省域相对贫困村多尺度空间格局与分异机制

Multi-scales Spatial Patterns and Differentiation Mechanisms of Relative-poverty Villages in Hunan Province

文章基于多尺度、人—地系统的理论视角,以湖南省8 000个相对贫困村为研究对象,利用平均最邻近指数、空间基尼系数和空间自相关方法,从市级和县级两个尺度揭示了相对贫困村空间分布格局差异,运用空间计量模型识别出相对贫困村空间分异的影响因素,最后基于"四层一体"模型提炼出相对贫困村空间分异的机制。结果表明:(1)相对贫困村在市级尺度上均呈现集群分布模式,而在县级尺度上呈现出集群、离散和随机三种不同的模式。(2)市级和县级尺度上相对贫困村均呈现显著的全局空间自相关性,但在两个尺度上局部空间自相关性特征存在差异。(3)城镇化率、第一产业增加值占GDP比重、第二产业增加值占GDP比重、人均财政支出、文盲率、坡度、平均海拔是影响县级尺度相对贫困村空间差异的显著因素。(4)基于"四层一体"模型揭示出相对贫困村的空间分异机制。

It is of certain theoretical significance to reveal the multi-scales effects and differentiation mechanism of spatial difference of poverty. Based on the theoretical perspective of multi-scales and human-earth system,this paper takes8000 relative-poverty villages in Hunan Province as the study object, uses average nearest neighbor index and spatial autocorrelation to reveal the differences of spatial distribution patterns of relative-poverty villages at the city and the county scales, takes spatial econometric models to explore the influential factors,and finally summarizes the mechanism of spatial differentiation based on the ’four layers-one whole’ model. The results show that: (1) the relative-poverty villages show a cluster distribution at the city scale,but there are three types of distributions at the county scale, which are cluster pattern, discrete pattern and random pattern. (2) There are significantly global spatial autocorrelation at the city and the county scales, but local spatial autocorrelations at the two scales are different. (3) Urbanization rate,proportion of primary industry in GDP, proportion of secondary industry in GDP,per capita fiscal expenditure,illiteracy rate,slope and average altitude are significant factors affecting the spatial difference of relative-poverty villages among counties.(4) Based on the’four layer-one whole’ analytical model, this paper outlines the spatial differentiation mechanism of relative-poverty villages.