以黄淮海地区为例,评价分析了该区2000-2008年乡村发展水平及其时空格局,并利用自组织特征映射(SOFM)人工神经网络聚类算法划分了区域乡村发展类型。结果表明:(1)研究时段内全区乡村发展水平明显提升,乡村综合发展指数年均增长7.71%;(2)县域乡村发展格局无明显变化,但乡村综合发展水平的区域差异仍在增大,乡村综合发展指数的变异系数由0.346增加到0.350;(3)县域人均粮食和主要农产品产量与乡村发展水平呈日益强化的显著负相关关系;(4)基于SOFM聚类算法,将黄淮海地区乡村发展类型划分为8大类。本文认为:在城乡转型发展新时期,黄淮海地区乡村发展应立足地域类型特征,突出核心优势、明确限制因子,因地制宜、分区推进;县域人均粮食和主要农产品产量与乡村发展水平关系“倒挂”的现象应引起足够重视,为保障国家粮食安全、促进农区乡村发展,亟需推进宏观层面的制度和政策创新。
China's agricultural and rural development has come into a new period of transformation since 2004, and this may provide new chances for rural development in less developed traditional agricultural areas. As such, geographical comprehensive studies on rural development in these areas are urgently needed. The Huang-Huai-Hai region, a represent- ative of China's traditional agricultural areas, is an important hinterland of Bohai Economic Rim and a major base of China's grain and cotton production. The middle- and low-yield farmland improvement and agricultural comprehensive development initiated in the early 1970s had significantly accelerated the agricultural development in this region. Agricultural production function of this region has been further strengthened. However, rural devel- opment was still at a low level. This paper established an indicator system for assessing in- tegrated level and spatial pattern of rural development in this region in 2000 and 2008 at county level. Furthermore, rural development types at county level were classified based on five indices including integrated rural development index, per capita output of major ag- ricultural product, the proportion of agricultural labor in total rural labor, scale industrial output value per capita and the proportion of tertiary industry in GDP using self-organizing feature maps (SOFM) network modeling. The results showed: (1) rural development of the Huang-Huai-Hai region in 2008 has been significantly improved compared with 2000, as evidenced by the integrated rural development index showing an annual increase of 7.71~~0; (2) however, regional differences of integrated rural development index is still large and its spatial pattern showed no significant change during the study period; (3) per capita grain output and per capita output of major agricultural product have significant negative correlation with integrated rural development index and other selected rural devel- opment indictors; (4) according to cluster analysi