Search

During the second half of the twentieth century, the Antarctic Peninsula was subjected to a rapid increase in air temperatures. This was accompanied by a reduction in sea ice extent, increased precipitation and a dramatic retreat of glaciers associated with an increase in heat flux from deep ocean water masses. Isotopic tracers have been used previously to investigate the relative importance of the different freshwater sources to the adjacent Bellingshausen Sea (BS), but the data coverage is strongly biased toward summer. Here we use a regional model to investigate the ocean's response to the observed changes in its different freshwater inputs (sea ice melt/freeze, precipitation, evaporation, iceberg/glacier melt, and ice shelf melt). The model successfully recreates BS water masses and performs well against available freshwater data. By tracing the sources and pathways of the individual components of the freshwater budget, we find that sea ice dominates seasonal changes in the total freshwater content and flux, but all sources make a comparable contribution to the annual-mean. Interannual variability is dominated by sea ice and precipitation. Decadal trends in the salinity and stratification of the ocean are investigated, and a 20-year surface freshening from 1992 to 2011 is found to be predominantly driven by decreasing autumn sea ice growth. These findings will help to elucidate the role of freshwater in driving circulation and water column structure changes in this climatically-sensitive region.

For the first time, we compute the sea-ice concentration budget of a fully coupled climate model, the Australian ACCESS model, in order to assess its realism in simulating the autumn–winter evolution of Antarctic sea ice. The sea-ice concentration budget consists of the local change, advection and divergence, and the residual component which represents the net effect of thermodynamics and ridging. Although the model simulates the evolution of sea-ice area reasonably well, its sea-ice concentration budget significantly deviates from the observed one. The modelled sea-ice budget components deviate from observed close to the Antarctic coast, where the modelled ice motion is more convergent, and near the ice edge, where the modelled ice is advected faster than observed due to inconsistencies between ice velocities. In the central ice pack the agreement between the model and observations is better. Based on this, we propose that efforts to simulate the observed Antarctic sea-ice trends should focus on improving the realism of modelled ice drift.

Several different classes of ocean model are capable of representing floating glacial ice shelves. We describe the incorporation of ice shelves into Fluidity-ICOM, a nonhydrostatic finite-element ocean model with the capacity to utilize meshes that are unstructured and adaptive in three dimensions. This geometric flexibility offers several advantages over previous approaches. The model represents melting and freezing on all ice-shelf surfaces including vertical faces, treats the ice shelf topography as continuous rather than stepped, and does not require any smoothing of the ice topography or any of the additional parameterisations of the ocean mixed layer used in isopycnal or z-coordinate models. The model can also represent a water column that decreases to zero thickness at the ‘grounding line’, where the floating ice shelf is joined to its tributary ice streams. The model is applied to idealised ice-shelf geometries in order to demonstrate these capabilities. In these simple experiments, arbitrarily coarsening the mesh outside the ice-shelf cavity has little effect on the ice-shelf melt rate, while the mesh resolution within the cavity is found to be highly influential. Smoothing the vertical ice front results in faster flow along the smoothed ice front, allowing greater exchange with the ocean than in simulations with a realistic ice front. A vanishing water-column thickness at the grounding line has little effect in the simulations studied. We also investigate the response of ice shelf basal melting to variations in deep water temperature in the presence of salt stratification.

Predictions from the first realistic model of the hydrodynamics of the riverine thermal bar in a medium-size lake are presented. Important features of field observations from Kamloops Lake, British Columbia are successfully reproduced, but the model adopts a generalised section, which is regarded as being representative of many other lakes. A study of the model sensitivity to various aspects of its formulation is also presented, particularly emphasising the important influence of Coriolis forcing on the thermal bar circulation. Plankton population dynamics within the thermal bar flow field is then studied by means of two ecological models of differing complexity. Differences between the predictions of the two models are explained with reference to intermediate simulations, and it is found that the simple ecosystem formulation used in previous work may give misleading results. The flow and stability conditions of the riverine thermal bar have a profound influence on ecosystem development, and support greater phytoplankton growth than in thermal bars resulting purely from radiative effects.