基于GF-5B卫星的随机森林模型反演京津冀地区PM_(2.5)  

Inversion of PM_(2.5) in the Beijing-Tianjin-Hebei region using a random forest model based on GF-5B satellite

作　　者：张瑞杰 周春艳 陈辉 周伟 谢慧真 陈睿智 王中挺 ZHANG Rui-jie;ZHOU Chun-yan;CHEN Hui;ZHOU Wei;XIE Hui-zhen;CHEN Rui-zhi;WANG Zhong-ting(Chinese Research Academy of Environmental Sciences,Beijing 100012,China;Satellite Application Center for Ecology and Environment,Ministry of Ecology and Environment,Beijing 100094,China)

机构地区：[1]中国环境科学研究院,北京100012 [2]生态环境部卫星环境应用中心,北京100094

出　　处：《中国环境科学》2024年第11期5961-5970,共10页China Environmental Science

基　　金：国家重点研发计划(2022YFE0209100);国家自然科学基金资助项目(41971324)。

摘　　要：本研究利用国产卫星GF-5B DPC载荷反演得到的气溶胶光学厚度(AOD)数据、ERA-5气象数据、NDVI数据、DEM数据、夜间灯光数据以及地面PM_(2.5)实测数据,基于随机森林模型,构建AOD-PM_(2.5)反演模型估算2022年京津冀地区PM_(2.5)浓度.结果表明,全年模型决定系数(R~2)为0.80,均方根误差(RMSE)为11.43μg/m^(3);春夏秋冬四组模型决定系数分别为0.84、0.71、0.88、0.87,均方根误差分别为9.50,7.37,9.71,11.32μg/m^(3),无论是在全年整体模拟还是分季节模拟均具有较好的模拟精度.2022年京津冀地区PM_(2.5)浓度呈现显著的季节变化规律,冬季的PM_(2.5)浓度最高,夏季最低.全年PM_(2.5)估算平均值为30.86μg/m^(3).夏秋两季PM_(2.5)浓度影响因素主要是AOD,春冬两季PM_(2.5)浓度影响因素主要是边界层高度以及相对湿度.This study utilizes aerosol optical depth(AOD)data derived from the domestic GF-5B DPC payload,ERA-5meteorological data,NDVI data,DEM data,nighttime light data,and ground-based PM_(2.5)measurements.Based on a random forest model,an AOD-PM_(2.5)inversion model is developed to estimate the PM_(2.5)concentration in the Beijing-Tianjin-Hebei region for the year 2022.The results showed that the model's coefficient of determination(R2)for the entire year was 0.80,with a root mean square error(RMSE)of 11.43μg/m^(3).For the seasonal models,the R2 values were 0.84,0.71,0.88,and 0.87 for spring,summer,autumn,and winter,respectively,with corresponding RMSE values of 9.50,7.37,9.71,and 11.32μg/m^(3).Both the annual and seasonal models demonstrated good simulation accuracy.The PM_(2.5)concentrations in the Beijing-Tianjin-Hebei region in 2022 exhibited significant seasonal variation,with the highest concentrations in winter and the lowest in summer.The estimated annual average PM_(2.5)concentration was 30.86μg/m^(3).During the summer and autumn,AOD was the primary factor influencing PM_(2.5)concentrations,while in spring and winter,the boundary layer height and relative humidity were the main influencing factors.

关 键 词：高分五号B星 随机森林模型 京津冀 气溶胶光学厚度(AOD) PM_(2.5) 

分 类 号：X513[环境科学与工程—环境工程]