The purposes of downhole physical properties measurements during Leg 178 were to provide near-continuous and submeter-scale records for hole-to-hole correlation and estimates of sediment properties, which can be used to reconstruct glacial and interglacial/proglacial depositional processes.
At the completion of coring operations at Sites 1095 and 1096, we logged the deepest holes (Holes 1095B and 1096C) using standard wireline techniques. The depths of investigation are sensor dependent, and data are recorded at intervals of 15.24 cm (0.5 ft). We completed two downhole logging runs in Hole 1095B (Shipboard Scientific Party, 1999a): a triple combination (TC) (lithodensity, porosity, resistivity, and natural gamma ray) log and a geological high-sensitivity magnetic tool (GHMT) log. Logging operations were hindered by the intermittent approach of icebergs and heavy ship heave, which was large enough to cause damage to some of the logging tools. Thus, we only ran the integrated porosity-lithology tool (IPLT) (natural gamma ray, porosity, and density) and the GHMT (natural gamma ray, magnetic susceptibility, and total magnetic field) strings in Hole 1096C (Shipboard Scientific Party, 1999b). As we encountered a blockage at 343 mbsf during the first IPLT run, we logged this hole in two sections.
Borehole caliper measurements showed that Hole 1095B was typically 40-45 cm (15.7-17.7 in) in diameter (Fig. F2), with some zones of wider washout beyond the maximum caliper extent of 47 cm (18.5 in). The washed-out zones resulted in poor contact with the borehole wall and hence poor quality for the density and porosity logs. The deeper penetrating logs, such as medium resistivity (IMPH) and magnetic susceptibility (MAGS), are much less affected by changing borehole diameter (Fig. F2). We can divide the sequence into two units on the basis of changes in the character of the downhole logs (Fig. F2). Within Unit 1 (100 [base of pipe] to 510 mbsf), we note a downhole increase in the base level of susceptibility variations, with a boundary marked by a (downhole) step decrease at 325 mbsf. The lower part of this unit contains a pattern that repeats about three times (Fig. F2). Each repetition is ~40 m thick and shows a steady uphole increase in resistivity and spectral gamma ray (HSGR), topped by a sharp decrease; magnetic susceptibility behaves in the opposite way. The transition from Unit I to Unit II (510-570 mbsf) is marked by a step increase downhole in resistivity, spectral gamma ray, and magnetic susceptibility, with a slight increase in variability for resistivity and magnetic susceptibility logs.
Hole 1096C was also very wide. Thus, the density and porosity logs should be regarded with caution, although neither displays the spikes that are characteristic of bad contact of the tool with the borehole wall (Fig. F2). Because of the interrupted logged depth intervals, we can not divide the formation into units on the basis of the logs alone. The logs are rich in variability, which can be related to the alternations seen in the cores.
The downhole logs were interactively depth-shifted with reference to the natural gamma ray by the Borehole Research Group (BRG) at Lamont-Doherty Earth Observatory (LDEO), and natural gamma ray data were corrected for borehole size and type of drilling fluid. As large and/or irregular borehole adversely effects recordings that require eccentralization and a good contact with the borehole wall, the porosity and density data measured in Hole 1095B were not processed with wavelets.
The physical properties measurements were made with a multisensor track (MST), which combines four sensors on an automated track for measuring whole-core magnetic susceptibility, bulk density (by gamma ray attenuation), P-wave velocity, and natural gamma ray emission. Whole-core magnetic susceptibility, bulk density, and P-wave velocity were measured every 2 cm, whereas natural gamma was measured every 15 cm. Color spectral reflectance (lightness parameter L*) was then measured on the split core at 5-cm intervals (Fig. F3) using a spectrophotometer (Shipboard Scientific Party, 1999a).
Before applying the wavelet analysis, the raw data were smoothed using a five-point running average. There are very large gaps in core measurements, due to the incomplete recovery of cores. For Hole 1096C, these gaps occur between 123 and 255 mbsf (Fig. F3). To avoid spurious features that could result from the interpolation over these gaps, we use only the data from below 255 mbsf. We then split the data into several files to eliminate the zones without core recovery over more than 4 m (e.g., Hole 1096C lightness parameter L* from 378 to 386 mbsf) (Fig. F3). The interpolation over these zones would have given spurious results with the wavelets analysis. The same process was performed for the data from Hole 1095B. Each interval was interpolated every 2, 5, or 15 cm to be later analyzed separately.
We also constructed a signal representing the bioturbated intervals and their localization in depth (Fig. F3). For that, we first digitized the master lithostratigraphic column showing dominant lithology of Hole 1095B (from Shipboard Scientific Party, 1999b) and then extracted the numerical components as a function of depth. The depth resolution of the resulting bioturbation log is controlled both by the space between the finest and closest bioturbated levels and by the number of pixels in the image. The estimated resolution here is ~30 cm. Even if this resolution is much lower than the actual core descriptions, we thought it would be interesting to analyze this signal because these bioturbated intervals might be representative of the sea temperature, currents, sediment supply, or some other property directly related to climatic variations.
The natural gamma ray (NGR) count in Hole 1095B shows an overall decrease with depth. Between 270 and 300 mbsf, it decreases sharply from ~13.5 to 7 counts per second (cps), which is not matched in the downhole log measurements (Figs. F2, F3). The gamma ray count in Hole 1096C shows an increase with depth in the first 60 mbsf, followed by a broad decrease with depth to the base of hole, possibly with a weak 50-m cyclicity (Fig. F3). There is also an inverse correlation between the NGR data and the biogenic component of the sediments (Shipboard Scientific Party, 1999c).
We can compute theoretical insolation received by the Earth from its orbital parameters and its obliquity and precession (e.g., Bretagnon, 1974; Berger, 1976, 1978; Berger et al., 1989; Laskar, 1988, 1990, 1993, 1999). The insolation curve is of great interest, for it represents the major input in terms of paleoclimatic changes such as alternations of glacial-interglacial periods and of global ice volume. We computed the insolation at 67°S (the latitude of Sites 1095 and 1096) to identify the main cycles present in this signal and compare them with the cycles registered in the sediments of Holes 1095B and 1096C. We obtained the insolation for each month of the year using the program La93 developed by J. Laskar (1993, 1999). The insolation, sampled every 1000 yr, was computed over the last 10 m.y., the time interval of our study (Fig. F4). We then converted the insolation from age to depth (using the geomagnetic reversals age model presented in the next paragraph) to be homogeneous with downhole and core data.
When we perform spectral analyses for a signal as a function of depth, we obtain wavenumbers in cycles per meter. The next step is to transform these wavenumbers into periods (kiloyears per cycle) using the sedimentation rates. As the variations in sedimentation rate for both sites are significant, the use of an average sedimentation rate over the entire section was not accurate enough. We therefore used the sedimentation rates inferred from geomagnetic reversal identifications on split cores or the GHMT log (Shipboard Scientific Party, 1999b, 1999c), as discussed in "Time-Frequency Analysis Results and Discussion". The overall trend for Site 1095 is an uphole decrease in sedimentation rate from ~11 cm/k.y. near the base of the hole to ~2.5 cm/k.y. at the top (Fig. F5A). Site 1096 sedimentation rates show two main intervals: from the bottom of the hole to a depth of 216 mbsf, the sedimentation rate averages ~18 cm/k.y., and above this interval, ~9 cm/k.y. (Fig. F5B). For Site 1095 we also used the sedimentation rates inferred from radiolarian and diatom data (Fig. F5A) to compare the results and estimate the discrepancies induced by the different age models.