The Role of Boundary-Layer Convergence Zones and Horizontal Rolls in the Initiation of Thunderstorms: A Case Study

The initiation of thunderstorms is examined through a combined observational and modeling case study. The study is based on Doppler radar, aircraft, mesonet, balloon sounding, and profiler and photographic data from the Convection Initiation and Downburst Experiment (CINDE) conducted near Denver, Colorado. The study examines the initiation of a line of thunderstorms that developed along a preexisting, quasi-stationary boundary-layer convergence line on 17 July 1987. The storms were triggered at the intersection of the convergence line with horizontal rolls where enhanced updrafts were present. The primary effect of the convergence line was to deepen the moist layer locally and provide a region potentially favorable to deep convection. The critical factor governing the time of storm development was apparently related to the attainment of a balance between horizontal vorticity in the opposing flows on either side of the convergence line. The effect was to cause the updrafts in the convergence line to become more erect and the convergence zone deeper, as discussed theoretically by Rotunno et al. Modeling results for this case also indicated that storm initiation was very sensitive to the depth of the convergence-line circulation. Storm initiation also frequently coincided with the location of misocyclones along the convergence line. Model results suggested this was because both events were caused by strong updrafts. The misocyclones resulted from stretching of existing vorticity associated with the convergence line. They tended to form where a convective roll intersected the convergence line leading to a local maximum in convergence and vertical motion. Some misocyclones suddenly deepened and strengthened when they became collocated with the deep, intense updraft of a convective storm. The updraft was responsible for advection and stretching of the vertical component of vorticity, leading in the most intense cases to the development of nonsupercell tornadoes, as discussed previously by Wakimoto and Wilson.