Research Of Spatial Identification Of Poverty In China Based On Nighttime Light Data

Poverty has emerged as one of the chronic dilemmas facing development of human society during the 21 st century. It was a complex and comprehensive social phenomenon, especially in the developing countries. So, eliminating poverty and narrowing the rural-urban gap were one of the main objectives of achieving sustainable development in every country. The poverty reduction effect of China which was the largest developing country in the world could provide reference for the implementation of poverty reduction in other countries. By improving the accuracy of the poor regions and poor population identification, the government could eliminate the obstacles encountered in the implementation process of poverty alleviation and improve the relevant issues. Besides, the accuracy of the poor regions identification could improve the pro-poor efficiency. So, precision measure and identifying the spatial distribution of poor regions were very significant for the implementation of the government's policy on poverty alleviation.This study which brought new NPP-VIIRS referenced the application in the poverty of the internationally comparable vulnerability-Sustainable Livelihoods Approach(SLA) model. And it constructed the corresponding multidimensional poverty index system. Taking Chongqing municipality as a sample and estimating the multidimensional poverty index(MPI) by using the Averaged Nighttime Light Index(ANLI) worked well. The regression model which R2 was 0.6544 was constructed by using the correlation of ANLI and MPI in the typical study region. The model was test in Shaanxi province and the estimates which the average relative error was only 12.51% were better. The MPI's spatialization was realized in the whole country by using the tested statistical model.In the county scale, we divided the MPI index of 2852 counties in China into extreme poverty region, poverty and vulnerable region, general region, advantage region, rich region and extremely rich region. There were 848 multidimensional poverty counties in China by the way of identifying the counties within extreme poverty regions or poverty region as multidimensional poverty counties. Spatial clustering effect of multidimensional poverty was analyzed by the method of Explore Spatial Data Analysis(ESDA) and it was found that the value of a global Moran's I was 0.6209. The result showed multidimensional poverty had a high spatial concentration effect on regional scale. It was found that the 598 counties concentrated in the poor regions designated by the state overlapped with the multidimensional poverty counties identified in this research by comparing the distribution of the multidimensionalpoverty counties identified, poverty alleviation counties designated by the central government and 14 contiguous poor regions. Based on that, the paper put forward the national key poverty alleviation counties adjustment suggestions. The 848 multidimensional poverty counties were identified, according to the absolute poverty line of 3000 Yuan, the relative poverty line of 4000 Yuan and non-income poverty line of 4853 Yuan standards. There were 254 absolute poverty counties, 543 relative poverty counties in China's multidimensional poverty counties; other 195 counties were multidimensional poverty county caused by non-income poverty.848 multidimensional poverty counties were divided into three zones in eastern, central and western through the study of spatial pattern and distribution of the multidimensional poverty counties. The first level poverty type region was also identified. The 17 of China's second level poverty type region was determined based on the similar nature, economic combination and regional leading influence factors such as poor division principle.