RECONSTRUCTING LATE PLEISTOCENE NORTH PACIFIC ICE-RAFTING HISTORY THROUGH TWO GLACIAL CYCLES

Analysis of ice-rafted debris (IRD) in cores from ODP Site 887 contribute to the reconstruction of Pleistocene-Holocene climate in the North Pacific. Early work by St. John and Krissek (1999) described the temporal pattern of IRD accumulation in Hole 887A over the last 1500 kyr using a low-resolution magnetostratigraphic age model. Since that time, a more robust age model was developed (Galbraith, 2008) using planktonic foraminiferal δ¹⁴C data from Galbraith (2007) and benthic foraminiferal δ¹⁸O data from McDonald (1997). The purpose of this study is to re-examine the last two glacial cycles in the Site 877 cores using this higher resolution age model and additional sediment samples to 1) more thoroughly understand the temporal relationships between ice rafting and climate change over the last 170 kyr and 2) determine the feasibility of grain size analysis as an estimator for IRD abundance. To date, 61 samples from Holes 887A and 887C were analyzed using a laser diffraction particle size analyzer and were wet sieved to separate the coarse sand fraction (250μm-2mm) to calculate IRD wt% and IRD mass accumulation rates (MAR). A composite depth scale was constructed based on magnetic susceptibility data in order to apply the updated age model. A moderately positive relationship between the coarse fraction calculated from grain size analysis and wet sieving, however grain size analysis is unreliable with the coarsest samples. The majority of IRD MAR pulses occurred during glacial Marine Isotope Stages (MIS); many at rapid cooling events, or rapid warming events near the end of a glacial cycle when ice sheets were already established on the Alaskan coast. IRD pulses are also synchronous with 5 of 6 North Atlantic Heinrich Stadial events which may indicate that hemispheric control on iceberg discharges were overriding local controls at these times. However, there is a noticeable difference in IRD MARs between the two glacial maxima: the LGM IRD fluxes behaved as expected, but the PGM showed low IRD MARs. This may indicate regional differences in source ice response to the orbital climate cycles, however, age model errors also need to be considered. An additional 60 samples are currently being processed to increase the study resolution; these will be integrated into this dataset, as will the original IRD data from St. John and Krissek.