Stable and strontium isotope ratios and age determination of ODP Site 130-803

Stratigraphic information from strontium, oxygen, and carbon isotopic ratios has been integrated with diatom and planktonic foraminifer datums to refine the Oligocene to early Miocene chemostratigraphy of Site 803. The Sr isotope results are based on analyses of mixed species of planktonic foraminifer and bulk carbonate samples. 87Sr/86Sr ratios of bulk carbonate samples are, in most cases, less radiogenic than contemporaneous seawater. Estimated sediment ages based on planktonic foraminifer 87Sr/86Sr ratios, using the Sr-isotope-age relation determined by Hess and others in 1989, are in moderately good agreement with the biostratigraphic ages. Chronological resolution is significantly enhanced with the correlation of oxygen and carbon isotope records to those of the standard Oligocene section tied to the Geomagnetic Polarity Time Scale at Site 522. Ages revised by this method and other published ages of planktonic foraminifer datums are used to revise the Oligocene stratigraphy of Site 77 to correlate the stable isotope records of Sites 77 and 803. Comparison of the Cibicidoides stable isotope records of Sites 77 and 574 with paleodepths below 2500 m in the central equatorial Pacific, and Site 803 at about 2000-m paleodepth in the Ontong Java Plateau reveals inversions in the vertical d18O gradient at several times during the Oligocene and in the early Miocene. The shallower water site had significantly-higher d18O values than the deeper water sites after the earliest Oligocene 18O enrichment and before 34.5 Ma, in the late Oligocene from 27.5 to at least 25 Ma, and in the early Miocene from 22.5 to 20.5 Ma. It is not possible to ascertain if the d18O inversion persisted during the Oligocene/Miocene transition because the deeper sites have hiatuses spanning this interval. We interpret this pattern to reflect that waters at about 2000 m depth were cold and may have formed from mixing with colder waters originating in northern or southern high-latitude regions. The deeper water appear to have been warmer and may have been a mixture with warm saline waters from mid- or low-latitude regions. No apparent vertical d13C gradient is present during the Oligocene, suggesting that the age difference of these water masses was small.

Age models for ODPO Leg 130 sites

The Neogene carbonate stratigraphy of five sites drilled on Ontong Java Plateau during Leg 130 reveals a number of patterns which are unexpected, and which we refer to as loss paradox, equatorial insensitivity, and climate paradox. They denote the following unresolved questions. 1 The loss of carbonate at depth (as derived from differences in accumulation rates) is much greater than suggested by the change in carbonate percentages (calculated under the assumption that carbonate dissolution is the cause of loss). This indicates an important role for redeposition processes, such as winnowing (bottom currents), sifting (resuspension and catabatic flow) and episodic sloughing or solifluction (presumably stimulated by earthquakes). 2 Accumulation rates are not markedly increased at the time a site crosses the equator. There are several possible reasons. Equatorial upwelling may be unimportant in controlling sedimentation rates this far in the western Pacific, or its output may be spread over a considerable distance from the equator. Alternatively, increased supply below the equator is compensated for by increased removal (e.g. from resuspension by bioturbation, combined with catabatic flow). It is conceivable that errors in the timescale could also produce the effect seen. 3 There is an overall tendency for agreement between the stratigraphic patterns of carbonate content and of accumulation rates, but neither pattern is readily explained by reference to changes in climate (represented by benthic delta18O) or in sea-level (as derived from sequence stratigraphy).

(Table DR1) Biogenic silica and chert in the Pacific Ocean, DSDP and ODP data

Evidence for the dissolution of biogenic silica at the base of pelagic sections supports the hypothesis that much of the chert formed in the Pacific derives from the dissolution and reprecipitation of this silica by hydrothermal waters. As ocean bottom waters flow into and through the crust, they become warmer. Initially they remain less saturated with respect to dissolved silica than pore water in the overlying sediments. With the diffusion of heat, dissolved ions, and to some extent the advection of water itself, biogenic silica in the basal part of the sedimentary section is dissolved. Upon conductively cooling, these pore waters precipitate chert layers. The most common thickness for the basal silica-free zone (20 m) lies below the most common height of the top of the chert interval above basement (50 m). This mode of chert formation explains the frequent occurrence of chert layers at very shallow subbottom depths in pelagic sections of the Pacific. It is also consistent with the common occurrence of cherts

Paleomagnetic properties of igneous rocks from several DSDP and ODP sites, western Pacific

