An important factor in projections of the Antarctic contribution to future sea level rise is the Southern Ocean. The Ice Sheet Model Intercomparison Project for CMIP7 aims to provide ice sheet projections, and here we introduce the protocol for handling the ocean in Antarctica. It describes how ocean changes from global atmosphere-ocean models can be used in Antarctic simulations and how ocean conditions can be translated into melt rates, taking into account oceanographic knowledge. 

Strong ocean-driven melting can create gaps in floating ice shelves. Because the pathways of meltwater beneath these gaps are poorly understood, we tested two contrasting scenarios: one in which meltwater loses heat and momentum within the gaps, and another in which it continues to flow across them. These different representations substantially alter projected ice mass loss, highlighting meltwater treatment across ice-shelf gaps as a key source of uncertainty.

As Antarctica loses more ice, increasing amounts of fresh water enter the ocean and can change Earth's climate. We used a climate model in which Antarctic ice loss responds to changing ocean conditions. We found that this feedback first increases warming near the ice shelves but later slows both Antarctic and global warming by changing the ocean, sea ice, and atmosphere. These results show that including these interactions is important for improving long-term sea-level and climate projections.

Computer models of the Antarctic ice sheet can project how much it will melt in response to climate change. That depends on the properties of the rock and the water beneath the ice, about which we know very little. That means the amount of sea-level in the future is very uncertain. Techniques to reduce that uncertainty do not always work as well as expected. We must get actual measurements of the bottom of the ice to improve our understanding of the ice sheet, and our knowledge of its future.

The future contribution of Antarctica to sea-level rise is strongly dependent on the melting of floating ice shelves by the underlying ocean. Here, we present version 2 of the two-dimensional ocean model LADDIE. We evaluate the model by comparing it to three-dimensional ocean models and satellite observations, showing good performance at low computational cost. With this open-source model, we hope to contribute to the evolution toward more realistic melting in ice sheet model simulations.