VERTICAL VELOCITY IN OCEANIC CONVECTION OFF TROPICAL AUSTRALIA

Time series of 1-Hz vertical velocity data collected during aircraft p enetrations of oceanic cumulonimbus clouds over the western Pacific wa rm pool as part of the Equatorial Mesoscale Experiment (EMEX) are anal yzed for updraft and downdraft events called cores. An updraft core is defined as occurring whenever the vertical velocity exceeds 1 m s(-1) for at least 500 m. A downdraft core is defined analogously. Over 19 000 km of straight and level flight legs are used in the analysis. Fiv e hundred eleven updraft cores and 253 downdraft cores are included in the dataset. Core properties are summarized as distributions of avera ge and maximum vertical velocity, diameter, and mass flux in four alti tude intervals between 0.2 and 5.8 km. Distributions are approximately lognormal at all levels. Examination of the variation of the statisti cs with height suggests a maximum in vertical velocity between 2 and 3 km; slightly lower or equal vertical velocity is indicated at 5 km. N ear the freezing level, virtual temperature deviations are found to be slightly positive for both updraft and downdraft cores. The excess in updraft cores is much smaller than that predicted by parcel theory. C omparisons with other studies that use the same analysis technique rev eal that EMEX cores have approximately the same strength as cores of o ther oceanic areas, despite warmer sea surface temperatures. Diameter and mass flux are greater than those in GATE but smaller than those in hurricane rainbands. Oceanic cores are much weaker and appear to be s lightly smaller than those observed over land during the Thunderstorm Project. The markedly weaker oceanic vertical velocities below 5.8 km (compared to the continental cores) cannot be attributed to smaller to tal convective available potential energy or to very high water loadin g. Rather, the authors suggest that water loading, although less than adiabatic, is more effective in reducing buoyancy of oceanic cores bec ause of the smaller potential buoyancy below 5.8 km. Entrainment appea rs to be more effective in reducing buoyancy to well below adiabatic v alues in oceanic cores, a result consistent with the smaller oceanic c ore diameters in the lower cloud layer. It is speculated further that core diameters are related to boundary layer depth, which is clearly s maller over the oceans.