The investigation of magnetic properties at Site 1083 included measurement of the bulk susceptibility of whole-core sections and the natural remanent magnetization (NRM) of archive-half sections and discrete samples. The Tensor tool was used to orient Cores 175-1083A-4H through 22H, Cores 175-1083B-3H through 22H, and Cores 175-1083D-3H through 21H (Table 7).
Measurements of NRM were made on all archive-half core sections from Holes 1083A, 1083B, and 1083D. Only one core was recovered from Hole 1083C; therefore, this hole will not be discussed further in this section. Sections from Hole 1083A were demagnetized by AF at 10 and 20 mT, and sections from Holes 1083B and 1083D were demagnetized by AF at 20 mT only. All discrete samples taken from Hole 1083A (one per section) were demagnetized by AF at 10, 20, 25, and 30 mT. Magnetic susceptibility measurements were made on whole cores from all holes as part of the MST analysis (see "Physical Properties" section, this chapter).
Magnetic susceptibility ranges between 0 and 5 x 10–5 (SI volume units), and the intensity of NRM after 20-mT demagnetization ranges between ~10–2 and ~10–4 A/m (Fig. 5). In general, they are inversely correlated with the total reflectance in spectrophotometry, which reflects the relative proportion of calcium carbonate to clay (see "Lithostratigraphy" section, this chapter). This correlation indicates that the concentration of magnetic minerals is low in calcium carbonate–rich horizons.
We encountered a curious secondary magnetization at this site. In each core that was recovered below ~80 mbsf, inclinations deviate downcore from the expected value toward steep negative values, indicating downcore acquisition of an upward magnetic overprint. Figure 6 shows an example of Cores 175-1083B-11H through 15H, which is within the Matuyama Chron, discussed below. The inclination expected from the geocentric axial dipole model during the Matuyama Chron is 37° at this site. The inclinations near the top of each core are close to the expected value, but they change toward negative values downcore. Furthermore, cyclic inclination fluctuations are superimposed on the downcore trend, which is inversely correlated to the total reflectance from spectrophotometry (Fig. 6). This suggests a relationship between the magnetic overprint and the sediment lithology. Discrete samples, however, preserved a primary magnetization and did not show the magnetic overprint (Fig. 7). The inclinations of the discrete samples are close to those expected from the polarity (Fig. 5). This indicates that the source of the upward magnetic overprint resides in the sediments along the rim of the cores, and thus deformation of sediments along the core liner could be responsible. Sticky calcium carbonate–rich horizons may suffer more extensive deformation during coring and thus acquire a more severe magnetic overprint. However, we do not understand the deformation mechanism that produces the upward magnetization; it cannot be explained by simple drag along the core liner because the overprint has a pervasive upward direction that is independent of the core orientation (see "Paleomagnetism" section, "Site 1077" chapter, this volume). The inclinations of surface sediments above ~30 mbsf in half cores are also significantly steeper than the expected value (Fig. 5); discrete samples did not show this tendency. This phenomenon may also be related to deformation of sediments along the rim of cores.
The magnetic overprint at this site is different from the coring-induced magnetization (CIM) that is radially inward and downward in APC cores and from that of –30° declination and downward in extended barrel (XCB) cores observed at other sites (see "Paleomagnetism" sections, "Site 1077" and "Site 1081" chapters, this volume). The downward direction of the CIM is opposite to the upward overprint here. Declinations of discrete samples (from working halves) generally agree with those of half-core (archive-half) measurements, which indicates that the radial-inward component is not significant. The upcore increase in intensity observed in the CIM, which indicates upcore growth of secondary magnetization, was not observed here. The CIM was reported from discrete samples as well as from half-core measurements (Shipboard Scientific Party, 1995, 1996).
The polarity identification was based on declinations and inclinations. The time scale used is that of Berggren et al. (1995). In the deeper part, polarity interpretation was hampered by the extensive magnetic overprint on the inclinations, as mentioned above. The hori-zons between 145 and 148 mbsf at Hole 1083A, between 154 and 158 mbsf at Hole 1083B, and between 154 and 158 mbsf at Hole 1083D are indicative of a radial-inward remagnetization because they showed pre-orientation declinations close to zero. As a result, the beginning of the Olduvai Subchron was poorly constrained.
The Brunhes/Matuyama boundary (0.78 Ma) occurs at ~67 mbsf at Hole 1083A, at 60 mbsf at Hole 1083B, and at 68 mbsf at Hole 1083D (Fig. 5). The termination and beginning of the Jaramillo Subchron (C1r.1n), 0.99 and 1.07 Ma, respectively, occur at ~85 and 89 mbsf at Hole 1083A, at 78 and 84 mbsf at Hole 1083B, and at 86 and 92 mbsf at Hole 1083D. The termination and beginning of the Olduvai Chron (C2n), 1.77 and 1.95 Ma, respectively, occur at ~135 and 144 mbsf at Hole 1083A, at 132 and 145 mbsf at Hole 1083B, and at 136 and 149 mbsf at Hole 1083D. The Matuyama/Gauss boundary (2.58 Ma) occurs at ~194 mbsf at Hole 1083A and at 192 mbsf at Hole 1083B. The bottom of Hole 1083D is estimated to be just at the Matuyama/Gauss boundary based on correlation of the remanent intensity between Holes 1083A and 1083D.