(Table 2) Aragonite, high and low magnesium calcite, major elements, bioclast and planktic foraminifera from ODP Hole 133-817A

Investigation of the Middle Miocene-Pleistocene succession in cores at ODP Site 817A (Leg 133), drilled on the slope south of the Queensland Plateau, identified the various material fluxes contributing to sedimentation and has determined thereby the paleogeographic events which occurred close to the studied area and influenced these fluxes. To determine proportions of platform origin and of plankton origin of carbonate mud, two reference sediments were collected: (1) back-reef carbonate mud from the Young Reef area (Great Barrier Reef); and (2) Late Miocene chalk from the Loyalty Basin, off New Caledonia. Through their biofacies and mineralogical and geochemical characters, these reference sediments were used to distinguish the proportions of platform and basin components in carbonate muds of 25 core samples from Hole 817A. Two "origin indexes" (i1 and i2) relate the proportion in platform and basin materials. The relative sedimentation rate is inferred from the high-frequency cycles determined by redox intervals in the cores. Bulk carbonate deposited in each core has been calculated in two ways with close results: (1) from calcimetric data available in the Leg 133 preliminary reports (Davies et al., 1991); and (2) from average magnetic susceptibility of cores, a value negatively correlated to the average carbonate content. Vertical changes in sedimentation rates, in carbonate content, in origin indexes and in "linear fluxes" document the evolution of sediment origins from platform carbonates, planktonic carbonates and insoluble material through time. These data are augmented with the variations in organic-matter content through the 817A succession. The observed changes and their interpretation are not modified by compaction, and are compatible with major paleogeographic events including drowning of the Queensland Plateau (Middle Miocene-Early Pliocene) and the renewal of shallow carbonate production, (1) during the Late Pliocene, and (2) from the Early Pleistocene. The birth and growth of the Great Barrier Reef is also recorded from 0.5 Ma by a strengthening of detrital carbonate deposition and possibly by a lack of clay minerals in the 4 upper cores, a response to trapping of terrigenous material behind this barrier. In addition, a maximum of biological silica production is displayed in the Middle Miocene. These changes constrain the time of events and the sequence-stratigraphy framework some components of which are transgression surface, maximum flooding surface and low-stand turbidites. Sedimentation rates and material fluxes show cycles lasting 1.75 Myr. Whatever their origin (climatic and/or eustatic) these cycles affected the planktonic production primarily. The changes also show that major carbonate variations in the deposits are due to a dilution effect by insoluble material (clay, biogenic silica and volcanic glasses) and that plankton productivity, controlling the major fraction of carbonate sedimentation, depends principally on terrigenous supplies, but also on deep-water upwelling. Accuracy of the method is reduced by redeposition, reworking, and probable occurrence of hiatuses.

Sediment color variation in laminated Holocene sediments of ODP Site 178-1098

Continuous sediment color records with a resolution of one measurement per millimeter were generated for Site 1098 (Palmer Deep, Antarctic Peninsula) from digital images of the core surfaces to test if the laminated intervals at this site will allow for analysis of high-frequency climate variability in the Circum-Antarctic. Long-term variation in color values correlates with gamma-ray attenuation bulk density. Darker colors are found in laminated intervals with lower bulk density, high biogenic silica, and high total organic carbon content. Darker color values result from the addition of dark laminae to background sediments that show little variation in color. The thicknesses of dark and light laminae were measured in the top 25 meters composite depth to determine the temporal resolution of the laminae. The alternation between dark, biogenic-rich laminae and background sediment essentially represents an annual cycle, but the sediment is not consistently varved. The modal thickness of light laminae is close to the long-term average annual accumulation rate, and results indicate that approximately half of the dark/light couplets in distinctly laminated intervals represent a single year. Missing biogenic laminae are interpreted to represent reduced primary productivity during cold years with delayed breakup of the sea-ice cover.

(Table T1) Correlations between marker beds and drill holes for ODP Leg 171B sites

More than 50 discrete volcanic ash layers were recovered at the five drill sites of the Blake Nose depth transect (Leg 171B, western central Atlantic). The majority of these ash layers are intercalated with Eocene hemipelagic sediments with a pronounced frequency maximum in the upper Eocene. Several ash layers appear to be deposited from volcanic fallout with little or no indication of secondary remobilization. They provide excellent stratigraphic markers for a correlation of the Leg 171B drill sites. Other ash layers were probably redeposited from volcaniclastic-rich turbidity currents, but they still represent geologically instantaneous events that can be used in stratigraphic correlation between adjacent drill holes. Additional nonvolcanic marker beds, like the suspect late Eocene impact event layer, were included in our hole-to-hole correlations. Stratigraphic and downcore positions of marker beds were compiled and plotted against existing composite depth records that were constructed to guide high-resolution sampling. Comparison of our correlation with the spliced composite sections of each drill site reveals several minor and some major discrepancies. These may result from drilling distortion or missing sections, from the lack of unambiguous criteria for the synchronism of ash layers, or from the systematic exclusion of marker-bed data in the construction of the spliced record. Integration of both correlation approaches will help eliminate most of the observed discrepancies.

A 2.8 Ma sedimentological, geochemical and palynological record of sediment core ICDP5011-1 in Lake Elgygytgyn, NE Russia

The reliability of Arctic climate predictions is currently hampered by insufficient knowledge of natural climate variability in the past. A sediment core from Lake El'gygytgyn (NE Russia) provides a continuous high-resolution record from the Arctic spaning the past 2.8 Ma. The core reveals numerous "super interglacials" during the Quaternary, with maximum summer temperatures and annual precipitation during marine benthic isotope stages (MIS) 11c and 31 ~4-5 °C and ~300 mm higher than those of MIS 1 and 5e. Climate simulations show these extreme warm conditions are difficult to explain with greenhouse gas and astronomical forcing alone, implying the importance of amplifying feedbacks and far field influences. The timing of Arctic warming relative to West Antarctic Ice Sheet retreats implies strong interhemispheric climate connectivity.

(Table 1) Stable oxygen and carbon isotopic ratios of tree rings, and tree ring width of white spruce (Picea glauca), Ennadai Lake

Stable isotope ratios from tree rings and peatland mosses have become important proxies of past climate variations. We here compare recent stable carbon and oxygen isotope ratios in cellulose of tree rings from white spruce (Picea glauca), growing near the arctic tree line; and cellulose of Sphagnum fuscum stems, growing in a hummock of a subarctic peatland, in west-central Canada. Results show that carbon isotopes in S. fuscum correlate significantly with July temperatures over the past ~20 yr. The oxygen isotopes correlate with both summer temperature and precipitation. Analyses of the tree-ring isotopes revealed summer temperatures to be the main controlling factor for carbon isotope variations, whereas tree-ring oxygen isotope ratios are controlled by a combination of spring temperatures and precipitation totals. We also explore the potential of combining high-frequency (annual) climate signals derived from long tree-ring series with low-frequency (decadal to centennial) climate signals derived from the moss remains in peat deposits. This cross-archive comparison revealed no association between the oxygen isotopes, which likely results from the varying sensitivity of the archives to different seasons. For the carbon isotopes, common variance could be achieved through adjustments of the Sphagnum age model within dating error.