Poster #128, Stress and Deformation Over Time (SDOT)

Anisotropic strength of fault zones suggested from geodesy and seismic imaging: Example from the Main Himalayan Thrust

Vera Schulte-Pelkum, Shaoyang Li, William D. Barnhart, & Marianne Karplus
Poster Image: 

Poster Presentation

2020 SCEC Annual Meeting, Poster #128, SCEC Contribution #10462
We combine InSAR, GPS, and seismic observations to jointly image the structure and behavior of the Main Himalayan thrust fault (MHT) in the vicinity of the 2015 M7.8 Gorkha earthquake. Data are geodetic observations of coseismic surface deformation spanning the earthquake from three InSAR tracks as well as GPS offsets, and structure based on azimuthal analysis of HICLIMB, NAMASTE, and HIMNT temporary experiment records.

Previous studies using receiver functions and surface waves proposed a several km thick low-velocity layer associated with the MHT. Such a low-velocity layer would imply a weak layer in terms of deformation, since seismic velocity and elasticity modules are li...
nked. On the seismic imaging side, the same low-velocity layer has also been proposed to host significant seismic anisotropy, in addition to or instead of low isotropic velocities.

Modeling and inversion of horizontal, vertical, and line-of-sight surface displacements suggests that the geodetic data are better fit when including an oriented strength heterogeneity within the low-velocity layer (i.e., strength anisotropy), rather than with a homogeneous rigidity structure or with a layered rigidity structure. We are systematically unable to fit both the vertical/line-of-sight displacements and the horizontal displacements unless we allow the rigidity of the low velocity zone to vary between the fault-parallel and fault-normal directions. We find that the low velocity zone is strong in the vertical/fault-normal orientation and weaker in the horizontal/fault-parallel orientation, suggesting an anisotropic strength of the layer.

Radial and transverse component receiver functions show conversions whose amplitudes vary sinusoidally as degree-1 in backazimuth with polarity reversals and negligible moveout, from depths corresponding to the top and bottom of the MHT-associated layer. Such arrivals are best matched with intermediate to steeply north-dipping foliation within the layer. Apparent low-velocity layer contributions to the signal may result from uneven azimuthal sampling or subvertical planes of symmetry.

These observations may suggest that significant fabric develops within sediments carried on the underthrusting Indian plate, creating an anisotropic strength profile that affects seismic observations as well as geodetic observations, modeling, and inversions. The isotropic assumption may therefore lead to bias in fault observations from both fields.