Su et al. (2025) Improving National Water Model Flood Forecast Skill over Coastal Western U.S. River Basins

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Journal of Hydrometeorology
• Year: 2025
• Date: 2025-07-30
• Authors: Lu Su, Ming Pan, Robert Hartman, Dennis P. Lettenmaier
• DOI: 10.1175/jhm-d-24-0072.1

Research Groups

• National Oceanic and Atmospheric Administration (NOAA)
• National Weather Service (NWS)
• River Forecast Centers (RFCs)

Short Summary

This study compares the flood forecasting accuracy of WRF-Hydro (the core of the National Water Model) against traditional NWS River Forecast Center methods across seven Pacific Coast watersheds. The results indicate that while WRF-Hydro performs comparably in northern basins, it is inferior in southern basins and tends to predict flood peaks too early.

Objective

• To evaluate the performance of WRF-Hydro-based flood forecasts compared to current National Weather Service (NWS) methods in the western United States.

Study Configuration

• Spatial Scale: Seven watersheds along the U.S. Pacific Coast (spanning Washington, Oregon, and California).
• Temporal Scale: Analysis of the largest flood events occurring over the past 7–20 years.

Methodology and Data

• Models used: WRF-Hydro (stand-alone mode) with the Noah-MP land surface model.
• Data sources: NWS quantitative precipitation forecasts (QPF) and archived forecasts from coastal western U.S. River Forecast Centers (RFCs).

Main Results

• Regional Performance: WRF-Hydro forecasting performance was comparable to RFC reforecasts in northern basins (Oregon and Washington) but was inferior in southern basins (California).
• Peak Accuracy: Both WRF-Hydro and RFC forecasts tended to underestimate flood peaks.
• Lead Time and Timing: WRF-Hydro discrepancies increased more rapidly with longer lead times and typically predicted flood peaks earlier than the RFCs.
• Extreme Events: For the largest floods, the magnitude of forecasts was largely comparable between the two methods, though WRF-Hydro continued to predict earlier peaks.

Contributions

• Provides a detailed comparative validation of the National Water Model's hydrologic core (WRF-Hydro) against operational NWS methods, specifically identifying regional performance gaps in the western U.S.

Funding

• Not specified in the provided text.

Citation

@article{Su2025Improving,
  author = {Su, Lu and Pan, Ming and Hartman, Robert and Lettenmaier, Dennis P.},
  title = {Improving National Water Model Flood Forecast Skill over Coastal Western U.S. River Basins},
  journal = {Journal of Hydrometeorology},
  year = {2025},
  doi = {10.1175/jhm-d-24-0072.1},
  url = {https://doi.org/10.1175/jhm-d-24-0072.1}
}