After measuring the natural remanent magnetization (NRM), all sections of the archive half of the core were partially demagnetized using alternating-field (AF) magnetization up to 20 mT in increments of 5 mT at 5-cm intervals. Additionally, the maximum demagnetization level was changed to 30 mT at 101.69 mbsf (Section 190-1173A-11H-7, 5 cm) to remove magnetic overprints.
Two oriented discrete samples were routinely collected from each section of the working half of the core primarily for shore-based analysis of the anisotropy of magnetic susceptibility. Additional measurements of polarity and basic magnetic character of selected discrete samples were used to aid in the interpretation of the archive long-core magnetization record. Most of the discrete samples were demagnetized up to 50 mT in 5-mT increments to permit principal component analysis.
The majority of NRM inclinations are strongly biased toward steep inclinations of ~60°-80° (Fig. F19A). These steep inclinations do not correspond to the direction expected (52°) for a geomagnetic axial dipole field at the current latitude (32°14.6634´N) of Site 1173. The steep inclinations are interpreted as magnetic overprints acquired during drilling, a problem identified on many previous DSDP and ODP legs. These overprints were successfully removed by AF demagnetization at 20 and 30 mT (Fig. F19B). This drilling-induced magnetic remanence (steeply positive with a low coercivity) is thought to be an isothermal remanent magnetization (Musgrave et al., 1993). After AF demagnetization, stable magnetic remanence inclinations were measured. These inclinations provide information about middle Miocene to Holocene magnetic polarity changes and were used in conjuction with the standard geomagnetic polarity time scale (GPTS) of Cande and Kent (1995) to date the sediments.
Declinations from APC cores (0 to ~160 mbsf) show stable directions after correction by the Tensor tool. Declinations of all XCB cores below 160 mbsf show rapid changes in direction within each section (Fig. F20). These anomalous declinations are interpreted to have been caused by rotation of individual core pieces in each section during XCB coring. However, the measurements taken from the rotated pieces of archive halves were used to correct the core orientation for structural analysis.
Discrete samples were AF demagnetized at fields up to 80 mT to isolate the characteristic stable magnetic component. Typical examples of demagnetization plots for discrete samples and archive-half sections are presented in Figure F21. Stable magnetization components were obtained for most sediments after removal of the steep drilling-induced overprint. After removal of this overprint, the directions of discrete samples and archive-half sections of corresponding horizons agreed very well.
The value of magnetic intensity is strongly associated with the magnetic mineralogy of the sediments. From 0 to ~160 mbsf, markedly high magnetic intensity is observed (Fig. F22). This is especially prominent in the lower part of lithostratigraphic Subunit IA from 60 to 80 mbsf (see "Lithostratigraphy") and is marked with a very high intensity peak. These high-intensity values gradually decrease downhole toward 160 mbsf. In contrast to the high-intensity peak, an anomalous low-intensity zone appears in Unit II (upper Shikoku Basin) at ~160 to 350 mbsf. These contrasting variations of intensities correspond to the changing character of the magnetic susceptibility (Fig. F22) measured with the multisensor track (MST) (see "Physical Properties"). These corresponding magnetic susceptibility and intensity values suggest a change in lithology and/or magnetic minerals in the sediments at this depth. A few high peaks in the low-intensity zone also correspond to peaks in the susceptibility values.
The boundary between high- and low-intensity values at the depth of 160 mbsf does not correspond to a lithostratigraphic boundary (Fig. F23). A previous rock magnetic study from Site 808 (Lu et al., 1993) also shows high- and low-intensity zones that do not correspond to lithostratigraphic boundaries.
Magnetic polarity was determined by measuring the magnetic inclination of archive-half samples after AF demagnetization at 20 and 30 mT. This was supported by AF demagnetization of discrete samples. As a result, many middle Miocene to Pleistocene magnetic polarity records were successfully identified using biostratigraphic datums (calcareous nannofossils) (see "Biostratigraphy") and correlated with the GPTS of Cande and Kent (1995) (Fig. F24). The identified chrons and subchrons are given in Table T14.
A change from normal to reversed polarity at 159.69 mbsf (Section 190-1173A-18H-1, 55 cm) is interpreted as the Bruhnes/Matuyama Chron boundary dated at 0.78 Ma (Cande and Kent, 1995). The Matuyama Chron (0.780-2.581 Ma), observed at 159.69-291.69 mbsf (Section 190-1173A-31X-6, 95 cm), is characterized by a predominantly reversed polarity. The Gauss Chron (2.581-3.58 Ma), interpreted to extend from 291.69 to 374.39 mbsf (Section 190-1173A-40X-4, 25 cm), is characterized by a change to a predominantly normal polarity. The magnetic polarity change at 374.39-423.84 mbsf (Section 190-1173A-45X-5, 20 cm) is interpreted as the Gilbert Chron. Subchrons C3An (5.894-6.567 Ma), C4r (8.257-8.699 Ma), and C4An (8.699-9.025 Ma) are observed at 423.84-439.44 mbsf (Section 190-1173A-47X-3, 0 cm), 478.94-491.49 mbsf (interval 190-1173A-51X-4, 10 cm, through 52X-5, 145 cm), and 491.49-500.34 mbsf (Section 190-1173A-53X-5, 60 cm), respectively. The boundary between normal and reversed polarities at 557.79 mbsf (Section 190-1173A-59X-5, 45 cm) is identified as the termination of Subchron C5n (10.949 Ma). The two characteristic normal polarities below Chron C5 (9.740-11.935 Ma) are clearly observed at 578.49-589.9 mbsf (interval 190-1173A-61X-6, 25 cm, to 62X-CC, 0 cm). However, it is difficult to identify magnetic events near the lower part of the hole because of poor core recovery. Based on biostratigraphic results, the two normal polarities at 635.89 and 662.69 mbsf (Sections 190-1173A-67X-5, 145 cm, and 70X-4, 75 cm) are estimated to be Subchrons C5ACn (13.703-14.076 Ma) and C5ADn (14.178-14.612 Ma), respectively.
Comparison of the magnetic stratigraphy and biostratigraphy (see "Biostratigraphy") shows a strong correlation in the interpreted age of the sediments. The marked change in sedimentation rate was estimated at the lithologic boundary between the upper and lower parts of Shikoku Basin at an age of ~3 Ma. The sedimentation rate for lithostratigraphic Unit II (upper Shikoku Basin) has been calculated at ~7.78 cm/k.y. This rate is in sharp contrast to lithostratigraphic Unit III (lower Shikoku Basin), which has a sedimentation rate of ~2.74 cm/k.y. (Fig. F25).