基于遥感和GIS的流域社会经济数据空间化方法研究

摘要

The importance of population and socioeconomic data and their spatial distributions to researches of natu-ral resources,environment and natural disaster assessment has been widely recognized.It’s a worthy exploration to solve the problem by combining remote sensing and GIS technology.According to the statistical population,GDP and land use data in Dongjiang watershed,we built multiple linear regression models by taking the statistical popu-lation or GDP as the dependent variable and per unit of statistical data carrying capacity as independent variables. Meanwhile,based on the raster land use data and 100m ×100m grid data,the constraint index layers of population and GDP’s spatial distribution were obtained.Furthermore,through the regression model and GIS interpolation, the spatialisation of the statistical population and GDP data of Dongjiang watershed in 2009 was completed.The simulated result was verified to be equal or superior in county-scale by comparing with the actual statistical data in the same scale.The result suggests that to combine remote sensing and GIS is an effective way to spatialising socio-economic data.%人口和社会经济数据及其空间分布，在资源、环境及自然灾害评估中的重要性已被广泛认知。把遥感和GIS技术相结合，是探讨解决社会经济统计数据空间化的重要思路。以东江流域为例，以多期人口、GDP、土地利用数据为基础，建立东江流域人口与土地利用、GDP和土地利用的多元线性回归模型；以土地利用数据和100 m × 100 m网格数据为基础，构建东江流域人口和GDP空间分布约束力指标图层；然后结合统计模型和面积内插，实现了东江流域2009年人口、GDP统计数据的空间化。在县域空间尺度上对模拟结果进行了验证，与同尺度研究工作进行了对比，结果显示模拟得到的人口和GDP空间分布数据，与同尺度的研究工作处于同一精度或者略高的精度水平，表明该方法是一种进行流域社会经济数据空间化处理的有效方法。