MODELLING THE CLIMATE EVOLUTION FROM THE LAST INTERGLACIAL TO THE START OF THE LAST GLACIATION

How seasonal and latitudinal variations of the incident solar radiation initiate internal feedbacks that produce a shift from an interglacial to a glacial mode is still a matter of debate. Until now, the last interglacial-glacial transition has been discussed either through interpretation of proxy records or through modelling studies. Studies based on Atmospheric General Circulation Models forced by modern sea surface temperatures (SSTs) or coupled to mixed layer ocean have not been successful in simulating perennial snow cover, while accounting for ocean and vegetation feedbacks has produced results in better agreement with proxy records. Nevertheless, beyond the Milankovitch theory that argues for an orbitally induced shift from an interglacial to a glacial climate, the available proxy records give important clues that have still not been fully explored through modelling studies. Recent work from North Atlantic sediment cores clearly identifies a cooling step correlated to a significant change in deep water circulation around 115 ky BP i.e. at the last interglacial-glacial transition [Shackleton et al., 2002; Chapman et Shackleton, 1999; Adkins et al., 1997; Hall et al., 1998]. There is more evidence, from both land and marine records, for an "intra-Eemian" cooling event in the high latitudes of the Northern Hemisphere around 120 ky BP [Maslin et al., 2001; Cortijo et al.,1999; Snchez-Goi et al., 1999; Tzedakis et al., 2002; Shemesh et al., 2001; Rioual et al., 2001]. The lag between insolation forcing and northern high latitudes cooling extrema at 120 ky BP emphasises the necessity of exploring the mechanisms responsible for the changes occurring through the last interglacial and their implication for the last glacial inception. We propose to explore this issue by combining an Earth system Model of Intermediate Complexity (EMIC), a coupled ocean-atmosphere GCM (OAGCM), and the comparison to paleoclimatic data. Using the insolation forcing alone, the EMIC model produces a first order response of SSTs and thermohaline circulation evolution which agrees with the available geological data. However, by comparing the OAGCM and the EMIC results we show that the orbitally induced shift from an interglacial to a pre-glacial climate is highly sensitive to the high northern latitudes moisture budget.