Previous   |   Next

doi:10.2204/iodp.proc.320321.105.2010

Paleomagnetism

We conducted a paleomagnetic study of archive-half sections of 48 APC and 12 XCB cores from Holes U1333A–U1333C, with the primary objective of determining the magnetostratigraphy of the site and providing chronostratigraphic age constraints. To accomplish this we measured the natural remanent magnetization (NRM) of each section at 5 cm intervals before and after AF demagnetization of 20 mT. When time permitted, an additional 10 mT demagnetization step was measured. We processed the extracted data from the Laboratory Information Management System (LIMS) database by removing measurements made within 5 cm of section ends and data from disturbed intervals (Table T11). Cleaned data are presented in Tables T12, T13, T14, T15, T16, T17, T18, and T19.

Core orientation was estimated from paleomagnetic declination data as described in "Paleomagnetism" in the "Site U1331" chapter. The azimuthal core orientation was determined by correlating distinct reversal patterns as recorded by the paleomagnetic declination in each hole with the geomagnetic polarity timescale (GPTS). When distinct correlatable patterns are not easily recognized, this method could lead to a magnetic polarity ambiguity in which one might be unable to differentiate between magnetic north and magnetic south. Such ambiguities can be resolved in most cases by using biostratigraphic age estimates to guide the mapping of identifiable reversals in each hole. Once we had confidently identified a unique, unambiguous reversals pattern, the mean paleomagnetic directions for each hole were calculated using Fisher statistics (Table T20). Subsequently data were reoriented so that normal and reversed polarity magnetozones had declinations of ~0° and ~180°, respectively.

Magnetic susceptibilities were measured for 106 discrete samples. The data were mass and volume corrected using sediment moisture and density data (MAD) (see "Physical properties") and are presented in Table T21. Of these samples, 72 were stepwise AF demagnetized and measured at 5 mT steps to a peak field of 40 mT and 10 mT steps to 60 mT. The remanence measurements and the characteristic remanent magnetization (ChRM) directions computed using principal component analysis (PCA) are given in Tables T22 and T23.

Results

Downhole paleomagnetic data for Holes U1333A–U1333C are presented in Figures F12, F13, and F14, respectively. NRM measurements indicate that the viscous isothermal remanent magnetization (IRM) drilling overprint (see "Paleomagnetism" in the "Site U1331" chapter) was weak for Hole U1333A; weak above 83 m CSF in Hole U1333B; and weak above ~107 m CSF in Hole U1333C. The increased strength of the drilling overprint in Hole U1333B below 83 m CSF and in Hole U1333C below ~107 m CSF coincides with the switch from nonmagnetic to standard steel core barrel. Figure F15 illustrates the effect of the steel core barrel and confirms the value of using nonmagnetic coring equipment. NRM inclinations before demagnetization reflected the patchy overprint with values of anywhere between –10° and 90°. The small and sometimes negative overprint may have been caused by the bottom-hole assembly (BHA) and drill string becoming magnetized by the local geomagnetic field, which is more or less horizontal. Declinations were typically less severely affected, and it was often possible to identify reversals before demagnetization. The patchy and sometimes shallow overprints also indicate that the BHA and drill string are probably contributing little to the drilling overprint at this site.

Demagnetization data from discrete samples (Fig. F16) indicate that the ChRM of the sediments is carried above 10–20 mT demagnetization steps and that in most cases 20 mT demagnetization effectively removed the drilling induced IRM. PCA directions of the ChRM component agree with measurements of coeval intervals from the archive halves (see Fig. F12), indicating that the magnetic directions after 20 mT demagnetization provide a reliable indicator of the ChRM of the sediments.

In a few isolated intervals, the inclinations remain steep after 20 mT demagnetization, indicating that the drilling overprint was not demagnetized fully. Between ~80 and ~105 m CSF in Hole U1333B is a noisy interval where declination and inclination vary considerably (Fig. F13). We remeasured the sections from this interval but found that the results were repeatable; therefore, the source of the noise is within the cores. Steel core barrels were used in Hole U1333B below ~83 m CSF. It is possible that the magnetization of the steel core barrels was passed onto the cored sediments or that the sediments were contaminated with rust or iron particles during coring.

