All core archive halves from Holes 1119A and 1119B and those between 131.80 and 494.80 mbsf of Hole 1119C were measured on the shipboard pass-through cryogenic magnetometer. Declination, inclination, and intensity of natural remanent magnetization (NRM) and a 20-mT alternating field (AF) demagnetization step were routinely measured at 10-cm intervals. When time permitted, a 10-mT demagnetization step was also measured and some core sections were measured at 5-cm intervals. In situ tensor tool data were collected at the tops of APC cores in an attempt to determine azimuthal orientation of the core to 160 mbsf. The measurements, however, proved inconsistent and therefore only inclination could be used to determine magnetic polarity for both the hydraulic piston and extended core barrel cores. At least two oriented discrete samples were collected from the working half of each core interval for progressive AF and thermal demagnetization and rock magnetic studies. Whole-core magnetic susceptibility was routinely measured on all cores, using a Bartington susceptibility loop on the automated multisensor track (MST).
NRM intensities were generally very low (~10-4 A/m on average) and ranged between 10-2 and 10-5 A/m (Figs. F13, F14). Intervals with higher NRM intensities are generally correlative with higher levels of magnetic susceptibility, suggesting that the larger order of magnetic susceptibility variation is caused by the ferrimagnetic component of the sediments. Smaller order variations in NRM intensity and magnetic susceptibility do not appear to be correlated.
Most NRM measurements displayed steeply positive (downcore) inclinations ranging between +60° and +80° , consistent with a drill-string overprint induced during coring (Stokking et al., 1993; Roberts et al., 1996; Fuller et al., 1998). The 20-mT AF demagnetization step was only partly effective in removing the drill-string overprint to the normal inclination of -60° expected for 45° S latitude in the vicinity of New Zealand. In general, the inclination signal at 20 mT of demagnetization was too noisy and random to reliably ascertain polarity, so a suite of discrete samples were stepwise AF demagnetized to 60 mT to determine if the noise level of the data could be reduced by increased levels of demagnetization (Figs. F15A, F15B, F15C). In some cases the positive drill-string overprint was observed to be effectively removed to a stable polarity with only 5- to 10-mT alternating fields (e.g., Fig. F15B) but, in most cases, AF demagnetization was ineffective in establishing a remanence direction for discrete specimens (e.g., Figs. F15A, F15C).
Discrete samples from ~290 mbsf in Hole 1119C were subjected to stepwise thermal demagnetization (e.g., Figs. F15E, F15F) to ascertain if the drill-string overprint was being effectively cleaned by the 20-mT step in the pass-through cryogenic magnetometer. This interval was chosen because it retained a positive inclination after 20 mT of demagnetization (Fig. F14) and it had high susceptibility and NRM intensity values relative to the rest of the core. Thermal demagnetization was 100% effective in removing the drill-string overprint (Fig. F15), demonstrating that AF demagnetization was generally ineffective. Unfortunately, mineral alteration at temperatures above 400° C (demonstrated by sharp increases in susceptibility) prevented the establishment of a stable endpoint or determining if the negative inclinations were caused by a Brunhes normal-field overprint. Saturation isothermal remanent magnetization (SIRM) and backfield SIRM were measured for several discrete samples from Hole 1119B to determine magnetic mineralogy. In all cases, remanence did not become saturated until applied fields of 300-500 mT and Bcr values were between 50 and 75 mT (Fig. F16). These values, combined with the low unblocking temperature (250° -300° C; Figs. F15E, F15F), suggest that the main carrier of remanence is a ferrimagnetic iron sulfide mineral present in relatively low concentrations (SIRM intensities are generally less than 10-2 A/m). This is not unexpected as the sediments have an ubiquitous pyrite presence (see "Lithostratigraphy"). Similar IRM acquisition values were found in late Neogene New Zealand sediments containing pyrrhotite (Roberts and Turner, 1993). Similar Pleistocene sediments from the Wanganui Basin, New Zealand, were found to contain reasonable quantities of titanomagnetite; however, SIRM values were <150 mT (Roberts and Pillans, 1993).
It was not possible to determine a magnetic polarity stratigraphy for the sequence cored by Holes 1119B and 1119C. From paleontologic and cyclic sedimentation information (see "Biostratigraphy" and "Lithostratigraphy") it was expected that the Brunhes/Matuyama boundary should occur at ~150 mbsf and that the interval between 150 and 495 mbsf ranges in age from mid-Pliocene at the base to middle Pleistocene at 150 mbsf. No consistent behavior was observed in the inclination of the cores to reliably identify any of the several reversals that should occur in this interval. Intensity of remanence was measurable throughout the core (Figs. F13, F14), and it is the magnetic mineralogy that has prevented isolation of primary remanence from shipboard measurements. However, it appears from shipboard experiments that targeted thermal demagnetization, guided by the paleontology, should allow identification of key magnetostratigraphic reversals in shore-based studies.