On the Antarctic surface mass balance in atmospheric GCMS

Increased surface mass balance of the Antarctic ice sheet is expected to significantly moderate sea-level rise through the 21st century. However, the disagreements between atmospheric circulation models used to predict climate change is larger than the expected change itself: The mean present Antarctic precipitation minus evaporation (P-E) in 7 high-resolution Atmospheric General Circulation Models (AGCMs) ranges from 118 to 158 mm/year, or 4.0 to 5.4 mm/year in sea-level equivalent. In addition, the models display common systematic biases when compared with observation-based maps of the surface mass balance, thus suggesting common deficiencies, missing processes or inadequate boundary conditions (Genthon and Krinner, 2000, The Antarctic surface mass balance and systematic biases in GCMs, submitted to J. Geophys. Res.). Observation-based maps are necessarily largely interpolated. Some of the model biases are likely to actually be observation-based map biases due to poorly constrained interpolation across observation-void regions.

Antarctic climate errors in models may be due to atmospheric circulation deficiencies around (the circumpolar trough and its variability) or on (e.g. moisture diffusion in the boundary layer, transport by the catabatic winds) the ice sheet. Forcing the circumpolar circulation of an AGCM should, in principle, help sort out the former contribution and focus on the latter. Forcing may be done by nudging analyzed dynamic variables (wind, pressure) at the lateral boundary of the Antarctic region, away from the ice sheet, just like lateral boundary conditions are provided to a regional circulation model. The impact of such lateral nudging on the surface mass balance simulated by the LMDZ AGCM is high. The model is run with a high resolution «zoom» over the Antarctic ice sheet and forced by the European Center for Medium-range Weather Forecast 1979-1993 reanalyses (ERA). As expected, lateral forcing results in convergence of the model towards the ERA-based (short term forecasts) distribution of surface mass balance, compared to the unforced model. This confirms the importance of the circumpolar circulation on hydrology on the Antarctic ice sheet.