The paper focuses on studies of snow-pit samples and shallow firn cores taken during the 1995-96 and 1996-97 field seasons at Amundsenisen, Dronning Maud Land, Antarctica. The dating of the firn is based on the artificial tritium distribution in the snow cover and on several reference horizons identified by electrical measurements. The early 1964 through 1965 horizon is marked by the deposition of sulfate released to the atmosphere during the eruption of the Agung volcano in March 1963; this horizon was detected by dielectric profiling and electrical conductivity measurements: the proof by chemical analysis has still to be seen. At the ten investigated sites on Amundsenisen the 1964-65 horizon was identified 4.1-5.7 m below the surface. The accumulation rates on Amundsenisen are 41-91kg m-2 a−1 The cores are up to 100 years old A relationship between isotope content and the mean air temperature on a regional scale can be based on measurements of firn temperature at 10 m depth at the drilling sites Between Neumayer station at the coast and Heimefrontfiella, the temperature gradient of the deuterium content is 9.6%0 K−1. South of Heimefrontfiella, on the Amundsenisen plateau, it is only 5.5‰ K−1. Time series of yearly accumulation rates show no significant trend For the isotope records a significant trend to higher values with gradients of 0.1-0.2 δ2H‰a−1 can be seen in five of the ten time series

The spatial variability in snow accumulation varies between different regions in Dronning Maud Land, East Antarctica. This pattern cannot easily be explained by the single action of parameters such as distance to open sea, surface elevation or slope. In 1996-97 we mapped snow-layer depths within the top 11 m of the snowpack with a ground-based radar along a 500 km traverse on the polar plateau in central Dronning Maud Land. The results showed that the general accumulation pattern could be described by three major characteristic sections: a pronounced trend of decreasing net accumulation with increasing altitude from 2400 to 2840 m a.s.l.; relatively high erosion rates and occurrence of areas with net erosion at 2840-3140 m a.s.l.; and a slight trend of decreasing net accumulation with increasing altitude from 3140 to 3450 m a.s.l. The spatial variability in snow-layer depths showed a marked change around 3080 m a.s.l., with high variability at lower elevations and low variability at higher elevations. We also determined the spatial representativeness of 11 firn cores drilled along the traverse. In general, the representativeness of the cores was high. However, the core with the lowest representativeness underestimated the mean accumulation rate around the coring site by 22%. This shows that snow-radar data on spatial snow distribution are important for the interpretation of accumulation rates obtained from firn and ice cores.

During 1996-97 a European Project for Ice Goring in Antarctica (EPIGA) pre-site surveying traverse worked in the area between 70° S, 5° E and 75° S, 15° E in Dronning Maud Land. We present data obtained from 10 and 20 m deep firn cores drilled between the coast and 600 km inland (to 3450 m a.s.l.). The cores were analyzed for electrical conductivity measurements and total β activity to obtain accumulation data between known time horizons. In addition, some of the cores were analyzed for oxygen isotopes. Annual accumulation varies from 271 mm we. at Fimbulisen to 24 mm we at 2840 m a.s.l. Accumulation at core sites 2400-3000 m a.s.l. has increased by 16-48% since 1965 compared to the 1955-65 period. However, the core sites above 3250 m a.s.l. and one core location on the ice shelf show a decrease during the same period. Furthermore, no change can be detected at the most inland site for the period 1815-1996. In all the cores the last few years seem to have been some of the warmest in these records.

This paper presents new data, derived from an analysis of 16 firn cores, on the distribution of the accumulation rate and 18O content of near-surface layers in the eastern part of the Ronne Ice Shelf, Antarctica. The firn cores were drilled along the traverse route of the Filchner-V-Campaign in 1995. The traverse followed an ice flowline of the Foundation Ice Stream and reached the margin of the inland ice, an area which has not yet been investigated. On the ice shelf the accumulation rates decrease with distance from the coast. Ascending to the inland ice the accumulation rates again reach almost coastal values. This regional distribution is in agreement with the temperature gradient along the traverse. The 18 content of the near-surface layers is closely related to the 10 m firn temperature. They strongly decrease from the grounding line towards the inland ice. At the southernmost site at 1100 ma.s.l, the mean δ18O value of the firn decreases to –40‰. Ice with that isotopic signature was found in cores from the central part of the Ronne Ice Shelf just above the marine ice layer, indicating that it originates from this area. All ice deposited as snow further south was melted beneath the ice shelf after passing the grounding-line area. The time series of accumulation rate and 18O content reveal no climatic trend during the last 30-50 years.

