Introduction

Greenland and Antarctica contain together almost all of the glacier ice on Earth, which if fully melted could add over 75 m to the level of the oceans. For time scales of a century or less, the response of the ice sheets to changes in climate is governed by surface processes, the accumulation of snow and meltwater formation, percolation, refreezing and runoff. Because of the very high surface reflectivity of snow and the dramatic changes in albedo which take place at the onset of melting, these regions can be expected to respond quite sensitively to changes in the climatic forcing, and to changes in temperature in particular. It is in fact in large part because of the snow/ice albedo feedback effect and changes in sea-ice extent that General Circulation Models predict that above average warming will take place in the Arctic and Antarctic regions (Houghton et al. (1996)).

The snow which accumulates on the Greenland and Antarctic ice sheets is for the most part gradually transformed into ice which then takes millenia to return to the ocean in the form of icebergs or meltwater. By providing this long term storage site for freshwater, the mass balance of Greenland and Antarctica also depends on the atmospheric circulation, and will be sensitive to changes in precipitation patterns.

Progress in estimating changes in accumulation and runoff has been slowed by two factors: the low resolution of most climate models which does not allow an accurate representation of the climatology over the ice sheets, and the lack of reliability in the calculation of meltwater formation and runoff. Both issues are addressed in this paper. A hierarchy of models which will be used to estimate the amount of melting and runoff from meteorological parameters is presented in section 2, with particular focus on a model of the snow cover developed for that purpose. The climate models and climate change scenarios used as forcing for the melt models are described in section 3. By comparing in section 4 the results obtained with different combinations of climate and melt models for the current climate and at the time of CO₂ doubling, the objective was to assess the reliability of the projected changes in mass balance and sea-level obtained with transient climate change scenarios. These results are discussed in section 5. The impact of the Kyoto protocol in reducing changes in sea-level is discussed in section 6.