29 August 2013

Dynamic Ice Sheets in the Community Earth System Model

Science

Climate scientists at Los Alamos National Laboratory and the National Center for Atmospheric Research have coupled the Glimmer Community Ice Sheet Model (Glimmer-CISM) to the Community Earth System Model (CESM). CESM is a complex global climate model used for many national and international assessments; it is the first such model to include a dynamic ice sheet model in a public code release. This paper describes and evaluates the initial model implementation for the Greenland Ice Sheet. The ice-sheet surface mass balance (SMB, the total accumulation minus the total ablation) is computed in multiple elevation classes in CESM's land model and downscaled to the ice sheet grid. As shown in a companion paper by Vizcaíno et al. (2013, Journal of Climate, in press), CESM is the first global climate model to simulate a realistic SMB for Greenland. 

Using this SMB to force the dynamic ice sheet model, the investigators ran a 100-member ensemble of Greenland Ice Sheet simulations, each with the same preindustrial (c. 1850) forcing but with different values of three key physics parameters. Ensemble members were ranked by comparing the simulated steady-state ice sheet geometry to the observed geometry. With reasonable parameter choices, the simulated geometry is broadly consistent with observations, but the modeled ice sheet is too thick and extensive in some coastal regions where the SMB has a positive bias. (This bias will be investigated in future work.) The top-ranking simulations were run forward using forcing from CESM simulations of the 20th and 21st  century. In these simulations the ice sheet loses mass, with resulting global mean sea-level rise of 16 mm during 1850-2005 and 60 mm during 2005-2100. This mass loss is caused mainly by increased melting near the ice sheet margin, partly offset by reduced ice flow to the ocean. Future work will focus on two-way coupling, with changes in ice sheet geometry feeding back on CESM’s land and atmosphere models.

Summary

Climate scientists at Los Alamos National Laboratory and the National Center for Atmospheric Research have coupled the Glimmer Community Ice Sheet Model (Glimmer-CISM) to the Community Earth System Model (CESM). CESM is a complex global climate model used for many national and international assessments; it is the first such model to include a dynamic ice sheet model in a public code release. This paper describes and evaluates the initial model implementation for the Greenland Ice Sheet. The ice-sheet surface mass balance (SMB, the total accumulation minus the total ablation) is computed in multiple elevation classes in CESM's land model and downscaled to the ice sheet grid. As shown in a companion paper by Vizcaíno et al. (2013, Journal of Climate, in press), CESM is the first global climate model to simulate a realistic SMB for Greenland. Using this SMB to force the dynamic ice sheet model, the investigators ran a 100-member ensemble of Greenland Ice Sheet simulations, each with the same preindustrial (c. 1850) forcing but with different values of three key physics parameters. Ensemble members were ranked by comparing the simulated steady-state ice sheet geometry to the observed geometry. With reasonable parameter choices, the simulated geometry is broadly consistent with observations, but the modeled ice sheet is too thick and extensive in some coastal regions where the SMB has a positive bias. (This bias will be investigated in future work.) The top-ranking simulations were run forward using forcing from CESM simulations of the 20th and 21st  century. In these simulations the ice sheet loses mass, with resulting global mean sea-level rise of 16 mm during 1850-2005 and 60 mm during 2005-2100. This mass loss is caused mainly by increased melting near the ice sheet margin, partly offset by reduced ice flow to the ocean. Future work will focus on two-way coupling, with changes in ice sheet geometry feeding back on CESM’s land and atmosphere models.

Contact
Acknowledgments

This work was supported by the Earth System Modeling program of the Office of Biological and Environmental Research within the US Department of Energy’s Office of Science.