APPENDIX

Table AT1 gives the data for Hole 1085A, located at 29°22.5´S, 13°59.4´E (water depth = 1713 m, penetration = 604 meters below seafloor [mbsf]).

The sample number column shows lab numbers assigned in the order of arrival from the Ocean Drilling Program Bremen repository. The rows are in ascending order of core and section number, starting with Core 175-1085A-7H. The depth entries are as given on the labels on the packaging in which the samples arrived. Revised depth, final depth, initial age, and final age estimate were obtained as described below. The remainder of the column entries contain the various data collected (see text). If no data were collected, it was either because the sample was not chosen for analysis or because the sample was unsuitable (e.g., absence of the species to be analyzed, in the case of C. wuellerstorfi). The comments column contains numbered comments referring to sample content such as clumps, fish tooth, and so on, as listed in the table notes.

The path from driller's depth to final depth has an intermediate step labeled revised depth. To fine-tune the age estimates for the samples, we first need to fine-tune the depths and reconcile the shipboard ages with the new depths. In the last step, we use spectral analysis to estimate the in situ sedimentation rate based on the assumption that 41-k.y. cycles are present in these early Quaternary sediments. We used the composite depth assignments made on board ship for a first estimate of expansion and gap size between cores.

The revised depth was calculated to compensate for core expansion. In estimating the effect on apparent depth in core, we keep the driller's depth for the top of each core. We assume core expansion of 10% and calculate new depths on this basis. This eliminates the problem of overlapping core depths due to expansion. A 10% expansion is chosen based on typical offset values when splicing the magnetic record of Hole 1085A with that of Hole 1085B (see Shipboard Scientific Party, 1998, table 8). (For comparison, in Leg 130 cores, which contain much less organic carbon, typical expansion values were near 6%; Berger et al., 1993). A 10% expansion also takes account of the fact that there are gaps between successive cores.

The final depth is based on the driller's depth for the top of Core 175-1085A-7H, the correction for expansion, and an adjustment for the gaps between Cores 7H to 10H to account for loss of sediment and the portions of core ends that were not sampled. This was done by using core recovery data and by estimating and connecting the (presumed) 41-k.y. cycles between cores. Core 175-1085A-8H apparently had substantial loss of well over a meter, most likely from the top of the section. We have assigned new depths to core top to account for the inferred loss: Core 175-1085A-8H top depth was reset downward by 1.22 m. Also, Core 175-1085A-9H top depth was reset upward by 0.35 m, and Core 10H top depth was reset downward by 0.84 m.

Leg 175 shipboard stratigraphic data are used for preliminary age assignments. Ages are calculated in a four-step process, initial age, revised age, approximate age, and final age estimate. The initial age is based on a simple interpolation between the biostratigraphic events listed in table 2 of the Site 1085 report (Shipboard Scientific Party, 1998). Revised age reconciles the initial age to the adjusted depths assigned to samples to compensate for core expansion and partial recovery. Approximate age is an adjustment of ages employing the final depth and position of the (assumed) 41-k.y. cycle extracted from ¹⁸O stratigraphy. Using spectral analysis based on this approximate age, a final age estimate is generated in sharpening the peaks for the 41-k.y. cycle (see Table T1).

Four fix-point ages were used for initial age as follows. Each sample was given an initial age based on connecting tie points, by a smooth-fit curve. The estimated depth is the mean between the upper and lower depth of constraint for the event reported; the suggested uncertainty is near 2 m. The fix-point ages are as follows:

1. Tie point = 0.96 Ma; initial depth = 45.57 mbsf.
2. Tie point = 1.25 Ma; initial depth = 55.65 mbsf.
3. Tie point = 1.67 Ma; initial depth = 70.74 mbsf.
4. Tie point = 1.95 Ma; initial depth = 90.99 mbsf.

The tie points are in excellent agreement with paleomagnetic data (Shipboard Scientific Party, 1998; table 7).

To obtain the revised age, initial depths were replaced by revised depths, keeping the initial age for the samples closest to each tie point and recalculating the sedimentation rate. After determining final depths for each sample, the reapplication of this same procedure yields approximate age. The final age estimate is calculated by running a spectral analysis on the ¹⁸O data, tagged by approximate ages. From the extraction of the (presumed) 41-k.y. cycle, a record of instantaneous sedimentation rates can be determined for each core. The core tops are given adjusted shipboard depth assignments (with the exception of Core 175-1085A-8H). The final age assignments are made by multiplying the approximate age by a correction factor. A second spectral analysis was performed using these ¹⁸O C. wuellerstorfi adjusted ages for verifying the fit in phase between cycles. Cores 175-1085A-9H and 10H ended up with overlapping cycles. This was corrected for by adding 35.72 k.y. to Core 175-1085A-10H ages. This had the effect of (1) diminishing the overlap, (2) keeping the tie points within the error bars, and (3) matching Cores 175-1085A-9H and 10H in such a fashion as to have the obliquity cycles at a matching phase for the end of one and the top of the other.

The label "final," of course, refers to the present study only. A more detailed reevaluation of magnetic properties and color sequences collected on board may make additional adjustments in age assignments advisable.