Adapt Precipitation Super-Resolution and Data Fusion Deep Learning Techniques for Operational Flood Forecasting

Researchers at the University of Alabama in Huntsville (UAH) Information Technology and Systems Center (ITSC) are developing a cloud-based analytic framework for improving the spatial resolution (super-resolution) of space-based precipitation datasets from the Global Precipitation Measurement mission by using deep learning models. These models, consisting of Convolutional Neural Networks (CNNs), are used to enhance the resolution of GPM Dual-frequently Precipitation Radar (DPR) data for refined identification of convective scale precipitation features and rain-rate estimates, particularly outside the coverage of ground-based weather radars. UAH/ITSC is supporting NOAA’s Cooperative Institute for Research to Operations in Hydrology (CIROH) by expanding its research using super-resolution precipitation imagery to enhance rain rate products. Results will be tested as an additional dataset for input into to NOAA’s National Water Model (NWM) for improved operational hydrologic forecasting. Positive impacts are expected to be the greatest in areas where ground-based radar precipitation data are lacking (i.e., CONUS mountainous regions, off-shore coastal areas, and NWM southern-Alaska domain).