The Southern Ocean is chronically under-sampled due to its remoteness, harsh environment and sea-ice cover. Ocean circulation models yield significant insight into key processes and to some extent obviate the dearth of data, however they often underestimate surface mixed layer depth (MLD), with consequences for water-column properties. In this study, a coupled circulation and sea-ice model was implemented for the region adjacent to the West Antarctic Peninsula (WAP), a climatically sensitive region which has exhibited decadal trends toward higher temperatures, a shorter sea-ice season and increasing glacial freshwater input, overlain by strong interannual variability. Hindcast simulations were conducted with different air-ice drag coefficients and Langmuir-circulation parameterizations to determine the impact of these factors on MLD. Including Langmuir circulation deepened the surface mixed layer, with the deepening being more pronounced in the shelf and slope regions. Optimal selection of an air-ice drag coefficient also increased the modeled MLD by similar amounts, and had a larger impact in improving the reliability of the simulated MLD interannual variability. This study highlights the importance of sea ice volume and redistribution to correctly reproduce the physics of the underlying ocean, and the potential of appropriately parameterizing Langmuir circulation to help correct for a bias towards shallow MLD in the Southern Ocean. The model also reproduces observed freshwater patterns in the WAP during late summer and suggests that areas of intense summertime sea-ice melt can still show net annual freezing due to high sea-ice formation during the winter.

The ocean coastal-shelf-slope ecosystem west of the Antarctic Peninsula (WAP) is a biologically productive region that could potentially act as a large sink of atmospheric carbon dioxide. The duration of the sea-ice season in the WAP shows large interannual variability. However, quantifying the mechanisms by which sea ice impacts biological productivity and surface dissolved inorganic carbon (DIC) remains a challenge due to the lack of data early in the phytoplankton growth season. In this study, we implemented a circulation, sea-ice and biogeochemistry model (MITgcm-REcoM2) to study the effect of sea ice on phytoplankton blooms and surface DIC. Results were compared with satellite sea-ice and ocean color, and research ship surveys from the Palmer Long Term Ecological Research (LTER) program. The simulations suggest that the annual sea-ice cycle has an important role in the seasonal DIC drawdown. In years of early sea-ice retreat there is a longer growth season leading to larger seasonally integrated net primary production (NPP). Part of the biological uptake of DIC by phytoplankton, however, is counteracted by increased oceanic uptake of atmospheric CO2. Despite lower seasonal NPP, years of late sea-ice retreat show larger DIC drawdown, attributed to lower air-sea CO2 fluxes and increased dilution by sea-ice melt. The role of dissolved iron and iron limitation on WAP phytoplankton also remains a challenge due to the lack of data. The model results suggest sediments and glacial meltwater are the main sources in the coastal and shelf regions, with sediments being more influential in the northern coast.