Quantifying fault displacement via optical image correlation using structure-from-motion orthomosaics in the 2019 Ridgecrest earthquake sequence

Alex E. Morelan, Kenneth W. Hudnut, & Andrea Donnellan

Published August 15, 2020, SCEC Contribution #10761, 2020 SCEC Annual Meeting Poster #018

Optical image correlation has proven to be a valuable technique for mapping faults and quantifying displacements in the aftermath of surface-rupturing earthquakes. The traditional application of optical image correlation relies on precisely orthorectified and coregistered pre- and post-event images of the same scene, ideally collected using the same platform to reduce distortion biases introduced by the sensor and lens. We demonstrate that useful optical image correlation results can be produced using a variety of source imagery combinations including orthomosaics derived from low-altitude aerial photographs. Images collected on UAV and helicopter platforms are an order of magnitude higher resolution than satellite images, enabling the ability to resolve smaller displacements than those possible using satellite imagery. The flexibility introduced by combining imagery across different platforms allows for higher spatial and temporal resolution correlation maps. We use helicopter photographs collected on 7/5/2019 (between the Mw 6.4 foreshock and the Mw 7.1 mainshock in the Ridgecrest earthquake sequence) and 7/12/2019 (after the Mw 7.1) to quantify displacements on fault strands activated in both events near the intersection zone of the NW- and NE-striking surface ruptures. Several UAV surveys were collected on portions of the surface rupture in the days after the earthquake before high quality satellite images were available, demonstrating the value of this platform in obtaining short temporal baselines. Using National Agriculture Imagery Program (NAIP) and WorldView imagery as pre-event imagery and UAV as post-event imagery resulted in correlation maps and slip distributions that are compatible with those derived from high-quality satellite images. Our work demonstrates the utility of analyzing imagery datasets collected across different platforms using optical image correlation in event response and further motivates the rapid collection and processing of low-altitude aerial photographs.

Key Words
Ridgecrest, optical image correlation, fault displacement

Citation
Morelan, A. E., Hudnut, K. W., & Donnellan, A. (2020, 08). Quantifying fault displacement via optical image correlation using structure-from-motion orthomosaics in the 2019 Ridgecrest earthquake sequence. Poster Presentation at 2020 SCEC Annual Meeting.

Related Projects & Working Groups
Earthquake Geology