PALEOMAGNETISM

Magnetic results from Hole 1272A reveal two intervals with distinctive remanence characteristics, corresponding to the two lithologic units identified for this site (see "Lithology and Stratigraphy" in "Igneous and Mantle Petrology"). Unit I (Cores 209-1272A-1R to 11R) includes dunite, harzburgite, gabbro, basalt, and sedimentary breccia. This unit is characterized by highly variable inclinations and demagnetization behaviors. It is difficult to establish whether the core from this unit is in situ, and thus the geological significance of the remanence from Unit I is uncertain. In contrast, the relatively uniform altered peridotites of lithologic Unit II (Cores 209-1272A-12R to 27R) show correspondingly little variation in magnetic properties. Unit II has uniformly positive inclinations (mean = ~45°) that are consistent with a normal polarity magnetization.

Pass-through remanence measurements from lithologic Unit I were sparse as a result of the low core recovery and the short average length (generally <10 cm) of the core pieces. From Core 209-1272A-13R to the bottom of the hole (lithologic Unit II) the number of pieces >10 cm in length increases and magnetic measurements are more continuous (Fig. F67). All archive halves were subjected to stepwise alternating-field (AF) demagnetization, typically in 5-mT increments to 30 mT followed by 10-mT steps to a maximum peak field of 50 mT. When few pieces of sufficient length were present in a section, these pieces were measured individually to avoid edge effects from adjacent pieces with different orientations (see "Paleomagnetism" in the "Site 1271" chapter).

The natural remanent magnetization (NRM) intensities from Site 1272 cores range ~0.2–5 A/m. Altered harzburgites and dunites from lithologic Unit I generally have intensities <1 A/m, whereas the gabbros, microgabbros, and basalts from this unit typically have NRM intensities >1 A/m. The altered peridotites of lithologic Unit II show a trend of decreasing NRM intensity with depth, paralleling a downhole decrease in magnetic susceptibility (Fig. F67). This trend is suggestive of lower degrees of serpentinization near the bottom of the hole, an inference that is supported by the increasing abundance of fresh olivine in the lowermost portion of Unit II (see "Igneous and Mantle Petrology" and "Metamorphic Petrology").

The initial inclinations for the archive halves are generally between 60° and 90°, reflecting the influence of a steep, presumably drilling-related overprint. This low-stability overprint is generally removed by AF demagnetization at 10–20 mT (Fig. F68B, F68D–F68F). Demagnetization of core pieces from lithologic Unit I reveals a range of stable inclinations from –45° to +60° (Fig. F61). The origin of this variability is discussed below in conjunction with the discrete sample results. After removal of the drilling remanence, a stable remanence direction with moderate positive inclinations (generally 30°–60°) is isolated for many samples from lithologic Unit II. However, in some cases the distinction between the drilling overprint and a final stable characteristic remanence (ChRM) direction is less obvious (Fig. F69D). Moreover, for the altered peridotites of Unit II, the stable remanence represents <10% of the NRM (average = 2.5%). Based in part on comparison with results from discrete samples (see below), we suggest that the archive-half data from Unit II provides a reasonable approximation of the stable ChRM.

The various lithologies of Unit I exhibit a range of behaviors during demagnetization (Fig. F68). A low-stability drilling overprint, generally removed by demagnetization at 10 mT, is evident in most lithologies. For most diabase/microgabbro samples (Fig. F68A, F68E) and the olivine gabbro from Core 209-1272A-4R (Fig. F68F), this low-stability overprint constitutes a large fraction of the remanence (the ChRM represents <10% of the NRM) (Table T9). In contrast, less drilling overprint is discernible in the weathered dunite from Core 209-1272A-2R (Fig. F68C) and the diabase/microgabbro from Core 7R (Fig. F68G). The final stable remanence direction for discrete samples from Unit I is highly variable (five samples have inclinations ranging from –38° to +52°).

The altered peridotites of Unit II exhibit much more uniform behavior during demagnetization (Fig. F69). All samples show a low-stability drilling overprint that is removed by demagnetization at ~10 mT. Although the final stable direction typically represents 1%–5% of the NRM, the ChRM direction is well isolated after demagnetization at 10–15 mT. A single sample from Core 209-1272A-27R has a ChRM representing 25% of the remanence (Table T9). Samples from Unit II also have low median destructive fields (MDF; the peak AF necessary to reduce the vector difference sum to 50% of its initial value) of ~2 mT. These low MDF values reflect the significant low-stability drilling overprint that is most likely carried by coarse-grained magnetite with low coercivities.