The Lambert Glacier basin is a region of the East Antarctic ice sheet with a distinctive topography. Accumulation data derived from cane measurements are available from recent inland traverses in the western part of the basin, together with data on stable isotopes and visible stratigraphy of shallow cores and snow pits. These data suggest that snow accumulation has decreased since the 1930s by 16-37%, but that since the late 1980s it has been increasing. Temperature records from the nearby coastal station, Mawson, show no clear trend. Although the data are from only the western part of the basin, these results are quite different from those in many other regions of Antarctica. Further investigations are needed in this potentially exceptional region.

As part of the pre-site survey in Dronning Maud Land for the European Project for Ice Goring in Antarctica (EPICA), the spatial variability of snow-layer thickness and snow chemistry was studied at two geographically different ice-core drill sites. The study aimed to quantify error bars on accumulation rates derived from firn and ice cores. One site is located on the polar plateau at Amundsenisen (76° S, 8° W) and the other in the coastal area at Maudheimvidda (73° S, 13° W). Medium-deep ice cores (100 m) and shallow firn cores (10-20 m) were drilled and snow pits (0.5-2 5 m) were dug at each site. At Amundsenisen a large (16 m x 6 m x 2.5 m deep) snow pit was dug. Snow structure in this large snow pit was mapped using optical surveying equipment, and photographically documented. Samples for analysis of nine ions and oxygen isotopes were collected along one depth profile. Density and in situ electrical conductivity measurements were made along three depth profiles! Snow-layer variability was studied in two different areas and at two different scales. At a regional scale, measured by snow-radar soundings, the variability was 8% on the polar plateau and 45% in the coastal area. The variability at a micro-scale in the large snow pit was 9%. The results indicate that ice cores from the polar plateau are more representative for a larger area than ice cores drilled in the coastal area There is no doubt that there are significant error bars on high-resolution accumulation data received from firn and ice cores, especially from the coastal area, but averaging over tens of years reduces the error in accumulation estimates.

Six snow-pit records recovered from Siple Dome, West Antarctica, during 1994 are used to study seasonal variations in chemical (major ion and H202), isotopic (deuterium) and physical stratigraphic properties during the 1988-94 period. Comparison of δD measurements and satellite-derived brightness temperature for the Siple Dome area suggests that most seasonal SD maxima occur within ±4 weeks of each 1 January. Several other chemical species (H2O2, non-sea-salt (nss) SO42-, methanesulfonic acid and NO3-) show coeval peaks with SD, together providing an accurate method for identifying summer accumulation. Sea-salt-derived species generally peak during winter/spring, but episodic input is noted throughout some years. No reliable seasonal signal is identified in species with continental sources (nssCa2+ nss Mg2+), NH4+ or nssCl-. Visible strata such as large depth-hoar layers (>5 cm) are associated with summer accumulation and its metamorphosis, but smaller hoar layers and crusts are more difficult to interpret. A multi-parameter approach is found to provide the most accurate dating of these snow-pit records, and is used to determine annual layer thicknesses at each site Significant spatial accumulation variability exists on an annual basis, but mean accumulation in the sampled 10 km2 grid for the 1988-94 period is fairly uniform.

