Inventory of EU Marine Climate Change Research

New search

EPICA - European Project for Ice Coring in Antarctica

Summary information

Funding:FP5 - Research project
Total cost:7058344
Ec contribution:2406164
Start date:2001-05-01
End date:2004-05-01
Duration:36 months
Coordinator:Heinrich Miller (
Organisation:Alfred Wegener Institute for Polar and Marine Research – Germany
Themes:Sea level rise; ice melting
Project name:EPICA - European Project for Ice Coring in Antarctica
Project summary:Abstract
The overall objective of EPICA is to reconstruct a continuous, highly resolved history of global climate and environmental changes from centuries to several hundred thousand years. For this, two ice cores are drilled in East Antarctica, one in the central part at Dome C (75°06'S, 123°21'E), the other in the Atlantic sector at Kohnen station in Dronning Maud Land (75°00'S, 00°04'E). The scientific objectives of the two complementary ice-core records are twofold:
• To achieve the best possible characterisation of the rapid changes during glacial and also during interglacial epochs.
• To provide a basis for understanding the mechanisms driving both global climate and the coupled biogeochemical cycles, and give a perspective for assessing current changes and associated feedbacks. The longest records from Antarctica will be recovered from central areas of East Antarctica and will be especially suitable for examining the forcing mechanisms driving the principal climatic cycles throughout the most recent geological period. Evidence on the associated responses of the ice sheet will also be obtained, which have implications for global sea-level changes.
Project outputs:Scientific achievements
The project recovered the planned two deep ice cores at Dome C and in Dronning Maud Land (DML). The age of the ice cores, which do not yet reach the bottom of the ice sheet, covers appr. 810.000 (the oldest ice core recovered so far) and 180.000 years, respectively. The preliminary δD record along the entire Dome C core provided the key climate profile to develop its chronology in combination with numerical modelling. Down to 3.139 m the core represents 740.000 years, including all of the interglacial marine-isotope stage (MIS) 11, which was not completed in the Vostok record, and running through a further three complete 100-kyr cycles. We first note a clear change in the amplitude of glacialinterglacial changes before and after MIS 12, and by consistently lower maxima. As in the marineisotope records, the most striking feature is this greater amplitude of glacial-interglacial change in the period after termination V (appr. 430.000 ago). The variations of the atmospheric CO2 and CH4 concentrations were also reconstructed through this termination. The overall characteristics of the DML and Dome C stable isotope records look very similar. The difference in δ18O or δD values between Last Glacial Maximum and the Holocene optimum is comparable, whereas the variations in the glacial are larger in the DML core than in the Dome C core, indicating the influence of the South Atlantic. Temperatures for the Dome C site and the precipitation source area were reconstructed resulting in different trends during the glacial.

The low frequency deuterium-excess fluctuations are strongly influenced by earth obliquity and show a remarkable similarity with a high-resolution southeast Atlantic sea-surface temperature record. By comparing the Dome C isotope record with Vostok and Dome Fuji records, it was found that the East Antarctic climate was homogeneous over the last 80.000 years. The new EPICA ice core records clearly indicate that each of the Dansgaard/Oeschger events found in the GRIP ice core has an Antarctic counterpart. This is also confirmed by CH4 data, which even show that those events were also characteristic for the previous glacial epoch. The relationship between many of the chemical parameters (and hence the environmental parameters they represent) and Antarctic temperature stayed similar over 740.000 years, but some subtle differences emerge. A comparison of calcium (representing dust, and hence a crude proxy for iron) with CO2 data for Dome C across warm events A1 to A4 within the last glacial period puts a limit on the extent of iron fertilization in the Southern Ocean. Use of highresolution sulfate profiles to tie the Vostok and Dome C cores together shows that the ratio of the snow accumulation rate at the two sites varies between glacial and interglacial periods. This unexpected result calls for a re-examination of the hypothesis that accumulation rates are determined only by the site temperature.

