Ma et al. (2025) Model Calibration and Data Assimilation for the National Water Model: A Case Study over California

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

Identification

• Journal: Journal of Hydrometeorology
• Year: 2025
• Date: 2025-12-05
• Authors: Xiaoyu Ma, Robert Hartman, Ming Pan, Dennis P. Lettenmaier
• DOI: 10.1175/jhm-d-25-0066.1

Research Groups

• National Oceanic and Atmospheric Administration (NOAA)
• Collaborators involved in the development of the National Water Model (NWM)

Short Summary

This study retrospectively evaluated and enhanced the Noah-MP hydrological model for flood forecasting in three California river basins, demonstrating that an enhanced Noah-MP produced forecasts comparable to, and often more accurate than, operational forecasts, particularly for lead times exceeding 2 days.

Objective

• To evaluate and enhance the Noah-MP hydrological model for retrospective flood forecasting in selected western United States watersheds, specifically focusing on the impact of model calibration, precipitation bias correction, and initial soil moisture adjustment through data assimilation.

Study Configuration

• Spatial Scale: Three Forecast-Informed Reservoir Operations (FIRO) watersheds in California: the Russian, Yuba–Feather, and Santa Ana River basins.
• Temporal Scale: Selected flood events occurring between 2004 and 2023 (a 19-year period).

Methodology and Data

• Models used:
  • Noah LSM with multiparameterization options (Noah-MP) as the core of the National Water Model (NWM).
• Data sources:
  • Observed (or forecasted) precipitation, surface air temperature, and other surface meteorological variables.
  • Discharge observations (for data assimilation to adjust initial soil moisture).
  • Archived flood forecasts from the National Weather Service California–Nevada River Forecast Center (CNRFC) for comparison.

Main Results

• In the two humid and semihumid basins (Russian and Yuba–Feather), Noah-MP reforecasts produced flood forecast results with accuracy comparable to archived forecasts from the National Weather Service California–Nevada River Forecast Center (CNRFC).
• When incorporating precipitation bias correction and adjustments to initial soil moisture (informed by forecasted peaks), Noah-MP reforecast results were slightly more accurate than CNRFC forecasts across all three river basins.
• The improved accuracy of the enhanced Noah-MP was particularly notable for lead times exceeding 2 days.

Contributions

• Provides a comprehensive retrospective evaluation of the Noah-MP model for flood forecasting in the western United States, an area previously with limited assessment.
• Demonstrates the significant improvement in flood forecast accuracy achieved through model enhancements, including calibration, precipitation bias correction, and data assimilation of discharge observations for initial soil moisture.
• Offers a direct comparison of enhanced Noah-MP performance against operational forecasts from the CNRFC, highlighting its potential for improved long-lead-time flood predictions.

Funding

• Not specified in the abstract.

Citation

@article{Ma2025Model,
  author = {Ma, Xiaoyu and Hartman, Robert and Pan, Ming and Lettenmaier, Dennis P.},
  title = {Model Calibration and Data Assimilation for the National Water Model: A Case Study over California},
  journal = {Journal of Hydrometeorology},
  year = {2025},
  doi = {10.1175/jhm-d-25-0066.1},
  url = {https://doi.org/10.1175/jhm-d-25-0066.1}
}