一种改进人口数据空间化的方法:农村居住地重分类  

An Enhanced Method for Spatial Distributing Census Data: Re-classifying of Rural Residential

作　　者：杨小唤[1] 刘业森[1] 江东[1] 罗春[1] 黄耀欢[1] 

机构地区：[1]中国科学院地理科学与资源研究所资源环境科学数据中心,北京100101

出　　处：《地理科学进展》2006年第3期62-69,139,共9页Progress in Geography

基　　金：国家自然科学基金项目(40471112)

摘　　要：人口(统计)数据空间化是解决统计数据与自然要素数据融合分析的有效途径。本文在论述已有人口空间化方法的基础上,认为遥感影像得到的居民地数据是表达人口分布的最好指标。为了使居民地数据更好地应用于人口空间化的研究,论文在分析各种与人口居住密度相关指标的基础上,确定了用农村居民地面积所占百分比对农村居民地进行重新分级,然后应用于人口空间化的计算。结果检验表明,人口空间分布数据的误差从分级前的17.4%降到分级后的12%,尤其是误差高于30%的乡镇个数从8个减少到1个,该方法有效地提高了人口空间数据的精度。Spatial distributing of census data is an effective way to integrate statistical data and natural factors. Land-cover and land-use change （LUCC） is the effect of human activities, and spatial distribution of population has close relationship with LUCC pattern both at regional and global scales. Population can be redistributed onto geo-referenced square grids according to this relation. Since there exist efficient approaches for monitoring LUCC with remote sensing and GIS, geo-referenced population data can also be updated conveniently. According to existing methods, it is found that the population density is directly related to land use types and the residential area is the best index for population distribution. Residential areas could be reclassified into three sub classes： urban residential, rural residential, and commercial-industrial. The paper presented an enhanced method for spatial distributing census data： re-classification of rural residential areas. On the basis of the relationship of various kinds of indexes and inhabitation density, several indexes were selected for re-classifying rural residential areas. Using these re-clas-sified rural residential data, the precision of census redistributing pattern was improved obviously. Methods and main algorithms used in these studies were presented in the paper. Characters and prospect of this study were also discussed.

关 键 词：人口 空间化 居民地 

分 类 号：C912.82[经济管理]