摘要
针对PM2.5浓度估算中全局平稳因素和局部非平稳因素同时存在的问题,该文以京津冀地区2015年1月—7月的PM2.5浓度为研究对象,人口密度、GDP、AOD、温度、相对湿度、风速和大气压强为影响因子,利用混合时空地理加权回归模型对PM2.5浓度进行估算。结果显示,考虑全局平稳特征时能有效地提升PM2.5浓度估算的精度,PM2.5浓度与风速、温度呈负相关关系,京津冀地区PM2.5浓度呈北低南高的趋势。
In the estimation of PM2.5 concentration,there is a phenomenon that both global stationary characteristics and spatial-temporal non-stationary characteristics exist at the same time.However,previous studies have been studied of spatial-temporal non-stationary characteristics.But there is no study on global stationary characteristics.MGTWR model can simultaneously analyze the influence of global stationary characteristics and spatial-temporal non-stationary characteristics.Therefore,it can obtain more accurate estimations of PM2.5 concentration by taking stationary characteristics into account.The results show that PM2.5 concentration is negatively correlated with wind speed and temperature.In addition,the spatial-temporal trends of PM2.5 concentration are analyzed.The trend of PM2.5 concentration is higher in the north and lower in the south.
作者
张小璐
刘纪平
梁勇
刘晓东
赵阳阳
董珍珍
ZHANG Xiaolu;LIU Jiping;LIANG Yong;LIU Xiaodong;ZHAO Yangyang;DONG Zhenzhen(School of Information Science＆ Engineering, Shandong Agricultural University, Tai＇an, Shandong271018, China;Chinese Academy of Surveying and Mapping, Beijing100830, China)
出处
《测绘科学》
CSCD
北大核心
2018年第5期33-39,49,共8页
Science of Surveying and Mapping
基金
国家重点研发计划项目(2016YFC0803108)
关键词
平稳特征
混合时空地理加权回归模型
PM2.5浓度
时空趋势
stationary characteristics
mixed geographically and temporally weighted regression model
PM2.5 concentration
spatial-temporal trends
