Extended Abstract (628K)1.8
Are temperature variations at Antarctic ice core sites representative of broad-scale climate variations?
John C. King, British Antarctic Survey, Cambridge, United Kingdom; and N. P. M. van Lipzig, W. M. Connolley, and J. C. Comiso
Stable isotope variations in ice cores are related to temperature variations at the core site through complex transfer functions. Even if the core record could be inverted to give a perfect local temperature history, it is important to know how representative temperature variations at the core site are of regional climate fluctuations when using the core to reconstruct regional palaeoclimates. The locations of core sites have generally been chosen for glaciological reasons (e.g. well-defined ice flow) and the climatological representativity of these sites cannot be guaranteed.
We have studied the spatial coherence of interannual variations in temperature at a number of existing and planned Antarctic ice core sites using a high spatial resolution (6.25 km) dataset of surface temperatures determined from satellite infrared radiometer measurements and a high-resolution (55 km) regional climate model. Spatial correlations are used to determine regions for which the core site may be considered representative.
The west coast of the Peninsula has warmed more rapidly than any other region on Earth over the past 50 years and palaeorecords from this region are urgently required to put this warming in a longer-term context. Our analysis shows that climate variations within the Peninsula region are spatially complex and only one of five existing ice core sites appears to be representative of the west coast. Furthermore, temperature variations in the Peninsula are not well correlated with those elsewhere in Antarctica.
Outside the Peninsula, temperature variations appear to be coherent over larger spatial scales. Temperature variations at Dome C are well-correlated with those across much of the East Antarctic Plateau and are reasonably well correlated with temperatures over much of the high latitude Southern Ocean. The temperature record from this core is thus likely to contain a strong regional signal. High correlations with temperatures at proposed drilling sites in Dronning Maud Land and on Berkner Island are confined to the Antarctic continent, suggesting that the temperature records from these cores will be dominated by local signals.
Session 1, Symposium on High-Latitude Climate Variations (Continued)
Monday, 12 May 2003, 10:45 AM-1:30 PM
Previous paper Next paper