JpGU-AGU Joint Meeting 2020

講演情報

[E] ポスター発表

セッション記号 M (領域外・複数領域) » M-IS ジョイント

[M-IS15] 南大洋・南極氷床が駆動する全球気候変動

コンビーナ:関 宰(北海道大学低温科学研究所)、岡 顕(東京大学大気海洋研究所)、野木 義史(国立極地研究所)、Robin Elizabeth Bell(Lamont Doherty Earth Observatory)

[MIS15-P05] Dense water production and downslope flow off Cape Darnley reproduced by a realistic numerical model

*Vigan Mensah1Yoshihiro Nakayama1Masakazu Fujii2Yoshifumi Nogi2Kay I. Ohshima1 (1.Institute of Low Temperature Science、2.National Institute of Polar Research)

キーワード:Ocean-ice modelling, East Antarctica, Dense Water production, Sea ice

The ability of the Massachusetts Institute of Technology general circulation model (MITgcm) to reproduce dense water production and its downslope outflow through a complex topography is evaluated through a series of simulations at different horizontal and vertical resolutions. The study domain is the region off Cape Darnley in East Antarctica, where intense sea ice production from March to October leads to the formation of Dense Shelf Water, previously observed to outflow periodically as a downslope current. Our experiments are based on a hydrostatic parametrization, using a coupled ocean-ice-atmosphere model for realistic forcing and sea ice production. Different vertical and horizontal resolutions are used to evaluate the relative importance of the model vertical grid and of the topography on the production of dense water and its outflow, and the model output is compared with the mooring observation data for evaluation purposes. The bottom topography data around the main canyon (Wild Canyon) are based on multibeam echo-sounder observations. While the outflow of dense shelf water is seen in the model outputs at the mooring location for each vertical and horizontal resolution, increasing the horizontal resolution leads to a start of the outflow and a periodicity more consistent with the observations. This suggest the dominant role of horizontal resolution for the accurate reproduction of dense shelf water production and outflow. We then proceed to determine the optimal setting allowing for reproduction of these phenomena at the lowest possible computational cost.