The first oceanographic measurements in the Ross Sea were made by its discoverer James Clark Ross, from the Erebus, on 18 January 1841. Since that time its continental shelf, seasonally ice free in most years, has proved a magnet to explorers and scientists, if not to fishermen and tourists. Nevertheless, our knowledge of this environment is rapidly being outpaced by our ignorance of its variability. For example, the Ross Sea contains two of the largest, most persistent polynyas on the Antarctic coastline, but its sea ice extent has increased over recent decades while its salinity has steadily declined. Are regional winds now stronger, the ocean circulation faster, and the ice thinner now than at the time of the IGY? Are its winter polynyas characterized more by upwelling driven by offshore winds, or downwelling due to brine release when sea ice is formed? How are polynya surface layers stabilized and iron-enriched, reportedly enhancing summer productivity, if the ice cover is blown away before it can melt in situ?

Changes in water mass distribution and horizontal circulation due to seasonal influences on the Ross Sea continental shelf are investigated using a circumpolar numerical model. An anticyclonic circulation cell that extends across the open shelf and into the ice shelf cavity is formed in the model. The increased east–west density gradient caused by the strong brine release in the Ross Sea polynya in winter results in an intensification of this anticyclonic cell from 1.5 Sv to 2.5 Sv. This supports the concept of a thermohaline-driven horizontal circulation on the Ross Sea continental shelf. In addition to a temporal change in the circulation strength, the changes in the density structure lead to complex temporal and spatial variability in the circulation around Ross Island. Due to seasonal variation in circulation strength and water temperatures, the area averaged basal melt rate of 25 cm a−1 shows a bimodal seasonal cycle ranging from 20–28 cm a−1, with maxima in March and August.

We applied a modified version of the Miami isopycnic coordinate ocean general circulation model (MICOM) to the ocean cavity beneath the Ross Ice Shelf to investigate the circulation of ocean waters in the sub-ice shelf cavity, along with the melting and freezing regimes at the base of the ice shelf. Model passive tracers are utilized to highlight the pathways of waters entering and exiting the cavity, and output is compared with data taken in the cavity and along the ice shelf front. High Salinity Shelf Water on the western Ross Sea continental shelf flows into the cavity along the sea floor and is transformed into Ice Shelf Water upon contact with the ice shelf base. Ice Shelf Water flows out of the cavity mainly around 180°, but also further east and on the western side of McMurdo Sound, as observed. Active ventilation of the region near the ice shelf front is forced by seasonal variations in the density structure of the water column to the north, driving rapid melting. Circulation in the more isolated interior is weaker, leading to melting at deeper ice and refreezing beneath shallower ice. Net melting over the whole ice shelf base is lower than other estimates, but is likely to increase as additional forcings are added to the model.

Spreading and sinking of the Deep Ice Shelf Water (DISW) in the Ross Sea are analysed using in situ observations and the results of a nonlinear, reduced gravity, layered numerical model, which is able to simulate the motion of a bottom trapped current over realistic topography. The model is forced by prescribing thickness and density of the DISW layer at the southern model boundary as well as ambient density stratification above the DISW layer. This density structure is imposed using hydrographic data acquired by the Italian PNRA-CLIMA project. In the model water of the quiescent ambient ocean is allowed to entrain in the active deep layer due to a simple entrainment parameterization. The importance of forcing the model with a realistic ambient density is demonstrated by carrying out a numerical simulation using an idealized ambient density. In a more realistic simulation the path and the density structure of the DISW vein flowing over the Challenger Basin are obtained and are found to be in good agreement with data. It is found that entrainment, which is particularly active in regions of strong topographic variation, significantly influences the pattern followed by the DISW layer. The evolution of the DISW layer beyond the continental shelf, i.e., in a region where the paucity of experimental data does not allow for a detailed description of the deep ocean dynamics, is also investigated.

Two new ocean tide models for the Ross Sea including the ocean cavity under the Ross Ice Shelf, are described. The optimum model for predicting ice shelf surface height variability is based on assimilation of gravimetry-derived tidal constituents from the Ross Ice Shelf. Synthetic aperture radar interferograms provide an independent test of model performance. The standard deviation of tide height variability is largest under the eastern ice shelf along the Shirase and Siple Coasts, where it can exceed 0.8 m. The maximum peak-to-peak tidal range in this region is ∼3 m. The best predictor for ocean tidal currents north of the ice front is a dynamics-based model that solves the depth-integrated shallow water equations with a linear representation of benthic friction rather than the more usual quadratic form. Tidal currents over the open Ross Sea are dominated by diurnal, topographically trapped vorticity waves. The strongest modelled currents exceed 1 m s−1 at spring tide in a narrow band along the upper continental slope in the north-western Ross Sea. Typical tidal currents in the central continental shelf area of the Ross Sea are 10–20 cm s−1. Under the ice shelf the typical currents are ∼5 cm s−1.

