Magnetic measurements, using the pass-through cryogenic magnetometer, were made on all archive halves of cores recovered from Hole 898A. Alternating field (AF) demagnetization of these archive halves was performed at 10-cm intervals using a peak field intensity of 15 mT. Cores 149-898A-4H through -14H (with the exception of Core 149-898A-12H, whose data were lost because of failure of a multishot tool) were oriented in situ using the tensor orientation tool (see "Explanatory Notes" chapter, this volume). Twenty-one discrete samples taken from the working halves of cores were progressively AF demagnetized and measured with the cryogenic magnetometer. The magnetic susceptibility of all cores was routinely measured at intervals of 3 cm on the Multisensor Track.
The majority of the whole-core pass-through measurements were of high quality and agree well with the discrete sample data. The mean inclination value is close to the expected inclination (59.7°) at the site from the Pleistocene to the present. In contrast to the overall excellence of the pass-through measurements obtained from APC cores, data from the XCB cores were less satisfactory. The NRM intensities for Cores 149-898A-20X through -36X are weak, and the data are highly scattered. Therefore, polarity interpretations for these cores were not attempted on board. The main features of our tentative magnetostratigraphic interpretation of the remaining cores at Site 898 are summarized in Figure 17. This interpretation is based on preliminary biostratigraphic ages (see Fig. 16 in "Biostratigraphy" section, this chapter).
The magnetic behavior of Cores 149-898A-1H through -3H is similar to that of their counterparts at Hole 897A. The stable component of remanent magnetization for all these cores is normal polarity. Declinations vary substantially within some individual sections of these cores, even after orientation using the tensor tool data. This indicates either an incomplete removal of the secondary magnetization, or rotation of the recovered material within the core liner or a combination of both. Biostratigraphic ages in these cores range from 0.3 to 0.8 Ma. Thus, in conjunction with the biostratigraphic data, we can assign these cores to the Brunhes Chron (<0.78 Ma).
The first evidence for a polarity reversal was found in Core 149-898A-4H at a depth of 28.2 mbsf (confirmed by a discrete sample measurement and by changes of about 180° in declination between normal and reversed intervals; Fig. 17). Thus, this magnetic polarity shift from normal to reversed may represent the Brunhes/Matuyama boundary (0.78 Ma). However, the next distinctive reversal did not occur until Core 149-898A-7H at 57.0 mbsf. The failure to reveal more reversed magnetizations from cores recovered from the interval (28.2-57.0 mbsf, i.e., a period of expected reversed polarity) may have resulted from an unrecognized sedimentary hiatus or from a large magnetic normal overprint during the Brunhes Chron. Preliminary planktonic foraminiferal dates suggest that sediments below a depth of 85.0 mbsf are of late Pliocene age. This information would suggest that the shift of polarity from reversed to normal at about 92.0 mbsf should correspond to the upper Olduvai boundary. However, the correlation of the three normal polarity intervals (63.0-66.0 mbsf, 98.0-107.0 mbsf, and a very short one at 107.0 mbsf) with Cobb Mountain, Reunion, and an excursion within the Matuyama Chron, respectively, is questionable (Fig. 17).
Pass-through cryogenic magnetometer measurements identified several magnetic reversals from 94.7 to 215.6 mbsf. Large uncertainty in the interpretation of key foraminiferal and nannofossil biostratigraphic markers (see Fig. 16 in "Biostratigraphy" section, this chapter) at this time do not permit a tentative correlation of these polarity intervals with the geomagnetic time scale. However, we noticed that the inclinations are dominantly normal between 110.0 to 133.0 mbsf, suggesting that this interval may correspond to a period of predominantly normal polarity.
Cryogenic magnetometer measurements also suggest several polarity reversals, which were recorded in Cores 149-898A-20X through -36X (177.4-339.7 mbsf). These polarity signals may not be reliable because of the weak magnetization of most of the sediments from 216.4 to 339.7 mbsf. We have not calibrated the measurements using corresponding discrete samples. Detailed studies of discrete samples and more accurate biostratigraphic markers will be needed to constrain a more reliable correlation of these intervals with the geomagnetic reversal time-scale.
Figure 18 shows the downhole profile of magnetic susceptibility for Hole 898A. Within the top 150 m, the magnetic susceptibilities have values that are consistently about 3 × 10-4 SI units. These relatively high susceptibility values correspond to Unit I (see "Lithostratigraphy" section, this chapter). The greatest susceptibility maximum occurs at about 98 mbsf. An examination of the corresponding core photograph reveals an interval of dark-colored soupy sand at the base of a turbidite sequence. Similar to Site 897, the high and low susceptibility peaks shown in Figure 18 generally correlate well with the turbidite layering. Lows correspond to the pelagic clays at the top of turbidite sequences and highs correspond to the terrigenous sands at the base of turbidite sequences. The susceptibility values decrease below about 163 mbsf (to about 1 × 10-4 SI units) in the pelagic sediments from Cores 149-898A-18X to -36X of Unit II. The low susceptibility values, the weak NRM intensities, as well as the relatively high percentage of quartz and feldspar observed in smear-slides from these cores, suggest that there are only minor amounts of magnetic material in these sediments.