High-Resolution Radar-Based Depth–Duration–Frequency Curves: A Case Study of Puerto Rico and the US Virgin Islands

Abstract Depth–duration–frequency (DDF) curves are essential tools in hydrologic and hydraulic engineering, used to estimate the magnitude and frequency of extreme precipitation events across varying durations. They play a critical role in flood risk assessment, infrastructure design, and urban drainage planning. Traditionally, DDF curves are derived from long-term rain gauge records; however, this method can be limited in regions with sparse observational networks or complex terrain, such as Puerto Rico (PR) and the US Virgin Islands (USVI). To address these limitations, this study presents updated DDF curves for PR and USVI using high-resolution ( 1 km × 1 km ) radar data from the Next Generation Weather Radar Level III archive spanning 2002–2023. Four radar-based precipitation products [digital storm total precipitation (DSP), one-hour precipitation (N1P), storm total precipitation (NTP), and dual-polarization one-hour precipitation (OHA)] were evaluated against gauge observations to identify the most suitable dataset; among them, the N1P product demonstrated the strongest agreement with ground-based measurements and was selected for further analysis. Using N1P, radar-derived DDF curves were developed at 1-km spatial resolution and compared with curves from the National Oceanic and Atmospheric Administration (NOAA), which are based on gauge records ending in 2004 for both PR and USVI. The analysis revealed notable differences, particularly for shorter durations (1–3 h), where radar estimates frequently surpassed the NOAA values. For longer durations, discrepancies exhibited greater variability. These differences may reflect a combination of factors, including differences between radar and gauge measurements, spatial averaging, record length, and radar retrieval uncertainty. The DDF curves developed in this study provide spatially detailed inputs that can support flood modeling, infrastructure resilience assessment, and climate adaptation planning.