METHODS

The age models for Sites 999 and 1000 are based on nannoplankton and planktonic foraminifer biostratigraphy (Sigurdsson, Leckie, Acton, et al., 1997; Kameo and Bralower, Chap. 1, this volume; Chaisson and D'Hondt, Chap. 2, this volume) and calibrated to the revised geomagnetic polarity time scale of Cande and Kent (1995). For comparison purposes, I have converted the ages of the Miocene ¹⁸O events from Miller et al. (1991) and Wright et al. (1992) and the ¹³C maxima from Woodruff and Savin (1991) to the same time scale (see Table 1).

The investigated intervals of Sites 999 and 1000 are dominated by pelagic carbonate, which shows at both sites a differential degree of lithification, increasing with depth and ranging from oozes to limestones. To maintain consistency in the stable isotope record throughout the interval, bulk samples were analyzed. More than 350 samples were analyzed for stable isotope composition. Samples were dried at 60°C, then, according to their degree of lithification, either broken and crushed or drilled to obtain sufficient material for analysis. Then they were reacted using orthophosphoric acid at 90°C and analyzed online using a PRISM mass spectrometer at ETH Zurich. Results are reported using the standard notation in per million () relative to the PDB standard. Reproducibility of replicate analyses was generally better than 0.1. The resolution of the stable isotope record is at an average of ~50 ka (up to ~35 ka) at Site 1000, and at an average of ~140 ka (ranging between 80 ka and 250 ka) at Site 999.

When comparing the data set generated at Sites 999 and 1000 with other existing records (e.g., Miller et al., 1998), a major difference to consider is that data generated in this study reflect bulk-rock analysis, rather than measurements on separate foraminiferal species. Bulk-rock analysis is uncommon in paleoceanographic and climate studies, which in most cases involve measurements on specific species of foraminifers. Even though bulk analyses reflect a mixing of carbonates from different sources, it has been shown that under certain circumstances, the isotopic composition derived from bulk analyses resembles closely the record derived from single foraminifer analyses (Shackleton and Hall, 1984; Shackleton et al., 1993). In some cases, bulk isotopes can be utilized when the lithologies are too lithified to allow separation of single foraminifers.

The isotopic composition of bulk samples is a function of the composition of the benthic and planktonic foraminifers and calcareous nannoplankton present, as well as their relative proportions. The ¹⁸O of foraminifers is a function of the seawater ¹⁸O value (_w) where the organism lived. Miocene oxygen-isotope events (Miller et al., 1991; Wright et al., 1992) reflect global changes, in _w, in both planktonic and benthic foraminifers, related to glacioeustatic fluctuations. In addition to glacioeustasy, changes in either temperature or salinity effect the ¹⁸O values.