In certain regions of the Southern Ocean, tidal energy is believed to foster the mixing of different water masses, which eventually contribute to the formation of deep and bottom waters. The Ross Sea is one of the major ventilation sites of the global ocean abyss and a region of sparse tidal observations. We investigated M2 tidal dynamics in the Ross Sea using a three-dimensional sigma coordinate model, the Regional Ocean Model System (ROMS). Realistic topography and hydrography from existing observational data were used with a single tidal constituent, the semi-diurnal M2. The model fields faithfully reproduced the major features of the tidal circulation and had reasonable agreement with ten existing tidal elevation observations and forty-two existing tidal current measurements. The differences were attributed primarily to topographic errors. Internal tides were generated at the continental shelf/slope break and other areas of steep topography. Strong vertical shears in the horizontal velocities occurred under and at the edges of the Ross Ice Shelf and along the continental shelf/slope break. Estimates of lead formation based on divergence of baroclinic velocities were significantly higher than those based on barotrophic velocities, reaching over 10% at the continental shelf/slope break.

Ship-based observations of sea ice thickness using the Antarctic Sea Ice Processes and Climate (ASPeCt) protocol provide information on ice thickness distribution at relatively low cost. This protocol uses a simple formula to calculate the mass of ice in ridges based on surface observations. We present two new formulae and compare these with results from the “Original” formula using data obtained in the Ross Sea in autumn and winter. The new “r-star” formula uses a more realistic ratio of sail and keel areas to transform dimensions of sails to estimates of mean keel areas. As a result, estimates of “equivalent thickness” (i.e. mean thickness of ice in ridged areas) increased by over 200%. The new “Probability” formula goes one step further, by incorporating the probability that a sail is associated with a keel underwater, and the probability that keels may be found under level surfaces. This resulted in estimates of equivalent thickness comparable with the Original formula. Estimates of equivalent thickness at one or two degree latitude resolution are sufficiently accurate for validating sea ice models. Although ridges are small features in the Ross Sea, we have shown that they constitute a significant fraction of the total ice mass.

During summer (2 December 1999–23 January 2000) an Ocean Surface Current Radar (OSCR-II) was used to provide surface current measurements within the Terra Nova Bay polynya, one of the most important coastal polynyas of the Ross Sea. This represents an important step towards a continuous monitoring of the area. Useful information is now available as a basis for future work in this field, although the two radar sites, necessary to calculate the total current vector, did not work together throughout the whole period of the experiment as one of the units was damaged. The results demonstrate the feasibility of this kind of measurement and suggest that very important dynamical characteristics of the polynya could be deduced from long term deployment of such a system.

This study looks at the feasibility of using Ocean Acoustic Tomography for long-term monitoring of polynyas using both observations in Terra Nova Bay polynya (Ross Sea) and simulations with a range dependent, multi-layered adiabatic normal mode acoustic propagation model. The summer sound speed profile is characterized by surface values of around 1450 m s−1, a minimum of 1441 m s−1 around 50 m depth and a linear increase with a 0.016 s−1 slope. Thus, the sound propagation is apparently ducted in the near surface layer and is refracted upward below it. During winter, due to water cooling and mixing processes, the subsurface minimum disappears, the surface sound speed is about 1440 m s−1 and no near surface layer ducted propagation occurs. Because of the specificity of the Terra Nova Bay seasonal sound speed profile and to cope with both deep and shelf water applicability, the feasibility study of acoustic inversion was undertaken using normal mode Match Field Tomography instead of the more classical travel-time inversion. The results from simulations demonstrate that ocean acoustic tomography is able to reproduce quite well the vertical sound speed profile, in particular the temporal evolution of summer stratification and winter mixing processes, thus providing information on the upper layer, where direct measurements are not possible.

The Southern Ocean plays an important role in the global carbon cycle and, as a consequence, in the planetary climate equilibrium. The Ross Sea is one of the more productive regions in the Southern Ocean, due to strong phytoplankton blooms occurring during summer. Satellite remote sensing is a powerful tool for investigating such phenomena, especially if the bio-optical algorithms are tuned with in situ data. In this paper, after having compared the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) and the ENEA Lidar Fluorosensor (ELF), the SeaWiFS chlorophyll a (Chl a) algorithm is tuned in the Ross Sea by means of the ELF measurements. The Chl a concentrations obtained in this way have been the basis for estimating productivity values and their evolution during summer 1997–98. Three primary production models have been used, providing information on their accuracy and performance in the Antarctic environment. Our investigations suggest that the primary production was lower than usual during the period 3 December 1997–16 January 1998.

