Previous   |   Next

doi:10.2204/iodp.proc.317.104.2011

Physical properties

Gamma ray attenuation (GRA) densitometer bulk density, magnetic susceptibility (loop sensor; MSL), NGR, and P-wave velocity measured with the P-wave logger (PWL) were measured on whole-round core sections from Holes U1352A–U1352C. Additionally, magnetic susceptibility (point sensor; MSP) and spectrophotometry and colorimetry were measured on section halves from the same holes. Whole-round measurements on XCB and RCB cores were degraded in quality because of drilling disturbance associated with coring and the slightly smaller diameters of XCB and RCB cores relative to the core liner. Discrete P-wave velocity (measured using the P-wave caliper [PWC] and P-wave bayonets [PWB]) and moisture and density (MAD) were measured on section halves from Holes U1352B and U1352C. Sediment strength was determined only on section halves from Hole U1352B. Unless otherwise noted, all depths in this section are reported in m CSF-A.

Gamma ray attenuation bulk density

GRA bulk density was measured with the densitometer at 5 cm intervals (measurement time = 5 s). Below 1361.4 m, the sampling resolution was increased to 2.5 cm (Section 317-U1352C-87R-1). The raw data range from –0.42 to 2.467 g/cm³ in Hole U1352B and from –0.417 to 2.606 in Hole U1352C (Fig. F34). Low values are assumed to represent cracks in the cores, gas, and/or incomplete filling of the cores with sediment. Thus, only bulk density values >1.4 g/cm³ were compared with MAD-derived bulk density and downhole GRA density estimates (Figs. F35, F36). GRA bulk density records from Holes U1352B and U1352C show the same long-term trends even where XCB cores from Hole U1352B overlap RCB cores from Hole U1325C.

A comparison of GRA densitometer and MAD data from samples in Hole U1352B (Fig. F35) yields results similar to those from Hole U1351B. That is, GRA data tend to be lower than, but roughly parallel to, MAD bulk density estimates. Lower bulk density estimates mainly reflect incomplete filling of the core liner as well as cracks in the core caused by gas expansion. These features do not affect MAD results. In general, higher GRA bulk density values are consistent with MAD estimates, and both methods reveal a trend toward higher bulk densities with depth. The increase and broadening of bulk density estimates across the lithologic Subunit IB/IC boundary (Fig. F35A–F35B) may reflect the sampling of increasingly calcite-cemented soft sediments.

In Hole U1352C, both MAD and GRA bulk density estimates generally increase with depth (Fig. F36). As the sediments become increasingly cemented, filled core liners become increasingly rare. In a few locations, GRA estimates reach those obtained by MAD analyses. The two trends of MAD data between 550 and 1200 m (Fig. F36A–F36B) are a result of cementation, with the higher MAD bulk densities reflecting more highly indurated strata (see "Moisture and density"). GRA data in this region (lithologic Subunit IIA; see "Lithostratigraphy") include measurements of both denser cemented sediments and less dense uncemented sediments. Lower GRA values (Fig. F36B) suggest that parts of the core liner were not filled. Below 1200 m (lithologic Subunit IIB), unlithified strata are rare and GRA results are consistently lower than MAD results by ~0.2 g/cm³ (Fig. F36A, F36C), which can be attributed to the ~10% smaller diameter of the RCB cores relative to the APC core diameter used for calibration.

A comparison of bulk density generated by the different laboratory methods and expanded to include downhole logging results in Hole U1352B is shown in Figure F35. For this comparison, only results from 100 to 500 m are considered because downhole density logs were acquired only in this interval. Very low densities measured by the downhole tool are spurious and may reflect hole enlargement generated by drilling (see "Downhole logging"). Beginning at ~280 m, density log values that exceed MAD values may be the result of the downhole tool sensing the more indurated beds not sampled via APC and XCB coring.

Magnetic susceptibility

Magnetic susceptibility (MSL and MSP) was measured at 5 cm intervals (measurement time = 2 s). Because of the excellent core quality and time-consuming RCB coring at greater depths, the sampling resolution for MSL data was increased to 2.5 cm below 1361.4 m (starting with Section 317-U1352C-87R-1) in order to provide a more detailed magnetic susceptibility record of the lowermost cores from Hole U1352C.

Raw MSL values are 3.7–39.0 instrument units for Hole U1352A, 0–97.2 instrument units for Hole U1352B, and 0–28.1 instrument units for Hole U1352C. The data were filtered using a Gaussian low-pass filter (30 passes; Fig. F34). The overall less noisy loop sensor data, lower amplitudes, and a shift to lower values in Hole U1352C compared to Hole U1352B may have resulted from the mainly cemented sediment and the smaller diameter of RCB cores relative to the core liner.

