THE MYSTIQUE OF ICE CORE CHRONOLOGIES

Chronologies from ice cores are one of the most important benchmarks for Quaternary dating up to 100 ka. In addition to being used to define the timing of specific events such as the Younger Dyras termination, the GISP2 ice core in particular has been critical for radiocarbon calibration prior to 10 ka, since its provides the basis of the chronology of the Cariaco basin varve sequence. Furthermore, paleoclimate records from ice cores, due to their high resolution and the general agreement that they are representative of climate over broad areas, are frequently used as benchmarks for the interpretation of other Quaternary records as far back as 400 ka. For all of these reasons, it is important that uncertainties in ice core chronologies be well understood and, where possible, improved upon.

In this talk, I will briefly review the layer-counting method used to develop the most widely-used benchmark ice core chronologies – GISP2 and GRIP in central Greenland – the trace-gas-matching technique used to transfer these chronologies to cores in Antarctica, and published estimates of the uncertainties. I will then discuss possible improvements to the latter chronologies, based on the requirement of self-consistency among time series of ice accumulation rate, temperature and the difference between gas age and ice age (\delta age). Finally, I will highlight our recent work on an ice core from Mt. Logan, Yukon, which demonstrates the reliability of the chronology of this particular core at the sub-annual timescale.