Evaluating How Well Active Fault Mapping Predicts Earthquake Surface Rupture Locations

Chelsea P. Scott, Rachel N. Adam, Ramon Arrowsmith, Christopher M. Madugo, Brian Gray, Rich D. Koehler, Stephen Thompson, Alexandra Sarmiento, Timothy Dawson, Albert R. Kottke, Elaine K. Young, Ozgur Kozaci, Michael E. Oskin, Reed J. Burgette, Ashley R. Streig, Gordon G. Seitz, Curtis W. Baden, & Malinda G. Zuckerman

Submitted September 10, 2023, SCEC Contribution #13248, 2023 SCEC Annual Meeting Poster #072

In surface-rupturing earthquakes, fault displacement can damage infrastructure that crosses or overlies fault zones. The uncertainty in earthquake surface rupture location is key in fault displacement hazard analysis and informs hazard and risk mitigation strategies. Geologists often predict future rupture locations from faults mapped based on landforms interpreted from remote sensing datasets. However, there has been little investigation into how well faults mapped from such geomorphic criteria predict future earthquake rupture locations. We assessed how well geomorphology-based fault mapping predicts surface ruptures for seven earthquakes: 1983 M6.9 Borah Peak, 2004 M6.0 Parkfield, 2010 M7.2 El Mayor-Cucapah, 2011 M6.7 Fukushima-Hamadori, 2014 M6.0 South Napa, 2016 M7.8 Kaikoura, and 2016 M7 Kumamoto. We trained geoscience students to produce active fault maps using topography and imagery acquired before the earthquakes (“pre-rupture mapping”). An experienced geologic professional completed a “control” map. Mappers used a new “geomorphic indicator ranking” approach to systematically rank fault confidence based on geomorphologic landform indicators. We determined which ruptures were predicted by the mapped faults by comparing the fault maps to published rupture maps (which were not consulted by the mappers). We define predicted ruptures as ruptures near a fault (50-200 m, depending on the fault confidence) that interact with the landscape in a similar way to the fault. The mapped faults predicted between 12% to 68% of the principal rupture length for the studied earthquakes. The median separation distances between predicted ruptures and strong, distinct, or weak faults are 15-30 m. We demonstrate an application of our research to define fault uncertainty zones and discuss ways to improve fault mapping with implications for site evaluation, regulatory zoning, and fault trenching. Our work highlights that mapping future fault ruptures is an underappreciated challenge of fault displacement hazard analysis—even for experts—with implications for risk management, engineering site assessments, and fault exclusion zones.

Key Words
Fault mapping, Fault hazard, Remote sensing

Citation
Scott, C. P., Adam, R. N., Arrowsmith, R., Madugo, C. M., Gray, B., Koehler, R. D., Thompson, S., Sarmiento, A., Dawson, T., Kottke, A. R., Young, E. K., Kozaci, O., Oskin, M. E., Burgette, R. J., Streig, A. R., Seitz, G. G., Baden, C. W., & Zuckerman, M. G. (2023, 09). Evaluating How Well Active Fault Mapping Predicts Earthquake Surface Rupture Locations. Poster Presentation at 2023 SCEC Annual Meeting.

Related Projects & Working Groups
Earthquake Geology