Regime-Dependent Differences in Surface Freshwater Exchange Estimates Over the Ocean

Abstract

Differences in gridded precipitation (P), surface evaporation (E), and the resultant surface freshwater exchange (P - E) among different products over the ocean are diagnosed as functions of moisture advection (Q(advt)) and moisture tendency by dynamical convergence (Q(cnvg)). Compared to the GPCP product, the TRMM3B42 product captures higher frequency of precipitation with larger extreme precipitation rates in regimes of deep convection and more light rain detections in regimes of frequent occurrence of boundary layer clouds. Discrepancies in E depend on moisture flux divergence, with the OAFlux product having the largest E in regimes of divergence. Discrepancies in mean P - E in deep convective regimes are highly influenced by differences in precipitation, with the TRMM3B42 product yielding P - E histograms closer to those inferred from the reanalysis moisture flux convergence. In nonconvergent regimes, observation-based P - E histograms skew toward positive values while the inferred reanalysis histograms are symmetric about the means. Plain Language Summary We quantify the relationship between the surface freshwater exchange and atmospheric water vapor transport and assess under what conditions current gridded products of precipitation and ocean surface evaporation have large differences. This article informs developers of these products where they should put more effort in the future to refine their products and users of these products what atmospheric conditions they would expect large discrepancies. The mean and extreme values as well as the discrepancies in precipitation and surface evaporation estimates are related to how the atmosphere transports water vapor. Since how the atmosphere transports water vapor is also related to large-scale dynamical conditions and cloud systems, the method used in this work further relates the discrepancies to large-scale environmental conditions. This is different from the traditional approach of reporting differences on geographical maps, in which each location may have gone through several scenarios of large-scale conditions in temporal averaging. This work has direct impact on scientists who perform estimations of precipitation and surface evaporation based on instruments on board of satellites. Since surface freshwater exchange is used as a boundary condition to force ocean models, this work reveals atmospheric conditions in which large differences in the boundary conditions exist.