Magnetic susceptibility was measured at 5 cm intervals on the Section Half Multisensor Logger (SHMSL) using the MSP (Fig. F34). These measurements were made on all sections unless drilling disruption or surface disruption precluded the collection of meaningful results. MSP measurements made on section halves from Holes U1352A and U1352B are noisier than MSL data; however, the key trends are still resolvable after 30-pass Gaussian filtering of the data, and these trends correlate well with MSL data. The noise level in the MSP data increases markedly at ~520 m in Hole U1352B. A faulty MS2F sensor was suspected to cause this deterioration in data quality, and the unit was replaced at Core 317-U1352C-4R. The replacement MS2F sensor measured slightly higher instrument unit values than the older unit, and data quality improved using this new sensor.

In general, higher magnetic susceptibility values from both instruments correspond to darker sediments, and lower values correspond to lighter sediments (see "Spectrophotometry and colorimetry" and Fig. F37). In the uppermost 275 m, both MSP and MSL data display a cyclicity, with peaks and lows (Fig. F34) that are broadly similar to the magnetic susceptibility record from Site 1119 (Shipboard Scientific Party, 1999b) for the last glacial cycle. For example, three peaks between 50 and 70 m in Hole U1352B (Fig. F38) correlate with similar peaks from Holes 1119B and 1119C (between 36 and 46 m) that were interpreted as representing marine isotope Stage (MIS) 5 (Shipboard Scientific Party, 1999b). Between 555 and 630 m, magnetic susceptibility, NGR, and GRA density show a conspicuous negative peak followed by a corresponding positive peak between 630 and 655 m (Fig. F34). Below ~1500 m, magnetic susceptibility and NGR signals decrease overall, whereas color reflectance increases, likely reflecting increasing carbonate content (see "Spectrophotometry and colorimetry").

Below the Marshall Paraconformity at ~1852.6 m (see "Lithostratigraphy"), magnetic susceptibility measured with both the MSP and MSL becomes negative, indicating the predominance of diamagnetic limestone.

Natural gamma radiation

NGR was measured on all core sections at 10 cm intervals down to 1333 m (Section 317-U1352C-82X-1). The values measured range from near zero to >60 counts per second (cps), with higher values associated with muddy lithologies and lower values associated with sands (Fig. F34).

A sustained downhole decrease in NGR from ~55 to ~40 cps occurs between the seafloor and ~1400 m, consistent with decreasing radiogenic terrigenous material and increasing carbonate content. This decrease is punctuated by a major NGR perturbation (with matching displacements in the magnetic susceptibility records) between 555 and 655 m, which comprises a negative excursion to <20 cps, followed by a positive excursion to >50 cps. No obvious lithologic explanation exists for this feature.

A further rapid decrease in NGR to ~15 cps occurs just above the Marshall Paraconformity. Below the unconformity, in the Amuri Limestone facies, NGR reaches its lowest levels of 3–5 cps, indicating very low terrigenous content.

A marked NGR cyclicity is apparent in the uppermost 275 m (Fig. F38), similar to that observed at nearby Site 1119. Postcruise analysis will likely permit the development of a timescale for this interval by comparing it with the Site 1119 age model, which Carter and Gammon (2004) constructed using an inverted NGR record that apparently forms a close New Zealand proxy for global MIS stages to at least 2.4 Ma. Assigning a provisional age of 127 ka to the inferred MIS 5e peak at ~63 m at Site U1352 gives an average sedimentation rate of ~50 cm/k.y. for the last glacial–interglacial cycle, compared with 37 cm/k.y. at Site U1351 and 20 cm/k.y. at Site 1119 over the same interval.

P-wave velocities

P-wave velocity measurements were recorded with different coverage on sections from Holes U1352A, U1352B, and U1352C using the PWL, PWC, and PWB (Figs. F39, F40). The differences in coverage depend on core condition. Poor results were caused by signal attenuation and sediment cracking resulting from high gas content in APC and XCB cores and sediment disturbance related to XCB coring. PWL measurements on whole-round sections from Holes U1352A and U1352B at 4 cm intervals yielded poor results below ~20.0 and 14.0 m, respectively (Fig. F39A–F39B). PWB P-wave velocities were measured only on the uppermost 20 m (y-axis) and 10 m (z-axis) of core, respectively, from Hole U1352B (Fig. F39C). Excellent PWC results were obtained from mainly cemented sediments in RCB cores (Fig. F40A). PWC measurements were conducted on the uppermost 20 m of core from Hole U1352B (Fig. F39C) and continuously on nearly every section half from Hole U1352C (Fig. F40A).

