Generating Fine-Scale Aerosol Data through Downscaling with an Artificial Neural Network Enhanced with Transfer Learning

Associate Professor Meredith Franklin published an article in Atmosphere entitled "Generating Fine-Scale Aerosol Data through Downscaling with an Artificial Neural Network Enhanced with Transfer Learning”. In this study, a spatiotemporal deep-learning method was developed to downscale satellite-observed aerosols over the Middle East. This work provides valuable indicators of air quality in a region of the world that experiences high anthropogenic emissions and dust storms but has very little infrastructure for pollution monitoring or control. The fine-scale aerosol information generated through this study is being used to support ongoing health effects assessments of military personnel who were deployed in the region during post 9-11 wars.