摘要
利用长江三角洲城市群中7个典型城市2019年1—12月中分辨率成像光谱仪(Moderate-resolution Imaging Spectroradiometer,MODIS)的产品数据,以深蓝(Deep Blue,DB)算法反演的气溶胶光学厚度(Aerosol Optical Depth,AOD)与大气污染物质量浓度(PM_(2.5)、PM_(10)、SO_(2)、NO_(2)、O_(3)及CO)为研究对象,结合AERONET站点数据验证AOD反演精度。同时,采用5种回归模型研究AOD与大气污染物质量浓度的相关性,结果表明:AOD反演结果与AERONET 2个站点数据相关性较好,拟合度(R^(2))分别为0.858和0.733,均方误差(MSE)分别为0.021 3和0.029 4,说明深蓝算法对长江三角洲城市群的AOD反演精度可靠;研究区域年度回归模型拟合中,AOD与O_(3)拟合的指数模型最优(R^(2)=0.53,概率(P)<0.05)。在月度回归拟合中,三次模型最优,R~2依次为0.67、0.65、0.52、0.50,其次是指数模型,与前2种模型相比,线性和幂次模型效果较差,其显著性水平均超过95%;在验证最优模型精度时发现,计算值与地面监测点实测值相关性显著,R~2为0.73~0.87,误差值(RMSE)≤15.93μg/m^(3),平均绝对误差(MAE)≤12.02μg/m^(3),证实了利用深蓝算法反演的AOD数据可用于长江三角洲城市群大气污染物质量浓度的估算和监测,为更好地防治区域大气污染提供科学依据。
Aerosol Optical Depth(AOD)was retrieved by Deep Blue(DB) algorithm by using the MODIS product data and air pollutant mass concentration (PM_(2.5),PM_(10),SO_(2),NO_(2),O_(3)and CO)in seven typical cities in the Yangtze River Delta Urban Agglomeration (YRDUA)from January to December in 2019,and the inversion accuracy was validated combined with the AERONET site data.Meanwhile,the correlation between AOD and atmospheric pollutant mass concentration was studied employing the five regression models.The major findings are as follows:firstly,AOD inversion results are highly correlated with AERONET site data,with values of the correlation coefficient R^(2) at 0.858 and 0.733,MSE at 0.021 3 and 0.029 4,respectively,indicating that DB algorithm is useful for aerosol inversion accuracy in the YRDUA;Secondly,the exponential model is with best fitting by AOD and O_(3)in the annual regression model of the study area (R^(2)=0.53,P<0.05).The cubic model is optimal in the monthly regression fitting,which R^(2) is 0.67,0.65,0.52 and 0.50 respectively,followed by the exponential model.The linear and power models are less effective than the first two models.All of them have passed the significance level of 95%or above;Thirdly,when verifying the accuracy of the optimal model,it is found that the calculated value is significantly correlated with the measured value of the ground monitoring stations,R^(2) ranging from 0.73 to 0.87,RMSE≤15.93μg/m^(3),MAE≤12.02μg/m^(3).It was proved that AOD retrieved by DB algorithm can be used as an effective means for estimating and monitoring the mass concentration of air pollutants in the YRDUA,and can provide scientific reference for better prevention and control of regional air pollution.
作者
李丹
陈优良
邹文敏
LI Dan;CHEN Youliang;ZOU Wenmin(School of Civil and Surveying&Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,Jiangxi,China)
出处
《江西冶金》
2022年第5期90-98,共9页
Jiangxi Metallurgy
基金
国家大坝安全工程技术研究中心开放基金资助项目(CX2019B07)。
关键词
大气污染物
气溶胶光学厚度
相关性分析
长江三角洲城市群
air pollutants
aerosol optical depth
correlation analysis
Yangtze River Delta Urban Agglomeration
