IODP Proceedings    Volume contents     Search

doi:10.2204/iodp.proc.317.206.2014

Results

NRM and directional measurements

Sample NRM intensity and directional data are given in Tables T1, T2, T3, and T4. NRM intensity ranges from 4.40 × 10–6 to 8.85 × 10–1 A/m with most samples between 10–3 and 10–5 A/m (median = 3.17 × 10–4 A/m). These intensities are relatively low but measurable with confidence on the shielded 2G cryogenic magnetometer. The intensity of some samples rapidly decreases to ~10% or less of the original NRM magnetization during demagnetization at low fields (below 20 mT), and reliability of measurement and stability at higher fields seems compromised (Fig. F1D). Samples of this kind and others for which erratic behavior upon demagnetization was observed have been ranked the lowest quality in the sample collection (Class 4), and no ChRM has been computed. Other samples display a more progressive decay upon demagnetization, and some of them do not fully demagnetize at the maximum field applied (100 mT). These samples usually demagnetize a steep downward component up to 20–25 mT and a second component at a higher field that usually trends toward the origin of the diagram that is considered the ChRM (Fig. F1A). Class 1 samples include those of higher quality trajectories that include several steps and usually have MAD angles <10 (i.e., Fig. F1A, F2A), whereas for Class 2 samples, the ChRM is less well defined and MAD angles are usually <20 (Fig. F1B). Class 3 samples include those for which a ChRM can be computed (usually not including consecutive steps), but they are classified not reliable and were not considered for the magnetostratigraphic determinations. Some samples notably increase in intensity above 60 mT (Fig. F2C) and define a horizontal direction along the x-y axis of the samples. This is the result of acquisition of a gyroremanence (a magnetic remanence acquired by magnetically anisotropic materials that are rotated within an alternating magnetic field) related to the static application of the three-orthogonal AF demagnetization, but the ChRM component can still be calculated in the range of 20–60 mT. This type of behavior has been described before and appears commonly associated with iron sulfides such as greigite (e.g., Fu et al., 2008). The presence of magnetic iron sulfides in the studied sediments can also be inferred from the thermal demagnetization of some samples (Fig. F3H), in which a typical decrease in intensity is observed around 360°C. However, most samples do not show the gyroremanence acquisition, and it is likely that the magnetic carriers are a combination of (titano)magnetites and iron sulfides (see below). It should be noted that dissolution of magnetic minerals and remagnetization of continental shelf sediments have been described in the literature (e.g., Oda and Torii, 2004; Rowan and Roberts, 2006).

The ChRM inclination for the 305 Class 1 and 2 samples form two populations with negative (normal) and positive (reversed) inclinations (Fig. F4) with average values of about –50°/50°, respectively, that are shallower than the expected geocentric axial dipole values for the site latitude. It is common for marine sediments to acquire a shallower magnetization related to depositional processes or compaction (Arason and Levi, 1990), and this may indicate the successful removal of the steep overprint for most of the Class 1 and 2 samples. However, the ChRM component of samples from the deep part of Hole U1352C below ~1600 m core depth below seafloor, Method-A (CSF-A) displays a single steep downward component (e.g., Fig. F3I), indicating unsuccessful removal of the pervasive drilling overprint in those cases.

