Research On Empirical Probability Model Of Earthquake Casualties Assessment In China

China is one of the countries with the highest risk of earthquake disasters,and the huge population base and the weak seismic resistance of engineering systems in some relatively backward areas have caused China to suffer huge losses of life and property in previous earthquakes.As it is often difficult to understand the situation in the disaster area as soon as possible after the earthquake,the government needs to determine the response level based on the preliminary assessment of the earthquake disaster situation,activate the emergency plan,deploy and arrange rescue forces,and carry out rescue work reasonably and efficiently.Among them,the earthquake casualties assessment is an important part of the earthquake disaster assessment,and the casualties situation is also the key basis for determining the level of earthquake emergency response.In order to establish an earthquake casualties estimating model suitable for China,this paper carried out the following work:1.115 seismic events with intensity VI or higher in the extreme seismic zone in China between 1990 and 2012 were selected,and their fatality data were counted and the mortality rate of each event was calculated.The fatality data of different intensity zones in 25 counties and cities in the 2008 Wenchuan earthquake in Sichuan were counted,and the number of fatalities in each intensity zone was determined according to the normalization weight,and then the mortality rate of each intensity zone of 25 cities and counties was calculated,filling the gap in the mortality data of different intensity zones in China.2.The two-parameter logarithmic cumulative normal distribution model in the PAGER system was selected,and the data were fitted by three methods respectively,and their prediction effects were verified based on actual earthquake cases in eastern and western China.The first is to establish the prediction model directly using the parameters for China in the PAGER,and the prediction results had large errors and order of magnitude differences from the true values;the second is a prediction model based on the maximum intensity in an earthquake,and it was verified that the prediction results still have order of magnitude differences from the true values;the third is a prediction model based on each intensity level,and it was verified that its prediction accuracy is significantly higher than the first two methods.The prediction results were in better agreement with the true values of the three western cases,but the number of fatalities in the eastern cases was overestimated to a certain extent.3.In order to extend the applicability of the prediction model based on each intensity level,the fatalities in 25 cities and counties in the 2008 Wenchuan earthquake were grouped according to two parameters,namely GDP per capita and population density,and the prediction model parameters were fitted separately to obtain the prediction model of earthquake fatalities in different per capita GDP and population density zones.The GDP per capita was divided into three intervals: above 15,000 yuan,between 9,000 yuan and 15,000 yuan,and below 9,000 yuan.The prediction accuracy for the eastern case was improved by selecting the corresponding prediction formulae based on the GDP per capita and population density.