Study On Statistical Data Spatialization In Support To Disaster Prevention And Mitigation

As the central area of population, economic activity and social development, China‘s coastal zone has a dense population and wealth. Meanwhile, due to the climate change and disturbance of human behavior, the ecological environment of coastal zone is rather vulnerable, and natural disasters such as typhoon, rainstorm and storm surge often take place there. It will enhance human‘s abilities of resistance, fitness and resilience to disasters, if the spatial distribution of social and economic resources, like population and property in coastal zone, could be provided. Based on the summary of research status at home and abroad, this paper created a statistical data spatialization model using the ?dynamic regionalization‘ method with GIS technique, and obtained the population and GDP spatialization data in 1 km2 of China‘s coastal zone in 2000, 2005 and 2010, which can strongly support the decision work of disaster prevention and mitigation there. The main works and conclusions of this research are as follows:(1) Building the population spatialization model based on the nighttime light data, because it could reflect lots of information, such as settlements, transportation networks and economic activities. Besides, analyzing the most relevant factors of each industry and it came out that: the first industry correlates closely with farmland, forest, grassland, water and mariculture, the second industry has high correlations with city, rural settlement, industrial-transporting site and salt field, and the tertiary industry nighttime light.(2) Using the land use data to extract the city and rural settlement as the populated area, and extract the city, rural settlement and industrial-transporting site as the distribution plots of the tertiary industry. This process can effectively alleviate the disturbance coming from the ?overglow‘ effect of nighttime light data.(3) Due to the prominent spatial heterogeneity of population and GDP distribution in China‘s coastal area, unitary spatialization model for the research area exhibits much poor precision. This research adopted the ?dynamic regionalization‘ method, which means setting a precision threshold of the regression model to dynamically divide the whole study area into several sub-regions, and building spatialization models for each sub-regions. The essence of this method is to divide the entire area, which has much spatially heterogeneous of population and GDP distribution, into several sub-areas, which have the similar feature of population and GDP spatial distribution internally. It can solve the problem of spatial heterogeneity simply, as well as assuage the adverseness of ?pixel saturation‘ in nighttime light data.(4) In this study, the population and GDP spatialization data were adapted into analyzing disaster-affected population and economical loss. Results indicated that these spatialization data could reflect the details of population distribution and economic activities in a fine scale, and they would be better than statistical data in the decision-making process of disaster prevention and mitigation.