At Site 1146, physical properties were measured on whole-round sections, split-core sections, and discrete samples from the latter. Whole-core logging with the MST included GRA bulk density, MS, NGR, and P-wave velocity logging on all cores. Sampling intervals were 5 cm for all cores in the three holes. The P-wave logger (PWL) data were bad because of instrument problems and/or cracks or voids in the sediment cores. The PWL data are not shown in this report but are available from the ODP JANUS database (see the "Related Leg Data" contents list). One thermal conductivity measurement per core was also performed on whole-round sections from Holes 1146A and 1146B as long as sediment conditions allowed. Color reflectance was measured on the archive halves of all split cores at 4-cm intervals. Moisture, density, and P-wave velocity were measured on discrete samples from split-core sections at intervals of one measurement per section (1.5 m) (see "Physical Properties" in the "Explanatory Notes" chapter).
Similar to the previous sites, one major feature affecting the core physical properties measurements is related to the change from APC to XCB coring (207.2 mcd in Hole 1146A, 216.30 mcd in Hole 1146B, and 170.5 mcd in Hole 1146C). The XCB cores are moderately disturbed by partial remolding and incorporation of drilling slurry, and the reduced diameter of these cores is probably the main component of the pronounced offset of GRA and somewhat slighter offset in MS and NGR values at those depths (Figs. F19, F20, F21). This effect is not compensated for because there was no time to perform a careful correction on board ship.
The other primary features in the physical properties data can be ascribed to changes in sediment composition. We can distinguish five major intervals, which are represented by clear changes in the physical properties of the sediment.
The first interval shows constant or slightly increasing values in MS and NGR (Figs. F19, F20). The GRA, bulk density (Fig. F21), and P-wave values (Fig. F22) increase normally with depth, having a steeper slope in the top 25 mcd and a distinct depression between 25 and 35 mcd. The porosity reveals an inverse pattern in this interval, and grain density displays a large scatter (2.62-2.75 g/cm3) (Fig. F23). This short feature in the first interval appears to be related to changes in the carbonate content (see "Organic Geochemistry") and may result from glacial-interglacial cycles. The L* reflectance record (Fig. F24) and NGR (Fig. F20) also denote the slightly increasing trend, with superimposed high-amplitude variations showing patterns very similar to stable oxygen isotope records. Much lower amplitude variations, which appear to correlate with glacial-interglacial cycles, are also displayed by MS, which does not exhibit a general downhole trend. The chromaticity ratio a*/b* reflects neither the cycles nor a general trend (Fig. F24).
The second interval is characterized by lower porosity and an increase in bulk density, MS, and NGR (Figs. F19, F20, F21, F23), indicating a major change in lithology and/or sedimentation rate. The most drastic increase is observed in the MS record, where values rise from ~20 (10-5 SI) above 110 mcd to >80 (10-5 SI) between 140 and 210 mcd. Several ash layers within this interval give high MS signals and can be found in all three holes. Both the carbonate content (see "Organic Geochemistry") and the L* value (Fig. F24) stay almost constant until 235 m.
The third interval is characterized by a dramatic increase in carbonate (see "Organic Geochemistry") together with a strong decrease in NGR (Fig. F20) and a slighter decrease in MS signals (Fig. F19) (probably a carbonate dilution effect). Bulk density starts to increase slowly from a low around 230 mcd, with little variation in the discrete moisture and density (MAD) values. The GRA bulk density values are slightly lower on average and display a very prominent sawtooth pattern (Fig. F21). This is likely to be an artifact induced by stretching and compression during coring and recovery. The lower GRA values result from reduced core diameter through the XCB coring; no correction has yet been applied. At ~410 mcd in Hole 1146A and 435 mcd in Hole 1146C, the sawtooth pattern becomes less obvious, indicating that the sediment is less sensitive to the coring influences. Porosity is characterized by long-period (tens of meters) fluctuations superimposed on a generally decreasing downhole trend, as is typical for the entire drilled interval. Grain density in this third interval varies around 2.72 g/cm3, with some low values and stronger scattering in the lower part until ~400 mcd (Fig. F23). The values of P-wave velocity in this third interval increase steadily with some scattering because of the core conditions (Fig. F22). Negative excursions in the chromaticity ratio at 325, 355, and 418 mcd present in Holes 1146A and 1146C indicate compositional changes in the sediment that are not always visible as color changes (Fig. F24). The relatively high L* values still reflect a high carbonate content. The strong negative chromaticity excursion observed in Hole 1146A at 408-416 mcd is not found in Hole 1146C. These negative chromaticity excursions are accompanied by drastic drops in MS at the same depths, indicating short but major changes in lithology/chemistry in these intervals.
