基于GTWR模型的3 km京津冀PM_(2.5)时空分布和影响因素分析  

Spatial and temporal distribution and influencing factors of PM_(2.5) in 3 km Beijing-Tianjin-Hebei region based on GTWR model

作　　者：王岩 刘纪平[2] 赵阳阳[2] 徐婧 WANG Yan;LIU Jiping;ZHAO Yangyang;XU Jing(School of Geomatics,Liaoning Technical University,Fuxin 123000,China;Chinese Academy of Surveying and Mapping,Beijing 100036,China;Faculty of Geomatics,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]辽宁工程技术大学测绘与地理科学学院,辽宁阜新123000 [2]中国测绘科学研究院,北京100036 [3]兰州交通大学测绘与地理信息学院,甘肃兰州730070

出　　处：《测绘通报》2024年第6期82-89,共8页Bulletin of Surveying and Mapping

基　　金：国家自然科学基金(42001343)。

摘　　要：PM_(2.5)与空气质量和公众健康密切相关,许多研究使用遥感结合其他辅助数据的模型反演PM_(2.5)的浓度,以捕捉各地区PM_(2.5)时空分布。本文针对京津冀地区数据分辨率较低的问题,采用3 km分辨率的气溶胶光学厚度(AOD)数据及12个辅助变量,建立时间地理加权回归模型(GTWR),估算3 km京津冀地区2020—2022年PM_(2.5)浓度分布。结果表明:①GTWR模型数据的R^(2)(0.86)均优于OLS模型数据的R^(2)(0.66)和GWR模型数据的R^(2)(0.78)。②在时空分布上,2020—2022年京津冀PM_(2.5)浓度的空间分布与地形呈负相关。低值区主要分布在地势较高的山区;高值区主要分布在地势较低的平原。③2020—2022年京津冀PM_(2.5)季节平均浓度差异显著,由高至低依次为冬季(60.88μg/m^(3))、秋季(37.78μg/m^(3))、春季(31.75μg/m^(3))、夏季(22.16μg/m^(3))。④PM_(2.5)浓度与AOD的相关性最强。研究得出3 km分辨率的AOD数据与GTWR模型相结合在反演PM_(2.5)浓度方面具有较好的适用性。PM_(2.5) is closely related to air quality and public health,and many studies use remote sensing combined with other auxiliary data models to invert PM_(2.5) concentration to capture the spatial and temporal distribution of PM_(2.5) in various regions.Aiming at the problem of low data resolution in the Beijing-Tianjin-Hebei region,this study adopts 3 km resolution aerosol optical depth(AOD)data and 12 auxiliary variables to establish a geographically and temporally weighted regression model(GTWR)to estimate the PM_(2.5) concentration distribution in the 3 km Beijing-Tianjin-Hebei region during 2020—2022.The results show that:①the R^(2) of GTWR model data(0.86)is better than that of OLS model data(0.66)and GWR model data(0.78).②The spatial and temporal distribution of PM_(2.5) concentration in the Beijing-Tianjin-Hebei region during 2020—2022 is negatively correlated with the terrain.The low-value area is mainly distributed in the high-lying mountainous area,and the high-value area is mainly distributed in the low-lying plain.③The seasonal mean concentration of PM_(2.5) in Beijing-Tianjin-Hebei from 2020 to 2022 was significantly different as follows:winter(60.88μg/m^(3)),autumn(37.78μg/m^(3)),spring(31.75μg/m^(3)),summer(22.16μg/m^(3)).④The correlation between PM_(2.5) concentration and AOD is the strongest.It is concluds that the combination of 3 km resolution AOD data and GTWR model has good applicability in retrieving PM_(2.5) concentration.

关 键 词：PM_(2.5) 气溶胶光学厚度 时空地理加权 时空分布 

分 类 号：P208[天文地球—地图制图学与地理信息工程]