The geographic origin of Antarctic precipitation is important for ice-core isotopic interpretation as well as ice-sheet mass-balance calculations. Here we estimate these moisture origins with the NASA/Goddard Institute of Space Studies atmospheric general circulation model, under different climatic conditions. This model reasonably simulates the broad features of the present-day observed hydrological cycle, and indicates a subtropical to subglacial (30-60° S) latitudinal origin for the Antarctic precipitation. We use different climatic reconstructions, all based on CLIMAP, for the Last Glacial Maximum (about 21000 years ago), which differ by the latitudinal sea-surface temperature gradient and seasonality. CLIMAP conditions increase the latitudinal gradient and the sea-ice extent, with the consequence of slightly enhancing the low-latitude origins. Shifting the seasonal cycle of oceanic prescribed conditions has an important effect on the hydrological cycle but less on the precipitation origin. Prescribing cooler tropical sea-surface temperatures, which decreases the latitudinal gradient, makes the latitudinal contributions closer to modern ones and increases the dominant oceanic sources. Globally the origins of Antarctic precipitation do not change significantly, either annually or seasonally.

In February 1995 an automatic weather station (AWS) was placed on Thyssen Hohe, the south dome of Berkner Island, Antarctica. A fairly complete 3 year meteorological dataset of hourly average data was obtained. The mean annual temperature is about –24°C. The annual mass balance is about +180 mm we. Summer temperatures stay below 0°C, which implies that no melt takes place. Because the AWS is located on a dome, katabatic winds are not active, the wind direction is variable (directional constancy 0.38) and the wind speed relatively low (4.5 ms−1). Annual average variables are compared with data from Recovery Glacier AWS and Halley station.

The measurements are used to evaluate the surface energy fluxes for the 3 year period by using a surface energy-balance model. The annual average gain of energy from the sensible-heat flux ( + 10.8 W m–2) is balanced by a negative net radiative flux (–9.1 W m−2) and a small negative latent-heat flux (-1.7 W rrT2). The annual subsurface flux is small.

Glaciochemical analysis of surface snow samples, collected along a profile crossing the Antarctic ice sheet from the Larsen Ice Shelf, Antarctic Peninsula, via the Antarctic Plateau through South Pole, Vostok and Komsomolskaya to Mirny station (at the east margin of East Antarctica), shows that the Weddell Sea region is an important channel for air masses to the high plateau of the Antarctic ice sheet (>2000 m a.s.l.). This opinion is supported by the following. (1) The fluxes of sea-salt ions such as Na+, Mg2 + and CF display a decreasing trend from the west to the east of interior Antarctica. In |eneral, as sea-salt aerosols are injected into the atmosphere over the Antarctic ice sheet from the Weddell Sea, large aerosols tend to decrease. For the inland plateau, few large particles of sea-salt aerosol reach the area, and the sea-salt concentration levels are low (2) The high altitude of the East Antarctic plateau, as well as the polar cold high-pressure system, obstruct the intrusive air masses mainly from the South Indian Ocean sector. (3) For the coastal regions of the East Antarctic ice sheet, the elevation rises to 2000 m over a distance from several to several tens of km. High concentrations of sea salt exist in snow in East Antarctica but are limited to a narrow coastal zone. (4) Fluxes of calcium and non-sea-salt sulfate in snow from the interior plateau do not display an eastward-decreasing trend. Since calcium is mainly derived from crustal sources, and nssSO42- is a secondary aerosol, this again confirms that the eastward-declining tendency of sea-salt ions indicates the transfer direction of precipitation vapor.

A combination of modelling techniques is used in conjunction with the limited available observational data to examine Antarctic sea-ice changes with global warming over the past century. Firstly a coupled global climate model is forced by prescribing the effect of increasing greenhouse gases from last century to the present. Secondly the GISST (U.K. Meteorological Office global sea-ice and sea surface temperature) observational dataset is used to force an atmosphere-sea-ice model to compute changes in the Antarctic sea ice from last century to the present. Thirdly the global sea-surface-temperature (SST) anomalies derived from the coupled model are used to force the atmosphere-sea-ice model over the same period. The change in the Southern Hemisphere annual mean surface temperature simulated by the coupled model with greenhouse-gas forcing is about 0.6°C, which is similar to the observed change. Over the Antarctic (poleward of 60° S) the corresponding simulated change is about 0.7°C, which also appears compatible with observations. The reduction in summer sea-ice extent simulated by the CSIRO (Commonwealth Scientific and Industrial Research Organisation) coupled model is 0.44° lat. which is, in general, less than the observed change. For the two SST forcing cases the changes are, in general, larger than indicated by the observations. It is concluded that future changes of reduced sea-ice extent from increasing greenhouse gases as simulated by the CSIRO coupled model are not expected to be overestimates.

