Dense water downslope flow and AABW production in a numerical model: Sensitivity to horizontal and vertical resolution in the region off Cape Darnley polynya

• Published in: Ocean modelling (Oxford) Vol. 165; p. 101843
• Main Authors: Mensah, Vigan, Nakayama, Yoshihiro, Fujii, Masakazu, Nogi, Yoshifumi, Ohshima, Kay I.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2021
• Subjects: Dense water production; Downslope flow; East antarctica; Ocean general circulation models (OGCM)/regional ocean models; Ocean-ice modeling; Polynyas; Ocean general circulation models (OGCM)/regional ocean models; East antarctica; Downslope flow; Polynyas; Dense water production; Ocean-ice modeling
• ISSN: 1463-5003; 1463-5011
• DOI: 10.1016/j.ocemod.2021.101843

The formation of Dense Shelf Water (DSW) and Antarctic Bottom Water (AABW) in the Southern Ocean is an essential part of the thermohaline circulation, and understanding this phenomenon is crucial for studying the global climate. AABW is formed as DSW flows down the continental slope and mixes with the surrounding waters. However, DSW formation and its descent remains a poorly resolved issue in many ocean models. We, therefore, simulated the formation and descent of DSW and investigated the model sensitivities to horizontal and vertical grid spacings. The Massachusetts Institute of Technology general circulation model (MITgcm) was used for the region off Cape Darnley in East Antarctica, one of the main AABW production areas, where historical and mooring data are available for comparison. Simulations with coarse horizontal grid resolutions of order (10 km) yielded high volumes of DSW on the shelf. However, the largest part of this DSW was transformed into intermediate water and advected westward. Horizontal model resolutions equal to or higher than 2 km were required to simulate the descent of DSW and a realistic AABW production. Simulated time series at a mooring located at a depth of 2,600 m showed periodic fluctuations in velocity and temperature of 0.3 m⋅s−1 and 0.5 °C, respectively, consistent with observations. We also found that high-resolution bathymetry datasets are crucial because the newly formed AABW volume was reduced by 20% when a smoother bathymetry was used on a 2-km resolution grid. Vertical resolution had little influence on model performance because the plume was much thicker (> 170 m) than the grids width. Therefore, reproducing the downslope flow of DSW and AABW formation in the Cape Darnley region can be achieved with a high horizontal resolution (≤ 2 km) and a relatively coarse vertical resolution (∼100 m on the continental slope). •We model AABW formation off Cape Darnley and test its sensitivity to grid spacings.•We simulate AABW formation in good agreement with observations in the region.•Fine horizontal resolutions ≥2 km yield realistic downslope flow, AABW properties.•Vertical resolution did not affect the simulation of AABW formation in the region.•Bathymetric data of O (100 km) is required to produce realistic amounts of AABW.