Examining Tropical Cyclone Rainbands as Detected by Ground- and Space-Based Radars

Tropical cyclones (TCs) contain large precipitation fields comprised of convective and stratiform clouds and undergo complex changes upon interacting with land which can trigger floods. Detecting these systems with remote sensing instruments provides valuable insights into structural changes that influence rainfall production. As TCs are synoptic-scale systems, the scan from a single radar cannot capture the entire storm either from space or the ground. Reflectivity values detected by space-based as compared to ground-based weather radars can differ due to multiple reasons including temporal offsets in scanning the same location and the time it takes to scan that location as well as the variation in signals emitted by ground-based radars. To improve our understanding of these differences in reflectivity during landfalling TCs, we compare the Dual-frequency Precipitation Radar (DPR) on board the Global Precipitation Mission’s satellite which predominantly scans in the vertical direction to returns from the Weather Surveillance Radar 1988-Doppler (WSR) units which emit energy on a more horizontal trajectory. We examine data from points matched in space and time between the DPR (V07A) and WSR beams as well as from WSR data mosaicked to a grid employing two different approaches: using a time-distance weight and retaining the maximum value for a grid cell. Results show that when a WSR location reports lower values than its neighbors that scan the same region, retaining the maximum value allows data from those neighboring stations to be utilized in the mosaic and improves the low bias that exists for WSR relative to DPR. We also perform spatial clustering to identify regions where differences between DPR and WSR are more homogeneous compared to the larger study area.