HayWired Scenario Mainshock Ground Motions

Abstract

Cover. Oblique aerial image showing how the main active traces of the Hayward Fault (red lines) cut through the urban landscape of the east bay part of the San Francisco Bay area (north is to the upper right; fault is oriented northwest-southeast). The image is overlain on a "hillshade" (artificially illuminated from the northwest) to show fault-zone ground-surface features, such as a right-laterally offset stream channel in the foreground. The fault trace roughly bisects California Memorial Stadium at the University of California, Berkeley (O-shaped feature in the middle of the image). The stadium was renovated and seismically strengthened at an expense of $321 million (see http://news. berkeley.edu/2011/09/01/memorial-stadium-renovated-with-help-of-berkeleys-own/) as an engineered countermeasure against fault-rupture hazard from a future earthquake such as that modeled in the HayWired scenario. Similarly, beneath the viewpoint of this image, the East Bay Municipal Utility District (EBMUD) mitigated the Hayward Fault crossing of the Claremont water-transmission tunnel by constructing a bypass tunnel at an expense of $38 million (see http://www.geoprac.net/geonews-mainmenu-63/64-project-related/333-the-claremont-tunneldesigned-to-survive-fault-rupture-on-the-hayward-fault/). Although notable risk-reduction measures such as these have already been taken along the Hayward Fault, issues with more than 300 buildings (mostly homes) that are situated directly on the fault, and dozens of major lifelines that cross it, have not yet been addressed.

A Simplified Method for Rapid Estimation of Emergency Water Supply Needs after Earthquakes

“…Earthquake-specific hazard intensity data are collected. In the case study, ground motion (e.g., PGV) and ground failure data (e.g., landslide, liquefaction and fault rupture) are provided from the HayWired Scenario [19].…”

“…Hazard intensity is based on the simulated ground-motion PGV data from the Hay-Wired scenario [19]. Porter [1] applies the break/leak ratio, identified by Ballantyne et al [25], which found that pipeline damage from earthquake-induced ground-failure in several earthquakes resulted in a 50/50 percent break/leak ratio.…”

“…Hazus loss estimation was used with liquefaction susceptibility maps [30] to estimate liquefaction probability in the rest of the study region [31]. Landslide probability [32] and fault rupture estimates [19] are from the HayWired scenario.…”

“…These service areas cover approximately 1.65 million people (615,000 service connections or households) or about 22% of the population in the nine counties adjacent to the San Francisco Bay (Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, Solano, and Sonoma Counties, in Figure 1-hereafter, the nine-county HayWired scenario study region, though this may include up to sixteen counties in other studies), which represents approximately 7.6 million people (2017 population estimate). Hazus (version 2.1) water system performance analysis was conducted by FEMA (Doug Bausch, formerly of FEMA Region VIII) for the rest of the HayWired scenario study region, using custom ground-motion data for the HayWired scenario mainshock [19]. In the case study, we extend applications of the CUWNet model to the nine-county HayWired scenario study region, using the simplified loss estimation approach to estimate water supply needs-without proprietary water pipeline data, and with increased precision.…”

“…In the simplified analytical method, Hazus water system performance analysis is adapted to use the new CUWNet water pipeline vulnerability function in Porter [1] to estimate average water pipeline break rates (per 1 km of pipe) by the ~90 m 2 (threearcsecond) raster grid, in conjunction with water service outage calculated from the default Hazus serviceability index by the ~250 m 2 (nine-arcsecond) raster grid. The water pipeline vulnerability function is conditioned on hazard intensity data using peak ground velocities (PGV) and ground failure probabilities (e.g., liquefaction, landslides, and fault offset) from the HayWired scenario [19]. The LandScan USA 2017 "conus_night" (~90 m 2 at the equator, three-arcsecond) nighttime population database (hereafter referred to as the LandScan population database) [20] provides ambient population estimates in the model, for targeted emergency water resource allocation.…”