Prototype of an ensemble radar and satellite data assimilation system for Warn-on-Forecast

One of the primary goals of the NOAA Warn-on-Forecast (WoF) program is to provide accurate short-range (0-1 h) probabilistic forecasts of severe convective storms. The prototype WoF storm-scale ensemble Kalman filter data assimilation (DA) system comprises of a 3 km convection permitting nest embedded within a larger mesoscale domain using 36 members. The 3 km nest is updated at 15 minute intervals by assimilating radar reflectivity and radial velocity, GOES Imager cloud water path (CWP) retrievals, and mesonet data, when available. This system was run for several events during the spring of 2013 and 2014 comprising multiple storm modes and environments. Results showed that 15 minute cycling was able to generate model analyses that verify well with observations and act as good initial conditions for short term forecasts. The addition of CWP generally improved the analysis early in the assimilation period by allowing the model to spin up storms at a faster rate. As a result, forecasts initiated during these periods were generally superior to those assimilating only radar data. As storms evolve and grow upscale, the positive impacts of satellite data appear to decrease. Overall, the positive results from this initial attempt have led to the development of a next generation model design to be used during spring 2015. Elements of the next version of this DA system will be discussed along with plans for implementation.