In this paper we first summarise major findings of recent atmospheric studies of nitrogen and sulphur species present in the boundary layer of coastal Antarctic regions. We then discuss the implications of such atmospheric data for the interpretation of nitrate, ammonium, methanesulphonate and sulphate records in deep ice cores extracted from central Antarctica in terms of past atmospheric chemistry changes.

Through comparison of snow-pit NO3– profiles from central East Antarctica and the northern Qinghai-Tibetan Plateau (QTP), we conclude that NO3– peaks in the uppermost surface snow layers in central East Antarctica are not related to an atmospheric signal and that they need to be accounted for by post-depositional effects. Such effects, however, are not found in the snow-pit NO3– profiles from the northern QTP. NO3– can be deposited as a gas (HNO3–) or as a neutral salt, particularly by reaction with ammonia to form NH4NO3, or fixed by sea salt or terrestrial dust (Mulvaney and others, 1998). Thus, a difference in speciation between NO3– in snow layers in East Antarctica and at the northern QTP is suggested as the reason for the different post-depositional processes of NO3– in the two areas.

Primary and secondary marine aerosol contributions to snow composition in Terra Nova Bay Antarctica, were evaluated by chemical analysis of surface and snow-pit samples. A seasonal pattern of marine inputs, very clear for secondary aerosol (mainly constituted by nssSO4- and methanesulphonic acid (MSA) coming from phytoplanktonic activity) and less evident for sea spray (Na+, CF, Mg2+) was shown by snow-pit samples. Altitude and distance from the coast were found to be the main parameters affecting seasonal change in the composition of snow collected within about a 200 km radius of the Terra Nova Bay Italian base. Using Na+ as a sea-spray indicator and MSA as a biogenic marker, fractionating aerosol effects (which are altitude-induced) as well as source contributions for Cl- and nssSO42- were found. A nssSO42- /MSA ratio of 2.7 (w/w), indicating their distribution from the common source dimethylsulphide, was calculated from Analysis of fresh summer snow collected in horizontal and vertical transects in northern Victoria Land.

A 2.7 m snow pit was sampled in the Lambert Glacier basin, about 650 km from the coast. Concentrations of methanesulfonic acid (MSA) and non-sea-salt SO42- (nssSO42-) show distinct variation with depth in the snow pit. The seasonal variation of MSA is used to date the snow-pit samples. MSA concentration is in the range 1.922.0 μgkg−1 with a mean of 7.0 μgkg-1. The mean nssSO42- concentration is 41.5 μgkg−1 with a range of 0-87 μgkg−1. The nssSO42- ∼ concentration was calculated by subtracting the sea-salt sulfate contribution using Nai as a conservative tracer. The nssSO42– contributes about 80% of the total sulfate in the snow pit. The mean mass ratio of MSA to nssSO42- 0.176, is much higher than that measured in the low to mid-latitudes, but is very close to the ratio in snow at the South Pole. El Chichon and Hudson volcanic eruptions do not significantly perturb the nssS042- concentrations at this site. MSA concentrations are anticorrelated with sea-ice area for the South Indian Ocean sector between 40° and 90° E, suggesting enhanced MSA production accompanying low sea-ice coverage.

As part of the European Project for Ice Goring in Antarctica, a new deep ice core is being drilled at Dome C. Two electrical methods have been used on the core drilled so far: a new design of electrical conductivity method (EGM) instrument, and a traditional dielectric profiler. The two profiles are very similar in both peaks and background, consistent with acidity being the dominant influence on both in this part of the record. The Dome C records have been compared to EGM records from Vostok, and a tentative match has been made between them. This suggests a long-term average ratio of accumulation rate of 1.36 between this Dome C core and Vostok, and that the Dome C core analyzed so far (to 358 m) probably includes the very end of the Glacial-Holocene transition.

