Assessing Radar Reflectivity in Tropical Cyclones: A Comparative Analysis from Satellite and Ground-based Radars

This study explores the advantages and limitations of space-borne and ground-based radar technologies in the context of tropical cyclone (TC) monitoring. Ground-based radars excel at detecting near-land and land-falling TCs, offering continuous atmospheric scanning and minimal temporal gaps in data acquisition. However, challenges such as beam spreading, limited effective range, and the need for data mosaicking exist. Additionally, ground-based radar instruments are susceptible to damage during TC passages, impacting data observations. In contrast, satellite-based radars provide extensive spatial coverage and global perspectives and are especially beneficial over the ocean, where TCs spend a significant portion of their lifecycle. Despite these advantages, space-borne radars have limitations in horizontal extent and temporal sampling.

Previous studies have compared the Global Precipitation Measurement (GPM) mission Dual-frequency Precipitation Radar (DPR) to ground radars, revealing acceptable performance but with some underestimation of heavy rainfall in convective scenarios. This study builds on existing comparisons by evaluating DPR during TC events. Observations from DPR are matched with U.S. National Weather Service’s ground-based NEXt Generation RADar (NEXRAD) Weather Surveillance Radar-1988 Doppler (WSR-88D) radars to compare reflectivity values during North Atlantic TC events making landfall in the contiguous U.S. The aim is to provide insight into measurement consistencies and biases of DPR reflectivity observations compared to the WSR ground-based radars, enhancing our understanding of DPR capabilities in observing TC rainbands over land. This is particularly relevant for relying on DPR observations over the ocean, where ground-based radars are unavailable.

Challenges in direct radar reflectivity measurement comparison are addressed by temporally matching and resampling data to a common grid. Metrics are then calculated to compare DPR reflectivity against WSR-88D measurements. 25 DPR overpasses that had at least 1000 jointly observed points with the WSR for TC events during 2014–2021 are used. The mean error (mean bias) of the TC cases ranged from -1.99 to 2.81 dBZ and the mean absolute error of the TC cases ranged from 2.43 to 4.56 dBZ. The correlation between the matched DPR and WSR for all TC cases ranged from 0.74 to 0.91. This study explores variables such as storm forward speeds, location of TCs, and TCs that underwent extratropical transition to account for variations in TC characteristics and their potential impact on measurement biases. By delving into these aspects, this study underscores the importance of a multifaceted approach, combining the strengths of both space-borne and ground-based radar technologies, to enhance understanding and forecasting of TCs, particularly in regions where ground-based radar observation may not be available.