Implementation of a brittle sea ice rheology in an Eulerian, finite-difference, C-grid modeling framework: impact on the simulated deformation of sea ice in the Arctic

• Published in: Geoscientific Model Development
• Main Authors: L. Brodeau, P. Rampal, E. Ólason, V. Dansereau
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2024
• Subjects: Geology; QE1-996.5; Sea ice
• Online Access: https://doi.org/10.5194/gmd-17-6051-2024; https://doaj.org/article/4847e7444b2e490fb3d79cfe72fb699f

We have implemented the brittle Bingham–Maxwell sea ice rheology (BBM) into SI3, the sea ice component of NEMO. After discussing the numerical aspects and requirements that are specific to the implementation of a brittle rheology in the Eulerian, finite-difference, Arakawa C-grid framework, we detail the approach we have used. This approach relies on the introduction of an additional set of prognostic stress tensor components, sea ice damage, and sea ice velocity vector, following a grid point arrangement that expands the C-grid into the Arakawa E-grid. The newly implemented BBM rheology is first assessed by means of a set of idealized SI3 simulations at different spatial resolutions. Then, sea ice deformation rates obtained from simulations of the Arctic at a 1 / 4 <svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="20pt" height="14pt" class="svg-formula" dspmath="mathimg" md5hash="f7b7a5d5bfd5c45cef6c295312dc1896"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="gmd-17-6051-2024-ie00001.svg" width="20pt" height="14pt" src="gmd-17-6051-2024-ie00001.png"/></svg:svg> ° spatial resolution, performed with the coupled ocean–sea ice setup of NEMO, are assessed against satellite observations. For all these simulations, results obtained with the default current workhorse setup of SI3 are provided to serve as a reference. Our results show that using a brittle type of rheology, such as BBM, allows SI3 to simulate the highly localized deformation pattern of sea ice, as well as its scaling properties, from the scale of the model's computational grid up to the basin scale.