9:35 AM - 9:50 AM
[ACG40-03] Impacts of CS2SMOS assimilation in an ocean and sea-ice forecast system and the diagnosed ice thickness response to sea level pressure
Keywords:Sea ice thickness, Response, Sea level pressure, Singular Value Decomposition
Skillfully dynamic sea ice prediction requires adequate precision for the ocean and atmosphere forcings and the concerned response processes. In recent years, the satellite-based sea ice thickness observations combined from Cryosat2 and SMOS have been assimilated into different ice model systems and show that the model bias of SIT can be dominantly reduced. However, the SIT variability and its response function to the atmosphere forcing, like sea level pressure, have not been noticeable. Aiming at this topic, the two parallel assimilation runs were done in the TOPAZ system with and without the assimilation of SIT during 2014-2017. Firstly, the differences in SIT variability incurred from the SIT assimilation will be investigated from spatial and time scales. The SIT variability results show a much longer timescale over one season, beating SIC variability. Further, the singular value decomposition (SVD) analysis shed light on the first three modes of sea ice thickness and sea level pressure. The first mode of SLP is an analogy to Arctic Oscillation by a vortex-dominated nature, which the model parameterizations for sea ice could overrepresent compared to the impacts of the rest modes. The following second and third modes of SLP show a dipolar pattern with an increased variance contribution through the DA. The results of this study further suggest a way to dig the physical information behind the data, which is helpful for future model development and even in present data-driven applications like machine learning to optimize the potential parameters.