Cordilleran Section - 121st Annual Meeting - 2025

Paper No. 30-3
Presentation Time: 8:50 AM

PROBABILISTIC FAULT DISPLACEMENT HAZARD ANALYSES FOR CALIFORNIA DWR DAMS INFORM SEMI-QUANTITATIVE RISK ASSESSMENTS


THOMPSON, Stephen1, LEWANDOWSKI, Nora1, MCGREGOR, Ian1, HOIRUP, Don2 and ILANKATHARAN, Lanka2, (1)Lettis Consultants International, Inc., 1000 Burnett Ave., Suite 350, Concord, CA 94520, (2)California Department of Water Resources, Division of Engineering, 3500 Industrial Blvd., West Sacramento, CA 95691

The California Department of Water Resources (DWR) has been performing semi-quantitative risk assessments (SQRAs) for its major dams and hydroelectric facilities to better understand the contributions to risk posed by various natural hazards and potential failure modes. In support of SQRA workshops for CDWR’s facilities along the East Branch of the California Aqueduct in San Bernardino County, we performed probabilistic fault displacement hazard analysis (PFDHA) at three locations where facilities were constructed across active or conditionally active faults. At Cedar Springs Dam, located east of Cajon Pass, strands of the West Silverwood Lake fault and/or Grass Valley fault zone were identified at the damsite during construction in the late 1960s and early ‘70s, and the potential for fault displacement was incorporated into the design of the embankment. The PFDHA explored alternative rupture mechanisms that could result in fault displacement beneath the embankment, outlet tunnel, and spillway, including principal fault displacement from earthquakes on the Grass Valley fault zone or West Silverwood Lake fault and distributed fault displacement from earthquakes on the nearby Cleghorn fault zone. To the south, along the southern front of the San Bernardino Mountains, the Waterman Canyon fault zone crosses the Devil Canyon Penstocks, and the San Andreas fault underlies a channel between the first and second Devil Canyon Afterbays. The PFDHAs for these facilities explored the hazard from principal and distributed surface-fault rupture from San Andreas and Waterman Canyon fault earthquakes. The PFDHA results for all three facilities consisted of fault displacement hazard curves (mean and fractiles) that provide the relationship between fault displacement amount and annual frequency of exceedance. With this information, participants in the SQRA workshops could estimate the risks of various failure modes related to fault displacement by estimating the likelihood of failure given various displacements along the hazard curve. The SQRA methodology is an ideal application of PFDHA, as it utilizes multiple points on the hazard curve and can readily consider large uncertainties in the hazard analysis.