摘要
利用植被指数、NPP-VIIRS、土地利用、DEM等多源数据,基于GIS/RS技术分析其与人口的密切程度,在多源数据下采用多元回归结合个案加权和最小二乘法加权进行人口建模,对比筛选最优模型后进行零误差调整实现500 m×500 m尺度大小的泉州市人口网格化。从DEM数据进一步提取坡度、坡向、地形起伏度、地形粗糙指数等派生数据,通过定量与定性分析人口空间网格化后与地形之间的关系。实验结果表明:人口空间网格化后单元网格值介于之间0~5135人/500 m2,主要分布于泉州市东南沿海环泉州湾相关县市。区内DEM值范围介于0~1828 m之间,坡度值介于之间0°~71.04°之间,RDLS1和RDLS2两种地形起伏度分别介于0~505 m和0~2.340之间,均与人口空间分布呈负相关,除与坡向不相关,其地形与人口关系散点图均呈“L”状分布,88.1%人口分布于20200 m海拔高度,77.8%人口分布于地形地幅度小的地方。
This paper study based on GIS/RS technology,analyze the relationship of multi-source data such as vegetation index,NPP-VIIRS,land use,DEM and population,and multi-source data was modeled by multiple regression combined with population case weighting and least squares weighting,after comparing,analyzed all kinds of models to select the best models,at last zero-error adjustment is made to realize the gridding of Quanzhou city population at the scale of 500 m×500 m.The relationship between the gridding population space and topography is analyzed quantitatively and qualitatively by extracting the derived data of slope,aspect,relief and TRI from DEM data.The results show that the grid values are between 0 and 5135 people/500 m2,which are mainly distributed in the counties and cities around Quanzhou Bay on the southeast coast of Quanzhou city.The range of DEM value is between 0~1828 m,the range of slope value is between 0°~71.04°,and the relief of RDLS1 and RDLS2 are between 0~505 m and 0~2.340,respectively,the scatterplot of the relationship between landform and population shows an"L"shape,the population about 88.1%distribu ted at 2~200 m altitude and 77.8%of the population distributed in areas with small landform and landform.
作者
黄耀裔
HUANG Yao-yi(School of Resources and Environmental Science,Quanzhou Normal University,Quanzhou Fujian 362000,China)
出处
《科技和产业》
2020年第11期134-140,共7页
Science Technology and Industry
基金
福建省自然科学基金项目(2016J01195)
泉州市科技项目(2018Z027)。
关键词
回归分析法
加权最小二乘估计
人口网格化
地形起伏度
泉州市
regression analysis
weighted least squares estimation
population gridding
topographic relief
Quanzhou city