Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer Julson, A., and van Andel, T.H. (Eds.), 1995 Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 138 14. SPATIAL AND TEMPORAL VARIABILITY OF LATE NEOGENE EQUATORIAL PACIFIC CARBONATE: LEG 138 1 T.K. Hagelberg, 2,3 N.G. Pisias, 2 L.A. Mayer, 4 N.J. Shackleton, 5 and A.C. Mix 2 ABSTRACT High resolution, continuous records of GRAPE wet bulk density (a carbonate proxy) from Ocean Drilling Program Leg 138 provide one the opportunity for a detailed study of eastern equatorial Pacific Ocean carbonate sedimentation during the last 6 m.y. The transect of sites drilled spans both latitude and longitude in the eastern equatorial Pacific from 90° to 110°W and from 5°S to 10°N. Two modes of variability are resolved through the use of Empirical Orthogonal Function (EOF) analysis. In the presence of large tectonic and climatic boundary condition changes over the last 6 m.y., the dominant mode of spatial variability in carbonate sedimentation is remarkably constant. The first mode accounts for over 50% of the variance in the data, and is consistent with forcing by equatorial divergence. This mode characterizes both carbonate concentration and carbonate mass accumulation rate time series. Variability in thefirst mode is highly coherent with insolation, indicating a strong linear relationship between equatorial Pacific car bonate sedimentation and Milankovitch variability. Frequency domain analysis indicates that the coupling to equatorial divergence in carbonate sedimentation is strongest in the precession band (19 23 k.y.) and weakest though present at lower frequencies. The second mode of variability has a consistent spatial pattern of east west asymmetry over the past 4 m.y. only; prior to 4 Ma, a different mode of spatial variability may have been present, possibly suggesting influence by closure of the Isthmus of Panama or other tectonic changes. The second mode of variability may indicate influence by CaCO 3 dissolution. The second mode of variability is not highly coherent with insolation. Comparison of the modes of carbonate variability to a 4 m.y. record of benthic δ 18 θ indicates that although overall correlation between carbonate and δ 18 θ is low, both modes of variability in carbonate sedimentation are coherent with δ 18 θ changes at some frequencies. The first mode of carbonate variability is coherent with Sites 846/ 849 δ 18 θ at the dominant insolation periods, and the second mode is coherent at 100 k.y. during the last 2 m.y. The coherence between carbonate sedimentation and δ 18 θ in both EOF modes suggests that multiple uncorrelated modes of variability operated within the climate system during the late Neogene. INTRODUCTION AND SCIENTIFIC BACKGROUND Because the oceanic carbon reservoir is 60 times the size of the atmosphere and is in direct exchange with the atmosphere, orbital scale atmospheric CO 2 changes must ultimately be explained by changes in the oceanic carbon cycle (Broecker and Peng, 1982). Evidence from ice cores indicates that the CO 2 increase at the ends of glacial periods occurred rapidly, on the order of a thousand years. To understand this process, it is necessary to understand the mechanisms controlling the carbon cycle on glacial/interglacial time scales. While ice cores pro vide a direct measure of changes in atmospheric CO 2 , examination of the marine sediment record is necessary to obtain a history beyond that of the last glacial cycle. Efforts to understand the mechanisms through which the oceanic carbon cycle changes atmospheric CO 2 have led to years of carbon cycle modeling. Hypotheses for glacial/ interglacial pCO 2 changes were grouped into four categories by Heinze et al. (1991) in a recent over view: three kinds of oceanic carbon pumps that are capable of influ encing atmospheric CO 2 (a solubility pump, a nutrient pump, and a CaCO 3 pump), and changes in the oceanic velocity field and ventilation rates. Evidence for oceanic carbon pool changes that is needed to evalu ate hypotheses is derived primarily from sedimentary δ 13 C and CaCO 3 records. δ 13 C indicates the efficiency of the ocean's biological pump and the nutrient content of deep and surface waters. The CaCO 3 concentra tion in sediments reflects productivity, dilution, and alkalinity changes. Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer Julson, A., and van Andel, T.H. (Eds.), 1995. Proc. ODP Sci. Results, 138: College Station TX (Ocean Drilling Program). College of Oceanic and Atmospheric Sciences, Oregon State University, Corvallis, OR U.S.A. 3 Now at Graduate School of Oceanography, University of Rhode Island, Narragansett, Rl U.S.A. 4 Institute for Ocean Mapping, University of New Brunswick, Fredericton, N.B., Canada. Subdepartment of Quaternary Research, Godwin Laboratory, Cambridge University, Cambridge, United Kingdom. The extent to which CaCO 3 sedimentation in equatorial Pacific Ocean sediments has been controlled by production or dissolution has been the subject of debate for more than 40 yr. Since Arrhenius (1952), studies have investigated the relationship of equatorial Pacific car bonate sedimentation to regional surface processes, global glacial/ interglacial changes, and external orbital (Milankovitch) influence (e.g., Hays et al., 1969, Moore et al., 1977, 1982; Farrell and Prell, 1989; Rea et al., 1991). In the equatorial Pacific Ocean, large changes in sedimentary calcium carbonate concentration have occurred on long time scales (millions of years), as a response to oceanic boundary condition and geochemical mass budget changes, and on shorter time scales (tens of thousands of years), as a response to climatic influence. Changing boundary conditions during the late Cenozoic have significantly influenced equatorial Pacific sediment composition, van Andel et al. (1975) estimated the initiation of the Equatorial Undercur rent (EUC) at about 11 to 12 Ma, when the northward movement of the Australia Plate cut off western Pacific Indian Ocean circulation. This development may have caused the onset of a narrow, equatorially symmetric zone of carbonate and opal sedimentation having steep gradients away from the equator. Widespread orogeny during the late Miocene and Pliocene may have had a large influence on oceanic alkalinity budgets through increased chemical weathering (Raymo et al., 1988). This uplift may have also influenced surface ocean circula tion in the equatorial Pacific through associated changes in atmos pheric circulation (Ravelo et al., 1992). A modeling study has sug gested that prior to closure of the Isthmus of Panama (3^4 Ma or earlier), North Atlantic Deep Water (N AD W) production was reduced, carbonate preservation was increased because of less undersaturated Pacific deep waters, and eastern equatorial Pacific surface upwelling was unchanged (Maier Raimer et al., 1990). Initiation of large scale continental glaciation in the Northern Hemisphere near 2.4 Ma may have altered oceanic alkalinity and productivity significantly. van Andel et al. (1975) demonstrated that similar spatial patterns are present in bulk sediment accumulation rates, carbonate accumula tion rates, and carbonate concentration of central equatorial Pacific 321