(Table 3) Li concentrations and isotopic compositions in foraminiferal shells, carbonate sediments, and associated pore waters from ODP Sites 131-806 and 138-851

An improved procedure for lithium isotope analysis using Li3PO4 as the ion source has been investigated for application to geological samples. The 7Li/6Li ratio is measured using double rhenium filament thermal ionization mass spectrometry in which isotopic fractionation is minimized at high temperatures. The method produces a stable, high intensity Li+ ion beam that allows measurement of nanogram quantities of lithium. This results in a reduction in sample size of up to 1000 times relative to that required for the established Li2BO2+ method while maintaining a comparable precision of better than 1? (1 sigma). Replicate analyses of the NBS L-SVEC Li2CO3 standard yielded a mean value of 12.1047+/-0.0043 (n=21), which is close to the reported absolute value of 12.02+/-0.03. Intercalibration with a wide range of geological samples shows excellent agreement between the Li3PO4 and Li2BO2+ techniques. Replicate analyses of seawater and a fresh submarine basalt display high precision results that agree with previous measurements. Taking advantage of the high ionization efficiency of the phosphate ion source, we have made the first measurements of the lithium concentration (by isotope dilution) and isotopic composition of calcareous foraminiferal tests and other marine carbonates. Preliminary results indicate that substantial lithium exchange occurs between carbonate sediments and their interstitial waters. In addition, a possible link between lithium paleoceanography and paleoclimate during the last 1000 ky may be derived from planktonic foraminiferal tests. This highly sensitive technique can be applied in the examination of low lithium reservoirs and thereby provide insight into some fundamental aspects of lithium geochemistry.

(Table 1) Stratigraphic datums from ODP Site 130-806

A 2 m.y. oxygen isotope record of Globigerinoides sacculifer from the Ontong Java Plateau, based on cores from Ocean Drilling Program Leg 130, is dated by matching variations to an orbital template. The procedure allows us to present the most complete Quaternary record available for the western equatorial Pacific. The template-generating algorithm describes a balance between growth and melting of ice. Following basic Milankovitch theory, ice growth is taken as constant, while melting is taken to depend on summer insolation, current ice mass, and average past ice mass. Template settings must be changed once, between 1 and 1.2 Ma, to reflect a major shift in climate. Template fits are strikingly good over much of the record and can be used to detect and fill gaps from core breaks and other disturbances. One result of template dating is an exact age for the Brunhes-Matuyama magnetic reversal boundary, at 790+/-5 ka, as well as several other precise dates (900 ka for the middle Pleistocene climate shift; 1070, 1240, and 1450 ka for isotope stages 31, 37, and 47, respectively). Sedimentation rates fluctuate between 18 and 28 m/m.y., a ca. 400 ka cycle being the most prominent. Major anomalies arise within the transitional regime (1.2 to 1 Ma). The origin of the cycles is unknown; we propose productivity variations in the western equatorial Pacific.