Postseismic deformation and stress evolution following the 2015 M 7.2 Sarez (Pamir) earthquake constrained by 5 years of Sentinel-1 and ALOS-2 observations

Abstract:

We investigate the effective rheological properties of the Earth’s crust and upper mantle in the Pamir orogen at the north-west (NW) margin of the Tibet Plateau using space geodetic observations of deformation due to the 2015 M7.2 Sarez (Tajikistan) earthquake. The earthquake occurred on the NNE striking Sarez-Karakul fault system in the central Pamir on Dec. 7^th, 2015, and ruptured an ~80 km long, sub-vertical, sinistral fault producing the maximum surface offset of ~3 meters. We processed Sentinel-1 and ALOS-2 Interferometric Synthetic Aperture Radar (InSAR) data and derive the timeseries of postseismic displacements due to the M7.2 earthquake. To mitigate challenging surface conditions that include ionospheric artifacts (primarily affecting ALOS-2 data) and highly elevated rugged topography and snow cover (affecting both Sentinel-1 data and ALOS-2 data), we used Persistent Scatterer and Split-Spectrum methods. Time series of line of sight (LOS) displacements were obtained for coherent pixels that were mostly confined to local valleys. The largest postseismic signal is localized within 10~20 kilometers from the rupture trace on the west (handing wall) side of the fault, primarily at and beyond the NNE end of the 2015 rupture. The observed LOS velocities of 10-30 mm/yr are highest immediately following the earthquake, and gradually decay over a characteristic time scale of 2-3 years. Both kinematic and stress-driven inversions show that the observed LOS displacements are well explained by the left-lateral afterslip NNE of the large-coseismic slip area. The data do not show any clear signature of viscoelastic or poroelastic rebound during the observation period. We investigate possible stress transfer and triggering relationships between the Sarez earthquakes and several M6+ events that occurred further north along the Pamir frontal thrust system within a few years after the M7.2 mainshock. We use a finite-fault slip model of the 2015 earthquake to compute co- and postseismic stress changes at the hypocenters of the M6 events.