Journal of Atmospheric and Environmental Optics, Volume. 18, Issue 3, 258(2023)
Improving the accuracy of NO2 concentrations derived from remote sensing using localized factors based on random forest algorithm
Figures & Tables(7)
Fig. 1. The comparison of the distributions of NO2 ground-level observed and satellite-derived concentrations. (a) Ground-level observed NO2 concentrations; (b) NASA satellite-derived NO2 concentrations
Fig. 2. The histograms of the cross validation concentrations from the three algorithms.(a) GWR_NO2; (b) MGWR_NO2; (c) RF_NO2
Fig. 3. Cross validation results of the GWR, MGWR and random forest. (a) GWR; (b) MGWR;(c) random forest; (d) random forest 2
Fig. 4. The comparison of the predicted concentrations from the four approaches for Tibet. (a) NASA_NO2; (b) GWR_NO2;(c) MGWR_NO2; (d) RF_NO2
Fig. 5. County-level NO2 concentrations predicted by the random forest algorithm. (a) Nationwide; (b) North China Plain;(c) Yangtze River Delta; (d) Guangdong Province; (e) Urumqi, Xinjiang
Table 1. Statistical description of observed，NASA and the cross validation concentrations
View tableView in ArticleTable 1. Statistical description of observed，NASA and the cross validation concentrations
Variable Count Mean Std Min 25% 50% 75% Max Obs_NO2 1494 31.850 12.504 5.435 22.329 31.421 40.205 76.919 NASA_NO2 1494 22.490 14.862 0.600 9.600 19.550 34.400 64.800 GWR_NO2 1494 31.087 10.988 7.860 22.891 30.003 38.151 80.722 MGWR_NO2 1494 31.853 12.022 -0.544 22.822 30.789 40.035 73.671 RF_NO2 1494 31.962 10.479 8.515 24.035 30.955 38.975 64.301
Table 2. The comparison of cross validation results of the models
View tableView in ArticleTable 2. The comparison of cross validation results of the models
Model CV r CV R2 Slope RMSE MAE MAPE Reg R2 NASA 0.6937 0.4813 0.8246 14.3809 11.8655 0.4066 NA GWR 0.8426 0.7100 0.7405 6.7895 5.2021 0.1904 0.7880 MGWR 0.8370 0.7005 0.8047 7.0156 5.3053 0.2002 0.9400 Random forest 0.8582 0.7365 0.7192 6.4223 4.9850 0.1953 NA Random forest 2 0.8413 0.7078 0.7065 6.7986 5.2283 0.1994 NA
Tools
Get Citation
Copy Citation Text
Miao FU. Improving the accuracy of NO2 concentrations derived from remote sensing using localized factors based on random forest algorithm[J]. Journal of Atmospheric and Environmental Optics, 2023, 18(3): 258
Paper Information
Category:
Received: Jan. 11, 2022
Accepted: --
Published Online: Jun. 29, 2023
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20