The fourth interval is marked at 420 mcd with a pronounced downhole increase in NGR and MS (Figs. F19, F20), most likely the result of the lower carbonate content below that depth. The NGR values increase more or less steadily down to 550 mcd, with a small peak at ~460 mcd. As in the interval above, the MS record again drops to near zero in some short intervals that are accompanied by negative excursions in the chromaticity ratio. The L* reflectance record (Fig. F24) decreases very similarly to the carbonate content (see "Organic Geochemistry"; Fig. F17). The GRA and discrete bulk density increase without major excursions (Fig. F21); the GRA values still exhibit a sawtooth pattern but to a far lesser extent than in the third interval. P-wave values vary between 1700 and 1800 m/s with a very small downhole increase to 515 mcd, where a sharp increase in value up to 1920 m/s is observed (Fig. F22). Below 515 mcd down to 532 mcd, no P-wave velocities could be measured because the response signal was too low. From 532 to 550 mcd the velocity values were again received, but they dropped to a lower level around 1800 m/s. Porosity decreases normally; grain density stays constant within a narrow bandwidth at ~2.74 g/cm3 down to 505 mcd. Below this depth the grain density values are significantly more scattered, and overall values decrease to a low of 2.68 g/cm3 at 550 mcd (Fig. F23).
The fifth interval is best documented in the chromaticity ratio a*/b*, which drops sharply between 540 and 550 mcd (Fig. F24). The change in sediment color is also observed visually (see "Lithostratigraphy"). The MS values also drop at this depth (Fig. F20), indicating that the drop in chromaticity may be accompanied by a change in mineralogy. Grain density increases from a low at 550 mcd to values around 2.74 g/cm3 and then stays fairly constant. P-wave velocities first increase to a high of ~580 mcd with values up to 1960 m/s, then decline. At 620 mcd, no measurements were possible because no signal could be received. We ascribe this to the high gas content prevalent in the lower part of the hole (see "Organic Geochemistry"). A similar archlike feature in the P-wave velocity record (Fig. F22) of the lowermost part of the hole was noted in the downhole logging data (see "Wireline Logging").
Thermal conductivity data from the APC and XCB cores range from 0.85 to 1.30 W/(m·K) (Table T16, also in ASCII format; Fig. F25). The values from XCB cores are compromised by poor core quality, particularly in the upper XCB interval. The values from APC cores show a gradual downhole increasing trend. A slight increase at ~80 mcd is observed, which corresponds to a downhole rise in bulk density at that depth.
Four downhole temperature measurements with the APC temperature tool were taken in Hole 1146A at depths of 32.4, 60.9, 98.9, and 150.0 mbsf, respectively. In addition, a bottom-water temperature measurement was taken before coring in Hole 1146B (Fig. F26). The objective was to establish the local heat flow. Original temperature records were analyzed using "Tfit" software to establish the equilibrium temperature at depth. Measurements at 32.4 mbsf seem problematic. The estimated errors in equilibrium temperature vary from 0.3° to 0.5°C, reflecting the amount of frictional heat introduced near the sensor by the ship's heave during the 10-min measurement. Depth errors are on the order of ±0.5 m. The measurements between 0 and 150 mbsf yielded a thermal gradient of 59°C/km (Fig. F27).