Towards improving short-term sea ice predictability using deformation observations

Abstract. Short-term sea ice predictability is challenging due to the lack of constraints on ice deformation features (open leads and ridges) at kilometre scale. Deformation observations capture these small-scale features and have the potential to improve the predictability. A new method for assimilation of satellite-derived sea ice deformation into the neXt generation Sea Ice Model (neXtSIM) is presented. Ice deformation provided by the Copernicus Marine Environmental Monitoring Service is computed from sea ice drift derived from Synthetic Aperture Radar at a spatio-temporal resolution of 10 km and 24 hours. We show that high values of ice deformation can be interpreted as reduced ice concentration and increased ice damage – scalar variables of neXtSIM. The proof-of-concept assimilation scheme uses a data nudging approach and deterministic forecasting with one member. Assimilation and forecasting experiments are run on example observations from January 2021 and show improvement of neXtSIM skills to predict sea ice deformation in 3–5 days horizon. It is demonstrated that neXtSIM is also capable of extrapolating the assimilated information in space — gaps in spatially discontinuous satellite observations of deformation are filled with a realistic pattern of ice cracks, confirmed by later satellite observations. The experiments also indicate that reduction in sea ice concentration plays a bigger role in improving ice deformation forecast on synoptic scales. Limitations and usefulness of the proposed assimilation approach are discussed in a context of ensemble forecasts. Pathways to estimate intrinsic predictability of sea ice deformation are proposed.