Project Abstract
|
The purpose of the current research effort is to evaluate the variation in annualized and probabilistic loss related to selection of ground motion attenuation relationship, and to time dependence. Using Hazus, annualized and return period loss estimates for all 58 counties within California have been generated from OpenSHA return period ground motion data for each of the four “Next Generation of Attenuation” (NGA) relationships. A total of 8 statewide loss analyses were conducted using Hazus-MH MR-4, evaluating each NGA relationship, for both time dependent and time independent hazard. These results may be compared to previous Hazus analyses executed using the weighted average of the 4 NGA relationships. The NGA relationship resulting in the smallest total statewide annualized building damage is the Campbell & Bozorgnia (C&B) relationship; losses were 15% less than those produced using the average of the 4 NGAs. The largest losses resulted when the Abrahamson & Silva (A&S) NGA relationship was used; annualized building damage for California was 11% larger than that produced by the 4 NGA average. Differences between time independent and time dependent losses are small. Statewide, time independent losses are slightly larger, with the difference ranging from 0.4% for A&S to 1.1% for C&B, and a net difference of 0.8% for the 4 NGA average. While statewide variation in loss between the different NGA relationships and their average may be modest, the geographic variation can be more significant, with differences as large as 59% for Amador County for A&S, and -25% for Sutter County for C&B, suggesting that the selection of the ground motion prediction equation can have a significant impact on loss estimates. |
