What was the surface temperature in central Antarctica during the last glacial maximum?

Abstract

The temperature increase at Vostok (Antarctica) from the last glacial maximum to the present warm period is about 8 ‡C based on the deuterium isotope profile. The bore hole temperature (temperature profile in the ice sheet) indicates that the temperature difference may have been much larger, about 15 ‡C. The temperature dependent gas occlusion process is the key to evaluate the two scenarios. Atmospheric air penetrates the porous firn layer of the ice sheet and gets trapped at the firn ice boundary. Consequently the air is younger than the surrounding ice when it gets enclosed in bubbles. This age difference (vage) between ice and enclosed gas is temperature and accumulation rate dependent. Therefore it is possible to estimate paleotemperatures from a known vage. We use the linkage between chronologies of CH 4 and water isotopes from Byrd station and Vostok to obtain an experimental vage for Vostok. This experimental vage is then compared to modeled vage for the two temperature scenarios. Our results indicate that the temperature reconstruction deduced from the water isotopic composition is the more probable one. ß

A brief history of ice core science over the last 50 yr

“…Thus, there is no perfect alternative to calibrating the isotopic paleothermometer there. Still, there are useful arguments coming from the isotopic composition of the air bubbles (Caillon et al, 2001), from constraints with respect to ice core chronologies Blunier et al, 2004) and from paleothermometry (Salamatin et al, 1998). As reviewed by , they converge towards the idea that the observed present-day spatial slope can be used to interpret Antarctic isotopic profiles with however a slight underestimation, of the order of ∼ 10 %, of temperature changes.…”

Glacial–interglacial dynamics of Antarctic firn columns: comparison between simulations and ice core air-δ<sup>15</sup>N measurements

“…Blunier et al, 2004;Bender et al, 2006;Loulergue et al, 2007). They assume a homogeneous snow material where the density profile, and thus the LID, is mainly dependent on the temporal scenarios of accumulation rate, surface temperature and surface density (Herron and Langway, 1980;Pimienta, 1987;Schwander et al, 1993;Arnaud et al, 2000;Goujon et al, 2003).…”

“…The coefficient 0.7 represents the ratio of column-averaged firn density to ice density, which is required to translate the firn equivalent DCH thickness to an ice equivalent DCH thickness. We adopt a 5 % uncertainty to account for variations with respect to firn density profiles as a function of temperature and accumulation rate and varying ice density (Blunier et al, 2004). We also used two different estimates of the thinning factor: one from the EDML glaciological model of Huybrechts et al, 2007) and one from the new AICC2012 chronology (Bazin et al, 2012), and we consider a 10 % uncertainty linked to this parameter.…”

Eight glacial cycles from an Antarctic ice core

“…Isotopic models predict that d values should vary linearly with temperature in mid-and high latitudes. There is now a series of arguments supporting the use of this present-day temperature/isotope spatial slope to interpret isotopic records from Antarctica 18,19 , at least for deep ice cores from the East Antarctic plateau.…”