Towards the Development of Sub-km Scale Warn-on-Forecast System at the NOAA National Severe Storms Laboratory

Current convective-storm numerical weather prediction (NWP) systems in the United States, including National Oceanic and Atmospheric Administration’s operational HRRR and the experimental Warn-on-Forecast System (WoFS), run models with a convection-allowing horizontal grid spacing of 3000 m. Future convective-storm NWP systems will employ cloud-resolving models with grid spacings of 100 m or less. This study investigates how convective-storm prediction changes as model resolution is increased with the eventual goal of developing cloud-resolving resolution WoFS. Several high-impact convective-storm events are simulated using WoFS ensemble Kalman filter data assimilation system with model horizontal grid spacings of 3000, 1000, and 500 m to generate multi-resolution initial conditions (ICs). Short-term 0-3 h ensemble forecasts from these varied model resolution ICs show effects on many aspects of the convective systems, including the storm intensity (e.g., low-level vertical vorticity), the location and coverage of storm hazards (e.g., strong wind) probability, and the speed of supercell and MCS motion. Results will be summarized and presented at the conference.