Prediction of Fort Worth Tornadic Thunderstorms using 3DVAR and Cloud analysis with WSR-88D Level-II Data

The impact of Level-II WSR-88D reflectivity and radial velocity data on the prediction of a cluster of tornadic thunderstorms are studied. Radar reflectivity data are used primarily in a cloud analysis procedure that retrieves the amount of hydrometeors and adjusts in-cloud temperature and moisture fields, while radial wind data are analyzed through a 3DVAR scheme that contains a mass continuity constraint in the cost function.

The 3DVAR system is briefly introduced, including aspects related to radial velocity analysis. The case studied is that of the March 28, 2000 Fort Worth tornado outbreak. The same case was studied by Xue et al. (2003) using the ARPS Data Analysis System (ADAS), WSR-88D Level-III (NIDS) data, and an earlier version of the cloud analysis scheme. Since then, several modifications in the cloud analysis procedure, including those to the in-cloud temperature adjustment and the retrieval of precipitation species, have been made and they improve the thunderstorm prediction in this case.

The Level-II data are used within assimilation cycles of 10-minute intervals for a one-hour period with the 3DVAR and cloud analysis. Three-hour forecasts are made starting from the assimilated initial condition. A 3-km resolution grid nested inside a 9-km grid is used for both assimilation and prediction.

It is found that best prediction is obtained when both reflectivity and radial velocity data are assimilated. The prediction is able to match up individual storm cells on the 3 km grid up to 2 hours into the prediction, and the supercell characteristics of the storm cell that spawned two individual tornadoes are well predicted, with timing errors of less than 15 minutes, and location errors of less than 10 km at the time of the tornadoes.

The forecast without using radar data completely failed to predict nearly all the storm cells within the 3 km domain. It is also found, with the current 3DVAR and cloud analysis procedure, reflectivity data has more positive impact on storm forecast than radial wind. The forecast using reflectivity but not radial velocity is able to reproduce most of the main characteristics of the observed storms, while the forecast storms with radial wind decay quickly. On the other hand, the use of radial wind along with reflectivity via cloud analysis does improve the storm forecast, especially in terms of the strong low-level vorticity centers associated with the tornadogenesis. Positive impacts of including mass continuity constraint in the 3DVAR cost function, and the modifications to the cloud analysis procedure will also be documented.