Peroxide speciation and formaldehyde measurements have been made on ice cores retrieved from Law Dome, Antarctica. Measurements were made for ice deposited during four different periods: modern, pre-industrial Holocene, early Holocene and Last Glacial Maximum (LGM). The data show modern peroxide levels >50% above pre-industrial levels (at ∼1.6 μmol L−1) and an absence of methyl hydroperoxide (down to a detection threshold of 0.003 μmol L−1). Formaldehyde levels show a 40% increase from pre-industrial to modern times (rising from ∼0.07 μmol L−1 to ∼0.10 μmol L−1), with a further increase and possible seasonality near the surface which we associate with post-depositional processes. Peroxide levels in LGM ice are low, but formaldehyde concentrations are high (at ∼0.13 μmol L−1) relative to modern levels. Similar high levels of formaldehyde are seen in early Holocene ice (∼6900 years BP).

Eastern Dome C, southern Talos Dome and northern Taylor Dome are drained by the Priestley, Reeves, David, Mawson and Mackay outlet glaciers, which flow into the Scott Coast on the west side of the Ross Sea, Antarctica. Airborne radar surveys were conducted on these glaciers to determine ice thickness and bed morphology along transverse and longitudinal profiles of the grounded and floating segments. A new analysis of a Landsat Thematic Mapper satellite image using a tracking technique was used to measure ice velocity at grounding lines and along ice tongues. The integration of radar and satellite data helped to locate grounding lines and to calculate the ice discharge. Changes in ice fluxes of floating glaciers were used to determine basal melting and freezing rates. The ice discharge calculated is less than half that required for a zero net surface mass balance according to the inputs given by the accumulation estimates widely adopted at present. The basal melting rates of meteoric ice represent 50% of the net ablation rate.

A multiparametric (chemical, isotopic and physical) study on three shallow firn cores sampled in northern Victoria Land was carried out to obtain glaciological information and climatic data in this Antarctic region. Sampling areas were accurately prospected to identify sites, located at different altitudes and distances from the sea, where the snow accumulation was not influenced by katabatic wind redistribution or summer melting. Stratigraphic, isotopic (δl8O) and chemical (H2O2, MSA and nssSO42−) profiles were mutually examined for dating purposes and to determine the mean snow-accumulation rates at three different stations. Annual accumulation rates of 85–420 kg m−2 a−1 were determined in the period 1971–92. An inverse pattern between accumulation rate and altitude was shown by the progression of the mean annual rates of 160, 203 and 260 kg m−2 a−1, respectively, in the highest, medium and lowest stations. The mean accumulation value of all northern Victoria Land data available, 170 kg m−2 a−1, represents a decrease of up to 35% with respect to the estimated value most widely used until now. Our accumulation value is very close to that required for a zero net surface mass balance according to ice discharge. A linear relationship with a gradient of 0.81‰ °C−1 has been found between mean δ18O values and mean annual surface temperature for different ice cores drilled in northern Victoria Land.

We have used ground-based radio-echo sounding (RES) profiles to reveal the spatial distribution of basal and internal ice properties across Siple Dome, West Antarctica, and under the dormant ice streams on its flanks. The RES-detected bed-reflection power, corrected for the effects of instrumentation and ice-thickness variation, is nearly constant across Siple Dome at a value suggesting spatially homogeneous basal properties of ice frozen to bedrock. Till, if present under the dome, must be thin (<0.1 m). The high basal reflectivity measured under now dormant “Siple Ice Stream” (SIS) and Ice Stream C suggests that they are underlain by either a thin (<0.05 m) water layer or a thick (>1 m) thawed or frozen till layer. The evidence that the dormant SIS is not frozen directly to underlying bedrock (but is separated by a water or till layer) is a further indication that it was once an active ice stream, and suggests that streaming motion may have ceased before the basal layer was frozen. The absence of a thick till layer beneath Siple Dome is consistent with its apparent stability as an inter-ice-stream ridge in the past and may suggest that it will remain as a stable limitation of ice-stream width in the future.

Empirical data suggest that the rate of calving of grounded glaciers terminating in water is directly proportional to the water depth. Important controls on calving may be the extent to which a calving face tends to become oversteepened by differential flow within the ice and the extent to which bending moments promote extrusion and bottom crevassing at the base of a calving face. Numerical modelling suggests that the tendency to become oversteepened increases roughly linearly with water depth. In addition, extending longitudinal deviatoric stresses at the base of a calving face increase with water depth. These processes provide a possible physical explanation for the observed calving-rate/water-depth relation.

A procedure is presented for using a simple flowline model to calculate the fraction of the bed that is thawed beneath present-day ice sheets, and therefore for mapping thawed, frozen, melting and freezing basal thermal zones. The procedure is based on the proposition, easily demonstrated, that variations in surface slope along ice flowlines are due primarily to variations in bed topography and ice–bed coupling, where ice–bed coupling for sheet flow is represented by the basal thawed fraction. This procedure is then applied to the central flowlines of flow bands on the Antarctic ice sheet where accumulation rates, surface elevations and bed topography are mapped with sufficient accuracy, and where sheet flow rather than stream flow prevails.

In East Antarctica, the usual condition is a low thawed fraction in subglacial highlands, but a high thawed fraction in subglacial basins and where ice converges on ice streams. This is consistent with a greater depression of the basal melting temperature and a slower rate of conducting basal heat to the surface where ice is thick, and greater basal frictional heat production where ice flow is fast, as expected for steady-state flow. This correlation is reduced or even reversed where steady-state flow has been disrupted recently, notably where ice-stream surges produced the Dibble and Dalton Iceberg Tongues, both of which are now stagnating.

In West Antarctica, for ice draining into the Pine Island Bay polynya of the Amundsen Sea, the basal thawed fraction is consistent with a prolonged and ongoing surge of Pine Island Glacier and with a recently initiated surge of Thwaites Glacier. For ice draining into the Ross Ice Shelf, long ice streams extend nearly to the West Antarctic ice divide. Over the rugged bed topography near the ice divide, no correlation consistent with steady-state sheet flow exists between ice thickness and the basal thawed fraction. The bed is wholly thawed beneath ice streams, even where stream flow is slow. This is consistent with ongoing gravitational collapse of ice entering the Ross Sea embayment and with unstable flow in the ice streams.