Stable isotopes in ice cores are used as a proxy for the temperature at the time of snow formation. Where net accumulation rate is relatively high, snow is buried quickly and initial isotopic values are preserved. However, in low-accumulation areas, snow is exposed to lengthy vapor exchange with the atmosphere. the original isotopic signature of this snow may be modified by equilibration with atmospheric water vapor in the boundary layer over the snow surface in summer. We estimate the characteristic times for equilibration by using an electrical resistor network analogue. Warm, windy summers and low accumulation rate enhance equilibration. Although equilibration of the complete snowpack is unlikely, significant post-depositional change may occur in some Antarctic environments.

We report on the first application of a new technique in ice-core research, based on direct absorption infrared laser spectrometry (LS), for measuring 2H, 17Oand 18O isotope ratios. the data are used to calculate the deuterium excess d (defined as δ2H– 8δ18O) for a section of the Dye-3 (south Greenland) deep ice core around the Bølling transition (14 500 BP). the precision of LS is slightly better than that of most traditional methods for deuterium, but not for the oxygen isotopes. the ability to measure δ17O is new and is used here to improve the precision of the δ18O determination. Still, the final precision for δ18O remains inferior to traditional isotope ratio mass spectrometry (IRMS). Therefore, deuterium excess was calculated from a combination of the LS and IRMS isotope determinations.

Stratigraphic dating of ice cores by identification and counting of annual cycles in, for example, chemical measurements requires skill and experience. the work presented here investigates a method of data enhancement which is a first step towards an automated and more objective method of annual-layer counting. the method of dynamical decorrelation is briefly introduced and is applied to data from Site D and NorthGRIP in central Greenland. With this method the measured data series are decomposed into a number of independent source series, one of which exhibits a more pronounced annual variation than the input data themselves. the annual variation is more regular in that (1) some double and triple peaks in the measured series are replaced by single peaks in the extracted signal, and (2) the resulting annual peaks have a much more uniform height. A simple method of determining the number of annual peaks in a series is set up. Using this method, it is shown that it is easier to determine the number of annual peaks in the series produced by dynamical decorrelation than in the original data series. Dynamical decorrelation may thus be used to improve data series prior to dating.

Old ice for palaeoenvironmental studies retrieved by deep core drilling in the central regions of the large ice sheets can also be retrieved from the ice-sheet margins. the δ18O content of the surface ice was studied at 15 different Greenland ice-margin locations. At some locations, two or more records were obtained along closely spaced parallel sampling profiles, showing good reproducibility of the records. We present ice-margin δ18O records reaching back to the Pleistocene. Many of the characteristic δ18O variations known from Greenland deep ice cores can be recognized, allowing an approximate time-scale to be established along the ice-margin records. A flowline model is used to determine the location on the ice sheet where the margin ice was originally deposited as snow. the Pleistocene–Holocene δ18O change at the deposition sites is determined by comparing the δ18O values in the ice-margin record to the present δ18O values of the surface snow at the deposition sites. on the northern slope of the Greenland ice sheet, the Pleistocene–Holocene δ18O change is about 10‰ in contrast to a change of 6–7‰ at locations near the central ice divide. This is in accordance with deep ice-core results. We conclude that δ18O records measured on ice from the Greenland ice-sheet margin provide useful information about past climate and dynamics of the ice sheet, and thus are important (and cheap) supplements to deep ice-core records.

Barnes Ice Cap, Baffin Island, Canada, is a remnant of the Laurentide ice sheet that separated from it about 8500 years ago. Owing to recession of the ice cap during the Holocene, Pleistocene-age ice is now exposed along the margin in a distinctive bubble-rich white band. δ18O variations across the white ice resemble those in Canadian Arctic ice cores, suggesting that Barnes Ice Cap preserves a climatic record through the last glacial period, possibly reaching back into the previous (Sangamon) interglacial. the δ18O shift at the Wisconsin–Holocene transition (15‰) exceeds that in other Canadian and Greenland records and cannot be explained solely in climatic terms. A steady-state model reconstruction of the Laurentide ice sheet during the Last Glacial Maximum suggests that Late-glacial strata in Barnes Ice Cap originated high up (>2400ma.s.l.) and far “inland” on the ice sheet, along a ridge that extended between the ancestral Foxe and Keewatin ice domes.

The isotopic and chemical signatures for ice-age and Holocene ice from Summit, Greenland, and Penny Ice Cap, Baffin Island, Canada, are compared. the usual pattern of low δ18O, high Ca2+ and high Cl– is presented in the Summit records, but Penny Ice Cap has lower than present Cl– in its ice-age ice. A simple extension of the Hansson model (Hansson, 1994) is developed and used to simulate these signatures. the low ice-age Cl– from Penny Ice Cap is explained by having the ice-age ice originating many thousands of km inland near the centre of the Laurentide ice sheet and much further from the marine sources. Summit’s flowlines all start close to the present site. the Penny Ice Cap early-Holocene δ18O’s had to be corrected to offset the Laurentide meltwater distortion. the analysis suggests that presently the Summit and Penny Ice Cap marine impurity originates about 500 km away, and that presently Penny Ice Cap receives a significant amount of local continental impurity.

