An interpretable deep forest model for estimating hourly PM10 concentration in China using Himawari-8 data

Estimation of Atmospheric PM10 Concentration in China Using an Interpretable Deep Learning Model and Top‐of‐the‐Atmosphere Reflectance Data From China’s New Generation Geostationary Meteorological Satellite, FY‐4A

“…Compared with PM 10 on haze days, PM 10 on dust days in China and northern China increased by 31.47 (50.5%) and 52.02 (73.7%) μg/m 3 , respectively. The results were similar to those of others (Chen, Song, Shi, & Li, 2022; Gobbi et al., 2013; Guan et al., 2019; Remoundaki et al., 2013). The source (originating from the Taklimakan Desert in China) and transmission path of the two dust weather processes were similar, and the contribution to atmospheric PM 10 concentration in China and northern China was the same, but the intensity of the second dust weather was weaker than that of the first dust weather.…”

“…The results of the FY-4A TOAR-PM 10 model showed that TOAR, BLH, RH, surface wind speed (U10 and V10), TM, and TIME contributed significantly to the model. The performance of the model was related to the contributions of these important features (Chen, Song, Shi, & Li, 2022). The performance of the model would be worse in areas with a large contribution to surface pressure (SP).…”

“…A 10‐fold cross‐validation method was used to test the model performance (Rodriguez et al., 2010). The parameters used to describe the model performance included the determination coefficient ( R 2 ), root mean square error (RMSE), and mean absolute error (MAE; Chen, Song, Pan, & Huang, 2022; Chen, Song, Shi, & Li, 2022). The expected error (EE, Equation ) was used to evaluate the accuracy of the TOAR‐PM 10 model.…”

“…The hourly cross validation R 2 estimated by the AOD‐PM 10 model ranged from 0.82 to 0.88. The R 2 for daily, monthly, seasonal, and annual averages of atmospheric PM 10 concentrations were 0.87, 0.91, 0.94, and 0.94, respectively (Chen, Song, Shi, & Li, 2022).…”

Estimation of the ground-level SO2 concentration in eastern China based on the LightGBM model and Himawari-8 TOAR

“…Its TOAR data covers a wide area of eastern China with high temporal resolution, including 16 bands ranging from visible to near-infrared light. Therefore, the Himawari-8 TOAR has great advantages in building a high spatial and temporal resolution estimation model of ground-level pollutant concentration, which has been widely used in many related studies (Zang et al, 2018;Wei et al, 2021;Xu et al, 2021;Song et al, 2022a;Chen et al, 2022c). However, to the best of our knowledge, the Himawari-8 TOAR has not been applied to ground-level SO 2 concentration estimation so far.…”