Testing Stochastic Physics Perturbation Methods in an Experimental Next-Generation 1-km Warn-on-Forecast System

Various stochastic physics perturbation methods (SPPMs) are used in most operational ensemble systems to optimally design ensembles for skillful forecasts. However, while the National Severe Storm Laboratory’s (NSSL) experimental Warn-on-Forecast System (WoFS), which is a regional, sub-hourly, multiphysics ensemble data assimilation and prediction system with 3-km horizontal grid spacing, has been proven to successfully provide useful probabilistic guidance of severe and hazardous weather to operational forecasters, it does not currently use any SPPMs. This study explores the impact of SPPMs on the WoFS using an experimental next-generation 1-km version of the WoFS (WoFS-1km). WoFS-1km consists of a 1-km ensemble system that is nested and simultaneously run within a 3-km ensemble system, so the impact of SPPMs is assessed for both 3-km and 1-km versions of WoFS. For this initial work, four SPPMs are implemented and tested using WoFS-1km for multiple severe weather events. These methods include the stochastic kinetic energy backscatter (SKEB) scheme, the physically-based stochastic perturbations (PSP) scheme for the planetary boundary layer schemes, a fixed perturbed parameters (FPP) method for the microphysics scheme, and the surface model stochastic blending (SMSB) method for the land-surface model. Both subjective and objective methods are used to assess the impacts of the SPPMs. Preliminary results from these experiments indicate the SPPMs can result in more skillful forecasts of storms, especially during periods of storm development and maturation. This result along with other preliminary results from individual and combination SPPM experiments will be presented.