Journal of Geophysical Research: Oceans

The δ18O record along the Greenland Ice Core Project deep ice core and the problem of possible Eemian climatic instability

Authors

  • Sigfús J. Johnsen,

  • Henrik B. Clausen,

  • Willi Dansgaard,

  • Niels S. Gundestrup,

  • Claus U. Hammer,

  • Uffe Andersen,

  • Katrine K. Andersen,

  • Christine S. Hvidberg,

  • Dorthe Dahl-Jensen,

  • Jørgen P. Steffensen,

  • Hitoshi Shoji,

  • Árny E. Sveinbjörnsdóttir,

  • Jim White,

  • Jean Jouzel,

  • David Fisher


Abstract

Over 70,000 samples from the 3029-m-long Greenland Ice Core Project (GRIP) ice core drilled on the top of the Greenland Ice Sheet (Summit) have been analyzed for δ8O. A highly detailed and continuous δ8O profile has thus been obtained and is discussed in terms of past temperatures in Greenland. We also discuss a three-core stacked annual δ8O profile for the past 917 years. The short-term (<50 years) variability of the annual δ8O signal is found to be 1‰ in the Holocene, and estimates for the coldest parts of the last glacial are 3‰ or higher. These data also provide insights into possible disturbances of the stratigraphic layering in the core which seems to be sound down to the onset of the Eemian. Spectral analysis of highly detailed sequences of the profile helps determine the smoothing of the δ8O signal, which for the Holocene ice is found to be considerably stronger than expected. We suggest this is due to a process involving diffusion of water molecules along crystal boundaries in the recrystallizing ice matrix. Deconvolution techniques were employed for restoring with great confidence the highly attenuated annual δ8O signal in the Holocene. We confirm earlier findings of dramatic temperature changes in Greenland during the last glacial cycle. Abrupt and strong climatic shifts are also found within the Eem/Sangamon Interglaciation, which is normally recorded as a period of warm and stable climate in lower latitudes. The stratigraphic continuity of the Eemian layers is consequently discussed in section 3 of this paper in terms of all pertinent data which we are not able to reconcile.

Ancillary