Based on correlation between magnetic susceptibility, NGR, GRAPE, P-wave velocity, and color spectral reflectance data, a record extending from late Miocene up to the present was obtained with ~85% recovery in the upper 200 mbsf of three holes drilled at Site 1095 (Hole 1095C was not used). A composite section to the base of the hole was developed (see "Composite Depths"  in the "Explanatory Notes" chapter). The base of the Site 1095 composite section is at 561.91 mcd (561.37 mbsf in Hole 1095B). Continuity of the stratigraphic section was confirmed from the mudline to 92.03 mcd (87.41 mbsf in Hole 1095A and 91.49 mbsf in Hole 1095B). Below this depth, successive cores from Hole 1095B were depth corrected by addition of the offset.

The depth offsets that were used for the composite depth section for Site 1095 are given in Table T36. Table T36 shows the depth offset values (in meters) that must be applied to the depths scale (mbsf) in order to optimally align all cores from every hole at a given site. Each depth offset value refers to the depth increment that must be added to (or subtracted from) each core so that a composite spliced record can be constructed. Depth offset values (Fig. F48) typically increase with increasing depth because they are cumulative depth corrections determined for each core.

At Site 1095, whole-core, high-resolution measurements were taken every 2 cm for magnetic susceptibility, GRAPE density, and P-wave velocity and every 15 cm for NGR for all cores in Holes 1095A, 1095B, and 1095D. P-wave data were not measured for XCB cores. GRAPE and magnetic susceptibility data, which proved to be the most useful for correlation, are displayed on the composite depth scales in Figure F49. Additional measurements of spectral reflectance on the core were done at a 5-cm interval with the Minolta color scanner. Initially, all data sets for each hole were visually and quantitatively compared to check their consistency. Subsequently, GRAPE density measurements (see "Physical Properties") and magnetic susceptibility (as the most continuous and consistent data sets) were used as the primary parameters to determine depth offsets for the composite depth section. All other parameters were useful to check the hole-to-hole correlation. No smoothing or culling was applied to the data.

Correlation between all data sets was very good, at least for all combined depths up to 480 mcd. The magnetic susceptibility was inversely (the first 55 mcd) and directly (the remaining data sets) correlated with the reflectance parameters (except for the L* parameter). Hole 1095A magnetic susceptibility correlated with chromaticity parameter a*, whereas Hole 1095B correlated better with chromaticity parameter b*. The spliced section, however, was built using the same chromaticity parameter a* for all holes for consistency.

The overlap between adjacent holes was documented throughout the comparative analyses (cross correlation) between the cores. It was found that the cores in Hole 1095D recovered a signal similar to those in Hole 1095A. The MST and reflectance record was, however, more expanded in Hole 1095A (at least in the upper two cores), which most likely reflects a change in sedimentation rate instead of a correlation stretching effect. This was confirmed by sedimentological analysis (see "Sedimentation Rates"). For this reason, the cores from Hole 1095D were predominantly used to build the spliced section.

In addition, it was noted (Fig. F50) that the magnetic susceptibility, the reflectance parameters (except for L*), and the GRAPE density showed a correlated cyclic pattern in the raw data, which is better preserved in the upper 200 mcd. The same cyclic pattern is observed in the P-wave data (values in the 1450-1700 m/s range). The relative agreement of different sedimentary features in adjacent holes was excellent. However, minor stretching and compression in the cores were unavoidable. Much of this distortion occurred on a scale of <1 m. Core stretching of the decimeter to centimeter scale would be required to align all sedimentary features. Because the Splicer software does not allow such adjustment, the cores in which the wavelength of the cyclical pattern was similar were chosen to build the spliced section (which must be considered preliminary; see "Composite Depths" in the "Explanatory Notes" chapter).

Following the construction of composite depth sections for the site, single spliced records were built as described in "Composite Depths"  in the "Explanatory Notes" chapter. The tie points for the Site 1095 splice are given in Table T37. Plots of spliced reflectance and magnetic susceptibility data are shown in Figure F51.