Thermomagnetic measurements

In addition to the shipboard measurement of isothermal remanent magnetization (IRM), IRM acquisition curves, and subsequent AF demagnetization (see individual site chapters), variations of low-field magnetic susceptibility versus temperature (thermomagnetic curves) were also measured. Figure F4 illustrates typical examples of thermomagnetic curves. For some samples, two successive heating/cooling runs to higher maximum temperatures were measured (Fig. F5A, F5B, F5F) to evaluate reversibility and the eventual formation of new mineral phases upon heating. In some instances heating to 400°C results in no reversibility (Fig. F5A), suggesting that a new magnetic mineral phase has formed. The new phase appears to be magnetite, as seen from Curie temperatures of 580°C on the successive heating run up to 700°C. Heating to 380°C produces mostly reversible curves in some samples (Fig. F5B), although magnetite with a Curie temperature of ~580°C can again be observed on the successive heating run. Therefore, the presence of magnetite in the original samples remains ambiguous. However, for some samples, lower Curie temperatures around 500°C have been measured (Fig. F5E), suggesting the presence of some sort of titanomagnetite. No direct evidence of magnetic iron sulfides has been observed in the thermomagnetic curves, although they appear to be ubiquitous in some intervals as deduced from NRM thermal demagnetization and inferences from gyroremanence acquisition. Collectively, all the experiments suggest the presence of a combination of magnetic iron sulfides (possibly greigite) and (titano)magnetite in the shelf/slope sediments of the Canterbury Basin.

Magnetic stratigraphy

Hole U1351B

Demagnetization data are found in Table T1. NRM intensity and ChRM inclination for discrete samples from the upper 270 m of Hole U1351B are plotted in Figure F6 along with shipboard data (see the “Site U1351” chapter [Expedition 317 Scientists, 2011c]). Despite the fragmentary nature of the record due to low core recovery, some magnetostratigraphic constraints emerge. Only Class 1 and 2 samples are used for magnetozone polarity determinations in Figure F6. Magnetic polarity appears to be mostly normal (negative inclination) to Core 317-U1351B-9H at ~63 m CSF-A, with the exception of three relatively thin intervals where single Class 2 reversed polarity (positive inclination) intervals exist (Fig. F1B). The normal interval can be assigned to the Brunhes Chron (C1n), whereas the thin reversed intervals could either represent short excursions within the Brunhes Chron or some sort of secondary component or remagnetization (we note a noisy signal in shipboard measurements within this interval). Core 11H contains Class 2 reversed polarity samples, thus loosely defining a reversed to normal boundary (R/N_1) between Sections 9H-6 and 11H-1 that must represent the Matuyama/Brunhes transition, even though part of Brunhes and/or Matuyama may be missing. Below Core 11H, recovery is very fragmentary, but all samples measured down to Core 18X appear to be reversed polarity (i.e., Fig. F2B, F2C). Core 19X includes high-quality samples of normal polarity (negative inclination) (e.g., Fig. F2D), whereas Cores 20X and 21X include Class 1 samples with reversed polarity. Consequently, the N/R_1 and R/N_2 polarity transitions can be constrained (Fig. F6). Another pair of reversal boundaries can be placed farther down in the hole (N/R_2 and R/N_3). These reversal boundaries are shown in Figure F7, which summarizes shipboard biostratigraphic constraints on an age-depth plot (see the “Expedition 317 summary” chapter [Expedition 317 Scientists, 2011a]) correlated to the geomagnetic polarity timescale (Lourens et al., 2004; Ogg et al., 2008). Consequently, the N/R_1 reversal may be correlated to the top of the Olduvai Subchron (C2n/C1r.3r, 1.778 Ma) and the RN_2 reversal to the Chron C2r.1r/C2n boundary (1.945 Ma) (Fig. F7). It is important to note that Surface U1351B-S6 (see the “Expedition 317 summary” chapter [Expedition 317 Scientists, 2011a]) is a sharp basal contact at Section 19X-2, 104 cm (144.14 m), between a 1.15 m thick shelly sand bed and an underlying mud with scattered shell fragments. A short normal polarity interval occurs in Sections 19X-3 and 19X-4 (Table T1), and therefore the upper part of the normal chron (C2n, Olduvai) and also the lower part of the reversed chron (C1r.3r) could have been truncated. Alternative reasoning without violating the biostratigraphic constraints could also correlate the normal chron in Core 19X to the older Reunion Subchron (C2r.1n, 2.128–2.148 Ma) (open stars in Fig. F7). Integrating additional biostratigraphic data could lead to refined correlation around this interval that relates to the classic Pliocene/Pleistocene boundary. Note that arguments to retain the base of the Quaternary system coincident with the base of the Calabrian stage of the Pleistocene series (1.806 Ma) failed when in 2009 the International Commission on Stratigraphy redefined the base of the Quaternary coincident with the base of the Gelasian stage (2.58 Ma), formerly the uppermost stage of the Pliocene. Finally, the N/R_2 reversal seems to correlate unambiguously with the top of Chron C2An.1n (2.581 Ma), whereas R/N_3 is best correlated to the base of Chron C2An.3n (the Gilbert/Gauss Chron boundary at 3.596 Ma).

