Paleomagnetic studies conducted on board the JOIDES Resolution during Leg 183 consisted of routine measurements of natural remanent magnetization (NRM) and of magnetic susceptibility of sedimentary and igneous rocks. NRM was measured on most archive-half cores of sediments and basement rocks and on discrete working-half samples of sediments and basement rocks. Stepwise, alternating field (AF) demagnetization for most archive-half cores and some discrete samples was adopted for magnetic cleaning. Some igneous rocks were thermally demagnetized to obtain their primary remanent magnetization. The anisotropy of magnetic susceptibility (AMS) was determined for discrete igneous rock samples to obtain information about the magnetic fabric. Magnetic susceptibility was measured for whole cores, archive-half sections, and discrete samples. Magnetic properties were compared with the observed lithostratigraphic and basement units.
Ages of sedimentary rocks were estimated by magnetostratigraphy to establish the sedimentary history following igneous activity on the Kerguelen Plateau and Broken Ridge. The paleolatitudes of the Kerguelen Plateau and Broken Ridge at the time of their formation were determined from the remanent magnetization of igneous rocks.
The remanence of all archive-half sections and all oriented discrete working-half samples was measured using a 2-G Enterprises (Model 760R) pass-through cryogenic magnetometer equipped with direct-current superconducting quantum interference devices. The sensing coils of the magnetometer have slightly different response curves for each axis. The width of the sensing regions is ~10 cm, which corresponds to ~100 cm3 of cored material. The practical limit on the resolution of natural remanence of the core samples is imposed by the magnetization of the core liner itself (~0.01 mA/m). An in-line AF demagnetizer capable of reaching peak fields of 80 mT (2-G Enterprises Model 2G600) was used with the cryogenic magnetometer. A Molspin spinner magnetometer was also available in the paleomagnetism laboratory for measurements of the NRM of discrete samples. For stepwise demagnetization of samples, the laboratory contains AF and thermal demagnetizers (Schonstedt Instrument Company Models GSD-1 and TSD-1) that are capable of demagnetizing discrete specimens to 100 mT and 700°C, respectively. Anhysteretic remanent magnetization was imparted to discrete samples by a DTECH, Inc., PARM-2 system. An Analytical Services Company Model IM-10 impulse magnetizer (capable of pulsed fields of 20 to 1200 mT) was available for studies of the acquisition of both stepwise and saturation isothermal remanence magnetization by discrete samples. For some discrete basement samples, the AMS was determined using a Kappabridge KLY/2 directional susceptibility meter and the program ANI20, provided by the manufacturer Geofyzika Brno (Czech Republic).
The magnetic susceptibility for whole-core and archive-half sections was routinely measured in 5-cm intervals using a susceptibility meter attached to the MST (see "Physical Properties"), and in 4- and 2-cm intervals using a susceptibility meter attached to the AMST (see "Spectrophotometer") for sediments and basement rocks, respectively. Susceptibilities of discrete samples were also measured. For the three types of susceptibility measurements (MST, AMST, and discrete sample), the same type of magnetic susceptibility meter (Bartington Instruments Model MS2) was used with a different sensor. The sensor for whole-core measurements (MS2C) has an inner diameter of 88 mm, and the core passes through the sensor coil. The AMST has a cylindrical tip probe (MS2F), and the sensor provides a depth of investigation approximately equal to its diameter (20 mm). The measurements were automatically recorded by the AMST, which permits measurements only at evenly spaced intervals along each section of core. Spacing of measurements varied from 2 to 10 cm. The MS1B (dual frequency) sensor was used for discrete-sample measurements. This sensor has an internal diameter of 36 mm and measures 2.54 cm of drill core at a time. General variations with depth were consistent among the three types of susceptibility measurements during Leg 183. However, the MST and AMST data occasionally had lower values than the discrete measurements, with the differences being more significant in the AMST measurements. The lower MST and AMST values were probably caused by gaps in the core and/or differences in the calibration of the sensors. The spatial resolution of the MS2C and MS2F sensors is 20 mm. For the MST measurements (MS2C), the sensor expects a volume of material 20 mm in length times the diameter of the ordinary core; thus, a gap in core material results in a lower value. For the AMST measurements (MS2F), the sensor expects a volume of material 20 mm in length times 15 mm in diameter. A gap in core material or between sensor and material thus results in a lower value. In addition, the calibration of the sensors also contributes to the differences between the three types of measurements.
