Previous   |    Next

doi:10.2204/iodp.proc.330.107.2012

Paleomagnetism

Archive-half core and discrete sample remanent magnetizations provide a consistent pattern of moderate to steep positive inclinations, reflecting Southern Hemisphere reversed polarity. Positive inclinations in the volcanic sandstone and limestone are similar to those in the underlying volcanic basement (including some breccia units). No magnetic reversals were observed for Site U1376 on Burton Guyot.

Archive-half core remanent magnetization data

Remanent magnetization of the archive halves of Cores 330-U1376A-1R through 23R was measured at 2 cm intervals using the cryogenic magnetometer. All data acquired within 4.5 cm of either piece end were filtered out prior to further processing, and thus only pieces longer than 9 cm were considered.

The natural remanent magnetization (NRM) intensity varies by >5 orders of magnitude (Fig. F50B) from a minimum of 2.51 × 10^–4 A/m (in sediments of Subunit IIA) to a maximum of 9.34 A/m (associated with lava lobes or fragments in Core 330-U1376A-19R; lithologic Unit 29). The most abrupt variation in NRM intensity and magnetic susceptibility occurs at the boundary between Unit I and Subunit IIA (Fig. F50B, F50C). The heterolithic volcanic breccia and volcanic sandstone at the base of Unit I are characterized by NRM intensities and susceptibilities that are ~3 orders of magnitude higher than those for the white algal limestone at the top of Subunit IIA. Both NRM and susceptibility increase systematically with depth throughout Subunit IIA and into the heterolithic conglomerate of Subunit IIB. Significant NRM intensity and magnetic susceptibility variations also occur in the volcanic basement (Units III and IV). Lava flows and dikes typically have NRM intensities in the range of 1–10 A/m, and this same range of variations is also apparent in the 33 m thick massive lava flow (lithologic Unit 15; 72.21–105.32 mbsf). The magnetization of volcanic breccia is quite variable. For example, the hyaloclastite breccia of lithologic Unit 21 (~114–127 mbsf) has NRM values that range from >1 A/m to low values (~10^–3 A/m), similar to that of the weakly magnetized limestone recovered at this site.

Best-fit principal component directions were calculated from alternating-field (AF) demagnetization data for each 2 cm interval of the archive halves using an automated routine that maximizes the percentage of remanence incorporated and minimizes the scatter about the best-fit direction and the deviation of this vector from the origin (see “Paleomagnetism” in the “Methods” chapter [Expedition 330 Scientists, 2012a]). On the basis of the distribution of misfit values, only directions with misfit values of <2.56 were considered, which represents the most reliable 40% of the total number of 2 cm interval principal component directions (see black and bright red circles in Fig. F50B, F50D). The significantly lower misfit value employed at this site suggests that the overall quality of the data set measured here is higher than that at Sites U1372, U1373, and U1374, where the misfit cutoff value was set at ~3.4. The resulting inclinations, intensities, and stability of remanent magnetization, as represented by the median destructive field (MDF′) of the vector difference sum, are shown in Figure F50D, F50B, and F50E, respectively.

Hole U1376A is dominated by positive inclinations, indicative of Southern Hemisphere reversed polarity (Fig. F50D). Only 1% of reliable data have negative inclinations. The uniform positive inclination is most evident in the 33 m thick lava flow (lithologic Unit 15), but a comparable signal is present in the volcanic breccia, dikes, and thinner lava flows in Units III and IV, including intervals of volcanic breccia and pillow fragments (e.g., Cores 330-U1376A-6R and 7R) with best-fit directions that have misfits exceeding the cutoff value of 2.56 (see light red circles in Fig. F50D). The algal limestone (Subunit IIA) shows the same consistent reversed polarity magnetization as the sequence stratigraphically below. The volcanic sandstone and volcanic breccia of Unit I also appear to have the same polarity, although inclinations are more scattered and many of the best-fit directions have misfit values that exceed the cutoff value.

Discrete sample remanent magnetization data

The remanent magnetization of 99 discrete samples from Hole U1376A was measured with the spinner magnetometer. NRM intensities range from 1.3 × 10^–4 A/m to 11.3 A/m (Fig. F51; Table T12) and are generally consistent with those determined from the same interval on archive halves. The lowest values are determined for hyaloclastite breccia samples from Core 330-U1376A-15R. The highest NRM intensities occur in lava flows, but only slightly lower magnetizations characterize many of the basalt clasts in volcanic breccia and heterolithic breccia units. Königsberger ratio (Qn) values of basalts (lavas, intrusive sheets, and clasts in volcanic breccia/conglomerate) are all >1 and typically ~5 or higher, indicating that the magnetization is dominated by remanent magnetization. Samples from hyaloclastite and the matrix of volcanic breccia display a wide range of Qn values, with the highest values corresponding to basalt samples and generally lower values (as low as ~0.01) corresponding to hyaloclastite samples and the matrix of volcanic breccia.