Spatial correlations between Alpine high-elevation and European low-elevation instrumental air temperatures are computed to assess the spatial representativity of a high-Alpine ice-core isotope proxy temperature record. the correlation analyses indicate that air-temperature records at Alpine ice-core drill sites are representative for central Europe, particularly in summer. While Alpine ice cores generally show a large scattering in the conserved section of the year, long-term records from low-accumulation sites consist almost solely of summer precipitation and thus reflect isotope proxy summer-temperature variability. However, correlation between seasonal and annual instrumental air temperature indicates that summer temperature variability provides an adequate approach to annual temperature variability. Comparison of long-term ice-core δ18O records from Colle Gnifetti (4450ma.s.l.), Monte Rosa, Western Alps, with local instrumental summer temperatures inferred from an instrumental network shows good agreement in the long-term scale. Thus Alpine long-term ice-core δ18O records are representative for central European air-temperature variability.

Ice cores recovered for paleoclimatic and/or paleoenvironmental reconstructions in the Tien Shan and Qinghai–Tibetan Plateau often encounter cracks. Although we expect that cracks opened to surface meltwater will inevitably change ice-core records, we do not know how and to what extent records are influenced. An ice core retrieved from glacier No. 1 at Ürümqi river head, Tien Shan, China, exhibits a crack nearly 2.5 m long that has admitted meltwater, forming secondary ice within the fracture. A small inclusion of the infiltrated ice in sampling is shown to reduce δ18O by an extent of Holocene vs Last Glacial Maximum while enhancing significantly the pH, conductivity and the following ionic species: CH3COO–, and CO(COO)22–. of the parameters increased, and HCOO– are the most affected, being enhanced nearly six-fold in the fractured section compared to the non-fractured sections, followed by CO(COO)22– and electrical conductivity measurement (ECM). Despite the alteration, primary fluctuations of some parameters are still recognizable. This suggests that if the infiltrated ice can be avoided in the sampling operation, ice cores with cracks may still provide authentic records. This shows the need to pay close attention to physical characteristics of ice cores in order to identify such secondary ice.

Sampling was carried out at Baishui glacier No. 1, the largest glacier on Yulong mountain, China, during the summers of 1999 and 2000, to investigate the spatial variations of oxygen isotopes in the atmosphere–glacier–river system. the results confirm that there is an apparent inverse relation between the oxygen isotopic composition of precipitation and air-temperature/precipitation amount in this region, with lower δ18O values when the amount of precipitation and air temperature in summer is higher, due to the influence of intense monsoon climate on the study area. There are marked differences in the δ18O values of winter-accumulated snow, glacial meltwater, summer precipitation and the glacier-fed river water. Spatial and temporal variations of isotopic composition are controlled by varied weather conditions at different altitudes. Isotopic depletion or fractionation occurs during phase changes, snow-to-ice and ice-to-meltwater transformations and runoff processes.Variations of stable isotopes in glacier runoff can indicate variations of sources of supply. Ionic changes occur as a result of meltwater contact with glacier bed materials.

To investigate recent variability in accumulation and δ18O, we synthesize data from five snow cores, covering the period 1932–96, from the sector 16˚38’ W–4˚48’ E in coastal Dronning Maud Land (DML), Antarctica. the δ18O records from the different sites are remarkably similar and suggest a common stable moisture source for this coastal section of DML. While the accumulation pattern is local, and specific features restricted to the individual sites, the overall accumulation pattern is related to the temperature variability as indicated by coastal instrumental records. Accumulation and δ18O correlate between 1955 and 1985 but deviate thereafter, with the proxy-temperature record showing a positive trend while accumulation decreased. This occurs at the same time as an increase in sea-ice extent in the area, which may have resulted in circulation changes and more northerly storm paths. Both stacked accumulation and δ18O records show that large-scale atmospheric signals, as well as some pronounced individual events, are recorded in DML coastal ice cores.

Continuous, detailed isotope (δD and δ18O) profiles were obtained from eight snow pits dug in the vicinity of Vostok station, Antarctica, during the period 1984– 2000. In addition, snow samples taken along the 1km long accumulation-stake profile were measured to determine spatial variability in isotope composition of recent snow. the stacked δD time series spanning the last 55 years shows only weak correlation with the mean annual air temperature recorded at Vostok station. Significant oscillations of both snow accumulation and snow isotope composition with the periods 2.5, 5, 20 and, possibly, ~102 years observed at single points are interpreted in terms of drift of snow-accumulation waves of various scales on the surface of the ice sheet.

