Accurately determining atmospheric CO2 concentration from ice cores can be compromised by the production of CO2 by chemical reactions between impurities in the ice. In addition an accurate chronology is sometimes difficult to establish making it challenging to see the detailed progression in the rise of CO2 at glacial terminations. As well the resolution of the ice core record is often too low to establish precise dating, for example, to determine whether temperature leads or lags rising CO2.

The Vostok (Antarctica) ice core record, which extends over past 420,000 years, has revealed the major features of the last four deglaciations. The record shows increases of the CO2 concentration between 80 and 100 ppmv for each of the glacial terminations. During the last deglaciation the concentration of CO2 in the atmosphere increased by ∼40%.


In this study the EPICA Dome C (Antarctica) ice core has been used to measure a high resolution record of CO2 and methane concentrations over the last deglaciation. The ice core was sampled between depths of 350 and 580 m, covering the period from 22,000 to 9,000 years before the present.

The chronology or age scale for the ice, as well as for the age of the enclosed air, is estimated to be reliable to within 200 years back to 10,000 years ago and to within 2000 years back to 41,000 years ago. The age of the ice in which air bubbles are captured is not the same at the age of the air in the bubbles, because air can diffuse within the surface snow/ice mixture called the firn. A fairly complex calculation was required to calculate the difference in ages between air and the encapsulating ice.

In addition the possibility of CO2 enrichment by chemical reactions between impurities in the Dome C ice core was carefully investigated. Because of the physical characteristics of the Dome C ice it was concluded that this was not an important factor and a correction for this effect was not required. This record is an accurate representation of the atmospheric CO2 concentrations.


The analysis reveals that atmospheric CO2 concentration increased from 189 ppmv between 18,100 and 17,000 years ago to 265 ppmv between 11,100 and 10,500 years ago. The total increase was found to be 76 ppmv. The rise in CO2 concentration was found to progress in four steps.

IntervalPeriod(years B.P.)CO2 at start(ppmv)CO2 at end(ppmv)CO2 rate(ppmv/1000 years)NorthSouth
I17.0 to 15.418921920
II15.4 to 13.82192398
III13.8 to 12.3239237–1Bølling-Allerød(B/A) warmingAntarctic Cold Reversal(ACR)cooling
IV12.3 to 11.223725920Younger Dryas(YD)coolingAntarctica warming

Within this framework there are two periods of very rapid CO2 increase. These periods could have been shorter, even less than a few centuries, because of the uncertainty in the chronology of the enclosed air.

Age (ky B.P.)Duration(years)Increase(ppmv)
13.8300 or less8
11.2200 or less6

Deuterium abundance (delta-deuterium) is a proxy for Antarctic surface air temperature. Comparing the CO2 record to the surface air temperature reveals a close correlation. The resolution is insufficient to determine whether there is a lag between the temperature and CO2 concentration, but it is possible estimate the times at which temperature and CO2 began to rise. Surface air temperature began to rise about 17,800 years ago. CO2 lagged and began to rise about 17,000 years ago. The start of the CO2 increase lagged the start of the temperature rise by 800 years. An uncertainty analysis reveals that the lag could be as low as 200 years or as much as 1400 years.

Temperature, CO2 and Methane over last deglaciation
Temperature, CO2 and Methane over last deglaciation
Temperature, CO2 and CH4 from Dome C ice core
Surface temperature (delta-deuterium) - solid curve
CO2 concentration - solid circles
Methane concentration - diamonds
Younger Dryas (YD) and Bølling-Allerød (B/A) events - shaded bars

The rise of CO2 and methane during the deglacial period are similar in some respects. They both start to rise at the same time. During the transition between intervals II and III there is a fast rise in both CO2 and methane. During Interval III, which corresponds to the Bølling/Allerød (B/A) warm phase in the North Atlantic region and to the Antarctic Cold Reversal (ACR), there is a small, slow decrease in the CO2 concentration, whereas the methane rises dramatically to a level near the concentration after the deglaciation. During Interval IV, which corresponds to the cooling Younger Dryas (YD) epoch in the North Atlantic region and to the warming interval after the ACR in Antarctica, there is a continuous CO2 increase terminated by a fast CO2 rise at the transition to the current warm period. During this period methane concentration drops ~200 parts per billion by volume (ppbv), returning to concentrations comparable to Interval II.

The parallel rise in methane and CO2 in interval I is interesting because the sources of CO2 and methane are different. The methane increase in Interval I is also seen in the Greenland ice core record (GRIP). The methane rise is generally ascribed to changes in the extent and activity of wetlands in northern latitudes and the tropics.

The fast increases of CO2 and methane concentrations between intervals II and III, about 13,800 years ago on the Antractica time scale, correspond to the fast warming in the Northern Hemisphere observed at 14,500 years ago using the the Greenland chronology. This warming is thought to be caused by enhanced formation of North Atlantic Deep Water and it has been suggested that the sudden CO2 increase could have been caused by a reduction in stratification in the Southern Ocean which resulted in upwelling and CO2 ventilation. The methane increase, on the other hand, is thought to have been caused by an expansion of wetlands in the tropics and northern latitudes.

CO2 decreased slightly during interval III and then increased during interval IV. Methane concentration follows the temperature evolution of the Northern Hemisphere in intervals III and IV. The accelerated CO2 increase at the end of interval IV is thought to be connected to the fast warming in the Northern Hemisphere rather a climate change in the Southern Hemisphere.

It is important to note that the CO2 increase in interval I, which occurred before any substantial warming in the Northern Hemisphere, supports the view that the Southern Hemisphere was the source of the CO2 increase.

The results of this study support the hypothesis that the Southern Ocean was the most important factor in regulating CO2 concentration during the last deglaciation. However, the fast increases between intervals II and III and at the end of interval IV show that additional mechanisms in the Northern Hemisphere influenced CO2, presumably through changes in North Atlantic Deep Water formation. The record shows that methane was regulated by a completely different mechanism probably related to the expansion/contraction of wetlands in the northern and tropical regions.

Common chronology for Antarctica and Greenland

Temperature and atmospheric methane
Temperature and atmospheric methane
Greenland and Antarctica Temperature and Atmospheric methane
EDC and EDML are Antarctica ice cores (blue and light blue)
NGRIP is a Greenland ice core (orange).
(A) Methane concentration
(B) Temperature proxies.

A new dating method based on inverse techniques was developed to enable consistent dating of ice cores. This method use regional or global markers such as methane spikes, volcanic ash, and other global markers to link chronologies on as common time scale. The method has been applied to one Greenland (NGRIP) and three Antarctic (EPICA Dome C, EPICA Dronning Maud Land, and Vostok) ices cores to produce consistent ice and gas chronologies over the last deglaciation.

Atmospheric CO2 Concentrations over the Last Glacial Termination, Eric Monnin, Andreas Indermühle, André Dällenbach, Jacqueline Flückiger, Bernhard Stauffer, Thomas F. Stocker, Dominique Raynaud, Jean-Marc Barnola, Science 05 Jan 2001: Vol. 291, Issue 5501, pp. 112-114 DOI: 10.1126/science.291.5501.112

Consistent dating for Antarctic and Greenland ice cores, Bénédicte Lemieux-Dudon, Eric Blayo, Jean-Robert Petit, Claire Waelbroeck, Anders Svensson, Catherine Ritz, Jean-Marc Barnola, Bianca Maria Narcisi, Frédéric Parrenin, Quaternary Science Reviews, 29, Issues 1–2, January 2010, Pages 8–20.