Ocean-Sea Ice Coupling

Model development to study interactions between sea ice and ocean turbulence

PI: Andrew Thompson (Divisions of Geological and Planetary Science and Engineering and Applied Science)
SASE: Skylar Gering, Scholar

Arctic sea ice extent and concentration continue to decline at rates that are commonly underestimated by climate projection models. Potential sources of uncertainty arise from an inaccurate representation of interactions between the ocean and sea ice within these climate models, as well as the simplification of sea-ice dynamics for the sake of reducing computational complexity. For example, climate models commonly represent sea ice as a continuous medium; however, in many regions, sea ice is comprised of many discrete, interacting pieces that span a large range of scales. Improvements in the representation of sea ice in models, as well as their interactions with the underlying ocean, could reduce uncertainty in the future evolution of Arctic sea ice.

The Office of Naval Research has funded a Multidisciplinary University Research Initiative (MURI), “Mathematics and Data Science for Physical Modeling and Prediction of Sea Ice,” to develop new models related to Arctic sea ice evolution. At Caltech, within the Ocean and Cryosphere, Theory and Observations (OCTO) Group, Dr. Mukund Gupta is working with a new ocean large eddy simulation (LES) model, Oceananigans, developed as part of the CliMA project (Caltech, MIT). Gupta has adapted this model to include the representation of simple, cylindrical sea-ice floes and has identified key coupled ocean-ice processes occurring at floe boundaries that impact sea ice melt rates. At the University of Washington, Dr. Georgy Manucharyan and his research team have developed the SubZero sea-ice model, which can rapidly simulate a large field of individual sea-ice floes that change in shape and mass over time.

The Schmidt Scholar is collaborating broadly with colleagues within the ONR-MURI and CliMA projects to further explore sea ice dynamics. Gering will translate the SubZero code from MATLAB to Julia, work to integrate SubZero and Oceananigans, and then test the integration to validate and optimize the coupled model.