Forecast Models
Mesoscale Convective Systems
A member of the Severe Weather Analysis and Prediction (SWAP) team at NSSL is working to examine the ability of computer models to produce accurate analyses and short-term predictions of thunderstorm systems, or mesoscale convective systems (MCSs). Accurate prediction of MCSs is an important part of the hazardous weather prediction problem because of their penchant for producing widespread severe surface winds and very heavy rainfall. However, the current capability of computer models that are configured for real-time forecasting is not enough to predict the complex evolution of MCSs consistently and accurately. Part of this problem is related to a lack of observations of the atmospheric state. Without observations at least every 1 to 2 km in and near the developing MCS, it is very difficult to tell a computer model how to begin its forecast. Therefore, the goal of this research is to examine the ability of Doppler radar observations, which contain atmospheric observations at a resolution well under 1 km, along with innovative data assimilation and numerical modeling techniques, to improve computer model forecasts of MCSs. The hope is that these techniques will provide the modeling system with an accurate representation of the fine-scale conditions during the development of MCSs and allow for accurate prediction of MCS evolution at least 2 to 4 hours ahead of time.
