Shipboard paleomagnetic measurements for Holes 1129C and 1129D consisted of long-core measurements at 5- to 10-cm intervals of the natural remanent magnetization (NRM) and the remanence after alternating field (AF) demagnetization at 20 mT, as described in "Paleomagnetism" in the "Explanatory Notes" chapter. Measurements were performed on archive halves of all APC and XCB cores, except for intervals affected by core disturbance. Long-core measurements established a magnetostratigraphy to a depth of ~550 mbsf, which includes the Brunhes and upper Matuyama epochs. In partially lithified materials below 250 mbsf, measurements on discrete samples were used to confirm polarity determinations obtained from long-core measurements. Discrete samples were also collected from representative core material and subjected to progressive AF demagnetization up to 30 mT. These samples were also used for anhysteretic remanent magnetization (ARM) and isothermal remanent magnetization (IRM) acquisition and demagnetization experiments.
The intensity of initial remanence is very low, with a median of ~2 × 10-4 A/m. High values occur in the uppermost 10 mbsf, as well as anomalous spikes observed at the top of several cores. The NRM is of shallow negative to moderate positive inclination (Fig. F10). After partial demagnetization (20 mT), magnetizations above ~340 mbsf are of moderately to steeply negative inclination, except for isolated outliers. This indicates that "blanket" demagnetization preferentially removes a magnetization of positive inclination. Inclinations are scattered, and the mean (I = -35°) is more shallow than expected for the present latitude of the site. The scatter is larger than expected for the natural variation of the geomagnetic field and larger than observed at other sites on the eastern shelf-slope transect (Sites 1127 and 1131).
We attribute most of the scatter observed in inclination values to instrument noise and partial contamination, reflecting the fact that intensity of the magnetization is significantly lower than at Sites 1127 and 1131. Differences in intensity between this and other sites are more significant in the upper 250 mbsf, with intensities at Site 1129 one order of magnitude lower than at Site 1127. Intensities at Site 1131 are only slightly higher than at 1129. Similarly, the lowest scatter around the mean inclination and the best estimate of expected value for the present latitude of the site is observed at Site 1127, with considerable scatter around the mean at Site 1129. Interestingly, below ~275 mbsf all three sites are characterized by similar intensities. This observation may suggest that diagenetic reactions that partially consume ferromagnetic phases within the sedimentary package are more active in the shallower sites of the transect.
Below depths of ~350 mbsf in Hole 1129C, long-core measurements are affected by persistent core disturbance (biscuits). Discrete samples obtained from disturbed sediments carry a magnetically hard remanence of high intensity, and thus polarity determinations are based on independent measurements on discrete samples and biscuits chosen to exclude disturbed sediment. Measurements of discrete samples reveal a characteristic magnetization that is isolated after removing a spurious overprint, which we interpreted as a drilling-induced remanence. Median destructive fields range between ~15 and 30 mT. In a biscuit from the interval 182-1129C-30X-1, 100-103 cm (Fig. F11A), a soft component with a steep downward inclination is removed first, isolating a characteristic magnetization of normal polarity and moderately high coercivity. In a biscuit from the interval 182-1129C-30X-3, 78-81 cm (Fig. F11B), a similar magnetization is isolated only after demagnetization with inductions of 20 mT. For samples obtained from Hole 1129D, the overprinting magnetization is of negative inclination. In the interval 182-1129D-3R-1, 24-28 cm (Fig. F11C), a prominent "soft" steep magnetization is removed with inductions of 10 mT, isolating a characteristic magnetization of reverse polarity. The different sign of the overprinting magnetization in XCB and APC cores and in RCB cores suggests that this is indeed associated with the BHA.
The magnetic Tensor tool was used in APC Cores 182-1129C-3H through 19H, although reorientation of declination data was unsuccessful. Uncorrected declinations of the NRM long-core measurements of APC cores from Hole 1129C are consistently biased in the direction of the fiducial orientation line (0° declination in sample coordinates). After partial demagnetization, declinations are more scattered, although the bias toward 0° declination persists. Furthermore, when declinations are corrected using Tensor tool data, between-core declinations are more scattered. For XCB cores where azimuthal orientation is not available, declinations are preferentially at 90° from the core fiducial line.
Normal polarity magnetizations are persistent in Cores 182-1129C-1H through 37X (0 to ~340 mbsf) (Fig. F10). Within this normal polarity, we observed intensity fluctuations in NRM after 20-mT demagnetization. These are particularly well developed at depths above 200 mbsf. Because these oscillations probably reflect secular variation of the geomagnetic field, they may provide an independent tool for between-hole stratigraphic correlation after normalizing to account for the effects of variations in concentration of ferromagnetic minerals.
Rock magnetism analysis consisted of measurements of weak-field susceptibility, IRM acquisition, and AF demagnetization of ARM. Rock magnetic properties are rather uniform within the cored interval, although there is an overall trend suggesting that coercivity of the NRM decreases downhole, an observation generally attributed to dissolution of fine-grained particles (Karlin and Levi, 1983). Decay of the NRM upon AF demagnetization is typical of a cubic phase, either magnetite or greigite. Representative samples were given a 400-mT IRM and subsequently demagnetized (Fig. F12). Inductions of 400 mT are not sufficient to reach saturation, suggesting that magnetic sulfides are present and may be important remanence carriers. AF decay of the IRM suggests that in some of the cored intervals, contributions from single-domain- size particles is less important than from multidomain particles (Cisowski, 1981). This is also manifest in the lower ARM:IRM ratios observed at this site. Magnetic susceptibility is of low negative values and is therefore dominated by the contributions of the diamagnetic carbonate matrix.
Moderately to steeply positive inclinations in discrete samples from Core 182-1129C-37X (at ~340 mbsf) are interpreted as reverse magnetizations, placing the Brunhes/Matuyama boundary between 338.8 and 347.2 mbsf. Long-core and discrete sample measurements in Hole 1129D define a long interval of predominantly reverse polarity magnetizations to a depth of ~540 mbsf, which is interpreted as the upper part of the Matuyama (C1r). The top of the Jaramillo Subchron (C1r1n) is recorded in both holes at depths of ~400-440 mbsf. Discrete samples place the upper boundary between 392.9 and 402.6 mbsf. Long-core measurements place the lower boundary of this subchron between Sections 182-1129D-9R-2 and 10R-1, at 440.8 mbsf. Normal polarities are recorded again in Hole 1129D at depths of 540-550 mbsf, possibly representing Chron C2n (Olduvai). This correlation to the Pliocene-Pleistocene geomagnetic polarity timescale implies a variable sedimentation rate, with an upper Pleistocene sedimentation rate nearly triple that of the lower Pleistocene. Correlation with the geomagnetic polarity time scale is nonunique, but is supported by the available biostratigraphic data.