Short Term Forecasting of Snowbands Using Radar Data and 4DVAR Assimilation

Data assimilation experiments using WSR-88D observations and MM5 are conducted for a snowstorm event in northeastern U.S. Simulation tests have shown that MM5 has quantifiable skill in forecasting the occurrence of snowstorms in northeastern U.S. which typically have strong synoptic forcing. However, the timing, duration, and amount of snowfall predicted by MM5 are much less accurate. The objective of this work is to develop assimilation techniques for improved precipitation forecasts in the 1-6 hour timeframe, by using multi-radar observations to provide information on the storm's wind and precipitation fields.

Several methods of incorporating radar observations of reflectivity and/or radial velocity are being explored and tested. Preliminary results show that it is important to adjust the modeled wind and thermal fields in addition to the precipitation fields during the data assimilation process. Further experiments will be performed using the MM5-4DVAR system as well as the Newtonian relaxation method. Forecast results using the two methods will be compared.