Illuminating co-seismic deformation and seismogenic structures by combined 3-D seismic imaging and InSAR analysis: a case study from the Qaidam Basin, northern Tibetan Plateau
Description
Understanding the relationship between co-seismic deformation and seismogenic structures is usually impeded by limited knowledge of underground deformation. We herein attempt to precisely establish the relationship through a case study of 2019 Mangya Earthquake in the Shizigou anticline, western Qaidam Basin (northern Tibetan Plateau) based on multiple datasets. Three-dimensional seismic imaging reveals complicated underground structures, including a shallow thrust, a deeply-buried reverse fault, and an intervening salt-bearing weak layer. The InSAR geodesy reveals co-seismic surface folding in the northeastern limb rather than the core of the anticline. Elastic inversion suggests a likely underground co-seismic slip surface at depth of ~2.1 km. They together indicate that Mangya Earthquake was caused by neither the shallow thrust nor the deep reverse fault; instead, it likely resulted from bed-parallel slip caused by local salt flow. Our findings demonstrate that complicated seismic behavior in structurally complex regions could be resolved by illumination of seismogenic structures.
Files
Steps to reproduce
The original SAR images of Sentinel-1 data are freely available data from European Space Agency (https://scihub.copernicus.eu/), and the SRTM and Landsat data are available from the U.S. Geological Survey (https://earthexplorer.usgs.gov/). Pre-stack migrated seismic data in depth domain were used to depict the underground structure and its along-strike variation of the Shizigou anticline (Figure 2). These data have a spatial resolution of ~50 m and a maximum probing depth of ~10 km, deep enough to capture the deformation of the entire Cenozoic successions. The co-seismic surface deformation of the Mangya earthquake was extracted from Interferometric analysis of two Synthetic Aperture Radar (SAR) images acquired a couple of days before and after the earthquake from Sentinel-1A. The details of these images, including the acquisition date, perpendicular baseline, temporal baseline and orbit types, are listed in Table S2. DEM data from the Shuttle Radar Topography Mission (SRTM) was used to remove the topographic phase component (Farr et al., 2007). The adaptive filtering method was employed to reduce the phase noise and enhance the signal-to-noise ratio (Goldstein and Werner, 1998). The minimum-cost flow algorithm (Costantini, 1998) was used to unwrap the InSAR interferograms, which are subsequently processed by orbit refinement and re-flattening (Figure 3). LOS (light of sight) displacement distributions of the two tracks were then obtained by conversion of phase to displacement (Figure S2). Detailed description of the inversion of seismogenic structure are included in Supporting Information Text S1
Institutions
- Zhejiang University School of Earth Science