Resolving off-fault damage in the 2019 Ridgecrest earthquake from satellite geodesy

Title : Resolving off-fault damage in the 2019 Ridgecrest earthquake from satellite geodesy
Advisors : James Hollingsworth, David Amitrano, Jerome Weiss

Contacts :
james.hollingsworth univ-grenoble-alpes.fr
david.amitrano univ-grenoble-alpes.fr
jerome.weiss univ-grenoble-alpes.fr

Surface ruptures produced by earthquakes provide valuable insights into the mechanics of faulting. Correlation of pre- and post-earthquake optical satellite/aerial images allow the detailed retrieval of 3-D displacements in the near-field of faults. Analysis of the 2019 Ridgecrest earthquake, California, using stereo WorldView satellite imagery reveals important features in the near-field displacement pattern. The majority of slip is accommodated on a complex network of fault segments which fail in shear. However, the full surface strain tensor can also be retrieved from the 3-D surface displacements, which reveals a narrow zone of inelastic deformation (strain >0.5%) accommodated along a <50 m wide zone surrounding the main fault rupture. Inelastic strain varies uniformly along-strike, independent of lithology, structural complexity, or rupture velocity, and contrasts with other proxies for inelastic strain such as off-fault deformation (measured from peak across-fault offsets measured from geodetic surface displacement profiles). Nevertheless, different methodological approaches for computing the strain tensor can lead to noticeable differences in the pattern of inelastic deformation resolved, raising concerns about how robust is our current view of inelastic strain for the Ridgecrest earthquake.

In this project, the student will take advantage of a new set of very high resolution optical images (30 cm ADS-100 images) spanning the co- and post-seismic phases of the Ridgecrest earthquake to produce the most comprehensive map yet of near-field 3D ground displacements using optical image correlation. This data will be integrated with precise far-field measurements from time-series processing of Sentinel-1 InSAR data. Both datasets will then be used to estimate the strain field produced by the Ridgecrest earthquake, from which a multi-scale analysis will be undertaken to investigate the scaling properties of surface strain. The analysis will be extended to a non-deforming region to verify that topography does not impact the scaling properties. The scaling analysis of strain patterns may then provide constraints for mechanical modelling of crustal materials ; e.g. failure models incorporating progressive damage are able to generate a scale invariant deformation field under very simple loading conditions. This means that such scaling properties are not necessarily inherited from pre-existing fault patterns, or from complex loading, but emerge from a combination of (1) disorder at the small scales (always present in natural objects), (2) threshold mechanics (rupture or velocity weakening friction), and (3) elastic stress redistributions. Using the strain field for the 2019 Ridgecrest earthquake, the student will then estimate the elastic properties of the crust, which will in turn be used as input for numerical modelling studies of failure using a progressive damage model. Comparison of the resulting strain field with that derived from optical correlation data will allow us to optimize our model parameters, thereby providing robust estimates of the elastic properties for the entire fault zone. The goal of this analysis will be to develop a more dynamic view of how fault zones deform during single earthquakes.

Dates : February to June 2020