Stepwise AF and thermal demagnetization were applied to 60 and 39 samples, respectively (Table T12). Demagnetization results from basalt samples reveal relatively simple behavior and are generally consistent with the steep positive inclinations observed in the archive-half data. Samples from volcanic sandstone, the relatively finer grained matrix of volcanic breccia/conglomerate, and limestone from Units I and II yielded positive inclinations, indicating reversed polarity (Fig. F52A). In several cases, particularly for samples with weak magnetization, the scatter of the best-fit direction was greater than the maximum angular deviation cutoff of 5°, and the characteristic remanent magnetization direction in these cases was not plotted or considered when evaluating the mean inclination for the site. In addition, a malfunction in the AF demagnetizer (DTech D-2000) resulted in five sediment samples (all from Units I and II) acquiring a spurious magnetization at the 5 mT step (see also “Paleomagnetism” in the “Site U1375” chapter [Expedition 330 Scientists, 2012d]).

AF and thermal demagnetization reveal the same magnetization component(s) for most basalt samples. This is true regardless of whether the basalt samples have high stability (e.g., vesicular pillow basalt samples from lithologic Unit 2; Fig. F52B) or more moderate stability characteristic of the 33 m thick lava flow of lithologic Unit 15 (Fig. F52C). In contrast, thermal and AF results from volcanic breccia and hyaloclastite units can yield different results (Fig. F52D). AF demagnetization typically reveals a single high-coercivity component of reversed polarity. This component apparently coincides with the lower unblocking-temperature magnetization component during thermal demagnetization, whereas at the highest unblocking temperatures an additional component is present. At present, it is not clear what the significance of this small, high-temperature component is; the better-defined low-temperature component is reported in Table T12.

The unblocking temperature spectra from thermal demagnetization studies provide some indication of the magnetic mineralogy of the samples. Many basalt samples, including those from the thick lava flow of lithologic Unit 15, have relatively low unblocking temperatures. In these samples, much of the remanence is removed by ~250°C, consistent with the dominance of Ti-rich titanomagnetite that has undergone little or no deuteric oxidation. In other samples, maximum unblocking is concentrated near 575°–600°C (Fig. F52B) or in a few cases extends well above 580°C and is accompanied by very high coercivity (Fig. F52E). These high unblocking temperatures probably indicate the presence of Ti-poor magnetite or (titano)hematite; both phases can be associated with high-temperature deuteric oxidation of titanomagnetite that is more common in subaerial lavas. A possible subaerial origin was suggested for highly vesicular lava fragments in lithologic Unit 26 (see “Igneous petrology and volcanology”) that occur only a few meters stratigraphically above the samples shown in Figure F52E. The magnetic mineralogy and the implications for the origin of the remanent magnetization will be the subject of shore-based studies.

Anisotropy of magnetic susceptibility

The anisotropy of magnetic susceptibility was determined for all discrete samples (Table T13). A significant fraction (~20%) of the samples, primarily those from sedimentary and hyaloclastite breccia units, are statistically isotropic. The remaining samples have an average degree of anisotropy (P′) of 1.02 and shape factors (T) ranging from –0.84 to 0.87 (where T = –1 if prolate, and T = 1 if oblate; Jelinek, 1981). The 33 m thick massive flow of lithologic Unit 15 does not have a consistent magnetic fabric (i.e., the eigenvectors associated with the minimum eigenvalues have a broad range of inclinations rather than the subvertical distribution expected for horizontal lava flows).

Discussion

The dominance of moderately to steeply dipping positive inclinations is further highlighted in Figure F53, which shows the average inclinations calculated using inclination-only statistics (Arason and Levi, 2010) for the 2 cm interval data and the Fisher piece-average data (see “Paleomagnetism” in the “Methods” chapter [Expedition 330 Scientists, 2012a]). In both cases, only a very limited number of lithologic units have shallow positive or negative inclinations (Fig. F53C, F53E) for which a low in situ confidence index (ISCI; see “Igneous petrology and volcanology”) of either 1 (lithologic Units 22, 31, and 39) or NA (lithologic Units 11 and 13) were assigned.

