Bivariate Conditional Sampling of Moisture Flux over a Tropical Ocean

Abstract

New applications of conditional sampling using the bivariate joint frequency distribution (JFD) and conditional mean distribution (CMD) are introduced to analyze time series of water vapor flux obtained from aircraft gust-probe vertical velocity and water vapor data collected during free convective conditions (zi/L≈?46) in the Atlantic trade winds. The JFD can provide estimates of fractional time as well as mean and variance estimates for the four ways vertical velocity/moisture fluctuations can combine (up-moist, up-dry, down-moist, down-dry). These combinations were interpreted using a subcloud layer model consisting of upward and downward moving convective cells and a mixing zone between them. The data suggest detrainment from the up-moving cell becomes the predominant mixing process above 0.08zi. The fractional time and mean values were used to parameterize the moisture flux. Three parameterization methods are introduced; two estimated the moisture flux to within 10% while a third, simpler version gave estimates to within 13%, The CMD analysis subpartitioned the flux into direct and indirect components associated with up and downdrafts. Though the majority of direct flux was associated with updrafts, downward moving dry air was a substantial part of the direct flux even at 0.03zi. Given the differences in approach, comparison with other conditional sampling analyses was good. Using saturation point analysis, downward moving dry air sensed at 0.25zi was shown to originate above zi, while the upward moving moist air sampled at 0.25zi was shown to originate near the ocean surface.