A Historical Streamflow Reanalysis Study for Puerto Rico

Abstract Submission: Reliable streamflow estimates are essential for effective water resource planning, flood risk management, hydropower resource evaluation, and hydroclimate impact assessment. However, many regions, such as Puerto Rico, are mostly ungauged or have limited streamflow observation, posing significant challenges for effective water resource management. To overcome this challenge, we developed a long-term retrospective reconstruction of streamflow across all NHDPlusV2 stream reaches in Puerto Rico using a hydrologic modeling framework that integrates the Variable Infiltration Capacity (VIC) model with the Routing Application for Parallel computatIon of Discharge (RAPID) routing model. The VIC-RAPID model is driven by the extended DaymetV4 for Puerto Rico, which provides gridded meteorological forgings at 1-km spatial resolution from 1950 to 2019. The VIC-RAPID models are calibrated and validated across multiple US Geological Survey streamflow gauges using the particle swarm optimization algorithm. This presentation will discuss model setup, calibration, validation and sensitivity analysis, and provide a comprehensive performance evaluation benchmarking against other available data for the region, such as NOAA’s National Water Model V3.0 Retrospective dataset and Global Reach-level Flood Reanalysis (GRFR), an updated version of Global Reach-scale A priori Discharge Estimates (GRADES) dataset. Plans to support further multi-model hydroclimate impact assessments will also be discussed. We anticipate that this new dataset will provide robust, process-based insights to support informed decision-making for water resource management, flood mitigation, and hydropower planning in Puerto Rico.

Learning Objectives/Expected Outcome (Optional) :