Impacts on Catastrophe Loss Modeling from Multi-Segment and Multiple Fault Ruptures in UCERF3

Date: 4/24/2019

Time: 09:30 AM

Room: Pike

Version 3 of the Uniform California Earthquake Rupture Forecast (UCERF3) is part of the current National Seismic Hazard Mapping Project model (Petersen, et al., 2014) and has been adapted for modeling financial losses from earthquakes. UCERF3 relaxes fault segmentation, allowing multi-segment and multi-fault ruptures, with lengths up to 1200 km on the San Andreas Fault. In catastrophe loss modeling, fault rupture length determines the area affected by shaking and drives the correlation of losses for geographically-distributed sets of properties. Longer ruptures produce more severe consequences. Recently, Schwartz (2018) examined surface ruptures observed worldwide from shallow crustal earthquakes since the mid-1800s and found that only a few ruptures have been longer than 300 km and none has exceeded 500 km. This suggests it may be more realistic to constrain rupture lengths for future earthquakes. In this study, we examine the impact of the UCERF3 ruptures with long rupture lengths on probabilistic losses, using a published residential inventory (Reinsurance Association of America, 2018) with more than 180,000 locations throughout California. We examine impacts from long UCERF3 ruptures on Average Annual Loss (AAL) and other loss parameters such as the Return Period Loss (RPL) as well as the Tail Conditional Expectation (TCE) at return periods of 100-, 250-, 500- and 1,000-years. Our results show that these long ruptures, even though rare, have significant impacts on probabilistic loss estimates. The impact increases as the loss return period of interest becomes longer. At the 250-year return period, about 30% of the expected loss contribution comes from ruptures with lengths greater than 500 km; at 1,000-year return period, the expected loss contribution reaches nearly 50%, even though such ruptures have not been observed from past shallow crustal earthquakes. Due to the impact the UCERF3 model on public policy and the financial industry, critical examination of the limits of potential rupture lengths is warranted in the future model update.

Presenting Author: Yajie Lee

Authors