倾角方向，以及它们相对于震源断层的位置。由于模型的扩展或缩短，所有模型都将包括重力、均衡和区域应力场。作为库仑应力变化的原因，将考虑地震震级、孔洞弹性效应、震后粘弹性松弛和震间应力积累。通过实施这一系列过程，我们将首次系统地评估它们对同震和震后库仑应力变化的相对重要性。此外，我们将提供不同过程的空间和时间尺度的详细分析，这将使得在自然库仑应力变化模式中更好地区分孔洞弹性效应和地震后粘弹性松弛。

Earthquakes on intra-continental faults pose a substantial seismic hazard to populated areas but their potential to cause major earthquakes has received less attention than the earthquake hazard at plate boundaries. Recent examples of devastating intra-continental earthquakes include the 2008 Wenchuan (China) earthquake and the 2009 L'Aquila and 2016 Norcia-Amatrice (Italy) events. In addition to the immediate damage, a large earthquake on a source fault causes stress changes on adjacent faults (= receiver faults), which may ultimately trigger or delay other earthquakes. The alteration of a receiver fault's stress state is typically expressed in terms of Coulomb stress changes, which are caused by the coseismic displacement and transient postseismic processes including poroelastic effects and viscoelastic relaxation. So far, many studies of Coulomb stress changes have focussed on individual earthquakes and faults in nature, for which the contributions of transient and non-transient processes often remain highly controversial. The main reason for this ambiguity is that the direct measurement of Coulomb stress changes is currently impossible and that Coulomb stress change analyses are therefore based solely on model calculations, which must use simplifying assumptions concerning fault geometry or the implementation of transient processes. For example, the widely used software Coulomb does not consider poroelastic or viscoelastic effects at all. In addition, there is a lack of conceptual models that provide information on theoretically expected stress change patterns and a base for comparison with stress changes induced by natural earthquakes.The goal of the proposed project is to advance our understanding how transient and non-transient mechanisms control co- and postseismic Coulomb stress changes on intra-continental thrust and normal faults. As a method, we will apply 3D finite-element modelling with the software ABAQUS. The model faults will be implemented as frictional contact interfaces, which accumulate slip according to the temporarily evolving stress and strain fields in the model. In different experiment, we will vary the dip and dip direction of the receiver faults as well as their positions with respect to the source fault. All models will include gravity, isostasy and a regional stress field owing to extension or shortening of the model. As causes of Coulomb stress changes, the earthquake magnitude, poroelastic effects, postseismic viscoelastic relaxation and interseismic stress accumulation will be considered. By implementing this range of processes, we will provide the first systematic evaluation of their relative importance for co- and postseismic Coulomb stress changes. Furthermore, we will provide a detailed analysis of the spatial and temporal scales of the different processes, which will enable a better differentiation between poroelastic effects and postseismic viscoelastic relaxation in natural Coulomb stress change patterns.