A composite depth scale places coeval, laterally continuous stratigraphic features into a common frame of reference by shifting the mbsf depth scales of individual cores to maximize the correlation between holes (Sacks, Suyehiro, Acton, et al., 2000). The individual cores are shifted vertically without permitting expansion or contraction of the relative depth scale within any core. Deviations of mbsf depths from true depths arise from factors such as uncertainties in depth measurements (most of which can be attributed to ship motion), core expansion, and incomplete recovery. Therefore, a horizontal feature present in recovered material from several holes will have, in the absence of local bathymetric variations, the same true depth but will likely have different mbsf depths. Errors in the mbsf depth scale range from a few centimeters to several meters, though rarely more than ~10 m (e.g., Acton et al., 2001, and references therein). After establishing a meters composite depth (mcd) scale, more complete stratigraphic records are spliced from the data from multiple holes.

The methods used during Leg 198 were similar to those used to construct composite depth sections during Legs 138 (Hagelberg et al., 1992), 154 (Curry, Shackleton, Richter, et al., 1995), 162 (Jansen, Raymo, Blum, et al., 1996), 167 (Lyle, Koizumi, Richter, et al., 1997), 178, (Acton et al., 2001), 186 (J. McGuire and G.D. Acton, unpubl. data), and 189 (Exon, Kennett, Malone, et al., 2001). At each site, closely spaced (2.5- or 3-cm interval) measurements of magnetic susceptibility and gamma ray attenuation (GRA) bulk density were made on the MST soon after the core sections had equilibrated to room temperature. The susceptibility values are presented as raw meter values, which can be converted to SI volume susceptibility units by multiplying by ~0.68 x 10^-5 (Blum, 1997). These measurements were entered into the shipboard Janus database. In addition, measurements of spectral reflectance were made at 2.5 cm resolution on the split cores (see "Lithostratigraphy"). Magnetic susceptibility values, total reflectance (L*) values from the spectral reflectance measurements, and GRA bulk density measurements from each hole were compared to determine if coring offsets were maintained between holes. Integration of at least two different physical properties allowed more reliable hole-to-hole correlations than would be possible with a single data set. In addition, magnetic susceptibility data collected from intervals that were disturbed by the drilling and recovery process were removed before making correlations for the mcd scale. This includes gaps caused by the removal of 5-cm-long interstitial water (IW) samples, which are taken from the ends of some sections prior to MST and other measurements.

The mcd scale was constructed using the program SPLICER (version 2.2, available on the Internet from Lamont-Doherty Earth Observatory Borehole Research Group [LDEO-BRG] at http://www.ldeo.columbia.edu/BRG/ODP). SPLICER allows data sets from several holes at a given site to be correlated simultaneously. Corresponding features in data sets from adjacent holes were aligned based on graphical and mathematical cross-correlations using an iterative process. Features were aligned by adjusting the ODP coring depths in mbsf, measured from the length of the drill string advanced, on a core-by-core basis. No depth adjustments were made within an individual core. Correlations were first made visually by selecting a tie point from data in one hole and comparing it directly with data from another hole. Ties are intended to correlate data based on matching patterns and amplitudes of the data sets.

Cross-correlation coefficients for all data sets were calculated within SPLICER. Depth adjustments were chosen that provided the best correlation within a preferred data set or the best compromise of correlation coefficients among all the data sets. The values of the cross-correlation coefficient vary from +1 to -1, with +1 indicating perfect correlation (such as would be obtained by comparing identical data sets) and -1 indicating anticorrelation (such as would be obtained by comparing a data set to its inverse). Values near zero indicate poor or no correlation. Each time a depth adjustment is made within SPLICER, the coefficient is recalculated, allowing the user to determine the preferred correlation. The window over which the coefficient is calculated is adjustable. The default window length of ±2.00 m on either side of the selected tie point was used for most correlations. This window was reduced to ±1.00 m as needed to focus on features of interest or to avoid spurious features such as those biased by coring disturbances.

