Extended Abstract (764K)JP2.6
NIDS-Based Intermittent Diabatic Assimilation and Application to Storm-Scale Numerical Weather Prediction
Donghai Wang, Hampton Univ. and NASA/LRC, Hampton, VA; and K. K. Droegemeier, D. Jahn, K. -. M. Xu, M. Xue, and J. Zhang
The lack of accurate moisture/cloud initial conditions is one of the major causes for the spinup problem in explicit cloud and precipitation forecasting models during the first a few hours. Although many studies have tried to improve the cloud/moisture initialization by using satellite or/and radar data, to reduce this spinup problem, the lack of detailed information on initial moisture, cloud water and latent heating fields is still a key problem. The NEXRAD data can provide the three-dimentional precipitation field in the high spatial and temporal resolution. The model other conventional variables (i.e., water vapor, temperature and wind), however, may not be consistent with the cloud and precipitation analysis fields. Therefore, the evaporation may quickly kill convective storms present at the start of the model forecast. To address this problem, a diabatic initialization scheme has been improved to provide a latent heat forcing in the model thermodynamic equation and to force vertical circulations and the associated divergence that consistent with the observed precipitation. A case of 28 March 2000 Fort Worth/Texas tornado storm was chosen to explore options for applying the diabatic initialization technique, which essentially involves forcing the model over some time period with a heating field based upon the NIDS radar reflectivity, and along with the intermittent data assimilation technique.
Joint Poster Session 2, Poster Session - Mesoscale Data Assimilation—with Coffee Break
Tuesday, 31 July 2001, 2:30 PM-4:00 PM
Previous paper Next paper