Assimilating Surface Water Ocean Topography (SWOT) Water Surface Elevations into the WRF-Hydro Modeling System in Alaska using HydroDART

The operational National Water Model (NWM) was implemented in August 2016 as an uncoupled instantiation of the Weather Research and Forecasting hydrological extension package (WRF-Hydro). Currently, the NWM only covers the contiguous United States, but will be expanded to include an Alaska domain in the near future. It is well known that Alaska presents several unique hydrological modeling challenges, including frozen soils and rivers during many months of the year, rapid snowmelt, braided rivers with variable geometries, and a severe lack of in situ or remotely-sensed observations for model initialization, all of which hinder operational forecasting efforts. In 2021, the Surface Water Ocean Topography (SWOT) mission will be launched to provide high-resolution measurements of water surface elevations (WSE) for rivers with widths greater than 50-100 m. SWOT measurements have the capability to improve model initialization, but few efforts are being made to prepare these measurements for assimilation into operational hydrologic models. Even though SWOT latency may preclude the use of SWOT measurements for real-time flood forecasting, assimilating SWOT measurements can improve baseflow estimates in the model analysis and provide a basis for assimilating observations from future follow-on missions with shorter latencies into operational hydrologic models. This presentation assimilates synthetic SWOT WSE into the WRF-Hydro modeling system using the National Center for Atmospheric Research (NCAR) HydroDART Ensemble Kalman Filter (EnKF) in a twin experiment methodology for the Chena River watershed in Alaska. For this study, WRF-Hydro is configured with a 1-km resolution Noah-Multiparameterization Land Surface Model (Noah-MP LSM). Terrain routing grids were generated at 100-m and 250-m resolutions for the control and corrupted WRF-Hydro simulations, respectively. During the assimilation experiment, synthetic SWOT WSE derived from the 100-m resolution control simulation are assimilated into the diffusive wave channel routing module of the 250-m resolution corrupted simulation, for which channel head is the prognostic variable. These experiments used Global Land Data Assimilation System (GLDAS) meteorological forcing, which was regridded to match the 1-km resolution of the Noah-MP LSM.