In the framework of the PNRA–ITASE (Programma Nazionale di Ricerche in Antartide–International Trans-Antarctic Scientific Expedition) project, during the field season 1998/99, surface snow (1m cores and pits) and shallow firn cores (10–50m) were collected along a traverse from Terra Nova Bay (northern Victoria Land) to Dome C (East Antarctic ice sheet). Results of chemical, tritium and stable-isotope composition are presented here for the 1 m cores, some snow pits and the first 2 mof some shallow firn cores. the δ18O values show a regular trend with altitude, and the regression line between δ18O and surface temperature is δ18O = 0.99T (˚C) – 0.67. Primary aerosol components such as Na+, Cl–, Ca2+,Mg2+ and K+ show high concentrations decreasing with increasing altitude in the first 250–350km from the coast. At greater distances, concentrations of these species remain more constant. NO3– concentration shows an irregular profile with a progressive decreasing trend as altitude increases. Non-sea-salt (nss) SO42– concentration decreases up to about 250 km from the coast, increases 250–770 km from the coast and remains relatively constant in the most remote stations. Methanesulphonate (MSA) concentration shows high variability. the MSA/nssSO42– ratio exhibits a decreasing trend 250–550km from the coast. With increasing distance, the ratio shows moderate oscillations. nssCl– concentration shows a progressive increase as distance from the coast increases, in agreement with the increasing influence of HCl on the Cl– budget of the inland Antarctic atmosphere. Post-depositional re-emissions of Cl– and NO3– were found at stations characterized at the surface by long-term accumulation hiatus (wind crusts). the chemical-species distribution is consistent with the presence in the studied area of local and long-range transport processes, post-depositional effects and snow-accumulation variations observed along the traverse.

The European Project for Ice Coring in Antarctica (EPICA) includes a comprehensive pre-site survey on the inland ice plateau of Dronning Maud Land. This paper focuses on the investigation of the 18O content of shallow firn and ice cores. These cores were dated by profiles derived from dielectric-profiling and continuous flow analysis measurements. the individual records were stacked in order to obtain composite chronologies of 18O contents and accumulation rates with enhanced signal-to-noise variance ratios. These chronologies document variations in the last 200 and 1000 years. The18O contents and accumulation rates decreased in the 19th century and increased during the 20th century. Using the empirical relationships between stable isotopes, accumulation rates and the 10 m firn temperature, the variation of both parameters can be explained by the same temperature history. But other causes for these variations, such as the build-up of the snow cover, cannot be excluded. A marked feature of the 1000 year chronology occurs during the period AD 1180–1530 when the 18Ocontents remains below the long-term mean. Cross-correlation analyses between five cores from the Weddell Sea region and Dronning Maud Land show that 18O records can in some periods be positively correlated and in others negatively correlated, indicating a complex climatic history in time and space.

Analyses of air extracted from polar ice cores are the most straightforward method of reconstructing the atmospheric concentrations of greenhouse gases and their variations for past climatic epochs. These measurements show that the concentration of the three most important greenhouse gases (other than water vapour) CO2, CH4 and N2O have steadily increased during the past 250 years due to anthropogenic activities (Prather and others, 2001; Prentice and others, 2001). Ice-core results also provided the first evidence of a substantial increase in the concentration of the three gases during the transition from the last glacial epoch to the Holocene (Raynaud and others, 1993). However, results from different cores are not always in agreement concerning details and small, short-term variations. the composition of the air enclosed in bubbles can be slightly changed by fractionation during the enclosure process, by chemical reactions and/or biological activity in the ice and by fractionation during the air extraction. We compile here several records with short-term variations or anomalies and discuss possible causes, taking into account improved analytical techniques and new results.

A compilation of nitrate (NO3–) data from Greenland has shown that recent NO3– concentrations reveal a temperature dependence similar to that seen in Antarctica. Except for sites with very low accumulation rates, lower temperatures tend to lead to higher NO3– concentrations preserved in the ice. Accumulation rate, which is closely linked to temperature, might influence the concentrations preserved in snow as well, but its effect cannot be separated from the temperature imprint. Processes involved in NO3– deposition are discussed and shown to be temperature- and/or accumulation-rate-dependent. Apart from scavenging of nitric acid (HNO3) during formation of precipitation, uptake of HNO3 onto the ice crystal’s surface during and after precipitation seems to contribute further to the NO3– concentrations found in surface snow. Post-depositional loss of NO3– from the top snow layers is caused by release of HNO3 and by photolysis of NO3–. It is suggested that photolysis accounts for considerable losses at sites with very low accumulation rates. Depending on the site characteristic, and given that the temperature and accumulation-rate dependence is quantified, it should be possible to infer changes in atmospheric HNO3 concentrations.

