Results from Vostok

However, when interpreting palaeo-excess records there are even more ambiguities to consider than just evaporation conditions and their exact influence on the double isotopic composition. Currently Antarctic deep-ice drillings cover several glacial-interglacial cycles. The Vostok ice-core provides us with a deuterium excess record spanning now the past ca. 400,000yr (Vimeux et al., 1999, 2001, 2002). Fully interpreting this long-term record, Vimeux et al. (2002) came to an important reevaluation of the Antarctic excess, which includes the combined effects of local site temperature and source condition changes. As mentioned before the deuterium excess is influenced principally by climate conditions prevailing in vapour source regions of the corresponding precipitation site. This feature led to a quantification of polar excess signals exclusively in terms of sea-surface temperature (SST) changes (Johnsen et al., 1989). Using simple single-trajectory distillation models the sensitivity of the excess to changes in SSTs in the corresponding source areas was estimated to 0.7-0.8°C %o-1. Such an approach has several fundamental premises. The most important one is probably the assumption that there is just one single and geographically stable source region to which deduced SST changes can be associated. Moreover this method assumes that the relative humidity—like the SST influence on evaporating conditions—remains unchanged even under drastically different climate conditions such as the last glacial maximum. These premises can be fully addressed only in future numeric simulations with general circulation models (GCMs) equipped with water isotope diagnostics (Hoffmann et al., 1998).

However, another important mechanism affecting synchronously the water isotope signal (S18O, SD) and the deuterium excess (d) is now included in the interpretation. Obviously SST changes influence the temperature gradient between the evaporation and the condensation site. This temperature gradient is the leading climate control on the water isotopes (and not just site temperature) and changes at the evaporation site are at least second order for the isotopic composition of precipitation. On the other hand condensation temperatures affect the S18O /SD

Difference Tsource-Td

Difference Tsource-Td

Vostock source temperature variations, Tsourc

Vostock source temperature variations, Tsourc

Low latitude contribution

High latitude contribution

Vostock source temperature variations with no correction, Td

Low latitude contribution

High latitude contribution

Vostock source temperature variations with no correction, Td

Figure 36.1 Vostok source region temperatures derived from the 400-kyr-old East Antarctic ice-core record (lower panel). Following the classic approach for interpreting the deuterium excess, Td assumes that the effective source temperature can be deduced from the excess alone. However, Tsource takes into account the combined effects of condensation-site temperature changes and source temperature changes. In the upper panel the difference between the two computations is shown indicating that these differences are growing and shrinking following the glacial-interglacial 100 kyr cycle.

slope and therefore the deuterium excess. A physically satisfying interpretation of the water isotopes (S18O, SD) and the deuterium excess therefore needs a synchronous computation of SST and site-temperature changes. In fact we are dealing with two unknowns (SST and site temperature) and two measured variables (S18O/SD and d) in a weakly non-linear system. The system is in fact only weakly non-linear because SST changes are second order for the interpretation of the water isotope records and vice versa site temperature (TSite) changes are second order for the interpretation of the deuterium excess in terms of SST changes.

Figure 36.1 shows the intriguing consequences of the simultaneous computation of SSTs and TSite by using a linearized version of a Rayleigh distillation model (Vimeux et al., 2002). The deuterium excess and therefore the source temperatures deduced from the excess are dominated by a strong obliquity periodicity of about 40kyr. Vimeux et al. (2002) argue that obliquity insolation modulates the temperature gradient between high- and mid-latitudes and affects not only directly the SSTs but also the intensity of meridional vapour transport. The 'classic' calibration of source temperatures, Td, assumes that evaporation conditions can be deduced from the deuterium excess alone Td = d/1.3°C %-1. Conversely TSource includes the effects of changing condensation site temperatures (see Fig. 36.1). The corrections due to the latter effect amount to up to 50% of the original signal (thus in fact more than just second order) and increases TSource changes during the last glacial to as much as 6°C. Such large changes are very unlikely for a single individual source oceanic region and are not observed in marine sediments for the time period covered by the Vostok core. Therefore the authors rather argue for insolation driven geographical redistribution of source areas (high versus mid-latitudes) combined with real SST changes ending up in an effective temperature change of 6°C.

Antarctic long-term excess records put strong constraints on the hydrological cycle in the Southern Hemisphere and its functioning in the past. They represent therefore a valuable and critical test for our understanding of past climate changes.

Changing glaciers and their role in earth surface evolution

Was this article helpful?

0 0

Post a comment