Supporting Data and Code for "Evaluating the U.S. National Water Model Retrospective Evapotranspiration Simulation using Eddy-Covariance Flux Tower Measurements"

This HydroShare resource contains the data, Python scripts, and documentation used in the study “Evaluating the U.S. National Water Model Retrospective Evapotranspiration Simulation using Eddy-Covariance Flux Tower Measurements” (Chapagain et al., 2025, Journal of Hydrology: Regional Studies).The study evaluates National Water Model (NWM) evapotranspiration (ET) estimates across the contiguous United States (CONUS) using multi-year comparisons with AmeriFlux tower observations at 72 sites spanning diverse hydroclimatic settings.

The analysis assesses NWM ET performance at the site level and across categorical groupings based on National Weather Service River Forecast Centers (RFCs), Köppen–Geiger climate zones, and land cover types. It also examines temperature forcing accuracy and contrasts ET behavior under water- and energy-limited (P/PET) regimes.

Results reveal wide performance variability, with stronger agreement in cold climates and forested basins and lower skill in arid and agricultural regions. ET estimates tend to perform better under water-limited than energy-limited conditions, while temperature forcings show strong agreement with observations.

The resource includes:

- Preprocessed NWM and AmeriFlux ET and meteorological datasets.
- Python scripts for computing performance metrics (e.g., Scaled KGE, PBIAS).
- Group-based and regime-based analysis notebooks.
- Visualization workflows reproducing key figures from the paper.
- ArcGIS files for site-location mapping.

Most analyses were performed in Python, with ArcGIS Pro was used for spatial processing and visualization. These materials enable reproducibility of the paper’s results and facilitate further analysis of NWM ET performance across CONUS.