This paper reports the lead concentration and flux (where accumulation rate is available) along the route of the 1990 International Trans-Antarctic Expedition. The lead concentration in Larsen Ice Shelf and Antarctic Peninsula, the western part of the route, was 7.4 ± 4.1 pg g−1. The lead concentration in East Antarctic snow (South Pole to Mirny station) was 2 3 times higher than that in West Antarctica (Larsen Ice Shelf to South Pole). Taking into account the difference in site conditions, the difference between the above value over this area in 1989 and the value of 6.3 ±3.3 pg g−1 at a site within this area in 1980 (Wolff and Peel, 1985) is not significant. Because the relative contribution of soil dust, volcanoes and the oceans to lead concentration in Antarctica is about 0.5 pg g−1 under modern climatic conditions (Boutron and Patterson, 1987), it is believed that the lead in Antarctic surface snow is dominated by pollution input. The lead-concentration increase from west to east over the trans-Antarctic route suggests that remote Antarctica has been impacted by anthropogenic activities. The lowest lead flux (0.064 ng cn-2 a−1) was on the Antarctic Plateau, mainly reflecting the background global pollution. The mean flux of 0.273 ng cnT2 a−1 in the western part of the route (Larsen Ice Shelf to the Ellsworth Mountains) may result from the pollution input from the Southern Hemisphere. In addition to the influence of global and/or hemispheric pollution, local activities (notably the use of leaded gasolene) appear to have affected the region from Pionerskaya to Mirny.

Snow samples collected in the 1996 austral summer at South Pole show that sulfate concentrations in snow and, by inference, sulfur aerosol concentrations in the Antarctic atmosphere were elevated from the end of 1991 to mid-1994 over a stable, non-volcanic background. The new data support earlier findings that the June 1991 Pinatubo eruption and the Hudson eruption in the same year deposited volcanic sulfate and tephra in South Pole snow, and provide strong evidence of the global distribution of volcanic materials from the Pinatubo eruption. In this study, snow samples were taken in six snow pits spatially distributed around the South Pole station in order to evaluate the local spatial variability of volcanic signals due to glaciological variables such as snow-accumulation rates and snow redistribution by wind after initial deposition. The results indicate that Pinatubo sulfate flux varies by as much as 20% throughout a 400 km2 area centered around the South Pole station. This glaciological variability probably represents the likely range of volcanic signals due to variations in snow deposition and post-depositional changes.

The Pinatubo eruption provides an unprecedented opportunity to estimate aerosol mass loadings by explosive volcanic eruptions found in Antarctic ice cores via a quantitative relationship between aerosol mass loadings and sulfate flux in Antarctic snow. Here the satellite-estimated Pinatubo SO2 emission and the measured volcanic sulfate flux in snow, with an assumed linearly quantitative relationship, are used to calculate SO2 loadings for several well-known volcanic eruptions in the past 300 years covered by a shallow (42 m) South Pole firn core drilled in 1996. The errors for the calculated mass loadings are estimated by means of the glaciological variability associated with Pinatubo volcanic flux.

A 42.2 m firn core was collected at the Hercules Névé plateau (100 km inland and 2960 m a.s.L), northern Victoria Land, during the 1994-95 Italian Antarctic Expedition. Chemical (Cl–, NO3–, SO42–’; δ18O δ18O δ18O; m-2a-1) and isotope (5180) analyses were performed to evaluate the snow-accumulation rate at this site. Tritium measurements were performed in the upper part of the core to narrow down the dating of the core.

High nssSO42- concentrations seem to be related to some explosive volcanic eruptions, such as Tambora (AD 1815) and the preceding event called "Unknown" (AD 1809), Coseguina (AD 1835), Makjan (AD 1861), Krakatoa (AD 1883) and Tarawera (AD 1886).

A comparison between the seasonal variations observed in the isotope and chemical profiles was carried out in order to reduce the dating uncertainty, using the tritium and the volcanic markers as time constraints. A deposition period of 222 years was determined.

The 3 year smoothed «5180 profile shows more negative values from the bottom of the core (dated AD 1770) throughout the 19th century, suggesting "cooler" conditions, in agreement with other East Antarctic ice-core records! Subsequently, a general increase in δ180-values is observed.

The calculated average snow-accumulation rates between the above-mentioned time markers are 111-129 kg m-2a-1.