Natural Geographical Environment Impacts On Spatial Differentiation Of Poverty At County Scale

Eliminating poverty is an important task for China to build a well·off society in an all-round way.The Outline for Development-oriented Poverty Reduction for China's Rural Areas(2011-2020)has divided the country into 14 contiguous poverty-stricken.These areas in China mainly distributed in mountainous and hilly areas and some regions even have large area of desert and Gobi,which are greatly affected by the geographical environment There are 65 poverty counties in Guizhou Province,belonging to Rocky Desertification of three provinces,Wuling Mountain and Wumeng Mountainou.Most of the poverty-stricken counties are belong to plateau mountainous tectonic topography,as we all as,these areas have large rocky desertification.Poverty caused by the restrictive role of the natural geographical environment is an important cause of poverty-stricken counties in Guizhou Province.Therefore,it is of great significance to reveal the impacts of geographical environment on the spatial differentiation of poverty,providing data support for effectively implement the targeted poverty alleviation.Based on the existing theoretical research on the impact of natural geographical environment on poverty,this study combines the poverty situation in Guizhou Province with the natural environment background,and takes the poverty incidence as the dependent variable,and selects 19 indicators as independent variable that reflect the topography,climate,water resources and road conditions,land conditions.Firstly,the spatial pattern of poverty incidence in poverty-stricken counties in Guizhou Province from 2011 to 2015 was explored by spatial autocorrelation analysis.Secondly,the impact factors of natural geographical environment on the spatial differentiation of poverty-stricken counties were then quantitatively measured by the Geo-detector model,which help to diagnose the dominant factors and explore the interactions among factors.Then based on the results of Geo-detector,the geographically weighted regression model was employed to analyze the spatial heterogeneity of dominant factors in their effect levels.Finally,according to the administrative division level,the multi-level detection was carried out by using the geo-detector model,according to the results,the degraded regional types of the study area were divided,and combined with the progress of poverty alleviation in poverty-stricken counties in Guizhou Province,the poverty alleviation suggestions are put forward in a targeted manner.Through analysis and research,the following conclusions were drawn:(1)The spatial distribution characteristics of the poverty incidence in the counties of Guizhou Province from 2011 to 2015 were consistent with the spatial distribution characteristics of the average poverty incidence.In the study period,the poverty incidence showed a relatively stable downward trend.Overall,the poverty incidence in the study area have a spatial heterogeneity characterized by decreasing from the south to the north and decreasing from the east to the west.(2)There was a strong spatial dependence on the poverty in the counties of Guizhou Province.During the study period,the high-high clustering areas mainly distribute in the southeast of Guizhou Province,and low-low clustering areas are distributed in the north and central area of Guizhou Province.Meanwhile,significant areas gradually increase during the study period,indicating the local dependence enhancement of each district;(3)The dominant factors of spatial differentiation are the percent of effective irrigated on cultivated land,per capita forest land area,road network density and slope.The interaction among dominant factors showed two types of nonlinear enhancement and bivariate enhancement,and the influence of each factor interaction on the poverty incidence was stronger than that of single factor;There was a significant difference of effect levels of each dominant factor on regional impoverishment,and the level of effect of different grades is distributed in the northwest-southeast zone;(4)According to the multi-level detection results of geodetectors,the study area can be divided into four types of impoverished area types:Water resource constraint type,land resource constraint type,traffic condition constraint type and topography condition constraint type,and according to the number of dominant factors,this area was further divided into single factor dominant area,two factor dominant area and multi-factor dominant area.Then accurate poverty alleviation classification guidance can be carried out for different types of areas.