ESSIC/CISESS Graduate Student Tianning Su was selected for a speaking award at the recent American Meteorological Society (AMS) Annual Meeting. He received the award from the judges at the 24th Conference on Integrated Observing and Assimilation Systems for the Atmosphere, Oceans, and Land Surface (IOAS-AOLS). His talk was entitled, “Retrieving Aerosol Optical Depth Retrievals over Land by Constructing the Relationship of Spectral Surface Reflectances through Deep Learning: Application in Himawari-8.”
His dynamic slide presentation took the audience through his training algorithm for an aerosol optical depth (AOD) satellite product, one of the most critical and often-used variables for air quality analysis. Su was able to improve an AOD algorithm based on its relationship to surface reflectance (SR) data. The new Deep Learning scheme (shown above) was able to untangle factors that make this a difficult parameter to retrieve (listed in the first box). The result was a significant drop in random noise and it improved results even in areas not used to train the Deep Learning scheme. This technique shows potential to improve AOD retrievals from other satellites as well.
Tianning Su is a doctoral student under the guidance of Prof. Zhanqing Li. His research interests include remote sensing techniques and the interaction between aerosols and boundary layer.
![Figure 3. Monthly TPW anomaly time series shown in red for ERA5 and blue for MiRS SNPP, 2012-2021. The straight lines are the linear regression fits. [a] is for global, [b] for Tropics, and [c] for mid-latitude regions, respectively.](https://essic.umd.edu/wp-content/uploads/2024/11/grassotti-zhou-150x150.png)
