摘要
基于夜间灯光人口校正指数(NANI)、城市夜间灯光校正指数(VANUI)、人居指数(HSI)3种指数修正NPP-VIIRS夜间灯光过饱和与溢出效应的修正模型,直接面向网格,采用按县域中有人口分布的网格数加权,使用最小二乘法回归对人口空间网格化,并随机抽取一定比例网格化后的乡镇域样本加以验证.结果表明:(1)3种修正模型改进后的网格化系数均通过p=0.001的统计学意义检验,拟合优度增加、预测标准误差大幅降低;(2) 3种修正模型的人口绝对误差(MAE)、人口纯方根误差(RMSE)、人口相对误差(MRE)值均有较大幅度降低;(3) 3种修正模型两两之间的夹角余弦系数为0.981~0.996,每个网格绝对距离为29.1~75.8人,同一性大,互证有效.说明3种修正模型的改进既考虑了县域规模差异的影响又避免了建模过程中行政单元与网格单元的尺度变换,显著提高了人口网格化精度.
Three NDVI-modified models based on NPP-VIIRS night light supersaturation and spillover effects were developed and applied directly to grids.The model was weighted by the number of grids with population distribution in the county,and the least squares regression was used to grid the population,and then a certain proportion of randomly selected town field samples were used for verification.The experimental results showed that:(1) the coefficients of the three models after the improvement of the grid method have passed the statistical significance test of p=0.001.This means that the fitting degree has increased and the prediction standard error has decreased significantly.(2) The MAE,RMSE and MRE of the three models have decreased significantly.(3) The angle cosine coefficients between the three models were between 0.981 and 0.996,and the absolute distance of each grid was between 29.1 and 75.8 person,which has high consistency and is mutually effective.In conclusion,the three models have taken into account the influence of the difference in scale of the county and avoided the scale transformation between the administrative unit and the grid unit,which significantly improved the accuracy of population gridding.
作者
黄耀裔
陈文成
陈显银
郑朝洪
赵益民
HUANG Yaoyi;CHEN Wencheng;CHEN Xianyin;ZHENG Chaohong;ZHAO Yimin(College of Resources and Environment Science,Quanzhou Normal University/Key Laboratory of Rural Environmental Remediation and Waste Recycling Fujian Province University/First-class Major in Human Geography and Urban&Rural Planning,Quanzhou Fujian 362000,China)
出处
《西南师范大学学报(自然科学版)》
CAS
2023年第3期94-102,共9页
Journal of Southwest China Normal University(Natural Science Edition)
基金
福建省社会科学基金项目(FJ2021B072)
福建省科技项目(2021J01974)
福建省中青年教师教育科研项目(JAT200543,JAT200549)。