Antarctic meltwater alters future projections of climate and sea level

Abstract

Imperfect understanding of ice sheet-climate interactions poses challenges for projecting the impacts of ice sheet mass loss on future climate and sea level. Here we couple a dynamic Antarctic ice sheet model and global climate model to simulate ice sheet-climate interactions. In our single-model, single-member modeling framework, we find sea level and climate projections are significantly modified from uncoupled simulations neglecting Antarctic meltwater under RCP8.5 and RCP4.5. Antarctic meltwater yields surface air temperatures up to 1.5 °C higher in parts of the Northern Hemisphere, while broadly dampening temperature rise in the Southern Hemisphere. Due to radiative feedback changes, both emissions scenarios have global mean surface temperature warming ~0.3 °C lower in the coupled scenario than the control by 2100, with a maximum anomaly of ~1 °C at 2200 under RCP8.5. This slows Antarctica's contribution to global mean sea level rise. Total Antarctic sea level contributions under RCP8.5 (2100: ~0.3 m, 2200: >3 m) include substantial contributions from East Antarctica, though not under RCP4.5 (2100: ~0.1 m, 2200: >1 m). Regionally, projected sea level is up to 0.9 m higher in the Pacific than the global mean Antarctic contribution under RCP8.5 at 2200.

Fig. 1. Ice sheet response and Antarctic sea level rise contribution.

a Change in Antarctic ice sheet thickness from 2005 to 2100 under RCP4.5. The ice sheet outline is the modeled grounding line position. b Same as a) but for 2200; c) Same as a) but for RCP8.5; d same as (b) but for RCP8.5. e Global mean sea level rise contributions from Antarctica, and mass loss in gigatons through 2100, for the West Antarctic Ice Sheet (WAIS), East Antarctic Ice Sheet (EAIS), and total Antarctic contribution; f same as (e) through 2200. Maps created with Cartopy.

Fig. 2. Southern Ocean climate change by 2100 for RCP8.5.

a Southern Ocean surface air temperature (SAT) change in the coupled simulation during the 21^st century, b The SAT difference between the coupled and control simulations averaged over the last 30 years of the 21^st century, c same as (a) but for precipitation, d same as (b) but for precipitation, e same as (a) but for 400 m ocean temperatures, f same as (b) but for 400 m ocean temperatures. Maps created with Cartopy with colormaps from.

Fig. 3. Effect of Antarctic meltwater input on future climate.

a The difference between the 2 m surface air temperature (SAT) in the RCP8.5 coupled experiment versus the control by the end of the century (2070-2099). b Same as (a) for RCP4.5. c Global mean surface temperature (GMST) above the 1850–1900 average with horizontal lines marking 1.5°, 2°, and 3 °C. d A time series of the temperature difference between the coupled and control simulations. e Difference in precipitation between the RCP8.5 coupled and control simulations at the end of the century. f Same as (e) for RCP4.5. g Zonal mean precipitation under the control simulations. h Same as (g) for the difference between the coupled and control simulations. Maps created with Cartopy with colormaps from.

Fig. 4. Change in overturning circulation strength and northward heat transport due to coupling.

a Meridional overturning circulation (MOC) strength in the Atlantic Ocean for the coupled and control simulations under RCP8.5 and RCP4.5, b northward heat transport in the Atlantic Ocean averaged over the period 2070-2099.

Fig. 5. Spatial change in sea level resulting from projected Antarctic ice sheet mass loss.

a The spatial pattern of sea level change resulting from gravitational, rotational, and Earth deformational effects at 2100 under RCP4.5. The purple band designates the global mean sea level contribution from the Antarctic ice sheet. b same as (a) for RCP8.5. c same as (a) for 2200. d same as (b) for 2200. Maps created with ArcGIS using Natural Earth land boundaries.