Specific Attenuation-Based Rain-Rate Applicability to Varying Rainfall Intensity in Complex Terrain

• Published in: Journal of atmospheric and oceanic technology Vol. 42; no. 8; pp. 889 - 907
• Main Authors: Cornejo, Ian C., Rowe, Angela K., Dixon, Michael, Romatschke, Ulrike
• Format: Journal Article
• Language: English
• Published: Boston American Meteorological Society 01.08.2025
• Subjects: Algorithms; Atmospheric precipitations; Atmospheric research; Automation; Clutter; Convective cells; Datasets; Deployment; Estimates; Estimation; Extreme weather; Heavy rainfall; Mountain regions; Mountainous areas; Mountains; Performance evaluation; Polarization; Precipitation; Radar; Radar attenuation; Rain; Rainfall; Rainfall forecasting; Rainfall intensity; Terrain; Topography; Taiwan
• ISSN: 0739-0572; 1520-0426
• DOI: 10.1175/JTECH-D-24-0094.1

Dual-polarization radar is a valuable tool for quantifying rainfall, including in remote and mountainous regions, but subject to partial beam blockage and ground clutter. Recent studies have shown that the utilization of specific attenuation by a radar for rain-rate estimations R ( A ) can overcome partial beam blockage impacts by relying on radial profiles of differential phase shifts. This study applies this R ( A ) algorithm to the National Science Foundation (NSF) National Center for Atmospheric Research (NCAR) dual-polarization S-band (S-Pol) radar deployment in northwest Taiwan for the NSF-funded 2022 Prediction of Rainfall Extremes Campaign in the Pacific (PRECIP) dataset to provide reliable rain estimates, including over the mountains. While operating for nearly 3 months, S-Pol measured multiple heavy rainfall events including isolated convective cells and widespread stratiform precipitation, thus providing an ideal dataset for evaluating R ( A ) performance sensitivities to local terrain, clutter, ray segmentation, and rain type. The R ( A ) performance improved over the more commonly used NSF NCAR hybrid rain algorithm when applying to an extreme PRECIP event after several modifications to the R ( A ) algorithm. Over the entire campaign, R ( A ) overestimated low-intensity rainfall and was challenged in overall performance owing to remaining clutter. A new algorithm R (Synth), developed to merge R ( A ) into the hybrid algorithm when it was most optimal, led to improved performance in lower intensity rainfall when compared to R ( A ) alone.