Oceanic basalts and other related igneous rocks are considered excellent recorders of the Earth's paleomagnetic field. Consequently, basalt core paleomagnetic data are valuable for the constraints they provide on plate tectonic motions, especially for oceanic plates such as the Pacific. Unfortunately, few Deep Sea Drilling Project (DSDP) and Ocean Drilling Program (ODP) boreholes have been cored very deeply into the ocean crust. The result is that there are only a few sites at which a large enough number of basalt flows have been cored to properly average secular variation (e.g., Kono, 1980, doi:10.2973/dsdp.proc.55.135.1980; Cox and Gordon, 1984, doi:10.1029/RG022i001p00047). Furthermore, there are a number of sites where basaltic core samples were retrieved but the cores were not measured. Often this occurs because leg scientists had more important sections to work on, or the section was ignored because it was too short to record enough time to average secular variation and obtain a reliable paleolatitude. Even though it may not be possible to determine a precise paleolatitude from such short sections, measurements from a small number of flows are important because they can be combined with other coeval paleomagnetic data from the same plate to calculate a paleomagnetic pole (Gordon and Cox, 1980, doi:10.1111/j.1365-246X.1980.tb02642.x; Cox and Gordon, 1984, doi:10.1029/RG022i001p00047). For this reason, I obtained samples for paleomagnetic measurements from eight Pacific sites (169, 170, 171, 581, 597, 800, 803, and 865), most of which have not been previously measured for paleomagnetism.

(Table 1) H2O and CO2 contents in submarine basaltic glasses from Ontong Java Plateau

The Ontong Java Plateau in the western Pacific is anomalous compared to other oceanic large igneous provinces in that it appears to have never formed a large subaerial plateau. Paleoeruption depths (at 122 Ma) estimated from dissolved H2O and CO2 in submarine basaltic glass pillow rims vary from ~1100 m below sea level (mbsl) on the central part of the plateau to 2200-3000 mbsl on the northeastern edge. Our results suggest maximum initial uplift for the plateau of 2500-3600 m above the surrounding seafloor and 1500+/-400 m of postemplacement subsidence since 122 Ma. Our estimates of uplift and subsidence for the plateau are significantly less than predictions from thermal models of oceanic lithosphere, and thus our results are inconsistent with formation of the plateau by a high-temperature mantle plume. Two controversial possibilities to explain the anomalous uplift and subsidence are that the plateau (1) formed as a result of a giant bolide impact, or (2) formed from a mantle plume but has a lower crust of dense garnet granulite and/or eclogite; neither of these possibilities is fully consistent with all available geological, geophysical, and geochemical data. The origin of the largest magmatic event on Earth in the past 200 m.y. thus remains an enigma.

Carbonate and bulk grain-size record for ODP Leg 130 holes

A major goal of Ocean Drilling Program (ODP) Leg 130 was to drill four sites down the northeastern flank of the Ontong Java Plateau to collect a series of continuous sedimentary sequences that would provide a depth transect of Neogene sediments. In particular, the study of the sediments recovered along the depth transect is expected to yield high-resolution stratigraphic, geochemical, and physical properties records across intervals of major paleoceanographic changes by evaluating variations of primary sedimentological and paleoceanographic indicators (e.g., carbonates, isotopes, grain size, microfossil assemblages, etc.). This data report presents the results of highresolution (3-5 Ka sample intervals) analyses of carbonate concentration and bulk sediment grain size at Sites 803-806 for the time interval from 2 Ma to the present.

Bulk calcite minor element composition of ODP Leg 130 samples

We investigated minor element ratios (Sr/Ca and Mg/Ca) in bulk sediment samples from Sites 803-807 using a recently optimized sample treatment protocol for calcium-carbonate-rich sediments consisting of sequential reductive and ion exchange treatments. We evaluated this protocol relative to bulk sediment leaching using samples from Sites 804 and 806, the two end-member sites in the depth transect, reporting as well Mn/Ca and Fe/Ca ratios for sediments from these two sites processed by means of both methods. The Sr/Ca ratios were only slightly affected by the sample treatment, with an average reduction of 6%-7% caused primarily by the ion exchange step. The reductive sample treatment, designed to be effective at removing Mn-rich oxyhydroxides, has a major effect on Mg/Ca ratios, with up to 50% reduction, whereas little effect occurred in ion exchange alone on Mg/Ca ratios. The Mn/Ca and Fe/Ca ratios were not consistently offset by the sample treatment, and these ratios do not appear to be representative of calcite geochemistry reflecting either ocean history or diagenetic overprinting. Celestite solubility appears to be an important control on interstitial water Sr concentrations in these sites, and it must be considered when constructing Sr mass balance models of calcite recrystallization. Calcite Sr/Ca ratios (range 1-2 mmol/mol) are similar from site to site when plotted vs. age, with a pattern comparable to that for well-preserved foraminifer tests over the past 40 Ma. Interstitial water Mg and Ca gradients appear to reflect basement character and the intensity of alteration; they can vary substantially over a small area. Calcite Mg/Ca ratios (range 1.5-4.5 mmol/mol) differ from site to site, with generally higher ratios for sites at a shallower water depth. Increasing calcite Mg/Ca ratios correlate with decreasing Sr/Ca ratios in the treated samples. No consistent pattern exists for calcite Mg/Ca ratios vs. age or depth, nor is any direct correlation to interstitial water Mg/Ca ratios present.