EGU General Assembly 2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Slip model of the 2013 April 16 Mw 7.7 Saravan intra-slab earthquake (Makran subduction zone) derived from InSAR, GPS, and Teleseismic P-wave modeling

Andrea Walpersdorf1, Meysam Amiri2, Erwan Pathier1, Zahra Mousavi2, Fatemeh Khorrami3, and Sergey V. Samsonov4
Andrea Walpersdorf et al.
  • 1CNRS/UGA, ISTerre, Grenoble Cedex 9, France (
  • 2Department of Ear¬th Sciences, Institute for Advanced Studies in Basic Sciences (IASBS), Zanjan 45137-66731, Iran
  • 3National Cartographic Center, Tehran 1387835861, Iran
  • 4Canada Centre for Mapping and Earth Observation, Natural Resources Canada, Ottawa, ON K1A 0G1, Canada

The 2013 April 16 Mw 7.7 Saravan earthquake, an intra-slab earthquake with a normal faulting mechanism at of 50 km depth, occurred in the western part of the Makran subduction zone, where the Arabian oceanic lithosphere subducts northward under Iran and Pakistan. This event was the first instrumental recorded earthquake with a magnitude larger than Mw 6 since the last century. Studying this earthquake using geodetic and seismological data brings a unique opportunity to measure surface displacement due to the earthquake and assess causative fault parameters. Furthermore, it enables us to address some problems in the Makran subduction zone including slab dip angle, depth of dip angle change.

We used interferograms generated from RADARSAT-2 Synthetic Aperture Radar (SAR) data and coseismic GPS velocity field to combine with teleseismic P-wave data to model source fault parameters. First, we apply uniform slip modeling using a Bayesian bootstrap optimization nonlinear inversion method to find causative fault parameters. We specify search grids based on the LOS displacement map and focal mechanism solutions for each fault parameter to find the best solutions. These parameters include length, width, depth, strike, dip, rake, slip, location of the fault plane, rupture nucleation point, and origin time. Based on some prior tests and seismological information of earthquake, we decreased the search area of each parameter: depth 30- 70 km, dip 40˚- 80˚, strike 200˚-250˚, length 50-120 km, width 30-50 km, rake -150˚ -80˚ slip 1-4 m and let rupture nucleation point and origin time to be wide enough implying that all possible and reasonable fault geometry and kinematics parameters can be explored. Synthetic static displacements and seismic waveforms in a layered medium were computed with the Green's functions calculated using QSSP and PSGRN/PSCMP, respectively (Wang et al., 2006; Wang et al., 2017). A Green's function store contains pre-calculated Green's functions on a grid for combinations of source depth and source-receiver surface distance. For the layered half-space medium, we used the velocity structure of the GOSH seismic station to derive the Green Functions (Sebastian et al., 2016). After 450,000 iterations, the waveform fits, subsampled surface displacements as observed, modeled, and residual maps based on the best model are resolved. The distributions and resulting confidence intervals indicate that the parameters were well constrained. The joint inversion's best result indicates that the Saravan 2013 causative fault is a North-dipping normal fault with a dip of ~ 67°. The earthquake source length and width are approximately 120 and 80 km respectively.  In the second step, we model the derived fault plane in the previous step to retrieve the distributed slip model, allowing the slip to vary across the fault plane. In this step, all the parameters assumed fixed except slip. We extend fault length and width to 150 km and 100 km to prevent unwanted slip in the corners. The slip variation along the causative fault is characterized by one significant patch at the depth between 30-65 km with a maximum magnitude of about 4 m at 42-52 km.

How to cite: Walpersdorf, A., Amiri, M., Pathier, E., Mousavi, Z., Khorrami, F., and Samsonov, S. V.: Slip model of the 2013 April 16 Mw 7.7 Saravan intra-slab earthquake (Makran subduction zone) derived from InSAR, GPS, and Teleseismic P-wave modeling, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7507,, 2022.