P-wave velocities (PWL, PWB, and PWC) from Holes U1352A and U1352B range from ~1400 to ~1800 m/s, whereas P-wave velocities (PWC) from Hole U1352C range from ~1600 to ~5900 m/s. The significantly higher velocities in XCB cores are probably caused by good conductive contact with the mainly cemented material. All velocity measurements from Holes U1352A and U1352B show an abrupt change from lower to higher values between 1.5 and 2 m. A similar change was observed in the uppermost ~2.0 m of Hole U1351B. Scattered values in the lowermost part of the PWL curves are attributed to decreased core quality (Fig. F39). Whole-round and discrete P-wave velocity measurements from Hole U1352B are positively correlated. PWC and PWB velocity measurements abruptly decrease between 12 and 14 m, approximately the same depth at which the PWL measurements terminate. At greater depth, PWC values from Hole U1352C are inversely correlated with porosity (Fig. F40C), showing a slight increase below 1255 m (average = ~2500–3500 m/s) and a large increase between 1500 and 1670 m (average = ~3500–4000 m/s) (Fig. F40B). Between 1670 and ~1795 m, velocity is continuously high (average = ~3900 m/s). Below ~1795 m, P-wave velocity increases again to ~5900 m/s. A slight increase in P-wave velocity was observed below the Marshall Paraconformity at ~1852.6 m (see "Lithostratigraphy"). The unexpectedly high velocities below 1255 m require a revision of the present traveltime/depth interpretation of seismic records in this area.

Spectrophotometry and colorimetry

Spectrophotometric measurements and associated colorimetric calculations were made on section halves at 5 cm intervals at the same positions used for MSP measurements. Color data were recorded as L*, a*, and b* variations. L* values are relatively constant in Holes U1352A and U1352B. Where Holes U1352B and U1352C overlap between 600 and 800 m, L* values are slightly but noticeably lower in Hole U1352C (Fig. F37). A clear increase in L* values occurs at ~1576 m. This increase can be attributed to a transition to paler, more carbonate-rich lithologies over this interval (see "Lithostratigraphy").

Notably, the degree of scatter in L* data also increases from Hole U1352B to Hole U1352C, and a further, more pronounced increase in scatter occurs downhole in Hole U1352C below ~1600 m. This change in L* between Holes U1352B and U1352C is due to the change from homogeneous, unlithified sediment in Hole U1352B to the more heterogeneous, lithified rock that characterizes Hole U1352C. The additional increase in L* variance downhole in Hole U1352C at ~1600 m is partly due to an increase in reflectivity difference (i.e., contrast) between the dominant pale marlstone and carbonate lithologies and the minor interbedded dark lithologies (e.g., glauconitic muddy sandstone layers; see "Lithostratigraphy"). Furthermore, the change to lithified rock in Hole U1352C is also accompanied by an increase in the occurrence of fractures within the split sections. Spectrophotometer measurements of these fractures resulted in anomalously low L* values.

The a* and b* values are less scattered than L* values, and trends are discernible (Fig. F37). b* values in Holes U1352B and U1352C increase over the uppermost ~1000 m before falling slightly downhole to the bottom of Hole U1352C. Several more abrupt changes also occur, and these are readily discernible in the filtered b* data. The position of abrupt changes in b* values matches the position of similarly abrupt shifts in a*. However, a* values do not show the same long-term trends as b*, and, as demonstrated at Site U1351, the a* record is not as variable as the b* record. In general, a* is negatively correlated with b*, and changes in these two parameters reflect changes from grayer (low b* and high a*) to more greenish lithologies (high b* and low a*).

a* and b* values for Hole U1352A do not correlate well with those for Hole U1352B between ~18 and ~54 m (i.e., the bottom of Hole U1352A). This difference may reflect a problem with instrument calibration in Hole U1352A for this depth interval because the minimum and maximum values of b* and a* in Hole U1352A are anomalous compared to values from the rest of the site. In addition, as observed in L* data, a* and b* data from Holes U1352B and U1352C where the two holes overlap have different absolute values. In particular, b* values at the top of Hole U1352C (i.e., between ~574 and 670 m) are anomalously high compared to b* values from the same depth interval in Hole U1352B (Fig. F37). As inferred for L* data, these differences reflect the fact that different coring techniques preferentially recover different lithologies that may have different color characteristics.

Moisture and density

Throughout Holes U1352B and U1352C, MAD measurements were made on approximately one sample per core for low-recovery cores and on 3–5 samples per core for complete cores with 6–8 sections. Samples were coordinated with thermal conductivity measurements, discrete P-wave (PWC) measurements, chemical analyses, and smear slide samples. Whenever possible, both soft-sediment samples and indurated sediment samples were taken (using syringe or minicorer, respectively). When soft sediments were sampled, the standard Method C analysis was employed. When the minicorer was used and the sample formed a measurable cylinder, the Method D calculation was used in addition to the Method C calculation (see "Physical properties" in the "Methods" chapter).

