Changes in Antarctic surface conditions and potential for ice shelf hydrofracturing from 1850 to 2200

Abstract. A mixed statistical–physical approach is used to emulate the spatio-temporal variability of the Antarctic Ice Sheet surface mass balance and surface melt rates of a regional climate model. We demonstrate the ability of this simple method to extend existing regional climate simulations to periods, scenarios, or climate models that were not originally simulated. This method is useful to quickly populate ensembles of surface mass balance and melt rates, which are needed to constrain ice sheet model ensembles. Here we apply this method to estimate (i) the changes in Antarctic surface mass balance over 1850–2200 and the associated effect on sea level and (ii) the changes in potential for ice shelf hydrofracturing. After weighting 16 climate models to obtain a realistic distribution of the equilibrium climate sensitivity, we find a likely contribution of surface mass balance to sea level rise of −2.2 to −0.4 cm from 1900 to 2010 and −3.4 to −0.1 cm from 2000 to 2099 under the SSP1-2.6 scenario versus −4.4 to −1.4 cm under SSP2-4.5, and −7.8 to −4.0 cm under SSP5-8.5. The contribution from 2000 to 2200 is highly uncertain: between −10 and −1 cm in SSP1-2.6 and between −33 and +6 cm in SSP5-8.5 depending on the climate model. Based on a criterion on the presence of liquid water beyond firn saturation in our emulated ensemble, we estimate the surface conditions that make ice shelves prone to hydrofracturing. Our results suggest that a majority of Antarctic ice shelves could remain safe from hydrofracturing under the SSP1-2.6 scenario, but all of them could become prone to hydrofracturing before 2150 under the SSP5-8.5 scenario.