Abstract. The air occluded in ice sheets and glaciers has, in general, a younger age (defined as the time after its isolation from the atmosphere) than the surrounding ice matrix because snow is first transformed into open porous firn, in which the air can exchange with the atmosphere. Only at a certain depth (tim-ice transition) the pores are pinched off and the air is definitely isolated from the atmosphere. The tim-ice transition depth is at around 70 rn under present climatic conditions at Summit, central Greenland. The air at this depth is roughly 10 years old due to diffusive mixing, whereas the ice is about 220 years old. This results in an age difference between the air and the ice of 210 years. This difference depends on temperature and accumulation rate and did thus not remain constant during the past. We used a dynamic tim densification model to calculate the tim-ice transition depth and the age of the ice at this depth and an air diffusion model to determine the age of the air at the transition. Past temperatures and accumulation rates have been deduced from the •5180 record using time independent functions. We present the results of model calculations of two paleotemperature scenarios yielding a record of the age difference between the air and the ice for the Greenland Ice Core Project (GRIP) and the Greenland Ice Sheet Project Two (GISP2) ice cores for the last 100,000 years. During the Holocene, the age difference stayed rather stable

No abstract

No abstract

In order to study in detail the pre‐industrial CO2 level (back to about 900 AD) and its temporal variations, several ice cores from Greenland and Antarctica were analysed in two laboratories, and compared with previous records. The agreement between the two laboratories and between the different cores of the same hemisphere is good. However, the comparison of the northern hemisphere (Greenland) and southern hemisphere (Antarctica) records shows values systematically higher in the north than in the south, ranging from 20 ppmv at the turn of this millennium to nearly zero around the 18th century. Based on our present knowledge of the carbon cycle, an inter‐hemispheric gradient of 20 ppmv is unrealistic. Thus, in the oldest part of the record, at least one profile should not represent the true atmospheric CO2 concentrations. A companion paper by Anklin et al. (submitted), discusses the possible processes which can alter the atmospheric CO2 once trapped in the ice. Due to the fact that the impurity content is one order of magnitude lower in the Antarctic than in the Greenland ice, we are much more confident in the Antarctic record. The new results from D47 and D57 (Adélie Land) presented in this paper, confirm the CO2 fluctuation of about 10 ppmv at the end of the 13th century, previously observed by Siegenthaler et al. (1988) on an ice core drilled at South Pole. This fluctuation corresponds to a small imbalance of the carbon cycle (∼ 0.3 GT C/ yr), but its duration led to a significant cumulative input into the atmosphere. The changes observed in the pre‐industrial level are discussed in terms of climatic noise and variability.

No abstract