基于GIS和空间自相关模型的耕地空间分布 格局及变化特征分析——以成都市龙泉驿区为例

耕地资源是人类生存和发展的基础性资源,科学判定耕地空间分布格局及其变化特征,对促进耕地保护及空间优化布局等方面具有重要意义。本文利用四川省成都市龙泉驿区2005年、2009年及2013年土地利用数据提取耕地数据,运用核密度计算、耕地集中度指数、空间自相关等研究方法开展耕地空间分布格局及其变化特征研究。研究结果表明:1)2005—2013年龙泉驿区耕地空间分布密度呈现出集聚态势,耕地整体呈现出西北及北部密集、中部及南部稀疏的空间分布特征,其中耕地分布密度高值区域呈现出由西北向西南扩张的趋势,低值区域则以城镇为中心有零星扩散的情况。2)从耕地面积分布情况来看,整体呈现出北高南低的空间分布特征,耕地分布相对集中,而且在时间序列上有进一步增强的趋势,但耕地空间变化区域差异较为显著。3)耕地空间分布存在显著的全局空间正相关性,但局部空间异质性有所增强,耕地面积比例较高的空间单元集中分布在北部、西北部一带,并呈现出向西缩减的趋势,耕地面积比例较低的空间单元分布在以城镇为中心的建成区及边缘地带,并经历扩张与收缩两个阶段;2005—2013年因城镇建设占用耕地和农村土地整治影响而出现局部异质性"热点"与"冷点"区域。该研究结果在一定程度上为掌握区域耕地空间变化趋势、制定耕地保护与优化布局政策提供理论方法和参考借鉴。

Analysis of spatial distribution pattern and evolutionary characteristics of cultivated lands based on spatial autocorrelation model and GIS platform*- A case study of Longquanyi District,Chengdu, China

Cultivated lands are among the most fundamental resources for the existence and development of mankind. Scientific analysis of the spatial distribution patterns and evolutionary characteristics of cultivated lands is highly useful in the protection and resource optimization of cultivated lands. Working on such research project, this paper extracted cultivated land data from the 2005, 2009 and 2013 land use data of the Longquanyi District of Chengdu, China, and calculated the kernel density, concentration index of cultivated land and conducted spatial autocorrelation analysis by using GIS platform. The results suggested that: 1) the evolution of cultivated lands in 2005-2013 in the study area was driven by aggregation. Spatially, cultivated land densities in the northwestern and northern regions were higher than those in the southern and central regions. Regions with higher cultivated land density apparently expanded from northwest to southwest, while regions with lower cultivated land density sporadically expanded from urban to rural area. 2) For distribution of cultivated lands, feature land density in the north was generally higher than that in the south. The distribution of cultivated lands was relatively concentrated, with more concentrated years in terms of the time series, but with more significant variation in the spatial differentiations among the regions. 3) There was a significant spatial autocorrelation for global distribution, with strengthening heterogeneity for local distribution of cultivated lands in the study area. Spatial units with higher proportion of cultivated lands was concentrated in the northern and northwestern regions and reduced westwards from year to year. The units with lower proportions of cultivated lands were more distributed at built-up areas and suburbs surrounding the urban centers, with two stages of expansion and contraction. Because of urban expansion and rural land consolidation in 2005-2013, there were hot spots and cold spots in different localities of the distributed cultivated lands. This study provided the theoretical methods and references for determining the dynamic evolution of regional cultivated lands and developing relevant policies for protection and resource optimization of cultivated lands.