The investigation of magnetic properties at Site 1082 included the measurement of bulk susceptibility of whole-core sections and the natural remanent magnetization (NRM) of archive-half sections. Discrete samples were taken from Hole 1082A. Because of time restrictions, only parts of the discrete samples taken from XCB cores were measured on the ship. The Tensor tool was used to orient Cores 175-1082A-3H through 14H, Cores 175-1082B-4H through 14H, and Cores 175-1082C-3H through 24H (Table 7).
Measurements of NRM were made on all archive-half core sections from Holes 1082A, 1082B, and 1082C. APC sections from Hole 1082A were demagnetized by AF at 10 and 20 mT; XCB sections from Hole 1082A and all sections from Holes 1082B and 1082C were demagnetized by AF at 20 mT only. Magnetic susceptibility measurements were made on whole cores from all holes as part of MST analysis (see "Physical Properties" section, this chapter).
Above 470 mbsf, the intensity of NRM after 20-mT demagnetization fluctuates between 10–2 and 10–4 A/m, except for the upper 80 mbsf where it ranges between 10–2 and 10–3. Below ~470 mbsf, it sharply decreases to between 10–4 and 10–5 A/m (Fig. 11, second panel). Variations of magnetic susceptibility, generally between 0 and 10 x 10–5 (SI volume units), do not follow those of the remanent intensity (Fig. 11, first panel). Susceptibility is low between 200 and 320 mbsf and high between 350 and 540 mbsf. Diamagnetic materials including water, opal, calcium carbonate, or the plastic core liner (void intervals) could be responsible for the slightly negative susceptibility values.
A magnetic overprint was generally removed by 20-mT demagnetization and a primary NRM was measured for all APC cores (Fig. 11). For XCB cores, however, a significant magnetic overprint remained after AF demagnetization. Declinations cluster around –30°, independent of the orientation of the sediments, even where the sediments are extensively biscuited (Fig. 11A). This phenomenon is similar to that observed at Site 1081 (see "Paleomagnetism" section, "Site 1081" chapter, this volume), where the magnetic overprint was attributed to the coring process. Inclinations, however, show two groupings with distinct polarity biases after 20-mT demagnetization, although the directions are strongly scattered. The scattering is most severe below an intensity decrease at about 470 mbsf. One grouping, with a bias range of 40° to 60° in inclination, is interpreted as reversed polarity, and the other grouping, with a bias range of –30° to 0° in inclination, is interpreted as normal polarity. Overall, however, the inclinations are biased downward, probably because of a downward-oriented magnetic overprint. Normal-polarity intervals (the Olduvai Subchron, discussed later) occurring from ~165 to 182 mbsf in XCB cores from Hole 1082A and from 164 to 182 mbsf in APC cores from Hole 1082C lend support to the polarity interpretations based on inclinations from XCB cores (Fig. 11). Preliminary measurements of discrete samples between 370 and 470 mbsf from Hole 1082A yielded results similar to the half-core measurements.
We identified the polarity of the NRM from the magnetic declinations and inclinations of APC cores and from the inclinations of XCB cores. The magnetostratigraphic interpretation is summarized in Table 8 and displayed in Figure 12.
There are two possible interpretations (models) for the polarity-chron assignment between ~300 and 460 mbsf (Table 8). The major difference between the two is the position of the top of Chron C2Ar (the Gauss/Gilbert boundary), which is at 390 mbsf in Model 1 and at 340 mbsf in Model 2. According to Model 1, the sedimentation rate was significantly higher in the earlier part of the Gauss Chron, between ~3.0 and 3.5 Ma, whereas it was lower between ~3.5 and 4.2 Ma (Fig. 12). In Model 2, the sedimentation rate is rather constant between ~2.6 and 4.2 Ma, but higher between ~4.2 and 4.8 Ma. Model 1 fits the biostratigraphy better than Model 2 (see "Biostratigraphy and Sedimentation Rates" section, this chapter) and is thus adopted here.
The Cobb Mountain event (1.201–1.211 Ma) was identified by an intensity decrease and a directional shift at all three holes. The Reunion event at Hole 1082C also accompanies an intensity decrease and a change in direction. Hole 1082B did not reach the depth at which this event would be expected. The Reunion event may also be recorded in XCB cores from Hole 1082A; a change in the inclination and a decrease in the intensity at ~200 mbsf is consistent with the depth observed at Hole 1082C.
A complete transitional record of the Brunhes/Matuyama polarity change was identified at Hole 1082C, where the measurement was carried out at 2-cm intervals instead of the routine 5-cm intervals. The transitional interval spans 0.4 m (70.5–70.9 mbsf). In the other two holes, part (at Hole 1082A) or all (at Hole 1082B) of the transition record unfortunately was missing because of a coring gap.
The intensity of NRM can be controlled by the strength of the geomagnetic field, the concentration of magnetic minerals that carry the NRM, and other rock-magnetic characteristics of sediments including composition, grain-size, and interaction of magnetic minerals. If the sediments are uniform rock-magnetically and the concentration of magnetic minerals is represented by some rock-magnetic parameters (e.g., magnetic susceptibility, anhysteretic remanent magnetization, and isothermal remanent magnetization), variations of remanent intensity after normalization by the rock-magnetic parameters can be interpreted as relative changes of past geomagnetic-field strength (paleointensity; King et al., 1983). The paleointensity during the last 200 k.y. is relatively well understood (Yamazaki and Ioka, 1994; Guyodo and Valet, 1996), and efforts have been made to establish a paleointensity curve for the Brunhes Chron (Kent and Opdyke, 1977; Valet and Meynadier, 1993; Yamazaki et al., 1995). Figure 13 shows the relative remanent intensity at Hole 1082A during the Brunhes Chron after normalization by the magnetic susceptibility. Variations of the relative intensity are similar to the documented characteristics of the paleointensity during the Brunhes Chron; that is, the quasi-cyclic intensity lows at ~100-k.y. intervals and a peak in the intensity just after the Brunhes/Matuyama polarity transition (Valet and Meynadier, 1993; Yamazaki et al., 1995). Our results suggest that sediments at this site could be useful for the study of detailed paleointensity changes, although the assumptions mentioned above must be thoroughly tested by rock-magnetic analyses before the variations can be interpreted as paleointensity of the geomagnetic field.