Integration of WSR-88D and GOES-8 data for detecting and forecasting storm initiation

Abstract

Several studies utilizing research radar have documented the importance of surface conver­gence boundaries in initiation of convection. Detection and tracking of these boundaries provides valuable information for forecasting thunderstorm occurrence. The presence of cumulus clouds above these boundaries is believed to increase the probability that thunderstorm. development could occur. With the advent of the GOES-8 satellite 15 min update rate and higher spatial resolution, it is now possible to detect and track cumulus cloud growth more closely. The focus of this thesis is to show how operational WSR-88O radar and GOES-8 satellite data can be used together to provide information on the growth of cumulus clouds into thunderstorms, based on radar­detected, boundary layer covergence features, cloud-top brightness temperatures, and change in reflectivity echo intensity with time. The evolution of cumulus clouds over a spectrum of radar-detected, boundary layer features have been examined. While thunderstorms did form above horizontal convective rolls in the absence of any additional surface forcing, the most intense storms initiated above gust fronts, gust front interaction with horizontal rolls, and above stationary convergence zones. In all cases pre­sented, cloud growth was evident in the satellite data in the form of decreasing cloud top tempera­tures, prior to the first detection of 10-20 dBZ radar echoes aloft. Sub-freezing cloud top temperatures were found to be precursors to the start of precipitation phase in clouds, with cloud top temperatures below 0° C occurring 15 min prior to the detection of 20 dBZ echoes, and 30 min prior to the detection of 30 dBZ echoes. The rate of cloud top temperature change was found to be important for discriminating between slow and rapid storm growth. Results of this thesis show that both satellite and radar data can be exploited to provide up to 30 min additional lead time in issuing forecasts of thunderstorm initiation, by monitoring the data in an automated way.