A late summer study of marine bacteria activities, and the interrelationships with the microbial loop and the microbial food chain was carried out from 22 January to 10 February 2000 in a coastal area of Terra Nova Bay (Ross Sea). The objective was to investigate the transition from the end of a phytoplanktonic bloom to the start of winter. Intense bacterial activities, comparable to those of temperate marine environments, were observed. The carbon potentially mobilized from proteinaceous matter was quantitatively the most important source of carbon for the bacterioplankton. The leucine aminopeptidase activity was higher in January samples and decreased towards 10 February whereas an opposite trend was observed for alkaline phosphatase and �-glucosidase activities. The bacterial production was supported by c. 0.2% of the amounts of dissolved organic carbon mobilised by hydrolytic activities and by 7% of inorganic phosphate mobilised by alkaline phosphatase activity. A sharp reduction in the bacterial biomass, possibly due to zooplankton grazing or viral lysis, was observed for the first time in Terra Nova Bay.

Adélie penguin (Pygoscelis adeliae) stomach contents, krill (Euphausia superba and E. crystallorophias) and silverfish (Pleuragramma antarcticum) from the Ross Sea were analysed to determine several persistent organic pollutants (POPs). In discussing the data, the prey-predator linkage between these species was taken into account. Sampling was carried out during the 1995/96 and 1999/00 Italian Antarctic Expeditions. Fifty four polychlorobiphenyls (PCBs) and p,p’-DDE and hexachlorobenzene (HCB) were quantified in stomach contents of penguins nesting at Edmonson Point (Victoria Land) and in whole specimens of silverfish and krill from the Ross Sea. Xenobiotic concentrations in organisms were low compared to data reported for many marine species of lower latitudes and ranged from 0.22 ng g−1 wet wt p,p’-DDE in krill to 161 ng g−1 wet wt PCBs in silverfish. Fingerprints and class of isomer patterns showed a predominance of low chlorinated PCBs, mainly in pelagic organisms. Average input of these POPs through the diet was also evaluated. Concentrations of the most toxic non-ortho PCBs, IUPAC nos 77 (3,3′,4,4′), 126 (3,3′,4,4′,5) and 169 (3,3′,4,4′,5,5′), were 1.63 pg g−1, 7.31 pg g−1 and 0.23 pg g−1 wet wt, respectively, in stomach content samples. Stomach contents had 0.037 pg g−1 wet wt TEQ (Toxic Equivalents) of which penta-CB126 accounted for most of the toxicity.

The analysis of the mixing processes involving water masses on the Ross Sea continental shelf is one of the goals of the CLIMA project (Climatic Long-term Interactions for the Mass balance in Antarctica). This paper uses extended Optimum MultiParameter analysis (OMP), which is applied to four datasets collected during the cruises of 1994/95, 1995/96, 1997/98 and 2000/01 in the Ross Sea (Antarctica). Data include both hydrological, (temperature, salinity, and pressure; T, S, and P, respectively) and chemical parameters (O2, Si(OH)4, PO4, and NO3+NO2). The OMP analysis is based on the assumption that the mixing is a linear process which affects all parameters equally so that each sample shows physical/chemical properties that are the result of the mixing of some properly selected Source Water Types (SWTs). OMP thus evaluates the best set of contributions by all SWTs to each sample, and allows the spatial distribution and structure of the water masses in a basin to be evaluated. Ocean circulation may subsequently be inferred by means of a deeper analysis of the spreading of the water mass. In this study, the “real” Redfield ratios observed in the Ross Sea were used to correct the equations referring to the chemical parameters in accordance with the extended version of OMP. The solutions include some physically realistic constraints. The results allow a detailed description of the water mass distribution, validated through comparison with some “canonical” thermohaline characteristics of the Ross Sea hydrology. In particular our results verify the spreading of the HSSW over the entire continental shelf and emphasize the key role it plays in the ventilation of the deep waters outside the Ross Sea. In addition a description is given of the intrusion of relatively warm waters coming from the open ocean and flowing at some specific locations at the continental shelf break.