Correlation began by selecting the core that had the most pristine record of the upper portion of the upper few meters of the sedimentary record, particularly the mudline. This first core is defined as the top of the composite section, and its mcd depth is the same as its mbsf depth. At Site 1209, the top of the composite section was Core 198-1209A-1H. A tie point that gives the preferred correlation was selected between data from this core and a core in a second hole. All of the data from the second hole below the correlation point were vertically shifted to align the tie points between the holes. Once the appropriate tie was determined and the depth adjustment was made, the shifted section became the reference section, and a tie was made to a core from the first hole. The process continued downhole, vertically shifting the data one core at a time, relative to data from the other hole. By tying points of different mbsf depths, SPLICER vertically adjusted the individual sections of the cores and brought the chosen anomalies into the common mcd depth scale. The tie points were added to the SPLICER "affine" table, which records all of the depth adjustments that define the composite depth scale, in mcd.

The composite depth section for each site is presented in tabular form in the "Composite Depths" section of each site chapter. The composite depth table of Site 1209 is given as an example in Table T5. For each core, the depth adjustment required to convert from the mbsf depth scale to the mcd scale is given. The last two columns in each table give, for each core, the cumulative-depth offset added to the ODP curatorial subbottom depth (in mbsf) and the composite depth (in mcd), respectively. The depth-offset column facilitates conversion of sample depths that are recorded in ODP curatorial subbottom depth (in mbsf) to composite depth (in mcd). The equivalent depth in mcd is obtained by adding the amount of offset listed to the depth in mbsf of a sample taken in a particular core.

The need for a composite section to verify stratigraphic continuity is illustrated in Figure F7. In the leftmost panel, magnetic susceptibility data from three holes at Site 1211 are shown on the mbsf depth scale. In the middle panel, the same records are shown after depth scale adjustment so that correlative features are aligned. The correlation of lithologic parameters between parallel holes and associated depth adjustments for individual cores was optimized in such a way that a single record could be sampled from the aligned cores without any additional depth scale changes. The right panel shows the resulting spliced record. Where the amount of offset necessary to align features was ambiguous or imprecise for all lithologic parameters or where multiple hole data were unavailable (e.g., for the sediments below the uppermost Maastrichtian), no additional depth adjustments were made. In these cases, the total amount of offset between mbsf depth and mcd is equal to the cumulative offset from the overlying cores. The composite depth section extends to the base of the core including the K/T boundary interval.

For each site, after the composite section was constructed, a representative spliced record was assembled. Because the missing intervals between successive cores in the sedimentary sequence could be identified, it was possible to splice in these missing intervals with data from adjacent holes. The resulting composite record ("splice") provides a single representative record of each lithologic parameter (i.e., magnetic susceptibility, spectral reflectance, and GRA bulk density) for a given site. Additionally, these records are ideally suited for serving as sampling schemes for paleoceanographic studies.

Splice tie points were made between adjacent holes at identifiable, highly correlated features. Each splice was constructed beginning at the mudline at the top of the composite section and working downward. Typically, one hole is chosen as the backbone for the record, and cores from other holes are used to patch in the missing intervals in the core gaps (Fig. F7). Intervals were chosen for the splice so that section continuity was maintained, whereas disturbed intervals were avoided.

Tables that give the tie points for construction of the spliced records are presented in each site chapter. By identifying intervals where features present in the multiple-cored holes were most highly correlated, it was possible to construct a spliced record that avoided duplication or omission of individual features or cycles. Splice tie points always connect features with exactly the same composite depths. As a result, the final alignment of the adjacent holes could be slightly different from the best overall visual or quantitative hole-to-hole correlation. Further adjustments to the composite depth section by expanding and compressing the depth scale within individual core intervals are required to align all features exactly. These additional adjustments will be made as part of normal postcruise studies.