The geological significance of the ChRM directions isolated from samples in lithologic Unit I is uncertain. For example, five small (~9 cm) pieces of diabase/microgabbro from Core 209-1272A-7R were analyzed in pass-through mode and yielded very stable remanence but incoherent inclinations ranging –25°–20° (Fig. F67). A discrete sample from this same core (Fig. F68G) shows only a small drilling overprint and a well-defined stable inclination of –39°. Moreover, both the archive-half and discrete sample data from Core 209-1272A-7R indicate high magnetic stability. The scattered remanence directions for samples with such high stability suggest that the core pieces are no longer in the orientations in which the ChRM was acquired. Other similarly aberrant directions are evident in the pass-through remanence data from other lithologies (e.g., dunite at Section 209-1272A-2R-1, 20 cm) (Fig. F68B).

Other intervals in Unit I have apparently coherent stable remanence directions that are generally compatible with the expected inclination at the site as well as with directions obtained from Unit II (see below). For example, some of the altered harzburgites and dunites of Cores 209-1272A-1R and 2R have ChRM directions with positive inclinations similar to the expected time-averaged inclination at the site (Fig. F68C). The gabbros from Cores 209-1272A-3R and 4R and breccias from Core 5R also have coherent, though significantly steeper, inclinations. The breccias (composed of angular clasts of peridotite) yielded coherent demagnetization results, with stable magnetization, positive and shallow inclinations, and no significant drilling overprint.

There are at least two possible explanations for the scattered remanence directions in portions of Unit I. First, many of these samples may represent rocks that have fallen from the walls of the borehole. This is a feasible scenario for core pieces that are smaller than the nominal diameter of the borehole (10–12 cm) as, for example, for the pieces from Core 209-1272A-7R that all have lengths <9.5 cm. If this explanation is valid for all pieces in Core 209-1272A-7R, then core pieces from higher levels might represent in situ samples. Alternatively, the directional heterogeneity in Unit I may represent sampling of a talus pile. In this case, individual core pieces (particularly for pieces longer than the width of the core liner and with parallel edges) may accurately reflect the vertical orientation of blocks within the talus pile. Because the origin of the inclination variation within Unit I remains uncertain, paleomagnetic data from this unit will not be considered further here.

Samples from lithologic Unit II, in contrast, have consistent positive inclinations that are broadly consistent with the expected normal polarity dipole inclination (28°) at the site (Fig. F70). Fourteen discrete samples from this unit yield an average inclination of 45.2° (+5.4°/–6.7° 95% confidence limits; = 42.4) using the inclination only technique of McFadden and Reid (1982). The distribution of inclinations obtained from archive-half measurements (Table T10) is slightly broader than that from the discrete samples. Nonetheless, the archive-half data yield a mean inclination (42.7° +1.4°/–3.5°; = 25.6; N = 139) that is nearly identical to the discrete sample mean inclination. The declination distributions from both discrete samples and archive halves are also similar, with a concentration of values in the northeast quadrant. This clustering of declination values presumably reflects the imposition of a relative orientation prior to splitting the cores.

The mean inclination for Site 1272 is statistically distinct (at the 95% confidence level) from the expected dipole inclination at the site. As suggested for previous sites, this inclination discrepancy suggests that the site has been tectonically rotated since acquisition of the remanence. Although counterclockwise rotations about a ridge-parallel (020°), horizontal rotation axis may initially steepen a normal polarity remanence, such a rotation would steepen the remanence by a maximum of ~8° and thus could not account for the observed inclination of 45°. Because Site 1272 was drilled on an inside corner bathymetric high (~10 km from the fracture zone) the rotation axes are poorly constrained and other rotation axis orientations are possible. A more complete discussion of tectonic scenarios that could explain the remanence data can be found in "Structural Geology".

Anisotropy of magnetic susceptibility (AMS) was determined for all discrete samples from Site 1272 (Table T11). Given the uncertainty in whether samples from Unit I are in situ, only the magnetic fabric data from Unit II will be considered. The peridotite samples from this unit show variable degrees of anisotropy (P = maximum/minimum eigenvalues [1.04–1.37]). Most samples have oblate magnetic fabrics, although the three eigenvalues are distinct in all cases.

The remanent declinations from these same samples can be used to provide a first-order reorientation of the magnetic fabrics into a geographic reference frame. A simple vertical axis rotation has been applied to each sample to restore the magnetic declination to 360° under the assumption that the ChRM in each sample reflects a normal polarity magnetization that approximates the time-averaged dipole declination at the site. No attempt has been made to account for the declination offset that would result from tilts about horizontal rotation axes.

After this reorientation, the poles to magnetic foliation (the minimum eigenvectors) are mostly clustered in the southwest quadrant. The corresponding maximum eigenvectors are subhorizontal, with a mean trend/plunge of 151°/12° (Fig. F71). The reoriented magnetic foliations are approximately parallel to the dominant northeast-dipping planes, which have been interpreted as fractures, evident in the downhole resistivity images from the site (see "Downhole Measurements"). The susceptibility anisotropy is measured on discrete samples (free of fractures), and thus the reason for the close correspondence with the fractures imaged during logging is not immediately clear. One possibility is that the AMS data reflect the distribution of magnetite in serpentinite veins and these veins, in turn, may constitute (or be parallel to) planes of weakness for fracture formation.