Report: ACCESS Ice-sheet Model (ISM) Selection

ACCESS-NRI has adopted the Ice-sheet and Sea-level System Model (ISSM) as its support Ice Sheet model after an open and rigorous community-led selection process.

The Cryosphere Modelling Working Group (CMWG) was established in February 2023, and one of its first tasks was to establish a process for selecting an Ice Sheet Model (ISM) to be included in the ACCESS framework, with the ultimate goal of coupling ice sheet processes into the ACCESS Earth system model. ACCESS-NRI received an additional funding under the 2023 NCRIS Funding round to support ice sheet and coastal ocean modelling as part of an NCRIS-wide Coastal Research Infrastructure (CoastRI) Initiative.

Ice sheets are large masses of ice grounded on land that grow due to the accumulation and compaction of snow into ice and shrink primarily due to ocean and atmosphere melting and iceberg calving. They are different from sea ice, which is frozen ocean water. We currently have two ice sheets on Earth – Antarctica and Greenland – but have had other ice sheets in the Northern Hemisphere during past glacial periods. Antarctica and Greenland play a fundamental role in regulating the amount of incoming solar radiation due to their high albedo, in ocean and atmosphere circulation, and are the largest drivers of global sea level change on centennial and longer timescales.

The primary objective of ACCESS-NRI’s 2023 funding is to couple an ISM of Antarctica and Greenland into ACCESS in order to improve decadal- to centennial-scale projections of global mean and regional sea level and climate change.

The Ice-sheet model selection process

Over the past year, the CMWG developed an ice sheet selection approach designed to meet this objective that was completed over three phases:

  1. Community Consultation
  2. Assessment
  3. Model Selection

The outcome of the first two phases was a shortlist of three candidate ISMs that were deemed suitable for coupling with the ACCESS modelling framework to meet the primary objective. Those models were: Elmer/Ice, IcePack, and the Ice-sheet and Sea-level System Model (ISSM).

The third and final phase, the Model Selection, included an expert panel review of the consultation and assessment materials developed in the first two phases. The role of the expert panel was to assess each of the three shortlisted candidate ISMs against the primary objective and make a formal recommendation on which ice sheet model to incorporate into the ACCESS framework. The panel convened on May 13, 2024 and consisted of 8 members from the cryosphere and ACCESS modelling communities. It included the ACCESS-NRI Director (panel chair), CMWG co-chairs (3), a CMWG early career representative (1), and experts in ice sheet (1) and Earth system modelling (2).

Outcome

The panel considered all the information available on the capability of each model, the input from community assessment and the risk profile of each model. At the conclusion of this deliberation it was clear that IcePack, while having some advantages, had less capacity and held greater risks than Elmer/Ice and ISSM. In most respects, Elmer/Ice and ISSM had similar capacity, similar existing Australian capability and an identical risk profile.

The committee determined that ISSM has critical advantages over Elmer/Ice in modelling surface hydrology, forward and inverse capabilities for sliding laws, and data assimilation/adjoint capabilities. ISSM also has minor advantages in Holocene scale simulations capability, sea level coupling, and the size of the international developer community, while Elmer/Ice has minor advantages in fracture capabilities, rheology and existing coupling work in Australia.  For these reasons, the committee recommended that ACCESS-NRI adopts ISSM as its supported Ice Sheet Model. This recommendation was approved by the ACCESS-NRI Advisory Board in June 2024.  

The full Report provides all the information relevant to assessing the shortlisted ISMs against the primary objective that was considered by the expert panel.

Acknowledgements

A large number of people put an enormous amount of time into supporting and developing this process. We particularly wish to thank all the members of the Cryosphere Modelling Working Group and the time that they all freely gave to collating information on all the ice sheet models we reviewed, and providing input at each step of the process. We also wish to thank the ice sheet model developers who provided feedback – especially where we asked detailed, technical questions that took a while to answer! Every one of the developers was incredibly generous and it is a privilege to be able to work with them as part of the international ice sheet modelling community.

While we ended up deciding on only one ice sheet model to support, the ice sheet models that we reviewed are all excellent models. We learnt a lot about each of them and are looking forward to seeing – and participating in – the developments that occur in this discipline over the coming decades.

Download the full report