Downward fluxes of particulate matter were investigated in the polynya of Terra Nova Bay (western Ross Sea) from February 1995 to December 1997. Biogenic silica made up the bulk of the sinking matter, reaching a maximum of 85% of the total flux in both the shallow and deep traps. Organic carbon accounted for 2.1 to 23.5% of the sample mass in the shallow trap and 2.8 to 7.2% in the deep trap. Biogenic silica and organic carbon peaks occurred in February–March over the three years and were always coupled in the top trap. Carbonate fluxes showed an uneven temporal trend, were low near the seafloor and contributed up to 69.8% of the material collected by the shallow trap. Vertical fluxes in the polynya showed large seasonal and interannual variability. Differences in primary productivity, ice cover dynamics and current energies accounted for the interannual changes in the flux of materials to the subsurface trap. Advection from sites a few hundred kilometres apart was the most likely reason for the decoupling between shallow and deep fluxes after October 1996. The seasonal pattern of downward fluxes was essentially linked to the seasonal cycle of primary production and spring–summer variations largely resulted from the composition of the phytoplankton assemblage in the overlying water column. The export of phototrophically produced materials to depth was low in late spring–early summer, at the time when Phaeocystis and small flagellates were the predominant autotrophs and the microbial food web prevailed over herbivorous feeding. Enhanced fluxes were linked to the predominance of large diatoms above the trap, in late summer. Temporal dynamics in this polynya were delayed by approximately one month as compared to the southernmost Ross Sea polynya, but showed the same temporal pattern: the time lag between production and accumulation of biogenic materials in the upper water column was a few weeks, while the delay in export from the surface layer relative to the maximum accumulation of biomass was approximately one month.

Flux data from moored sediment trap experiments and mass accumulation rates in sediments were obtained for three sites in the Ross Sea which are currently studied for the formation and transit of High Salinity Shelf Water and Ice Shelf Water. These two data sets were compared to obtain inferences on the coupling between water column processes and sedimentary records. The depth distribution of physical features and concentrations of organic carbon and biogenic silica in box cores and gravity cores were studied. Mass accumulation rates, established on the basis of two conventional 14C dates for each core, range between 7.64 and 19.46 g m−2 yr−1. Although these are productive areas, downward fluxes measured by sediment traps are low: 7.5–25.6, 2.4–17.9 and 0.5–0.9 g m−2 yr−1 for particles, biogenic silica and organic carbon, respectively. The concentrations of biogenic components in surficial sediments are correspondingly low. Simple mass balances were calculated assuming the conservative behaviour of the lithic fraction of sinking materials and sediment. Lateral advection of suspended particles is needed to balance the fluxes at the three sites. Furthermore, the model suggests that the preservation of biogenic components is lower than at other sites of the Ross Sea, probably due to the low accumulation rates that imply a high residence time of biogenic materials at the sediment-water interface.

The structure and variability of Southern Ocean fronts south of New Zealand are described based on fifteen summer expendable bathythermograph (XBT) sections obtained between 1994 and 2001. The temperature variability north of 60°S is dominated by meanders and meridional shifts of the Sub-Antarctic Front (SAF), which often bifurcates to form northern and southern branches. The northern branch follows the southern edge of the Campbell Plateau, while the southern branch is found over the abyssal plain of the south-west Pacific Basin. The northern and southern branches of the SAF can be separated by as much as 900 km. Intense eddies or meanders of the SAF displace isotherms by as much as 5 degrees of latitude from their positions when such features are absent. The Polar Front (PF) position is more stable in time, although cold-core features associated with eddies or meanders of the front are occasionally observed between the southern SAF and the PF. The position of the southern ACC front is extremely stable, consistently overlying the 3000 m isobath on the northern flank of the Pacific-Antarctic/south-east Indian Ridge.

The Weddell Polynya, a large expanse of water that originated over Maud Rise (a bathymetric protrusion centred near 64°30′S, 3°E) and remained open during winter in the late 1970s, may have manifested a mode of deep ocean convection where despite large heat loss at the surface, sustained heat transport from below prevents lasting ice formation. In a different dominant mode (the present one), sea ice forms early in the winter and subsequently provides a thermal barrier that quickly quells incipient deep convection, thus preventing wholesale destruction of the ice cover. A possible mechanism for overcoming the thermal barrier is thermobaricity, the pressure dependence of the thermal expansion factor for seawater. An idealized, two-layer version of actual temperature and salinity profiles from the Weddell illustrates that thermobaric mixing can persist for extended periods in an ice-covered ocean, provided realistic melt rates (controlled by salt exchange at the ice/ocean interface) are specified. This furnishes a possible explanation for transient winter polynyas sometime observed in the ice-covered Southern Ocean. Thermobaricity may provide a trigger for widespread convection with possible climate impact.