H111-0028
Moisture condition impact and seasonality of National Water Model performance under different runoff-infiltration partitioning schemes
Friday, 11 December 2020
Poster
Rachel McDaniel, NOAA/NWS/OWP/National Water Center, Tuscaloosa, AL, United States, Yuqiong Liu, NOAA/NWS/OWP, Silver Spring, MD, United States; LEN Technologies, Oak Hill, VA, United States, Prasanth Valayamkunnath, National Center for Atmospheric Research, Boulder, United States, Michael Barlage, NOAA/NCEP/EMC, College Park, MD, United States, David Gochis, National Center for Atmospheric Research, Boulder, CO, United States, Brian A. Cosgrove, NOAA/NWS Office of Water Prediction, Silver Spring, MD, United States and Trey Flowers, NOAA/NWS Office of Water Prediction, National Water Center, Tuscaloosa, AL, United States
Abstract:
NOAA’s National Water Model (NWM) is an operational model that was developed through a collaborative effort between academic and federal research partners. The NWM is a physics-based, distributed hydrologic model based on the community WRF-Hydro modeling system. It produces high-resolution hydrologic forecasts across the continental United States (CONUS) and Hawaii that are used to support an increasing array of water management areas including emergency management via streamflow forecasts. A critical component of NWM’s streamflow forecast is partitioning rainfall between runoff and infiltration. The current NWM runoff-infiltration partitioning scheme uses a method described by Schaake which is similar to the SCS Curve Number method developed by the USDA/NRCS in the 1950’s and is best suited for estimating total storm runoff accumulation. The goal of this study is to develop and implement alternative runoff-infiltration partitioning schemes into the NWM as well as evaluate the impact of these schemes on modeled streamflow performance.
To achieve this goal, a set of watersheds representing a variety of hydrologic regimes was selected for analysis. Watersheds were divided into ten groups based on hydrologic regimes defined by a cluster analysis using hydrologic signatures. Two watersheds from each cluster were semi-randomly selected for analysis resulting in a total of 20 watersheds across CONUS. Calibration was performed over water years 2008-2013 using procedures from the NWM 2.1 operational model. Three alternative infiltration-runoff schemes, including VIC, will be compared to the existing scheme. To assess model performance under a variety of conditions, streamflow performance will be evaluated for different hydrologic regimes, different moisture conditions, and seasonal trends using both long term (e.g. PBIAS, KGE, NSE) and event-based (e.g. peak flow and timing) metrics. Evaluating the model under different conditions, including moisture conditions and seasonality, can provide insights into the strengths and weaknesses of different schemes as well as highlight areas for future development.
