In this project, an ensemble back-propagation neural network technique will be developed and used to estimate the spatial Aerosol optical depth (AOD) in the Greater Bay Area by combining the satellite remote sensing data, ground observation data, and numerical model data. The estimated AOD field in relatively high accuracy will then be assimilated into the 3D chemical transport model to further improve the photolysis rate simulation. The outcome of this project can assist us in better understanding the aerosol impacts on the radiation-related atmospheric chemistry reactions. Students should seek instructor's approval to take this course.