Department of Geo-informatics, Central South University, Changsha, 410000, China; School of Civil and Surveying Engineering, Jiangxi University of Science and Technology, Ganzhou, 341000, China.
School of Civil and Surveying Engineering, Jiangxi University of Science and Technology, Ganzhou, 341000, China.
Chemosphere. 2022 Dec;308(Pt 2):136301. doi: 10.1016/j.chemosphere.2022.136301. Epub 2022 Sep 2.
The AOD derived from the MODIS deep blue(DB) algorithm and AQI were used to investigate the correlation between AOD and AQI in seven major cities of Yangtze River Delta (YRD) from January to December 2019. The accuracy of MODIS AOD was validated by AERONET. Moreover, the AOD and AQI were studied to explore the annual and seasonal distribution characteristics, and the correlation analysis was carried out using five regression models. It was found: Ⅰ) There was a significant correlation between AOD and AERONET data (R ˃ 0.80, RMSE = 0.106, and MAE = 0.089). Ⅱ) The highest AQI was observed in winter (83), followed by spring (76), autumn (74), and summer (72). Ⅲ) The monthly average AOD showed noticeable seasonal variations, which reached the highest in summer (0.91) and the lowest in winter (0.69), followed by spring and autumn. Ⅳ) Among the five models, the cubic model obtained the best results with R ˃ 0.55. In the sub-seasonal regression model, the cubic model outperformed other models in spring (R ˃ 0.57), summer (R ˃ 0.76) and autumn (R ˃ 0.38). However, in winter the composite model outperformed others (R ˃ 0.68). Ⅴ) Considering annual data, the AOD can predict over 70% of the variations in AQI (0.41＜R ＜0.81). These results demonstrate the feasibility of AOD derived from the MODIS DB algorithm in AQI prediction. The method used in this study can be applied as an aid for air pollution control programs in different regions.
利用 MODIS 深蓝天域算法（DB）反演的 AOD 与 AQI 数据，研究了 2019 年 1-12 月长三角地区（YRD）7 个主要城市 AOD 与 AQI 的相关性。利用 AERONET 观测数据对 MODIS AOD 数据的精度进行了验证。同时，研究了 AOD 和 AQI 的年、季分布特征，并采用 5 种回归模型进行了相关性分析。结果表明：①MODIS AOD 与 AERONET 观测数据相关性较好（R ˃ 0.80，RMSE=0.106，MAE=0.089）；②AQI 冬季最高（83），其次是春季（76）、秋季（74）和夏季（72）；③AOD 月均值呈明显季节性变化，夏季最高（0.91），冬季最低（0.69），春季和秋季次之；④5 种回归模型中，三次模型拟合效果最好（R ˃ 0.55）。在亚季节回归模型中，三次模型在春季（R ˃ 0.57）、夏季（R ˃ 0.76）和秋季（R ˃ 0.38）的拟合效果好于其他模型，而在冬季复合模型的拟合效果较好（R ˃ 0.68）；⑤考虑年际数据，AOD 可较好地预测 AQI 的 70%以上变化（0.41＜R ＜0.81）。结果表明，MODIS DB 算法反演的 AOD 可用于 AQI 预测，该方法可为不同地区的空气污染防控提供决策支持。
