Estimated casualty risk for disaster preparation in five scenario great earthquakes, Sichuan-Yunnan region, China

For earthquake disaster preparation and risk reduction, quantitative estimates of casualties caused by potential future ruptures is crucial. We estimate the numbers of casualties and the population strongly affected for five “worst-case scenarios” of devastating earthquakes in the Sichuan-Yunnan region, China. These earthquake scenarios are assumed to rupture the entire segments of the Xianshuihe, Anninghe, Zemuhe, Daliangshan and Xiaojiang faults. The source parameters of the scenario earthquakes are derived from fault models. Before calculating scenario losses however, we validate the algorithms and data sets of Quake Loss Assessment for Response and Mitigation (QLARM) for the 13 earthquakes of the mainland of China since 2003, which have caused fatalities, by matching observed intensities and casualty counts with calculated values. These near real-time fatality estimates are reviewed and found to have been correct and, for people injured, six are 100% correct and three have been within a factor of 2.5. We use an average attenuation and assume that earthquakes occur at 03:00 local time, which expects assumes maximum occupancy of buildings. The greatest disaster is derived from the scenarios of the Xiaojiang fault rupture, where we calculate the possible numbers of fatalities on average as 286,000 and injured as 751,000. The scenario of the maximum rupture along the Daliangshan fault results in approximately twice the casualties reported in the 2008 M_W 7.9 Wenchuan earthquake. The three scenarios along the Xianshuihe, the Anninghe and the Zemuhe faults result in estimates of casualties similar to those in the Wenchuan case. These estimates are significant not only for disaster preparation and relief by the government, but also for risk prevention and mitigation by the public.