The climatic significance of the deeper part of the GISP2 ice core, below
2790 m depth and 110 kyr age, is a matter of considerable investigation and
controversy. The isotopic temperature records and electrical conductivity
records of GISP2 and GRIP, so similar for ice <110 kyr in age, are very
different in the lower part [ Grootes et al., 1993; Taylor et al.,
1993a]. Ice in GISP2 below 2790 m depth is folded and tilted, and shows
evidence of unconformities [ Gow et al., 1993]. The 
O of
O
in GISP2 above 2790 m matches almost perfectly with the Vostok
record [ Sowers et al., 1993]; below that depth, it is far noisier and
cannot be aligned with the smoothed Vostok signal [ Bender et al., 1994].
These features all suggest that ice age changes discontinuously in the deepest
part of GISP2 as a result of folding, extensive boudinage (squeezing out of
layers of ice), and/or intrusion. Bender et al. [1994] concluded that the
bottom 
200 m of ice at GISP2 may be correctly ordered but
discontinuous and extremely condensed, perhaps extending back to several
hundred kyr BP. Alternatively, the core may contain a disordered sequence of
much younger ice, perhaps largely from about 115-130 kyr BP.
GRIP (1993) interpreted the climate proxy records from the deepest part of their core as being properly ordered and continuous. They observed large and rapid changes in isotopic temperature with depth and concluded that these features represented rapid changes in climate during marine isotopic Stage 5e, the warmest part of the previous interglacial period. Such a conclusion has extremely important implications for climate because, together with the Dansgaard-Oeschger events, it suggests that rapid cooling events are possible during the current interglacial period.
Evidence for rapid climate change during Stage 5e rests on the assumption that the deep part of the GRIP ice core is continuous, in contrast to GISP2. GRIP is located over the present ice divide, while GISP2 is 30 km to the west. Therefore GRIP may in fact be more likely to be continuous, although a recent modeling study has shown that the divide itself has probably migrated [ Anandakrishnan et al., 1993]. Continuity of the deep part of the GRIP core has not yet been definitively demonstrated. Until this is done, by showing that gas composition records at GRIP are identical with those at Vostok or by some other approach, the evidence for rapid climate change in Greenland during the last interglacial remains equivocal.