The natural remanent magnetization (NRM) of whole-round sections from Site 1194 was measured at 5-cm intervals using the pass-through cryogenic magnetometer. A 30-mT alternating-field (AF) demagnetization was used. The Tensor tool was deployed to orient the APC cores from Core 194-1194A-4H to 13H. Tests were run to investigate the effects of a superposed laboratory field on the cores by storing the core sections in a mu-metal can prior to measurement. In the one case where the Tensor tool produced an independent test of the magnetization orientation, the 0° declination artifact was not seen in the whole-round measurement of the core section stored in the mu-metal can. When the archive half was measured afterwards, the 0° declination artifact had reappeared. In addition, there was good agreement between the orientation of the whole round and Tensor tool.
Discrete samples were collected from Holes 1194A and 1194B at an average sample rate of two per core. These were used to aid the interpretation of the long-core record of magnetization by providing additional measurements of polarity and basic magnetic characterization. Most of these samples were demagnetized at 5, 10, 15, 20, 40, 60, and 80 mT to permit principal component analysis, but the signal was frequently too weak to permit reliable analysis. For rock magnetic characterization, anhysteretic remanent magnetization (ARM) was measured in 0.2 DC and 200-mT AC fields, and isothermal remanent magnetization (IRM) was measured in a DC field of 1 T. Samples were also progressively magnetized in fields up to 1.0 T to study the acquisition of the IRM.
The long-core measurements for the uppermost 100 mbsf were seriously corrupted by core-top magnetic anomalies (Fig. F25). Unlike at Hole 1192A, where the intensity of magnetization in the uppermost 100 m away from core-top anomalies was ~10-3 to 10-4 A/m, the intensity was much less at Site 1194. The core top anomalies in the first sections of cores at Site 1194 were 10-3 to 10-4 A/m. In most APC cores, the intensity decreased systematically through Cores 194-1194A-2H and 3H, so that in Cores 194-1194A-4H and 5H, it reached 10-5 A/m. These core-top anomalies were often accompanied by magnetic susceptibility peaks, indicating that new magnetic material had been introduced at the tops of cores and then magnetized in the direction of the downhole overprint. Thus, the true intensity of the uncontaminated sediments is ~10-5 A/m, and the contamination reaches 10-3 A/m. Iron oxide (rust) within the pipe string is a well-known contamination problem, and a small number of such particles would swamp the signal from the carbonates. The magnetization is acquired in the same field that gives the downward-directed drilling overprint, which likely comes from the drill bit in the bottom-hole assembly.
In the presence of this contamination, magnetostratigraphy becomes problematic at the very top of the borehole instead of from ~100 mbsf, as it did at Site 1192. However, low-pass filtering removes these high-intensity values associated with core tops (using an arbitrary cut-off intensity of 10-4 A/m) (Fig. F26). The inclinations are adjusted using the intensity cut off by simply deleting the inclinations associated with the high-intensity amplitudes (Fig. F26). Furthermore, it is clear that from ~50 to 110 mbsf, a sequence of inclination changes gives an indication of the polarity. The inclination is not symmetrical about the zero baseline because of the positive overprint. Between ~85 and 95 mbsf, there is a prominent reverse-magnetized interval. Assuming that the biostratigraphy is correct, this interval can be assigned to Chron C3Ar (see "Biostratigraphy and Paleoenvironments").
When overprint is a problem, it is sometimes useful to compare the z-component intensity with inclination. A simple filtering approach is to balance the positive and negative z intensities by biasing the zero line (Fig. F26). Doing this confirms the earlier suggestion of a negative interval between ~85 and 95 mbsf, which is assigned to Chron C3Ar. Beneath is an interval of 15 m, which shows a normal-reversed-normal sequence in less strongly magnetized material. This could then represent the Chron sequence C3Bn-C3Br1r-C3Br1n. Higher in the section, above the hiatus, the magnetostratigraphy is less clear. The age assignment of the reversal sequence is not an independent determination from the paleomagnetic record; however, it is dependent upon the biostratigraphic age constraints. Between 100 and 300 mbsf, recovery was insufficient to produce a useful magnetic record, but between 330 and 370 mbsf, recovery improved and reversals could be recognized so that a sequence could be constructed (Fig. F27). However, no useful magnetostratigraphy is possible with such a short sequence.
Overall, the record from the long cores does not allow a clear magnetostratigraphic interpretation. In the upper part of the section, where we had good recovery, the magnetization of the sediments was weak, so that core-top anomalies obscured the record. Downcore, poor recovery precluded useful magnetostratigraphy.
Discrete samples were used to try to improve the NRM determinations and to attempt to correlate rock magnetic properties to the lithologies observed in the cores, so that magnetic proxies could be used to interpret environmental changes.
The NRM of most of the Site 1194 sediments was too weak for useful shipboard analysis, but could be successfully measurable onshore with magnetometers designed to measure discrete samples. Only two samples gave convincing evidence of the isolation of a single direction of magnetization that appeared to be converging to the origin of the Zijderfeld diagram (Fig. F28). Sample 194-1194B-34R-2, 45-47 cm, from near the basement, is strongly magnetized. It has a soft component with a declination of 215°, an inclination of -18°, and a maximum angular deviation angle of 1.7°. The apparent characteristic remanent magnetization is shallowly inclined down to the east-northeast (declination = 67°, inclination = 14°, and maximum angular deviation = 4.9°). The basement Sample 194-1194B-34R-5, 12-14 cm, is strongly magnetized with a steep downward direction of magnetization (declination = 242°, inclination = -61°, maximum angular deviation = 2.4°). Further analysis onshore should better determine these directions and may permit useful comparison with the Australian apparent polar wander path to date the magnetization and associated alteration events.
Figure F29 is a remanent magnetization characterization plot. The IRM acquisition shows almost complete saturation by 100 mT, which is a strong indication of the dominance of magnetite in the sample. The intersection of the IRM demagnetization and acquisition curves shows weak negative interactions. The stability of IRM against AF demagnetization can be measured by the ratio of IRM after demagnetization to 30 mT to saturation IRM. Values between 60% and 40% for magnetite are suggestive of bacterial magnetite, as is the high value of the ratio of normalized ARM to IRM.
Environmental magnetists (e.g., Maher and Thompson, 1999) have developed a method to identify magnetic phases based on simple measurements similar to those made aboard the JOIDES Resolution. Magnetic stability measured by the fraction of IRM remaining after demagnetization to 40 mT is plotted vs. the ratio of ARM susceptibility to IRM. Such a plot calculated for Sample 194-1194B-4R-1, 22-24 cm (Fig. F29), shows that the values fall in the field of disaggregated chains of magnetite (fine detrital magnetite field). Downhole variations in NRM, ARM, and IRM show that at Site 1194 most magnetic material occurs in the basement and immediately above it (Fig. F30). The ratio of IRM acquisition at 100 mT to 1T, which is near 1 with the exception of the hardground that is near 110 mbsf, indicates that magnetite dominates the magnetic mineralogy. In the region of the hardground there is evidence for a much higher coercivity phase, possibly hematite, which would be in concert with the red staining of the samples.