) notation and expressed in per mil (
).
Hole 1098C (Fig. F1) was advanced hydraulic piston cored during ODP Leg 178. The shipboard party measured whole-core MS at Site 1098 at 2-cm intervals (averaged over 2 s) (Fig. F2) (Shipboard Scientific Party, 1999). The late Holocene (0-9 mbsf) contains high-amplitude fluctuations that have an average value of ~50 × 10-5 SI. From 9 to 25 mbsf the magnitude of MS values drops, but high-frequency cycles are still present.
Based on the variability of the shipboard MS record (Shipboard Scientific Party, 1999), we selected four intervals for the detailed high-resolution analysis of this study. The uppermost interval A (2.82-4.26 mbsf) contains two cycles of high- to low-amplitude MS variations that are characteristic of the late Holocene portion of the Palmer Deep sediment record (Fig. F2). Interval B (6.0-7.75 mbsf) records the transition from the relatively low MS values that characterize the middle section of Hole 1098C into the interval of higher values and high-amplitude cycles that characterize the upper record (Fig. F2). Intervals C (14.42-15.81 mbsf) and D (23.47-24.12 mbsf) are from the strongly laminated sediments with low MS values and low-amplitude fluctuations that characterize the early and middle Holocene record.
The chronology of Hole 1098C is based on accelerator mass spectrometry (AMS) 14C dating of bulk sediments and foraminifers from three sediment columns (Hole 1098C, core PD92-30, and core LMG98-02-KC1) (Domack et al., 2001). The AMS dates were corrected for a reservoir effect of 1260 yr and calibrated using INTCAL98 (Stuiver et al., 1998). The composite depth scale for the three cores was determined using the SPLICER program (Acton et al., Chap. 5, this volume). Based on 54 14C dates, the age model proposed for the upper 25 mbsf uses a third-order polynomial to regress the age. The resulting polynomial is a better fit than a simple linear trend (Domack et al., 2001). Applying the Domack et al. (2001) equation, the sedimentation rate in Hole 1098C varies between 170 and 340 cm/k.y., with the highest sedimentation rates occurring in the middle Holocene. Most certainly, the sedimentation rate was variable within each of the four intervals studied for this paper, especially between the laminated and homogenous intervals with each cycle, but such small-scale variability is beyond the limits of radiocarbon dating.
Significant differences in the ages of identifiable MS events (i.e., the large transition in interval B between Hole 1098C and the published record of core PD92-30 (Leventer et al., 1996) is probably due to the fact that the chronology for the earlier core PD92-30 research was based on noncalibrated ages. The most recent chronology (Domack et al., 2001) uses corrected and calibrated ages and is supported by more dates than the previously published version.
All cores from Site 1098 were split and described on board the JOIDES Resolution in March 1998. Restricted shipboard sampling prevented the examination of core-catcher samples. The 130 samples from Hole 1098C for our study were obtained from the ODP Bremen Core Repository in August 1998, ~6 mo postcruise.
The 3- or 5-cm sampling interval for paleontological and isotope analysis was designed to retain the details of the MS cycles. Each sample consists of 13 cm3 of sediment spanning ~1 cm of core depth and are spaced between 7 and 20 yr, based on the sedimentation rate (Table T1) (Domack et al., 2001). Approximately 1 cm3 of material was reserved for diatom analysis. Based on the foraminiferal results from 130 samples, 38 samples were identified for diatom analysis.
Samples for foraminifer and isotope analyses were oven dried at <60°C and weighed (Table T1). Samples were soaked in 200 mL of distilled water to which 5 mL of 10% hydrogen peroxide solution was added. The samples were placed on a shaker plate for 1 hr then wet-sieved at 63 µm. The sieved fraction was oven dried at <60°C and examined. Because of the large amounts of diatom frustules and the high total organic carbon (>1 wt%) (Shipboard Scientific Party, 1999), the samples were very difficult to disaggregate. Usually the wet-sieving process was repeated up to four times before a final weighing. The sample was examined microscopically after each washing to assure that no foraminifers were lost. Benthic foraminifers in the >63-µm fraction were studied. The entire sample was examined or split to contain ~300 foraminifers (Table T1).
In order to describe the most pronounced changes in the benthic foraminifer assemblage, a cluster analysis was performed. A total of 125 samples and 24 species were clustered using the Pearson correlation coefficient and complete linkage with Systat version 5.2. Only samples collected in August 1998 were included in the cluster analysis, and five samples (marked with an asterisk in Table T1) were not included because of low foraminiferal numbers.
Oxygen and carbon isotopic analyses were done at the Woods Hole Oceanographic Institution using a Finnigan MAT 252 mass spectrometer with a Kiel automated carbonate preparation device. The occurrence of calcareous foraminifers was sporadic throughout Hole 1098C, and B. aculeata was the only calcareous species that occurred in sufficient numbers to be analyzed. In general, ~8 specimens of B. aculeata were analyzed in 75 samples. Isotopic values are reported relative to the Peedee belemnite (PDB) standard in delta () notation and expressed in per mil ().
Thirty-eight samples representing various MS values from Hole 1098C were selected for diatom analysis based on the preliminary foraminifer results. Slides were prepared according to a settling method described by Scherer (1995), in which a known mass of sediment (average = 10 mg) is settled onto coverslips of a known area (22 mm2) placed in beakers. Accordingly, quantitative abundance data can be acquired if diatom counts are completed for known areas of the coverslip.
