Millennial-Scale Variations in North Atlantic Sediments in the Late

Pliocene and Early Pleistocene

K. Mc Intyre (Ocean Sciences/Institute of Marine Sciences, University

of California, Santa Cruz, CA 95064; ph. 408-459-3123; Internet: kmci@cats.ucsc.edu); A.C. Ravelo, M.L. Delaney (Ocean Sciences/Institute of Marine Sciences, University of California, Santa Cruz, CA 95064)

Research on millennial scale climate variations has focused on the last 150 kyrs because this is the interval over which there are high resolution records both on land and in the oceans, and the interval over which there are good age constraints. As a consequence, mechanisms to explain these millennial scale variations have evolved that account for them in this late Pleistocene environment. In order to characterize millennial scale events under different boundary conditions we focus on an interval of the late Pliocene and early Pleistocene (~1.4-2.0 Ma). During this interval, maximum glacial ice sheets were smaller, as small as a third of the last glacial maximum, while the pattern of incoming solar insolation was similar to that of the last 600 kyrs. In order to characterize millennial scale variability during this time period we have measured the concentration of ice rafted debris (IRD) in sediments as a proxy for ice berg flux and we have measured the of % planktonic foraminiferal species N. pachyderma sinistral as a proxy for sea surface temperature and salinity. To look at variability during specific intervals in detail, we generated records of IRD concentration for relatively short intervals of the late Pliocene (~1.75-1.92 Ma) at the Gardar drift (60N 23W, 1983 meters depth). This drift site had a sedimentation rate of ~22 cm/kyr during this interval, yielding a sample resolution of 200-500 kyrs. Our data suggest that there were large inputs of ice rafted debris reoccurring every 5-8 kyrs during maximum glacial intervals. During glacial build up and decay, inputs of ice rafted debris were more frequent, reoccurring every ~1 kyr, but were also lower in magnitude. There was no deposition of IRD during maximum interglacial intervals. To look at the pattern of variability over the course of the entire 1.4-2.0 Ma interval, we generated measures of IRD concentration and the % of planktonic species N. pachyderma sin. at the Feni drift (55N 15W, 2173 meters depth ). This drift site has a sedimentation rate of ~6 cm/kyr, and our record has a resolution of ~2 kyrs during this interval. The strongest evidence for millennial scale variability in the lower resolution Feni drift record is between 1.62 and 1.75 Ma. During this 120 kyr period low amplitude peaks of IRD occur approximately every 8 kyrs, and are accompanied by peaks in the %N. pachyderma sin. Before and after this time, the highest concentrations of IRD are limited to de-glaciations, and major variations in the %N. pachyderma sin. occur at the 41 Kyr glacial- interglacial frequency. Overall our results demonstrate that millennial scale variability in IRD input and sea surface temperature is by no means limited to the late Pleistocene, and that it may be a pervasive feature of geologic record. Synthetic mechanisms to explain the origin of millennial scale events must take into account how these events could have been generated this far in the past, under different boundary conditions of ice volume.