中国福彩快三网站 Almost all remote sensing atmospheric PM2.5 estimation methods need satellite aerosol optical depth (AOD) products, which are often retrieved from top-of-atmosphere (TOA) reflectance via an atmospheric radiative transfer model. Then, is it possible to estimate ground-level PM2.5 directly from satellite TOA reflectance without a physical model? This study established the relationship between satellite TOA reflectance, observation angles, meteorological factors, and satellite NDVI using a deep belief network model, and the ground-level PM2.5 has been effectively generated. The results show that the model has achieved a great performance (cross-validated R2 = 0.87, RMSE = 9.89 μg/m3). This study provides an alternative technique to estimate ground-level PM2.5. The model structure and results are presented in Figure 1 and 2.
Figure 1：The structure of deep belief network for satellite reflectance based PM2.5 estimation
Figure 2：Daily estimates of PM2.5 on some specific days. From top to bottom: (a) 20160115, (b) 20160516, (c) 20160725, and (d) 20161127. Left column: top-of-atmosphere reflectance based estimation of PM2.5. Middle column: aerosol optical depth-based estimation of PM2.5. Right column: ground station measurements.
Read More>> H. Shen, T. Li, Q. Yuan, and L. Zhang, “Estimating Regional Ground‐Level PM2. 5 Directly From Satellite Top‐Of‐Atmosphere Reflectance Using Deep Belief Networks,” Journal of Geophysical Research: Atmospheres, vol. 123, no. 24, pp. 13–875, 2018.