During the past 220 years, prominent signals of non-sea salt sulfate ion (nssSO42–) concentration exceeding the background level, including both marine biogenic and anthropogenic SO42–, were found in shallow ice cores from site H15 in East Antarctica and Site-J in southern Greenland. They were mostly correlated with past explosive volcanic eruptions. on the basis of this result and published results of shallow ice cores and snow pits at various locations on the Antarctic and Greenland ice sheets, eight common signals were found, of which six were assigned to the following explosive eruptions: El Chichόn, Mexico, in 1982; Agung, Indonesia, in 1963; Santa Maria, Guatemala, in 1902; Krakatau, Indonesia, in 1883; Cosiguina, Nicaragua, in 1835; an unknown volcano between 1831 and 1834; Tambora, Indonesia, in 1815; and an unknown volcano in 1809. Volcanic eruptions which have a potential to imprint their signals in both the Antarctic and Greenland ice sheets were characterized by (1) location in low latitudes between 20˚N and 10˚ S, and (2) eruption column height ≥25 km, corresponding to a volcanic explosivity index (VEI) ≥5.

Optical microscopy, X-ray topography, scanning electron microscopy and X-ray spectroscopy have been used for microstructural analysis of ice from 94 m depth from Greenland Ice Sheet Project 2. the ice had a high density of air bubbles, a 2 mm grain-size and contained a high dislocation density in most grains (>1×1010 m–2), although a few lower-dislocation-density grains were present. the main impurities, Na and Cl, were found in several grain boundaries and triple points, often as filaments. Na and Cl were also found in the lattice, but to a lesser extent, along with S, Mg, K and Ca. These observations are compared to previous studies of dislocations and impurity locations in natural ice.

The ice depth–age relationship used for the Dye 3 (Greenland) core implies strong deformation of the silty ice present at the bottom. This paper focuses on the study of the relationship between carbon dioxide and methane concentration in the silty ice. In a CO2– CH4 diagram, sample points are well aligned along different straight lines with striking differences in slopes and concentration ranges, depending on their spatial location in the core. These slopes are interpreted as resulting from distinct mixing events between ice-sheet ice flowing from upstream and ice of local origin. Occurrence of these various mixing events suggests a complex deformation history for the basal part of the ice sheet. It involves a progressive build-up of the high-deformation layer along the basal flowline up-glacier from Dye 3, reaching 425 mthickness at the site.

A novel flow-through microparticle detector was deployed concurrently with continuous flow analyses of major ions during the North Greenland Icecore Project 2000 field season. the easy-handling detector performs continuous counting and sizing. In this deployment the lower size-detection limit was conservatively set to 1.0 μm equivalent spherical particle diameter, and a depth resolution of ≤1cm was achieved for microparticle concentrations. the dust concentration usually followed the Ca2+ variability. Here results are presented from an inspection of the Ca/dust mass ratio in 23 selected intervals, 1.65 m long each, covering different climatic periods including Holocene and Last Glacial Maximum (LGM). A (Ca2+)/(insoluble dust) mass ratio of 0.29 was found for the Holocene and 0.11 for LGM. Changes to the Ca/dust ratio occur on an annual to multi-annual time-scale exhibiting the same pattern, i.e. a lower Ca/dust ratio for higher crustal concentrations. Moreover, the Ca2+/dust ratio may increase significantly during episodic events such as volcanic horizons due to enhanced dissolution of CaCO3. This calls into question the notion of deploying Ca2+ as a quantitative mineral dust reference species and shows the importance of variable source properties or fractionating processes during transport and deposition.

Ion chromatography (IC) is a widely used technique for analyzing ice cores for ions like Na+, NH4+, K+, Mg2+, Ca2+, F–, MSA–, Cl–, NO2– and SO42– that are present in polar ice cores at ppb level. By using sample preconcentration and an optimized separation technique, we have been able to detect Li+ in ice-core samples in concentrations as low as 0.0001 μeq kg–1 or 0.7 ppt by IC. During routine analysis of ions in ice cores, the lithium content has been evaluated and recorded. the IC technique used in these measurements and some exemplary IC data from the Greenland Icecore Project (GRIP) and the North Greenland Icecore Project (NorthGRIP) ice cores are presented. By these data we introduce Li+ concentration as a new parameter in the analysis of ice cores. Like other ions, Li+ reflects climatic changes and shows seasonal cycles. on the basis of the geochemistry of lithium, we suggest that Li+ measured in the Greenland ice cores is derived from mineral dust. However, data from the NorthGRIP ice core representing the 8.2 kyr BPHolocene cold event show a strong Li+ signal that does not correlate with any other ionic component measured. This means that the lithium content in ice cores is a signal with its own pattern, which is not yet understood.