CONCLUSIONS

At Site 975, four major periods—Intervals I-IV—have been defined by the amplitude of variations of ¹⁸O values. The lower interval boundary at 1.7 Ma occurs above the Pliocene/Pleistocene boundary; the interval boundary at 0.9 Ma, corresponds to a time of significant ice volume increase in the Northern Hemisphere; the upper interval boundary at 0.42 Ma, is synchronous to the time when the permanent ice sheet in the Northern Hemisphere reached its maximum dimensions. The glacial-interglacial cycles show the typical 100 k.y. eccentricity period after 0.9 Ma and the 40 k.y. obliquity period before 0.9 Ma. The link between the western Mediterranean and the open ocean is clearly demonstrated because the major paleoclimatologic events, as well as the astronomical periodicity, are recorded by the planktonic ¹⁸O values.

The specificity of the Mediterranean is revealed by the higher amplitudes of the ¹⁸O oscillations at Site 975, compared to those of the open ocean. Starting from the lower Pleistocene, part of this increase is explained by the progressive cooling of the surface waters during the glacial times by 2°C at 0.9 Ma and then again by 2°C at 0.42 Ma. The other part of the ¹⁸O increase during glacial times after 0.42 Ma, is better explained by the increase of excess evaporation over freshwater inputs, which resulted in an increase by 2 of the surface salinities, rather than by a local additional cooling by 2°C in the West Balearic Basin.

The variations of the ¹³C values of G. bulloides record both the ¹³C global change, part of which is related to variations of ocean productivity, characterized by high glacial levels and low interglacial levels, and the local modifications of the productivity, which are controlled by the thermohaline stratification of the surface waters.

The sapropel layers were mostly deposited during interglacial episodes where high levels of surface productivity existed. The synchronism of these events with the sapropel deposition in the eastern Mediterranean is verified, but a few layers are not present at the same time in the western and eastern Mediterranean. A precise comparison of the recent results obtained from ODP Legs 160 and 161 is thus necessary to define the interconnections which existed between the western and eastern Mediterranean during the Quaternary.