SETAS: A Spherical Version of the Space–Time ETAS Model

Abstract

AbstractTo analyze and forecast seismicity globally or in high-latitude regions, and particularly in relation to remote seismicity interactions on a worldwide scale, we reformulate the space–time epidemic-type aftershock sequence (ETAS) model from a planar to a spherical version and implement it in a FORTRAN software package, spherical ETAS (SETAS). This article explains the details of the related algorithms and the usage of the package, which can fit the SETAS model to seismicity data from regions specified by three types of boundaries: no, quadrilateral, and spherical polygon boundaries. The software is tested with the Global Centroid Moment Tensor (Global CMT) catalog and compared to the original planar ETAS model by fitting both the versions to local seismicity in Japan and Alaska. The results show that the spherical model has a high computational efficiency, even with a large range of global data, and is consistent with the well-established planar version at the local scale.