基于GIS和人工神经网络的区域贫困化空间模拟分析——以贵州省猫跳河流域为例

发布时间：2018-08-13 13:49

【摘要】：我国西南喀斯特地区山高坡陡,地形破碎,生境脆弱,水土流失严重,是我国典型的极贫困代表区域之一。本文选择贵州省猫跳河流域作为研究区,以乡镇为基本单元,应用GIS和ANN技术,模拟区域自然致贫因子和消贫因子的空间分布,计算各乡镇的贫困度,揭示区域贫困的空间分布格局,以期为指导研究区早日脱贫及生态重建提供科学依据。结果表明,地形、土壤侵蚀等自然要素是主要的致贫因子,而社会经济要素是缓解贫困的因子。贫困度较小的乡镇主要分布在研究区的中部和东部,贫困度较大的乡镇主要分布在研究区的南部和北部边缘。可见,应用人工神经网络模拟区域贫困化简便、实用,避免了传统的单纯依靠统计数据进行贫困化研究的做法,是一种可行的方法与技术途径。
[Abstract]:Karst region in southwest China is one of the typical poor regions in China because of its steep mountain slope, broken terrain, fragile habitat and serious soil erosion. In this paper, the Maodiao River Basin of Guizhou Province is selected as the research area, and the township is taken as the basic unit. The spatial distribution of natural and anti-poverty factors is simulated by using GIS and ANN techniques, and the poverty degree of each township is calculated. The spatial distribution pattern of regional poverty is revealed in order to provide scientific basis for guiding the research area to get rid of poverty and rebuild the ecology at an early date. The results showed that natural factors such as topography and soil erosion were the main factors of poverty, while social and economic factors were the factors to alleviate poverty. The towns with low poverty degree were mainly distributed in the middle and east of the study area, and the towns with higher poverty degree were mainly located in the southern and northern edges of the study area. It can be seen that it is simple and practical to use artificial neural network to simulate regional poverty, which avoids the traditional method of simply relying on statistical data to study poverty. It is a feasible method and technical approach.
【作者单位】：中国农业大学资源与环境学院北京大学环境学院北京大学环境学院
【基金】：国家自然科学基金资助重点项目(40335046) 高等学校博士学科点专项科研基金资助课题(20040001038)资助
【分类号】：K901.8

【参考文献】