Rainfall Assimilation and Numerical Forecast of the 25 January 2000 Snow Storm

On 25 January 2000, a heavy snow storm paralyzed Washinton, D. C. and most part of Maryland. This storm was initiated with the cyclogenesis in the southeastern US and moved to Maryland along the coast. The observed 24-h maximum snowfall in that area is around 16 inches on 25 January 2000. Most operational models failed to predicte the storm.

In this study, the predictability of the snow storm with respect to model initial conditions is studied. The non-hydrostatic, mesoscale model MM5 and its 4D-Var system is used to carry out the experiments. In order to improve the quality of the model initial conditions, the observed precipitation information by the rain gauges and satellite are assimilated into the model initial conditions. The forecast is initialized at 0000 UTC 24 January 2000. Multi-nested domian simulations are performed using MM5 V2. By comparing the results between control and data assimilation experiments, the predictability of the snow storm with respect to the model initial conditions is explored.