Revisit of the Global Surface Energy Balance Using the MEP Model of Surface Heat Fluxes

Abstract:

The recently proposed model of surface heat fluxes, based on the theory of maximum entropy production (MEP), was used to estimate the global evapotranspiration (ET) and heat fluxes. Compared to bulk transfer models, the MEP model has several remote-sensing-friendly features including fewer input variables, automatic closure of surface energy budget, being independent of bulk gradients of temperature and water vapor, not using wind speed and surface roughness as model parameters, and being less sensitive to uncertainties of input variables and model parameters. The MEP model is formulated for the entire range of soil moisture from dryness to saturation over the land surfaces and has even more advantages over water-snow-ice surfaces compared to traditional methods due to its independence of surface humidity data. The MEP model provides the first global maps of water heat fluxes at ocean surfaces as well as conductive heat fluxes at snow/ice covered polar regions. Ten years of Clouds and the Earth's Radiant Energy System (CERES) earth surface radiation fluxes, surface temperature data products supplemented (when needed) by the Modern-Era Retrospective analysis for Research and Applications (MERRA) surface specific humidity data are used to test the MEP model by comparing the MEP based global annual ET and heat fluxes with existing products. The MEP based fluxes over land surfaces agree closely with previous studies. Over the oceans, the MEP modeled ET tends to be lower than previous estimates while those of sensible heat fluxes are in close agreement with previous studies. A counterpart, “off-line” analysis is also carried out using the NOAA GFDL climate model output from a control experiment and a “warming” experiment. Substantial differences in the warming-related changes of ET and Bowen ratio are found over regions such as North Africa and the southwestern U.S. The implications of these differences for understanding trends and variability in regional energy and water cycle will be discussed.