Dawei Li, tenure track associate professor at the School of Oceanography, Shanghai Jiao Tong University, has recently published a research article in Science Advances as the first and corresponding author. The work, entitled “Climate model differences contribute deep uncertainty in future Antarctic ice loss”, was done at Shanghai Jiao Tong University in collaboration with the University of Massachusetts-Amherst and the Pennsylvania State University.
The Antarctic Ice Sheet (AIS) has an ice volume equivalent to 58m global-mean sea level (GMSL) rise. The West Antarctic Ice Sheet portion (5.3m sea level equivalent), sitting on the sea floor, is subject to marine ice sheet instabilities and displays a remarkable sensitivity to climate change. As the climate warming due to anthropogenic greenhouse gas emissions, the response of the AIS and the consequence on global mean sea level rise are subject to substantial uncertainty, posing a great challenge to future projection of sea level rise and enaction of mitigation policies for coastal regions. Part of this uncertainty is attributable to poorly understood physical processes represented in ice sheet models and their parameterization schemes. On the other hand, uncertainty in future climate boundary conditions fed to ice sheet models adds another layer. Climate models display a substantial spread for projected changes in global-mean surface air temperature (GMSAT), even with identical future greenhouse gas concentrations. The “equilibrium climate sensitivity”, essentially the change in GMSAT as the climate stabilizes after doubling CO2, ranges from 1.8°C to 5.6°C among the current generation CMIP6 climate models. Existing ice sheet intercomparison projects are mainly focused on ice sheet processes, but have paid little attention to the uncertainty in future ice sheet evolution originating from climate models.
Fig. 1 Simulated equilibrium thickness of the Antarctic Ice Sheet under preindustrial climate conditions from 36 CMIP6 models
For a systematic evaluation of the uncertainty due to climate model fields used as input to the ice sheet models, Li and coauthors drove a 3-D model of the AIS with the output from 36 climate models to simulate past and future changes in the AIS. Simulations show that a few climate models result in partial collapse of the West Antarctic Ice Sheet under modeled preindustrial climates, implying that biases in simulated polar climate from state-of-the-art climate models are large enough to drive the AIS to equilibrium states distinctly different from the present-day, although the ISM simulates a realistic AIS with observational climate data. Under the SSP5-8.5 future scenario, with simulated climate fields bias-corrected against present-day observational data, the spread in simulated future changes in the AIS’s volume is comparable to the structural uncertainty originating from differing ice sheet models. These results highlight the need for improved representations of physical processes important for polar climate in climate models.
Fig. 2 Simulated changes in the Antarctic Ice Sheet’s area and volume in the SSP5-8.5 scenario.
Climate modeling has now entered the “Earth system model” phase, where the most sophisticated models have added biogeochemical cycles and land ice sheets to the atmosphere-land-ocean system. Results from this study, however, warn of substantial uncertainty among Earth system models with interactive ice sheets for the evaluation of future sea level rise. While progress has been made in ice sheet modeling, the uncertainty in future changes of the AIS and associated impacts on GMSL have not been reduced to a level suitable for straightforward decision making, and more work is required. Current greenhouse gas emissions put the climate on track of a 3°C warming by 2100, and the time window is shrinking for reducing carbon emissions to avoid rapid and unstoppable sea level rise. For more robust sea level projections, improved understanding of processes important for polar climate, including cloud radiative forcing and deep ocean circulations and mixing, is urgently needed.
Article information and related links
Li, D., DeConto, R. M., Pollard, D., Climate model differences contribute deep uncertainty in future Antarctic ice loss. Science Advances, 9, 7, 2023.
https://www.science.org/doi/10.1126/sciadv.add7082
