The natural remanent magnetization (NRM) of archive halves or whole rounds of APC cores and RCB core sections from Site 1198 were measured at 5-cm intervals with the pass-through cryogenic magnetometer. Many sections were not measured in their entirety because they contained fragments that were small and thus could have rolled over in the core barrel during the coring and recovery process. The NRM and 5- and 30-mT alternating-field (AF) demagnetization values were measured except when time constraints precluded the complete treatment.
Representative discrete samples were collected and measured from Holes 1198A and 1198B. These were used to aid the interpretation of the long-core records of magnetization by providing additional measurements of polarity and basic magnetic characterization.
As at other Leg 194 sites, the sediments at Site 1198 had poor recovery in seismic Megasequence C and weak magnetization in Megasequences B and D. In addition, problems were encountered in measuring whole-round core sections in the 1-2 m heave conditions experienced at Site 1198. Apparently, the increased mass associated with whole-round sections in the magnetometer, coupled with the accelerations due to the ship's heave, stressed the plastic trackway, which affects the detection system of the magnetometer. This resulted in low-quality data in some of the cores. Inclinations of 0° ± 10° were excluded from the record (Fig. F12). This arbitrary cutoff is used because the mechanical problem in the magnetometer is known to produce near zero inclination. The cutoff is low enough that its elimination does not impair the recognition of true magnetozone boundaries.
The expanded Pleistocene section mitigates the effect of the difficulties and some magnetostratigraphy was obtained with the aid of the biostratigraphic datums (see "Biostratigraphy and Paleoenvironments"). In the magnetostratigraphy columns, alongside the inclination records, the hatched intervals in Figure F12 indicate indeterminate polarities. The reversed polarity at ~5 mbsf is probably a record of one of the excursions, or events, in the Brunhes Chron. The next good reversed interval is between 52 and 57 mbsf. Assigning this to the onset of Chron C1 is consistent with the biostratigraphic data, but with the coincidence of a core-top anomaly even this assignment is questionable. Down to a depth of 205 mbsf, only few intervals show a well-defined polarity. An exception is the normal interval from 140 to 148 mbsf, which, based on biostratigraphic datums, can be interpreted as the Olduvai Subchron (C2n). Finally, around 200 mbsf, there appears to be a partial record of the Gauss Chron (C2An).
Recovery was so poor in Megasequence C and the top of Megasequence B (200-400 mbsf) that no magnetostratigraphy was attempted. Recovery improved between 400 mbsf and the bottom of the hole, but the record has a strong residual normal overprint in the interval of good recovery between 400 and 500 mbsf that could not be demagnetized to separate the characteristic magnetization (Fig. F13).
The interval immediately above the basement basalt has a strong reversed magnetization, as does the basalt at the bottom of the hole. The basalt has a strong signal compared with the sediments, which, however, only amounts to one-tenth of the mean intensity for subaerial and submarine basalts given by Prévot and Grommé (1975). A stable primary magnetization appears to have been isolated by AF demagnetization.
Representative discrete samples were subjected to standard rock magnetic analysis (see "Paleomagnetism" in the "Explanatory Notes" chapter), and principal component analysis was carried out on samples with a sufficiently strong intensity of magnetization.
Figure F14 illustrates the remanent magnetism characteristics of two samples. Sample 194-1198A-7H-4, 110-112 cm, is weakly magnetized and representative of the skeletal wackestone/grainstone from Megasequence D. The dominant magnetic phase appears to be magnetite, but a minor higher coercivity contribution, possibly a sulfide, is indicated. The anhysteretic remanent magnetization (ARM) is approximately one order of magnitude smaller than the isothermal remanent magnetization (IRM), which is consistent with a detrital origin for the magnetite. The NRM is weak and at the limit of the sensitivity of the magnetometer. In contrast, Sample 194-1198B-34R-1, 66-68 cm, from the basement basalt, contains magnetite as the dominant magnetic phase. A strong NRM, about two orders of magnitude smaller than the IRM, is consistent with a primary thermoremanent magnetization that was acquired as the rock initially cooled.
The downhole variation of rock magnetism parameters is shown in Figure F15. In lithologic Unit I, the ratio of IRM (100 mT:1T) departs from 1, attesting to the presence of a harder phase in addition to the magnetite. Given the presence of pyrite, some of the magnetite may have been destroyed and partially replaced by sulfides, among which greigite could account for the slightly harder phase. IRM, ARM, and NRM are all weak in Unit I, as is to be expected from the difficulty experienced in measuring the NRM on long cores. The ratio of normalized ARM:IRM is low, suggesting that the grain size includes a coarser fraction. No representative samples from lithologic Unit II are available. The deep shelfal facies of lithologic Unit III has a strong hard component that is probably hematite, whereas the basalt is dominated by magnetite and shows a high intensity of magnetization.