The synchronisation of the EPICA cores with the Taylor Dome core using CO2 concentrations shows that there are accumulation rate changes on some Antarctic locations. Significant advances were made in explaining the meaning of sea salt concentrations in ice cores. The sea ice surface, rather than open water, is a significant source of the sea salt seen in the cores. This opens up the possibility that, on long timescales, the sea salt concentration in ice cores can indicate changes in sea ice productivity. An interpretation of sea salt in the DML core has allowed an estimate of the persistence of the Antarctic Circumpolar Wave to be made over a 2000 year time period. For the greenhouse gases CO2, CH4, and N2O detailed records between 220.000 and 200 years BP are available in unprecedented resolution and precision. The CO2 record over the transition from the last glacial epoch to the Holocene is so detailed that it provided important information about the mechanisms responsible for the CO2 increase. Further remarkable are the reconstruction of the CH4 minimum at 8.200 years BP. It was also possible to evaluate a more precise interpolar gradient of the atmospheric CH4 concentrations. The Dome C record for δ13C on CO2 exhibits, besides a general increase of about 0.15 ‰ over the Holocene, three millennial scale variations of the order of a few tenths of a permil, which are unexpected, based on model calculations. It appears that the δ13C results are much less affected by fractionation during incomplete gas extraction than CO2 concentration measurements. The record suggests that δ13C could be correlated to the atmospheric 14C concentration.

The δ15N measurements on N2 provide information on the thickness of the diffusive firn layer. As has been observed in other cores from low accumulation sites, the δ15N trend from Dome C for the transition from the last glacial to the Holocene epoch is opposite from that predicted by firn models. The pore space of snow, firn and ice was reconstructed by X-ray micro-computer tomography showing that the original grain size determines the densification rate of polar firn. The grain-size profiles of the two cores show a strong correlation with climate as a result of grain boundary pinning by insoluble (dust) particles. A line scanning system documented the stratigraphy of the ice cores. Scanning electron microscope studies revealed the complexity of impurity distribution in the ice. The climate record on a depth scale from ice cores, has to be transformed to an age scale. As the DML core is drilled on a flank position, effects of horizontal advection have to be taken into account. With a newly developed 3-D thermo-mechanic nested ice flow model the DML core was dated by Lagrangian backtracing carried out by consecutive cubic spline interpolation of a particle’s location using the reversed 3-D velocity field obtained from forward experiments. According to the model results the last three glacial cycles are contained within the uppermost 97% depth. The general properties of atmospheric transport towards Antarctica under varying climatic conditions were studied by analysing tracer transit time climatologies based on the concept of tracer age in a LMDZ general circulation model. A state of the art regional atmospheric climate model has been used to model the climate and mass balance over Antarctica for the period 1957-2002.

Socio-economic relevance and policy implications
This project has contributed to a comprehensive study of how climate has changed over the last 740.000 years. This, and the detailed studies that underlie it, will provide critical input to improving climate models that will be used to predict future climate. One output that directly relates to policy has been an assessment of the role of iron fertilization in past CO2 change, clearly suggesting that iron fertilisation is likely to have only a limited role in mitigating future CO2 increases. EPICA has a triggering impact to the international collaboration within the Antarctic Treaty system.

Most parts of the project reached or even surpassed the promised milestones and deliverables. Good quality ice cores older than 800.000 years are available. The measuring methods are well developed. Important inferences about the nature of climate changes have already emerged and this project will contribute significantly to our understanding of linkages between components of the climate system.

Dissemination of results
The results were presented at national and international scientific conferences, including Euresco- Conferences, and published in peer-reviewed scientific journals. 62 papers were published, 30 are in press and another 17 presently submitted. National and international press releases and press conferences attracted the interest of the public media, which reported frequently about EPICA and paleo-climate studies with ice cores. Selected data are stored in public accessible databases dedicated to paleo-climate or ice cores.

A list of EPICA related publications is available at