Precipitation Structure of the Cloud Clusters in a Tropical Easterly Wave

Abstract

This paper describes the statistics of precipitation observed over the 42.5 ? 104 km2 area covered by all four of the GATE digitized radars fair the period 2300 GMT 4 September 1974 to 2400 GMT 5 September 1974. The period coincided with the GATE double cloud cluster of 5 September, the latter stages in the life cycle of the squall line system of 4?5 September and the passage of the trough of an easterly wave in the midtropospheric flow. Hourly precipitation amounts over 4 km ? 4 km data bins were stratified by time, data source, individual mesoscale precipitation feature (1 of 12), type of precipitation (convective or stratiform) and, if convective, cell top height. Each subtotal contained a total precipitation amount and the number of 4 km ? 4 km bins which contributed to the total. From these subtotals a variety of precipitation statistics were compiled. The precipitation statistics of the 5 September double cloud cluster confirm the importance of deep convection in producing observed tropical mesoscale precipitation patterns. They also underscore the importance of stratiform precipitation in non-squall cloud clusters as well as in squall line systems. The evolution of the total, convective and stratiform amounts of rainfall in the 5 September double cloud cluster is similar to that of the mesoscale precipitation features of which it is composed and also to the evolution of squall line systems consisting of a single intense mesoscale precipitation feature. The time lag in the development of stratiform precipitation suggests that storage of water condensed in convective cells after transfer to the anvil may be an important component of the water budget of a mesoscale precipitation feature. When precipitation from the 4?5 September squall line system and the 5 September double cloud cluster are combined to show the evolution of precipitation over the large mesoscale in and near the trough of the easterly wave, the total rainfall is nearly constant for over 24 hours. This result is consistent with the statistical quasi-equilibrium assumption Arakawa and Schubert used to provide a closed parameterization of cumulus convection for use in prognostic models of large-scale atmospheric motion.