The first three cemented samples were tested to evaluate the need to soak samples in seawater for 24 h before obtaining reliable wet sample measurements. These test samples were weighed immediately, soaked in seawater for 24 h, and then weighed again. In all cases, sample weight after soaking was greater than that before soaking. However, the difference was never >0.55% of sample weight. Because this is well below other measurement errors, subsequent rock samples were not soaked in seawater for 24 h before their wet mass was determined.

MAD porosity results were calculated using Methods C and D (Fig. F41) with a correlation coefficient of 0.98. Method D, in which wet volume is measured and wet mass is calculated, is best used when pore water may have been lost during sampling, because it tends to slightly overestimate porosity (Fig. F41A). Although occasional chips caused cylinder volume to be overestimated, more often the top and base or sides of the cylinder were not parallel. In either case, the calipers tend to measure the largest diameter or height, resulting in a tendency to overestimate wet volume.

The porosity trend for sediments that were hard enough to be sampled with the minicorer is readily distinguished from that of sediments that were soft enough to be sampled using the syringe method (Figs. F41B, F42) between ~600–1750 m. Indurated sediments within that interval have a relatively constant porosity downhole, whereas the softer sediments follow a decreasing porosity trend downcore until their trend merges with that of the indurated samples at ~1750 m.

Overall, density and porosity trends are clear in Holes U1352B and U1352C (Fig. F42), and the data are consistent over the interval from 575 to 821 m where the two holes overlap. A few semi-indurated samples from Hole U1352B show cementation beginning at ~500 m, whereas softer sediments tend to more gradually decrease in porosity and void ratio and increase in bulk density. Grain density slightly decreases with depth and shows no relationship with cementation, which supports the interpretation that the lithologic makeup of these rocks is consistent throughout.

Downhole logging of Hole U1352B was successful between ~100 and 500 m (Fig. F43). An overall decrease in porosity with depth is overshadowed by very large excursions that are indicated by downhole logging tools and missing from MAD porosity estimates. Despite a relatively wide hole near the top, porosity values derived from density logs are consistent with those obtained by MAD analyses, whereas neutron porosity log values follow similar, though consistently higher, trends (Fig. F43B). Slight offsets between MAD results and downhole log measurements may be related to differences in depth measurement between CSF-A and wireline log matched depth below seafloor (WMSF). Deeper in the hole (Fig. F43C), very high downhole log porosity values (70%–100%) occur where the hole diameter is ≥20 inches, suggesting that the sediments surrounding the hole may have been washed out at that level, generating unrealistically high porosities. In general, where this is not the case, porosities from downhole logging are near or in some cases lower than those obtained from MAD analyses. Again, this could reflect the presence of indurated horizons that were not sampled by the XCB system, despite the apparent near-100% recovery at these depths. The very high porosity excursion seen between ~240 and 300 m in the downhole logs (Fig. F43A) is likely a result of drilling disturbance in the surrounding sediments.

Sediment strength

Sediment strength measurements were conducted on working section halves from Hole U1352B using automated vane shear (AVS) and fall cone penetrometer (FCP) testing systems (Fig. F44). A comparison of both measurement methods is shown in the cross-plot in Figure F44D. Shear strength indicates that sediments range from very soft (0–20 kN/m²) to very stiff (150–300 kN/m²). Vane shear and fall cone shear strength correlate well in very soft and soft sediments, but AVS values are about three times lower in firm to very stiff sediments (standard deviation = 24.5 kN/m²) than FCP values (standard deviation = 81.8 kN/m²). These findings suggest that the applicability of vane shear in firm to very stiff sediments is limited and that sediment strength from vane shear tests is underestimated. Overall, vane shear and fall cone strength data from Hole U1352B are positively correlated (Fig. F44). Between 0 and ~295 m, shear strength increases continuously, indicating a change from very soft to firm sediments. The increase and range of values seen in vane shear tests in the uppermost ~170 m of Hole U1352B are comparable to those from Hole 1119B (Fig. F44C). A pronounced cyclicity like that seen in holes at Site U1351 was not observed. The generally lower sediment strength below ~295 m in cores from Hole U1352B coincides with the change from APC to XCB drilling at 297 m.

Below ~295 m, sediment cracking resulting from high gas content and sediment disturbance associated with XCB coring often prevented optimal insertion of the vane blade. Repeated downcore increases in sediment strength were also observed within longer cores of uniform lithology by both AVS and FCP tests. These downhole increases were interpreted as an effect of secondary sediment compaction caused by drilling stress. Similar drilling effects were observed in magnetic susceptibility data. We caution, therefore, that interpretation of sediment strength using raw data from XCB cores from Hole U1352B requires careful analysis of both the drilling technique used and also of the in situ sedimentology.

Top of page   |   Previous   |   Next