Two massive tabular icebergs calved from the Ross Ice Shelf in 2000 (B-15) and 2002 (C-19) and perturbed regional ocean processes for several years. Here we document the ocean's response in McMurdo Sound to the icebergs using in situ data collected before, during and after the icebergs’ residence in the Ross Sea. Departures from typical McMurdo Sound seasonal oceanography included the non-appearance of Antarctic Surface Water in summer, a cooler and more homogeneous water column during winter and ‘super-fresh’ High Salinity Shelf Water that gradually recovered its salinity. We found that each iceberg triggered a distinct response to regional ocean processes. B-15a, the largest piece of iceberg B-15, restricted surface circulation, cooled and freshened the upper water column and reduced melting near the ice shelf front for four years. Iceberg C-19 interrupted the operation of the Ross Sea polynya, from which McMurdo Sound took three to four years to recover, and was responsible for a geographic shift in the dense water formation region for the south-western Ross Sea. These results differ from earlier modelling studies and highlight the challenges of modelling the polar ocean. We also show that one pathway previously thought to supply dense water to the Ross Ice Shelf cavity was not operating at that time.

The production of Antarctic Bottom Water is influenced by Ice Shelf Water which is formed due to the modification of shelf water masses under huge ice shelves. The coupling of inflow conditions, thermohaline processes at the ice shelf base and the sub-ice shelf circulation is studied with a two-dimensional thermohaline circulation model which has been developed for a section perpendicular to the ice shelf edge. Different boundary conditions appropriate to the Filchner Ice Shelf regime are considered. The model results indicate that, in general, shelf water is transported toward the grounding line, where at the ice shelf base melting occurs with a maximum rate of 1.5 my−1. Accumulation of ice takes place at the end of the melting zone close to the ice shelf edge with a rate on the order of 0.1 my−1. The location of this accumulation zone determines whether or not the density increase by salt rejection causes an upper circulation cell and the separation of the modified water mass from the ice shelf base at mid-range depth. At the ice shelf edge the simulated temperature, salinity, helium and δ18O values for the temperature minimum layer are typical for Ice Shelf Water. However the sub-ice shelf circulation is highly variable as well as sensitive to changes in boundary conditions. Moderate changes in the characteristics of the inflowing water or in sea-floor topography may double the intensity of the circulation. Non-linear processes in the accumulation zone cause variabilities which can be described by an ice shelf edge oscillator influencing the entire circulation regime.

In the Weddell Sea oceanographic data and numerical models demonstrate that Ice Shelf Water, one ingredient in the production of Weddell Sea Bottom Water, is formed by thermohaline interaction of High Salinity Shelf Water with the base of the Filchner-Ronne ice shelves. South of Berkner Island a passage with a water column thickness of about 300 m linking the Filchner and the Ronne regimes is important for the ventilation of the sub-ice shelf cavities. To simulate the flow we tested a two-dimensional thermohaline circulation model on several sections which approximate different geometries of a sub-ice shelf channel bounded by the ocean bottom and the ice shelf base. Temperature and salinity profiles measured in front of the Filchner-Ronne ice shelves are used to force the model. The results indicate that the circulation is sensitive to both salinity (density) forcing and depth of the shelf bottom prescribed at the open boundary representing the Ronne Ice Shelf edge. Where the shelf is shallow, 400 m deep, a closed circulation cell within the Ronne cavity acts like an ice pump with accumulation rates of marine ice at the ice shelf base up to 1.5 m y−1. The total outflow at the Ronne Ice Shelf edge is supported by an inflow from the Filchner regime. Where the shelf is deeper, a flow from the Ronne into the Filchner cavity develops if the bottom salinity at the Ronne Ice Shelf edge exceeds a critical value of 34.67. Seasonal variability imposed at both edges modifies the circulation pattern at the Filchner Ice Shelf edge such that the depth and magnitude of Ice Shelf Water outflow correspond with observations in the Filchner Depression.

The paper presents two years of observations (1995 and 1996) for water velocity and temperature near the Ross Ice Shelf edge, where a mooring, composed of three current meters, five temperature and three conductivity sensors, one turbidimeter, two sediment traps and one Acoustic Doppler Current Profiler, was deployed. The mooring site was chosen to study the behaviour of the Ice Shelf Water, within the framework of the Project on the Climatic Long-term Interaction for the Mass balance in Antarctica of the Italian National Programme for Antarctic Research. The Ice Shelf Water outflow from below the Ross Ice Shelf is a persistent phenomenon in this area, but it is characterized by impulsive injections often associated with recirculation processes. Two years of data records of physical variables are investigated here in order to analyse the Ice Shelf Water outflow behaviour and to give an estimate of this feature. Our results show that the Ice Shelf Water events have variable duration (from one to seven days) and present an interannual variability in their timing. In spite of this, the mean Ice Shelf Water outflow during the overlapping period of the two time series is quite similar: 0.15 Sv for 1995 and 0.28 Sv for 1996.

This paper presents the results of the analysis of hydrological data of a 5-day mesoscale experiment (53 CTD casts) conducted during the XIIIth Italian Expedition to Antarctica (1997–98 cruise) in the framework of the CLIMA (Climatic Longterm Interaction for the Mass balance in Antarctica) Project of the Italian National Programme for Antarctic Research (PNRA). The experiment site was chosen for studying the dense water overflow in relation to the shelf-break in the central Ross Sea, after a large-scale synoptic survey, aimed to detect the general hydrological characteristics of the basin. A classical θ/S analysis was carried out for better understanding of the shelf-slope connection and the interactions between the water masses of this zone: the Circumpolar Deep Water (CDW) coming from the oceanic domain and the Ice Shelf Water (ISW) spreading from the Ross Ice Shelf (RIS) edge. Our results show the evidence of an overflow of dense water, originated on the continental shelf, on the shelf-break. This supercold water signal is found on the continental slope down to 1200 m depth. The shape of this tongue of modified ISW, whose thickness reaches up to 100 m, is very narrow, suggesting that the overflow occurs in very localized areas.