Applications of HRRR ensembles for ensemble hydrologic prediction using the WRF-hydro and SACSMA models as testbeds

The primary goal of this project is to advance the application of ensemble precipitation forecasts for short-term probabilistic streamflow prediction using the High Resolution Rapid Refresh Ensemble (HRRRE) outputs.  Under this goal, we identified five supporting tasks. 

  1. Evaluate the use of state-of-the-art PQPF based on HRRR ensembles to create ensemble streamflow information by using different probability thresholds of POP from HRRRE (e.g., every 10%) to input into the hydrologic models. 
  2. Compare the hydrologic forecast skill when using POP to hydrologic forecast skill when directly using HRRRE members’ as input into the models to create probabilistic streamflow forecasts.
  3. Examine spatial shifting of HRRRE member QPF to generate streamflow ensembles to improve streamflow forecast errors associated with displacement error in the QPF when forecasting for individual basins. 
  4. Investigate the mesoscale and synoptic scale environment under which various spatial shifts in QPF occur to identify relationships that can be used to improve hydrologic forecasting by shifting the QPF appropriately based on weather conditions.  
  5. Quantify the sensitivity of the streamflow ensemble to hydrologic model parameter and state perturbations.