IODP Proceedings    Volume contents     Search
iodp logo

doi:10.2204/iodp.proc.302.202.2008

Revised depth offsets

Based upon these misalignments, the published composite splice has been revised, resulting in a better fit among the three records (Fig. F5; Tables T1, T2). One of the biggest changes occurs where Cores 302-M0004C-2H and 3H overlap with the base of Core 302-M0003A-1H. In the original mcd scale, Cores 302-M0004C-2H and 3H overlapped, and given that both had >100% recovery, this was reasonable. However, the amount of overlap could not be unambiguously determined because Core 302-M0003A-1H ended at a similar depth to Core 302-M0004C-2H and there was a gap before the start of Core 302-M0003A-2H (Fig. F6). Increasing the amount of overlap between Cores 302-M0004C-1H and 2H removed a short interval that contained both a density maximum and a negative geomagnetic feature that could not be reconciled with the stratigraphy of the neighboring cores. The removed feature also looked identical to one that sat immediately above it. Based on the following lines of evidence, it was decided that this was a splicing error:

  • The interval could not be correlated with the stratigraphy in the neighboring sites.

  • The interval occurred at a splice point with minimal overlap.

  • The interval had the same magnetic and physical characteristics as a feature directly above it in the spliced record.

The error was corrected by increasing the overlap between Cores 302-M0004C-2H and 3H.

The revised offsets introduced in these shallow cores are carried over through Cores 302-M0004C-4H and 5X, 302-M0003A-2H and 3H, and the tie made to Core 302-M0004A-1H. Below Core 302-M0004A-1H a different approach was adopted in constructing the rmcd scale that results in differing offsets for Cores 302-M0002A-6X through 13X. Instead of relying on the material recovered from Hole M0002A, where XCB coring provided very poor recovery in the upper tens of meters beneath the seafloor, the composite section is extended to the base of Hole M0004A. This is possible because Cores 302-M0004A-1H, 2X, and 3X all partially overlap, implying duplicate recovery within the same hole. Again, an overall deterioration in core quality accompanied the switch to XCB for Cores 302-M0004A-2X and 3X, and the best available measurements for integrating these cores are environmental magnetic measurements and X-ray fluorescence (XRF) scanning results from U-channel samples (see Figs. F7, F8, F9). Hence, a continuous composite section from the series of well-constrained ties to the top of Core 302-M0004A-1H can be extended to the base of Core 302-M0004A-3X at 26.69 rmcd.

Because of the uncertainty in the positions of shallow cores from Hole M0002A (Cores 302-M0002A-1X through 4X), they are not included in the new spliced record and their original mcd depths are retained. However, the ties between Cores 302-M0002A-5X and 6X with material from Hole M0004A were adjusted using results from XRF scanning of U-channel samples (Fig. F10). The new offsets for these cores are derived partially from the adjustments made to Cores 302-M0004A-2X and 3X and partially from the new ties linking these cores to the revised composite depth scale. Although these offsets provide the best estimate for the relative positions of these cores, their depths remain less constrained than the material above in Holes M0003A and M0004C. Based solely on the paleomagnetic inclination data (Fig. F7), the sections do not appear to be coeval; however, considering the other available data and accepting the poorer core quality and overall reduction in a clear paleomagnetic inclination record below ~20 rmcd, these ties provide the best depth estimate currently available. Tying cores 302-M0002A-5X and 6X into the revised composite depth scale is critical because below Core 302-M0002A-6X material from this hole accounts for the remainder of the Neogene sediments included in the ACEX composite record. The new offsets for Cores 302-M0002A-5X and 6X place these cores 0.08 and 0.76 m deeper in the rmcd scale, respectively, than in the mcd scale (Table T1).

Following the adjustment to Core 302-M0002A-6X, the largest change in the revised offsets occurs at Core 302-M0002A-7X. Originally there was a 4.57 m offset applied to this core, which was difficult to reconcile with the high recovery seen in the surrounding cores (Table T1). Gradually this offset was removed, and the mcd scale was the same as the mbsf scale for cores deeper than Core 302-M0002A-14X. In the revised depth scale, no attempt is made to stratigraphically correlate Core 302-M0002A-7X with material from Cores 302-M0004C-8X and 9X. Instead, the top of Core 302-M0002A-7X is aligned with the base of Core 302-M00002A-6X. To do this requires an offset of 2.62 m. Below Core 302-M0002A-7X, this offset is gradually removed so that by Core 302-M0002A-9X the rmcd scale is the same as the mbsf scale.

The positions of Cores 302-M0004C-8X and 9X are poorly constrained, as they fall after a core with 0% recovery (302-M0004C-7X). In generating the mcd scale, Core 302-M0004C-6X was aligned with Core 302-M0002A-6X, and Cores 302-M0004C-8X and 9X were aligned to Cores 302-M0002A-7X and 8X, respectively. No further improvements were made to these ties, and in the rmcd scale the relative positions of these cores are maintained. Revisions to the mcd offsets of Cores 302-M0004C-6X, 8X, and 9X are determined by the coeval adjustments made to the corresponding cores in Hole M0002A.

Using the revised composite depth scale offsets, a new splice is developed for the upper 27 rmcd (Fig. F6; Table T2) that only incorporates material from Holes M0003A, M0004C, and M0004A. Although the spliced record extends to the base of the cored section, the first tie to Hole M0002A at 26.69 rmcd marks the end of the “continuous spliced record.” Because there is no overlapping recovery, there is no direct check on the stratigraphic positions of cores below Core 302-M0002A-6X, and the revised composite depths are determined by trying to maintain the closest match to the mbsf depths. Based upon the above discussion, it is also clear that no clear ties exist between Cores 302-M0002A-3X and 4X; 302-M0004C-6X, 8X, and 9X; and 302-M0004B-1X and the composite spliced record, and any data taken from these cores must be used with caution.