Uncertainties in estimating moisture fluxes over the Intra-Americas Sea

Abstract

This study estimates discrepancies in moisture flux divergence in the Intra-Americas Sea (IAS; including the Gulf of Mexico and the Caribbean Sea) calculated using sounding observations, the NCEP Eta high-resolution regional analysis, and the NCEP-NCAR coarse-resolution global reanalysis. The main purpose of this exercise is to quantify the uncertainties in the global reanalysis when it is used to calculate annual and interannual variability of moisture flux divergence in the region. An accurate estimate of moisture flux divergence is crucial to evaluate whether the IAS serves as a water vapor source for rainfall over the adjacent land. Using the three datasets, the uncertainties of calculated moisture flux divergence due to the design of the boundary of the area, mathematical algorithms, and spatial and temporal resolutions are quantified. The results show that the large seasonal and interannual variability in moisture flux divergence estimated using the NCEP-NCAR reanalysis is not compromised by these uncertainties. Therefore, NCEP-NCAR reanalysis, with its global coverage and long-term record, can be used to provide the best estimate of short climate variability of moisture flux divergence available to date. Further comparisons are made of the moisture flux divergence based on the NCEP-NCAR reanalysis with previous estimates using single-year sounding observations, as well as with multiyear estimates based on global datasets of surface evaporation and precipitation. It is shown that the previous estimates using single-year sounding observations bear large uncertainties because of interannual variability. Large uncertainties also exist in datasets of surface global evaporation and precipitation. © 2005 American Meteorological Society.