Previous   |    Next

doi:10.2204/iodp.proc.339.104.2013

Stratigraphic correlation

The meters composite depth (mcd) scale for Site U1386 was based primarily on correlation of magnetic susceptibility between holes. NGR and GRA density data provided confirmation of the correlation in some intervals and were used as constraints in other intervals where the magnetic susceptibility was low or lacked correlative features. GRA density provided the primary constraint for the first core in each hole. Both cores recovered the mudline, but Core 339-U1386B-1H was shifted (expanded) on average downhole by ~11 cm relative to Core 339-U1386A-1H. The density anomalies could only be correlated well with each other downhole to ~30 mcd.

The correlation is relatively straightforward downhole to 70 mcd but has several uncertain links between cores below this depth (Fig. F55). The first of these uncertainties is with the connection between the base of Core 339-U1386A-8H and the top of Core 339-U1386B-8H. The lack of distinct susceptibility or density anomalies and the relatively broad NGR anomaly in this junction only constrain the correlation to within about ±50 cm. The connection between the base of Core 339-U1386B-10H and the top of Core 339-U1386A-11H is uncertain because of coring disturbance, which is severe over most of Core 339-U1386A-11H. Correlation below ~110 mcd, which is just below the interval in Cores 339-U1386A-12H and 339-U1386B-11H that contains a sequence of very distinctive cyclic susceptibility anomalies, is much more uncertain. Part of the uncertainty arises from real coring gaps and part from coring disturbance, as virtually all XCB cores are partially disturbed, with the disturbance often consisting of the ubiquitous occurrence of coring biscuits (see “Lithostratigraphy”).

To achieve a composite depth scale beyond 110 mcd, we first shifted the cores with a constant growth factor of 1.11 for the three holes at Site U1386, indicating an 11% increase in mcd values relative to meters below seafloor (mbsf) values (Fig. F56). This growth factor fits the upper part of the sequence. Subsequently, we shifted the individual cores upward or downward for the best correlation between both holes, taking into account that successive cores in each hole do not overlap (Table T24). When no apparent correlation existed between the base of one core and the top of the next in the parallel hole, the lower core (the one deeper in the mbsf scale) was appended to the base of the one above. Given the lack of evidence for overlap, appending cores to each other improves the odds that the spliced section will contain a complete section, albeit with the possibility of some duplication. Some serious difficulties are posed, however, by incomplete cores or cores with no recovery (e.g., Cores 339-U1386A-25X and 339-U1386B-20X) and conflicting magnetic susceptibility values between parallel cores (e.g., Cores 339-U1386A-24X, 339-U1386B-23X, and 339-U1386C-4R). In these cases, the comparison between NGR and logging HSGR data was used to improve construction of the composite record.

The three holes cored at Site U1386 provide enough material to produce a splice with relatively few gaps within the upper 390 m of the section (Table T25). Below this depth, most of the section is single-cored, with coring gaps undoubtedly occurring between each core (Fig. F55). Where possible, we avoided including intervals in the splice that were disturbed significantly by drilling, voids, and interstitial water samples (Table T14). In addition, all magnetic susceptibility data were cleaned for the top of each section and measurement outliers (Table T26).

To facilitate a direct comparison of the highly expanded spliced record of Site U1386 with the original mbsf scale, we transferred the mcd scale to mbsf values (i.e., mbsf*) by applying both a linear regression fit through the core tops of all holes (Fig. F57) and forcing the fit line through the origin (i.e., at zero mcd and mbsf*), revealing an average compression factor of 0.9006472906.

Finally, to test the robustness of our spliced record, we compared the NGR core data in Holes U1386A and U1386B with the logging HSGR data of Hole U1386C (Fig. F58). Because of its relatively low resolution (~20 cm), we included all NGR measurements in both holes and plotted them on the mbsf* depth scale, using a five-point moving average. The HSGR data are plotted on the logging mbsf scale, which was directly derived from the tool string. Characteristic features within both records between 100 mbsf (top of the logging operations) and 380 mbsf clearly correlate to one another, suggesting that the spliced core data contains an almost continuous stratigraphic succession. The minor misalignment between NGR and HSGR anomalies are caused by very small differences between the logging mbsf and the mbsf* depth scales, which can be resolved using a few tie points (Figs. F58, F59). Evidently, the comparison of NGR and HSGR data illustrates that the spliced section is virtually complete downhole to 380 mbsf*, providing a useful guide for postcruise studies to plan sampling strategies.

Top of page   |    Previous   |    Next