Investigating the regional scale rock strength distribution and its self-similar characteristics from a drone-based high-resolution digital elevation model in Mecca Hills, CA.

Nairong Du, Hiroki Sone, Ken Ferrier, & Randolph Williams

Published September 8, 2024, SCEC Contribution #13948, 2024 SCEC Annual Meeting Poster #147 (PDF)

Rock properties can be measured by lab experiments at specimen scales and inferred from surveys at geophysical scales. However, there exists a gap in observations between these scales which hinders us from populating physical properties in our models with realistic heterogeneities. We aim to fill this gap and characterize the heterogeneity of rock properties at a mesoscopic scale by inferring the distribution of rock strength parameters from a high-resolution digital elevation model based on limit-equilibrium slope stability.

We characterized the distribution of rock strength (i.e. apparent cohesion) along a ~12-km long and ~2-km wide segment of the San Andreas Fault Zone in Mecca Hills, California. A DEM with ~6 cm/pixel resolution was generated from over 7000 photographs recovered from an unmanned aerial vehicle (UAV) processed using photogrammetry software. Apparent cohesion values were inferred for each 10 m x 10 m cell based on the Culmann limit-equilibrium slope stability analysis.

The distribution of apparent cohesion values correlates with the local geology, which includes several sedimentary units and strands of fault gouge exposures. The fault gouge shows significantly lower rock strength than the sedimentary units. Within the same sedimentary unit, rocks sandwiched by adjacent strands of the SAF appear weaker, whereas regions identified as structurally less deformed appear stronger. We however do not observe an obvious trend with fault normal distance.

To study the self-similar characteristics of the distribution of apparent cohesion values, we took profiles of apparent cohesion along and across the SAF and calculated the power spectral density (PSD) of the strength profiles. The PSDs appear linear in the log-log space suggesting a fractal nature of the distribution of rock strengths along these profiles. Preliminary results show that the slope of the best-fit lines through the PSDs in the fault parallel direction ranges between -1 to -0.2, whereas in the fault normal direction, the slope ranges from -1.5 to 0.1

Citation
Du, N., Sone, H., Ferrier, K., & Williams, R. (2024, 09). Investigating the regional scale rock strength distribution and its self-similar characteristics from a drone-based high-resolution digital elevation model in Mecca Hills, CA.. Poster Presentation at 2024 SCEC Annual Meeting.

Related Projects & Working Groups
Fault and Rupture Mechanics (FARM)