After alternating-field (AF) demagnetization, the natural remanent magnetization (NRM) of whole-core sections from Site 1171 was measured at 5-cm intervals using the pass-through cryogenic magnetometer. An exception was made for cores whose liners or end caps were deformed, because these could damage the magnetometer. These deformed cores were measured as archive-half cores. The AF of 20 mT applied for Holes 1171A and 1171B did not completely remove the strong drilling overprint, so for Hole 1171C an AF of 30 mT was used, which provided a relatively good magnetic record. In view of these results, demagnetization of 30 mT was used for Hole 1171D and the remaining sites. The nonmagnetic core barrel assembly was again employed for even-numbered cores in Holes 1171A and 1171C and for odd-numbered cores in Hole 1171B, starting with Core 3H. The comparison between results from cores collected with the nonmagnetic corer and with those from the standard corer is discussed in the "Appendix" chapter, as are results of experiments investigating the effect of core splitting on magnetization and other coring-related magnetic experiments. The Tensor tool is usually used to orient the APC cores beginning with the third core at each hole, but the poor determination of the declination values of the cores at this site precluded the orientation of cores.
Discrete oriented samples were routinely collected from Holes 1171A, 1171C, and 1171D. These samples were used to aid in the interpretation of the long-core magnetization record by providing additional measurements of polarity and basic magnetic characterization. Most of them were demagnetized at 5, 10, 15, 20, 30, 40, and 50 mT to permit principal component analysis. For rock-magnetic characterization, anhysteretic remanent magnetization (ARM) was given in 0.2-mT DC and 200-mT alternating-current fields and isothermal remanent magnetization (IRM) in a DC field of 1 T. Discrete samples were also progressively saturated up to 1 T to study the hardness of the IRM.
The long-core measurements for the cores are presented in Figures F24 and F25, which show inclination and intensity for Holes 1171A, 1171B, and 1171C. A new, larger drill bit was used in the APC BHA and appears to have induced a strong magnetic overprint, recognizable by abnormally high intensities. Thus, all measurements with an intensity higher than 10-4 mA/m or lower than 10-6 mA/m (the sensitivity limit of the cryogenic magnetometer) were excluded. In fact, the intensity of magnetization was very constant between 10-5 and 10-4 mA/m. Even if measurements approached the noise level of the instrument and the background noise from core liners, a record could be interpreted for most of Hole 1171C. GRA bulk density measurements (see "Physical Properties") allowed us to obtain a good depth correlation that showed a very small shift among Holes 1171A, 1171B, and 1171C. Furthermore, some individual features on the magnetic records could be recognized from one hole to another, which confirmed the absence of a major depth shift.
In the upper part of Hole 1171C, the onset of the Brunhes Chron (C1n), the Jaramillo Subchron (C1r.1n), the Olduvai Subchron (C2n), and the termination of the Gauss Chron (C2r.1r) are clearly identified providing a useful magnetostratigraphy down to 35 mbsf. The section between 35 and 80 mbsf was marked by a succession of sequences of reversals that could not be matched to the geomagnetic polarity time scale.
Between 83.5 and 107.75 mbsf, a long normal magnetozone was identified as Subchron C5n.2n with two short reversed magnetozones above and below.
Below 120 mbsf, where there was only core from Hole 1171C, we found the onset of Subchron C5An.1n at 130.06 mbsf, the onset of Subchron C5An.2n at 134.5 mbsf, the onset of Subchron C5Acn at 159.35 mbsf, and the onset of Subchron C5ADn at 167.2 mbsf. At deeper depths, between 207.2 and 223.5 mbsf, the particularly distinctive long normal Chron C6n was recognized (Table T14).
In Hole 1171D, poor recovery precluded any further interpretation of magnetostratigraphy until a depth of nearly 450 mbsf was reached. Between 450 and 950 mbsf, the magnetic record (Fig. F26) was marked by long magnetozones suggesting a high sedimentation rate. However, suspicious sequences of stable inclination around -40°, and weaker values of reverse inclination compared with normal, suggested a strong remagnetization in the normal direction. This remagnetization was not caused by the standard drill overprint, which was largely removed by 30 mT of demagnetization but was probably carried by the soft material surrounding and between biscuits. To minimize this effect, the data were filtered and only sections that ranked <2 in the 0-5 scale of soft intervening material between the biscuits (see "Lithostratigraphy") are shown in Figure F26. However, it was very difficult to identify these magnetozones reliably in the absence of biostratigraphic control (see "Biostratigraphy"). In our interpretation, we recognized the onset of Chron C20r at 523 mbsf, the onset of Chron C21n at 571 mbsf, and the onset of Chron C24r between 773 and 889 mbsf. This interpretation is consistent with the biostratigraphic constraints of 43.7 Ma at 482 mbsf and 53.5 Ma at 908 mbsf from the nannofossils, but requires a significant hiatus between Chrons C21r and C24r. The normal-moment bias may have obscured the magnetic record, so shore-based investigations will be required to determine if a reliable magnetostratigraphy of the Eocene can be definitely established.
The discrete samples were routinely subjected to NRM, ARM, and IRM demagnetization and progressive IRM acquisition. NRM demagnetization results were, for the most part, disappointing. Figure F27 shows the variation of the intensity of ARM and IRM downhole. There is progressive increase of about an order of magnitude in both, which reflects the increase of terrigenous input (see "Lithostratigraphy"). Some anomalous values near 200 mbsf are still not interpreted and do not appear to correlate with any major features observed in the core description. The normalized difference between IRM (20 mT) to IRM (Fig. F27) increases downcore, indicating an increase in the softest magnetic fraction and suggesting an increase of grain size of the magnetic fraction downhole. The normalized difference between IRM at 200 and 500 mT does not vary systematically downcore and suggests that the magnetic mineralogy does not change significantly downcore. However, there must be some component present other than magnetite to account for the very high coercivity of the hardest fraction, which is far greater than that of single domain magnetite.
Overall, the rock magnetic characteristics are similar from one lithostratigraphic unit to another and seem to characterize the same detrital source. In particular, they do not show the strong change of sedimentary environment marked by the nannofossil ooze/silty claystone contact.
The relatively good magnetic record of Hole 1171C (Table T14) allowed us to generate an age-depth curve (Fig. F28). From the Holocene down to early Pliocene and from the early Miocene down to the late Miocene, the sedimentation rate is relatively constant (10 m/m.y.). However, the position of Chron C5n in the record suggests the presence of a hiatus from the early Pliocene until the late Miocene, which is confirmed by biostratigraphic datums. Unfortunately, the absence of reliable magnetostratigraphy in this part of the record precludes the determination of its exact duration. Chron C5n represents a condensed section just before the hiatus, where the sedimentation rate reaches 25 m/m.y.
In comparison with the upper part of the record, the sedimentation rate is high in the Eocene section, as recorded in Hole 1171D. Whatever the interpretation of the magnetic record may be, these long magnetozones give a sedimentation rate of at least 25 m/m.y. If we base our estimation upon Chron C24r between 773 and 889 mbsf, we obtain a sedimentation rate of 45 m/m.y. The final estimates of these sedimentation rates will be made in conjunction with the biostratigraphic data (see "Biostratigraphy").