Abstract

The Earth’s ice sheets and glaciers are retreating and raising global sea level. Century-scale sea-level projections are uncertain and controversial, with discrepancies between semi-empirical approaches and physically based models. We are beginning to use a global climate model, the Community Earth System Model (CESM), to simulate land-ice retreat and sea-level rise. Version 1 of CESM includes a dynamic ice sheet model, the Glimmer Community Ice Sheet Model (Glimmer-CISM). The ice sheet model is forced by a surface mass balance (SMB) computed in multiple elevation classes in CESM ’s land model and downscaled to the ice sheet grid. For most of the Greenland ice sheet, CESM generates an SMB in good agreement with observations and regional climate models. Members of the CESM Land Ice Working Group are now working to (1) improve Glimmer-CISM by including higher-order ice-flow dynamics and more realistic physical parameterizations, (2) simulate the evolution of small glaciers and ice caps, (3) implement two-way coupling between the ice sheet and land models, with land surface types that change over time, and (4) incorporate ice-sheet/ocean coupling beneath ice shelves.

In the second half of the talk we focus on this last area of research. Fully coupled ice-sheet/ocean interactions involve dynamic boundaries between model components, a prospect that was not incorporated in CESM ’s design (or that of other Earth System Models). We are making rapid progress in bridging this gap. Recently, we have added the ability to simulate static sub-ice-shelf cavities into CESM ’s ocean model, the Parallel Ocean Program (POP). We are now in the process of validating the model under present-day conditions. We are also performing a series of idealized ice-sheet/ocean simulations that are coupled offline (asynchronously)—each model is run for a coupling time step based on boundary conditions from the other model’s previous time step. Building on our experience with offline coupling, we are in the early stages of implementing dynamic boundaries in both POP and the CESM coupler.