Abstract:

Space identification in poverty-stricken areas is an important basis for the formulation of accurate poverty alleviation policy. The spatial evolution process and characteristics of rural poverty index in 86 counties (cities and districts) were studied by using the variation coefficient, the Tel index, the Markov chain and the exploratory spatial analysis method from 1984 to 2015. The result shows that the difference of rural poverty index in the county of Gansu province presents the stage characteristics of "narrowing - enlarging - narrowing" in the trend of overall decline, and there are obvious differences in its spatial distribution, and the process of evolution tends to be stable. And the rural poverty in Gansu province has a spatial clustering image, showing a special distribution and stability. The poor degree of rural poverty in the Hexi Corridor and some counties in Lanzhou area is low, and there is a stable spatial agglomeration phenomenon, and the agglomeration range is extended continuously. The degree of poverty in rural areas of Dingxi, Longnan, Gannan and Linxia is relatively deep, and it also shows a strong spatial agglomeration phenomenon, however, the agglomeration range is very strong but shrinking and it is showing a fragmented characteristics.

Key words: rural poverty, spatial evolution, Gansu Province, county

摘要：

贫困地区的空间识别是精准扶贫政策制定的重要依据。运用变异系数、泰尔指数、马尔科夫链和探索性空间分析方法对甘肃省1984~2015年86个县（市、区）域农村贫困指数差异的空间演化过程及其特征进行了实证分析。研究表明：甘肃省县域农村贫困指数差异总体下降，并表现出“缩小—扩大—缩小”的阶段性特征，演化过程趋于稳定；同时，甘肃省农村贫困存在空间集聚现象，呈现出集中分布和稳定性特征。其中，河西走廊地区和兰州地区的部分县域农村贫困程度较低，存在空间集聚现象，且集聚范围不断延伸；定西市、陇南市、甘南州、临夏州等地区县域农村贫困程度较深，也表现出较强的空间集聚现象，但集聚范围在不断缩小，且显现出碎片化特征。

关键词: 农村贫困, 空间演化, 甘肃省, 县域