Empirical seismic vulnerability curve for mortality: case study of China

Most seismic casualty estimations are based on the vulnerability of various categories of structure and facilities in the region concerned. These approaches require detailed inventory database of the structures and facilities in the region. In China, however, the data is not always available, especially for some underdeveloped regions. This motivates the development of empirical models that use historical casualty data for earthquakes and provide a country- or region-specific earthquake death rate as a function of ground shake. However, diminishing divergences in fitting method and obtaining higher accuracy are still a great challenge. Building on the findings of previous research, the present paper employs vulnerability curves to express the relation between seismic intensity and mortality. From data of the death rate by intensity, curves are established for the whole of China and western China. Some methodological improvements are also discussed. To validate the curves, data from four recent strong earthquakes occurred in western China are used, and the best curves for estimation are selected and compared with similar models. Results indicate that the established curves for western China give a better estimation for the four earthquakes and the logistic functions show higher accuracy. They can be used to clarify the magnitude of seismic deaths in western China when structure and facility data are unavailable. Uncertainties and the application of the models are also analyzed and discussed in the conclusion.