The most striking observation from Figure F53 is the coherence of inclinations in lithologic Unit 15, a single 33 m thick lava flow that represents a short interval of time. Figure F54 focuses on the depth interval 65–110 mbsf and highlights the limited variation in all the magnetic parameters measured. What limited variation does exist is not obviously related to the mapped flow boundaries. Nearby lithologic Units 14 and 17 have similarly high ISCIs, but these units exhibit relatively more scatter in intensity, stability (MDF′), and inclination than lithologic Unit 15, although less scatter in magnetic susceptibility. The average inclination for lithologic Unit 15 is 71.9° ± 0.4° (N = 909) when calculated from 2 cm interval data and 70.1° ± 0.9° (N = 95) when calculated with Fisher piece-average data (Fig. F54E, F54F; Table T14).

The upper and lower boundaries of lithologic Unit 15 have a distinct magnetic signal. Three Zijderveld diagrams from lithologic Unit 14 are shown in Figure F55B. These samples show a progressive change from high to low coercivity with decreasing distance from the top of the massive flow. This change is also illustrated graphically by the systematic decrease in MDF′ (light blue circles in Fig. F55A) and supports the interpretation that the contact is gradational (see “Igneous petrology and volcanology”). The base of lithologic Unit 15 occurs in Section 330-U1376A-13R-4, 59 cm. In the lowermost 20 cm of the massive flow unit, MDF′ begins to increase again to values representative of lithologic Unit 16 stratigraphically below (light blue circles in Fig. F55C). Over this same interval there is a pronounced increase in magnetic susceptibility. The Zijderveld diagrams associated with measurements toward the base of lithologic Unit 15 (Fig. F55D) show a progressive increase in coercivity to values characteristic of lithologic Unit 16 (Fig. F55E). Together with magnetic data, the presence of hyaloclastite material similar to lithologic Unit 16 at ~45 cm in the section suggests that the base of lithologic Unit 15 might be more appropriately located at ~40 cm. This area will be studied in more detail, and a baked contact test will be carried out as part of shore-based research.

Inclination data from archive halves and discrete samples from Hole U1376A provide a remarkably consistent picture of moderate to steep reversed polarity magnetization (Fig. F56). The most abundant inclination data are from the unfiltered 2 cm archive-half measurements. These data show a pronounced peak near 70° (median = 67.7°) but a significant number of shallower positive to negative inclinations (Figs. F53A, F56A). The highest-quality archive-half data (misfit ≤ 2.56) have an even narrower distribution, with a median value of 70.7° (Fig. F56B). If intervals with low (<2) ISCI values are excluded, the distribution changes relatively little (Fig. F56C). The remaining data (ISCI = 3 or 2) yield an inclination-only mean of 72.2° ± 0.5° (α₉₅). Fisher piece-average directions from individual archive-half core pieces should provide a more robust estimate of average inclination because pieces with heterogeneous magnetization can be recognized by their high circular standard deviation (CSD) values (with CSD > 20° being excluded). These piece-average inclinations, after removal of three negative inclinations, result in an inclination-only mean of 69.4° ± 1.2° (α₉₅) (Fig. F56D). Finally, inclinations from discrete samples also show a similar inclination value. The inclination-only mean for all discrete samples with positive inclinations is 66.7° ± 4.4° (α₉₅) (Fig. F56E) and is within error of the inclination-only mean derived from the piece-average data.

Summary

Our shipboard measurements suggest that in addition to the volcanic materials recovered, the sedimentary rocks from the uppermost ~40 m of this hole likely possess paleomagnetic directions that may be used for establishing the magnetic polarity stratigraphy at this site and also for estimating the paleolatitudes at the time of deposition. Although we encountered trouble during stepwise AF demagnetization, as noted above, some samples from limestone and volcanic sandstone indicated linear remanent magnetization components with moderate coercivities (MDF′ = 10–30 mT) and high maximum unblocking temperatures (>500°C), as shown in Figure F52. These components are not a recently acquired viscous remanent magnetization because they have positive inclination (i.e., reversed polarity in the Southern Hemisphere). However, care needs to be taken when interpreting sedimentary inclinations in determining paleolatitudes because they may have been affected by compaction-induced inclination shallowing. This subject will be addressed by detailed shore-based investigations.

Top of page   |    Previous   |    Next