Comprehensive Earth system models (ESMs) and climate models are the main tools available for quantitative projections of future climate change and likely physical outcomes. However, diagnosing Southern Hemisphere model performance is difficult because of the spatial sparseness of field data and remaining uncertainties in reconstructions of recent real-world climate conditions. These factors limit the evaluation of ESMs and thus the reliability of their projections, especially at high spatial resolution.
To address this need, scientists from more than 17 countries, including 29 early-career scientists, gathered last October at the Scripps Institution of Oceanography for the #GreatAntarcticClimateHack, a workshop funded by the Scientific Committee on Antarctic Research’s Antarctic Climate Change in the 21st Century (AntClim21) initiative. Attendees’ intent was to decide on metrics to evaluate ESMs to improve the next generation of Intergovernmental Panel on Climate Change projections for Antarctica and the Southern Ocean. Participants included leading experts in oceanography, glaciology, atmospheric research, aquatic biogeochemistry, and biology working on past reconstructions, modern observations, and future projections.
The principal outcome of the workshop is a community agreement on an ensemble of metrics. These metrics were produced and prioritized using a bottom-up approach that allowed contributors from different disciplines to identify key aspects of model evaluation that are most important for their area of science. At the workshop, in-depth sessions were conducted on the atmosphere, ocean, sea ice, ice sheets, paleoreconstructions, ecosystems, and biogeochemistry. Discussions finalizing diagnostic tools for implementing the range of metrics are ongoing.
Key multidisciplinary insights that emerged from the workshop include the following:
- Ocean subpolar gyres around Antarctica influence key aspects of coupled systems. For example, these gyres transport water masses to the Antarctic coastline, where they can interact with the ice sheets. The gyres are also critical for the dispersal of nutrients, the transport of sea ice, and the vertical mixing of water masses. Penguins that rely on these gyres for their seasonal migration will help researchers evaluate the representation of such gyres because it is now possible to outfit the penguins with global location–sensing (GLS) biologgers, which use solar cues to determine location.
- Ice-ocean interactions at the grounding line (where a glacier on land extends into the water, becoming a floating ice shelf) are a principal challenge for determining ice mass loss from marine-based ice sheets. Emerging integration of grounding line behavior in coupled ice sheet models and the recent successes in drilling projects to access water masses near the grounding line provide unprecedented opportunities to assess model performance and refine Antarctic contributions to sea level rise.
- Sea ice connects many disciplines through its interaction with the atmosphere (winds), ocean (temperature and currents), and ecosystems (nutrients and light). Including new data metrics, such as those from GLS tags attached to penguins and the deployment of under-ice Argo floats, provides exciting new constraints to improve model performance.
In the upcoming World Climate Research Programme’s Climate Model Intercomparison Project Phase 6 (CMIP6), routine benchmarking and evaluation will be a key advance on previous CMIP exercises. Meeting participants agreed that the Earth System Model Evaluation Tool (ESMValTool) will play a valuable contributing role in CMIP. ESMValTool aims to facilitate the evaluation of comprehensive ESMs, raise the standard for model evaluation, and facilitate participation in, and analysis of, CMIP6 and related initiatives. Subsets of the metrics discussed at the workshop are being developed for implementation as part of an Antarctic and Southern Ocean contribution to ESMValTool.
More details can be found at the workshop’s website.
—Alia L. Khan (email: [email protected]; @AliaLaurenKhan), National Snow and Ice Data Center, Cooperative Institute for Research in Environmental Sciences, University of Colorado Boulder; Thomas J. Bracegirdle, British Antarctic Survey, Cambridge, U.K.; and Joellen L. Russell, University of Arizona, Tucson