During Leg 183, we could not program the AMST software to ignore measurements from voids or disturbed intervals of core (e.g., drilling slurry or biscuits). Consequently, these spurious measurements should be discarded before the data set is used. Moreover, the software (AMST v. 1.1) aborted the measurement each time the susceptibility sensor reached a core gap previously detected by the laser displacement sensor. To overcome this problem, gaps were covered during laser displacement measurements. This, however, led to systematically lower susceptibility measurements over core gaps. It is recommended that these values be discarded from the data set by comparing them with the core photographs. For rock cores, it was not always possible to measure planar, horizontal surfaces. Measurements from nonplanar, nonhorizontal surfaces produced slightly lower values (e.g., when the sensor touched the rock surface at an oblique angle). In some cores, the sensor stopped a few millimeters above the surface, also resulting in slightly lower values. Finally, a software error led to a slight shift of the starting point every few cores. This produced an offset of the actual length scale, which must be taken into account when comparing these measurements with the core sections.
The standard ODP core orientation convention (fig. 8 in Shipboard Scientific Party, 1991; fig. 8 in Shipboard Scientific Party, 1997) was used for paleomagnetic measurements. This convention is defined as follows: the z-axis points downhole and is parallel to the core, and the x-axis forms a line perpendicular to the split face of the core and is directed into the working half, toward a reference mark (a double line) along the center of the core half. Discrete sample cubes and minicores were marked with an arrow in the negative-z (uphole) direction on the plane representing the split surface of the working half. The plane marked with the arrow is the y-z plane. Core orientation with a Tensor tool mounted on the APC core barrel was not achieved because only the RCB was used for drilling. Because all cores were drilled by the RCB method, only inclination values were useful in paleomagnetic studies during Leg 183.
Discrete samples were taken from soft sediment using standard plastic cubes (7 cm3), with an arrow pointing in the uphole direction. Samples were not taken from portions of the core that were highly disturbed by drilling. To reduce deformation of the sediment while taking discrete samples, the core was cut using a thin stainless-steel spatula before pressing the cubes into the sediment. Cylindrical minicores (12 cm3) were drilled from hard sediment and crystalline rocks using a water-cooled, nonmagnetic drill bit attached to a standard drill press. Minicores were oriented in the same manner, with an arrow pointing in the uphole direction. Typically, we took one or more discrete samples per section, and some were used for pilot demagnetization studies. In addition, discrete samples were taken from sedimentary rocks at smaller increments to examine geomagnetic reversals, key geologic boundaries, or other intervals of interest. We took several discrete samples in each flow unit of igneous rock.
The NRM of the archive-half sections was analyzed on the cryogenic magnetometer at 5- and 2.5-cm intervals for sediments and basement rocks, respectively. To isolate characteristic magnetizations, archive halves were AF demagnetized at 20 and 40 mT for sediments and basement rocks, respectively. To confirm the reliability of the remanent magnetization, one section per each core was stepwise AF demagnetized up to 30 and 60 mT for sediment and basement rocks, respectively. For magnetic cleaning of most discrete igneous rock samples, stepwise thermal demagnetization up to 620°C was applied.
For determining the AMS, the induced magnetization of the specimens was measured at 15 positions in a 300 A/m field and the susceptibility tensor was calculated from the data. The principal magnitudes and directions of the tensor are obtained by the diagonalization of the tensor; the principal directions are given by the eigenvectors and the principal magnitudes by the eigenvalues.
The stable or primary component of remanence was obtained from sediments after AF demagnetization. The stability of remanence within the archive cores and discrete samples was determined by both Zidjerveld (1967) plots and equal-area stereographic projections. Magnetozones were defined by selecting directional data from the cryogenic magnetometer. The selection criteria were slightly different at each site. The following selection criteria were typically used during Leg 183:
Positive and negative inclinations are defined as reversed and normal magnetic polarities, respectively. We interpreted the magnetic polarity stratigraphy, using constraints from the biostratigraphic data (see "Biostratigraphy"). The time scales of Berggren et al. (1995) and Gradstein et al. (1995) (see "Biostratigraphy") were used for Cenozoic and Mesozoic polarity boundaries, respectively.
The characteristic inclinations of remanence from igneous rocks were obtained by component analysis (Kirschvink, 1980), and stability was examined in the same manner as for sediments. Paleolatitudes of each site at the time of eruption were calculated from mean inclination values based on the assumption of a geocentric dipole field using the equation:
tan (paleolatitude) = ½ tan (inclination). (1)The results from low-field susceptibility and AMS measurements were compared with lithologic units and/or geologic structures based on sedimentary, petrologic, and structural features (see "Lithostratigraphy," "Igneous Petrology," "Physical Volcanology," and "Structural Geology").