Intercomparison of the U.S. National Water Model with OpenET over the Bear River Basin, U.S

Authors: NASSAR, A - Utah State University; TARBOTON, D - Utah State University; Anderson, Martha; YANG, Y - Mississippi State University; SENAY, G - Us Geological Survey (USGS); FISHER, J - Chapman University; PURDY, A - California State University; BAIG, F - University Of Illinois; GOCHIS, D - National Center For Atmospheric Research (NCAR); MELTON, F - National Aeronautics And Space Administration (NASA)

Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/23/2025
Publication Date: 2/8/2025
Citation: Nassar, A., Tarboton, D., Anderson, M.C., Yang, Y., Senay, G., Fisher, J.B., Purdy, A.J., Baig, F., Gochis, D., Melton, F. 2025. Intercomparison of the U.S. National Water Model with OpenET over the Bear River Basin, U.S. Journal of Hydrology. 656. Article e132826. https://doi.org/10.1016/j.jhydrol.2025.132826.
DOI: https://doi.org/10.1016/j.jhydrol.2025.132826

Interpretive Summary: The U.S. National Water Model (NWM) is used by the National Weather Service to forecast streamflow, flooding, snowpack, evapotranspiration (ET) and other hydrologic states and fluxes across the United States at a resolution of 1 kilometer. While the NWM has been evaluated at points where ground-based measurements are available, these measurements are sparse in some regions, particularly in the western U.S. where hydrologic information is critically needed. This paper describes the use of a satellite remote sensing platform, OpenET, as an alternative method of NWM evaluation. OpenET is composed of six ET remote sensing models running as an ensemble on Google Earth Engine to provide routine information on water use at 30-m spatial resolution. Serving as an independent, and densely spatially distributed, measure of ET, OpenET was compared with NWM over the Bear River Basin near the Great Salt Lake of Utah. ET comparisons were reasonable at the seasonal level, but with some distinct differences in areas of irrigation, in riparian vegetation, and in wetlands around the Great Salt Lake. These are areas where the process-based NWM may miss ancillary moisture available for evaporation that are captured by the satellite imagery. The comparison was also able to identify a mis-calibrated sub-watershed within the NWM. In turn, the NWM provides valuable water balance constraints on the OpenET water use estimates. This study provides a valuable prototype for broader assessment of both NWM and OpenET via intercomparison.

Technical Abstract: The U.S. National Water Model (NWM) is a hydrologic modeling framework that uses the Weather Research and Forecasting Hydrological modeling system (WRF-Hydro) to simulate land surface hydrology and energy fluxes at moderate spatial resolution (1 km). Understanding the performance of the operational NWM in simulating evapotranspiration (ET) is necessary to identify problems and biases in streamflow forecasts that may result from poor partitioning of runoff and ET. In this study, we compared NWM ET fluxes against OpenET, a satellite-driven dataset that provides interpretive or diagnostic information on actual ET at 30-m spatial resolution. Monthly ET simulations from the NWM version 2.1 (NWM V2.1) retrospective analysis over the Bear River Basin, U.S. were compared against OpenET products from 2017 to 2020 for different months and seasons. Comparisons showed that there was general agreement between the ET assessments at the 1 km scale, but with notable discrepancies for some land cover types, such as irrigated agriculture and riparian areas. The NWM showed less spatial variability and tended to predict lower ET fluxes compared to OpenET, particularly from June to August in this basin. In comparison with water balance estimates of ET derived from precipitation and USGS streamflow observations within the Bear River Basin, OpenET modeled ET was biased high while NWM ET had better agreement. The results from this study provide useful information for both NWM and OpenET developers, demonstrating the power of comparing predictive and interpretive modeling systems. This study serves as a prototype for broader assessment of both NWM and OpenET via intercomparison.