| description abstract | Upper air and surface data from the GARP Atlantic Tropical Experiment (GATE) are used to examine the interrelationships between convective-scale precipitation and the larger scale wind field. The upper air winds from the inner (B) and outer (A/B) hexagonal observational arrays are fit with second-order polynomials to provide smooth estimates of the vorticity, divergence and vertical motion in the observational array. In these analyses we examined archived validated data from all three phases of the experiment and we formed averages based on the radar-estimated precipitation rates. Mean profiles for 19-day periods during each of the three observational phases establish the basic similarity of the kinematics during each phase. Strong boundary-layer convergence balanced, for the most part, by upper tropospheric divergence, is common to all three phases. Radar-estimated precipitation rates are used to define suppressed (precipitation rates <0.1 mm h?1) and highly disturbed (precipitation rates >0.5 mm h?1) states over the observational array. Mean profiles for the disturbed states in each phase show weaker easterly winds and much larger upward vertical velocities than do the mean profiles for the suppressed states. The mean vorticity profiles for each state do not show such clear-cut differences. Time series of 12 h averages indicate that the precipitation events in Phase III corresponded very closely to the cyclonic maxima of the 700 mb relative vorticity, reflecting the influence of the easterly waves described by Reed et al. (1977). During Phases I and II, when easterly waves were poorly organized, the precipitation events did not correspond closely to the cyclonic vorticity maxima. On the other hand, precipitation events showed good correspondence with the large-scale (A/B) 700 mb upward vertical velocity maxima and surface meridional convergence ?v/?y during all three phases. This shows that the precipitation is clearly related to events on a larger scale. The effects of convective activity on the large-scale flow are examined through the vorticity budget. The vorticity budget residual profiles were similar from phase to phase with cyclonic production maxima in the mid and upper troposphere. The upper tropospheric residual maximum is as strong during the suppressed state as it is during the highly disturbed side. At the surface, individual values of the residual are almost always opposite in sign to the vorticity. The mean vorticity budget for the A/B array shows the tipping term to have magnitudes comparable to other terms in the vorticity budget. | |
