Sea ice motion as a stochastic process

TL;DR

This paper develops a stochastic physics-based model for Arctic sea-ice floe drift, deriving a Langevin equation and analyzing velocity distributions, with implications for understanding sea ice motion dynamics.

Contribution

It introduces a Langevin-based stochastic model for sea-ice floe motion, incorporating Coulomb friction and deriving stationary velocity PDFs.

Findings

Stationary velocity PDFs are Laplace distributions near high ice compactness.

The model aligns with observed sea ice velocity distributions.

Discussion of extending the model to include thermal and mechanical effects.

Abstract

We use tools from statistical physics to develop a stochastic theory for the drift of a single Arctic sea-ice floe. Floe-floe interactions are modelled using a Coulomb friction term, with any change in the thickness or the size of the ice floe due to phase change and/or mechanical deformation being neglected. We obtain a Langevin equation for the fluctuating velocity and the corresponding Fokker-Planck equation for its probability density function (PDF). For values of ice compactness close to unity, the stationary PDFs for the individual components of the fluctuating velocity are found to be the Laplace distribution, in agreement with observations. A possible way of obtaining a more general model that accounts for thermal growth and mechanical deformation is also discussed.