Abstract
Diagnostics of energy and moisture transport and disposal over the Antarctic polar cap (70°S to the pole) and ice sheet are extracted from the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis archive over the 1979–93 period. Transport across 70°S is obtained from the 6-hourly analyses of wind, temperature, moisture, and geopotential, whereas top-of-the-atmosphere energy balance and surface energy and water fluxes are evaluated from 6- and 12-h forecasts. A full decomposition of transport is made and tabulated in terms of seasons, dynamic components (mean meridional, stationary eddy, transient eddy), and type of energy (sensible, latent, geopotential). For instance, in terms of type of energy, about 50% of the total converged to the polar cap is geopotential, which is almost entirely advected by the mean meridional circulation. Even though atmospheric moisture is very low, latent heat transport accounts for almost 20% of the total energy import, mostly by the transient eddies. In terms of dynamic components, transient eddies alone import about 50% of the total energy in the form of sensible and latent heat. Some components actually export energy from the polar cap, and the variety of signatures exhibited by the transport decomposition may prove useful to check the dynamics of climate models in the very high southern latitudes. According to the analyses, the total annual mean energy input to the polar cap south of 70°S by the atmospheric circulation is 80.8 W m−2 of horizontal surface. The short-term forecasts suggest that the oceanic import is much smaller, of the order of model and analysis uncertainties. The interannual variability of atmospheric energy convergence is unreasonably large, and it is partly, yet not quite convincingly, correlated with the El Niño–Southern Oscillation. No convincing correlation is found either between moisture convergence from analyses or surface water budget from forecasts and the El Niño–Southern Oscillation. This result contradicts a previous study using the ECMWF operational analyses, which are more prone to spurious variability than the reanalyses and associated forecasts used here. The interannual variability of moisture convergence is large but reasonable, about 25% of the annual mean. It might be useful as a control against which to check the dynamics of the hydrological cycle of climate models in the high southern latitudes.
Corresponding author address: Dr. Christophe Genthon, LGGE/CNRS, 54 rue Moliere, DU BP 96, 38402 Saint Martin d’Heres, Cedex, France.
