2007 GSA Denver Annual Meeting (28–31 October 2007)

Paper No. 5
Presentation Time: 9:20 AM

QUANTIFYING GLOBAL EARTHQUAKE RISK EXPOSURE AND ROLE OF GEOLOGY


ZOBACK, Mary Lou and SENEVIRATNA, Pasan, Risk Management Solutions, 7015 Gateway Blvd, Newark, CA 94560, marylou.zoback@rms.com

We assess relative global earthquake risk exposure by combining estimates of seismic hazard, population density, and structural vulnerability. Two seismic disaster-related outcomes are considered: earthquake mortality risk (due to building collapse) and the magnitude of potential building damage losses. We utilized the Global Seismic Hazard Project (GSHAP) peak ground acceleration (pga) values for the standard 10% chance of exceedance in 50 years, corresponding to a return period of 475 years (Giardini and others, http://www.seismo.ethz.ch/GSHAP/global/; data are available online in a 10 by 10 km grid). Gridded population values were obtained from the LandScan global population database (http://www.ornl.gov/sci/landscan/; available in a 30” by 30” grid, aggregated to match the the seismic hazard grid). We combined the hazard and population values through seismic-intensity based damage functions. To bracket a wide variability in construction quality and style, two seismic vulnerability end members were considered: India-type construction and Taiwan-type construction. Building damage was computed cell by cell from seismic intensity-based mean damage ratios, with population density as a proxy for building value. In a similar manner, earthquake mortality risk due to building collapse was computed as the product of population and the fraction of damage exceeding the collapse threshold in each cell. Both the building damage and mortality risk cell by cell values were summed country-wide and normalized by total population. The countries of highest risk exposure, both in terms of potential building damage and mortality risk are in Soviet Central Asia and South and Central America. Because the GSHAP hazard values assumed rock conditions everywhere except the US and Canada (where rock/firm solid was the reference), shaking amplification expected in large cities located on basins increases the risk exposure.