Rock Traits from Machine Learning: applications to precariously balanced rocks and fault scarps in Southern California

Tyler R. Scott, Zhiang Chen, Chelsea P. Scott, Jnaneshwar Das, & Ramon Arrowsmith

Published August 15, 2019, SCEC Contribution #9712, 2019 SCEC Annual Meeting Poster #120

Rock traits (size distribution, shapes, orientation, composition of pebbles, cobbles, clasts) distinguish many geologic features important in earthquake geology research. These include alluvial fans (fault activity, geomorphologic intensity), rock damage (fault zone rupture processes), precarious rocks (ground motion indicators over kyr-time scales), and fault scarps (fault zone evolution, fault slip rates, earthquake timing, fault segment linkage). Machine learning (ML) has revolutionized data intensive computing problems for many scientists. We apply ML to isolate desired rock traits within active fault zones near Bishop, CA, and will use this framework to develop a workflow for application to precariously balanced rocks (PBRs) and other fragile geologic features in Granite Dells, AZ and other locations. Our preliminary work on fault scarps formed in the Bishop tuff on the Volcanic Tablelands, California shows that deep neural networks trained on expert annotation of UAS-acquired rock imagery from a geological site leads to accurate and fast segmentation of individual rock characteristics. This capability of segmenting and classifying aerial imagery, primarily in 2D facilitates the estimation of distributions of rock traits such as angularity, size, and orientation. Rock size distributions reflect both the initial cooling joint fracture geometry, as well as faulting-induced fracturing and will vary with position as a function of strain magnitude and linkage characteristics. In addition, rock orientations indicate the degree of downslope transport along the fault scarps, enhancing our understanding of the fault scarp erosional processes and providing future insight on PBR toppling reconstruction. The ML network, once trained, significantly reduces manual efforts that typically coincide with big data. Large datasets of spatially explicit PBR fragility with detailed geomorphic and geologic context provide valuable assessment of sensitivity to ground motions. Instead of manually collecting a few fragilities, it should be possible to remotely capture a fragility spectrum for the fragile geologic features in 3D at a site. We are investigating safe and efficient imaging with UAS for mapping PBRs with emphasis on intelligent observation of basal (pedestal) contact relationships.

Key Words
Bishop, fault, scarp, machine learning, 3D, Structure from Motion, Drone, Rock trait, precariously balanced rock, PBR

Citation
Scott, T. R., Chen, Z., Scott, C. P., Das, J., & Arrowsmith, R. (2019, 08). Rock Traits from Machine Learning: applications to precariously balanced rocks and fault scarps in Southern California. Poster Presentation at 2019 SCEC Annual Meeting.

Related Projects & Working Groups
Earthquake Geology