Because of the general lithified nature of the sediments, Sr isotope analyses were done on bulk material. All samples were powdered and dried. Approximately 5 mg of sample was dissolved in 500 µL of 1-M acetic acid and centrifuged. The supernatant was decanted and dried. The residue was then dissolved in 250 µL of 5-M HNO3, loaded on a column containing 50 µL of Eichrom SrSpec resin, and washed with 600 µL of 5-M HNO3. The Sr was eluted in 1 mL of H2O. Total procedural blanks were ~200 pg Sr. Samples were loaded in TaCl5 on single Re filaments and analyzed at the University of North Carolina at Chapel Hill on a Micromass Sector 54 thermal ionization mass spectrometer in dynamic mode. Fractionation was corrected using 86Sr/88Sr = 0.1194.
During this study, 87Sr/86Sr analyses of 70 aliquots of Sr carbonate standard SRM 987 yielded an average value of 0.710263 ± 9 (2 standard deviations). Five to eight SRM987 aliquots were analyzed with each group of ~15 samples, and the average value for these replicates within each run was adjusted to 0.710250 to correct for possible instrument variation and to adjust results to those used to construct the seawater Sr curve (McArthur et al., 2001). For example, if the values for the standards analyzed with a group of samples averaged 0.710264, we subtracted 0.000014 from the 87Sr/86Sr ratio for each sample in that group. A total of 48 samples were analyzed, and 2 samples were analyzed in duplicate. Internal precision for Sr isotope analysis was typically 0.0006%-0.0009% standard error, based on 100 dynamic cycles of data collection.
Ages of samples are determined by plotting Sr isotope ratios on the correlative (determined using preliminary biostratigraphic data [Mahoney, Fitton, Wallace, et al., 2002]) portion of the LOWESS version 3 curve (McArthur et al., 2001). In the Santonian-Maastrichtian part of the section, Sr isotope values increase consistently with decreasing age and estimating ages from Sr measurements is straightforward. In the Aptian-Coniacian part of the section, however, where Sr isotope values both increase and decrease with age, application of biostratigraphy is often ambiguous and alternative age interpretations must be closely considered. For this interval, ages predicted by Sr isotope stratigraphy were compared in detail with those from nannofossil biostratigraphy. Selected datums that are well established and reliable are taken from Sikora and Bergen (submitted [N1]). Ages of these nannofossil datums were taken from Bralower, Premoli Silva, Malone, et al. (2002).