The mass balance correction reduces the moisture transport estimation dramatically compared to the previous Slonaker and Van Woert results using the same dataset. In Slonaker and Van Woert's study, the zonally-averaged, vertically-integrated moisture flux across 50S is -31.8 kg m-1 s-1; while the value of this study is only -14.1 kg m-1 s-1. The net precipitation estimation also substantially differs between the two studies. Between the 50S-60S latitude belt, the zonally averaged net precipitation from Slonaker and Van Woert is 636 mm yr-1, while in this study the estimation is 233 mm yr-1.
The reason for the difference is that the wind derivation approach without conservation of mass used by Slonaker and Van Woert leads to unrealistically strong annual mean wind speeds in the lower troposphere. Their strong mean wind estimates resulted in exaggerated mean moisture transport compared to other studies. The conservation of mass corrects the problem and produces mean moisture flux consisted with historical estimates.
The eddy moisture flux of the satellite derived in both Slonaker and Van Woert's study and the current study is underestimated by about 45% compared to radiosonde and analyses studies. This underestimation is most likely due to the lower spatial and temporal resolutions of the satellite observations.
The detailed characteristics of moisture flux across different sectors of the Southern Oceans and comparisons with analyses and radiosonde data will also be discussed.
Topic Area: Climatology and long-term satellite data studies