Holes U1352B and U1352C

Demagnetization data are listed in Table T2. NRM intensity and ChRM inclination for discrete samples from Holes U1352B and U1352C are plotted in Figures F8 and F9, respectively. The upper 240 m of Hole U1352B consists of normal samples (e.g., Fig. F2A), with the exception of one or two intervals with samples that clearly depict a reversed ChRM component (e.g., Fig. F3B). Shipboard data suggested reversed polarities at Core 317-U1352B-28H and below, but because at this level magnetic core barrels were introduced and because of the limited single 20 mT demagnetization step, ambiguity remains as to whether the reversal is meaningful (see the “Site U1352” chapter [Expedition 317 Scientists, 2011d]). Demagnetization of discrete samples from Core 29H (data from Core 28H look unreliable) clearly depicts reversed polarities (Fig. F3C), and hence a polarity reversal exists that must correlate to the Brunhes/Matuyama transition. Further downhole some Class 1 and 2 samples do exist with either normal or reversed ChRM, but the succession does not seem to clearly outline reasonable magnetozones because samples with different polarities succeed each other rapidly along the interval from 300 to 500 m CSF-A, delimiting an interval with zones depicted as ambiguous polarity. All samples measured below 520 m CSF-A from Hole U1352B contain a negative (normal) ChRM direction. In Hole U1353C, there seems to be a poorly constrained N/R reversal at ~705 m CSF-A with only one discrete Class 1 reversed sample in Section 317-U1352C-10R-2 (Fig. F9). Downhole, the succession of normal and reversed samples does not allow definition of a clear picture of magnetic zonation. The interval below 1500 m CSF-A to the Marshall Paraconformity) displays high-quality demagnetization data (i.e., Fig. F3I) with relatively steep downward inclinations and a single component of magnetization that suggests that a pervasive drilling overprint has not been removed.

Shelf Holes U1353B, U1354B, and U1354C

Demagnetization data are reported in Tables T3 and T4. NRM intensity and ChRM inclination data for Holes U1353B, U1354B, and U1354C are plotted in Figures F10, F11, and F12, respectively. Holes U1353B and U1354B both present normal polarities down to ~64–65 m CSF-A. The polarity change in shipboard data is clear for Hole U1354B (Fig. F11) (see the “Site U1354” chapter [Expedition 317 Scientists, 2011e]). Discrete sample data confirm the reversed polarities in Section 317-U1354B-13H-2, whereas only a Class 3 normal sample was measured from Core 12H. Collectively, the Matuyama/Brunhes transition can be established at 64.75 m CSF-A, although a lithologic boundary at this level suggests a sedimentary hiatus and a possibility that the upper part of the Matuyama Chron (and possibly the lower part of the Brunhes Chron) is missing, as indicated by biostratigraphic constraints (see the “Site U1354” chapter [Expedition 317 Scientists, 2011e]). The Matuyama/Brunhes transition is less obvious in Hole U1353B but has to be placed below Core 317-U1353B-10H, for which normal Class 1 and 2 samples were measured. Hole U1354C displays reversed polarity from 66 to ~110 m CSF-A. Below that depth some normal samples were measured. However, these normal samples are intercalated with good-quality reversed samples within short intervals (Fig. F12), and the pattern does not permit straightforward correlation. Data below 180 m CSF-A become too fragmentary and it was not possible to recover good magnetostratigraphy below that depth.