Magnetostratigraphy

The relatively clear polarity reversal pattern and detailed biostratigraphic framework of key nannofossil, radiolarian, and foraminifer datums from core catcher and additional samples (see "Biostratigraphy") allowed a relatively uncomplicated correlation of the magnetostratigraphy with the GPTS. The reversal depths for each core are provided in Table T24. The polarity interpretations for the three holes are provided in Figures F12, F13, and F14, and the summary of the magnetostratigraphy for this site is given in Figure F17.

At the top of Hole U1333A, our polarity assignments are constrained by clear declination and inclination records and by biostratigraphic data that indicate an age of ~21.5 Ma at the base of Core 320-U1333A-1H. Chron C6Ar was recovered twice (in Cores 320-U1333A-1H and 2H), which initially complicated our correlation. Most of Chron C7An is lost at a core break between Cores 320-U1333A-3H and 4H, Chron C8n.1r is absent, and the top of Chron C8r is lost in a break between Cores 320-U1333A-3H and 4H. Correlation of magnetozones with the GPTS is unambiguous for the remainder of the APC-cored portion of Hole U1333A to the base of Core 320-U1333A-10H, which records the upper part of Chron C12r. Extensive deformation and biscuiting of sediments associated with XCB coring below ~95 m CSF prohibits further identification of magnetozones.

The upper portion of Magnetozone N1 in Holes U1333B and U1333C is correlated with Chron C1n; however, biostratigraphic ages indicate ~21–22 Ma at ~10 m CSF. Therefore we correlate the lower portion of Magnetozone N1 with Chron C6n. Below this interval our interpretation is straightforward with one to one correlations with the GPTS to Chron C13r with a few exceptions where reversals were lost in core breaks.

In Hole U1333B, correlation of the magnetostratigraphy with the GPTS is relatively simple above Magnetozone N21. Chron C6Bn.2n is absent, and Chrons C11n.1r, C16n.2n, and C17n.2n are lost in core gaps between Cores 320-U1333B-8H and 9H, 13H and 14H, and 14H and 15H, respectively. Between ~80 and ~104 m CSF is an interval with unstable magnetic directions probably affected by a drilling-induced magnetic overprint. The reversal between Magnetozones R22 and N23 occurs within Section 320-U1333B-12H-6 and is correlated with the reversal between Chrons C12r and C13n. The Chron C13n/C13r boundary occurs in a break between Cores 320-U1333B-12H and 13H, so the true thickness of Chron C13n can not be determined in Hole U1333B.

At the base of Hole U1333B, correlation with the GPTS is more difficult because of coring gaps and infrequent reversals. Thick normal polarity Magnetozones N28 and N29 are correlated with Chron C18n.1n. The thin (~25 cm) reversed polarity interval R28 represents what is probably a newly recognized cryptochron within Chron C18n.1n. Our correlation to the bottom of Hole U1333B is aided by the lowest occurrence of R. umbilicus, which has an age of 42.5 Ma (see "Biostratigraphy") and occurs between 158.3 and 162.94 m CSF. Consequently it is our preference to correlate Magnetozone N32 with Chron C20n, as this agrees most closely with the biostratigraphy and the polarity boundaries that are more completely recovered in Hole U1333C. XCB coring below ~162 m CSF prevents further interpretation of the magnetostratigraphy.

In Hole U1333C, Chron C7An is lost in a 3 m core break between Cores 320-U1333C-4H and 5H; therefore, the upper portion of Magnetozone R13 is correlated with Chron C7r and the lower portion with Chron C7Ar. The Magnetozone R20/N21 boundary is in a gap between Cores 320-U1333C-13H and 14H and is correlated with Chron C12r/C13n boundary. The Chron C13n/C13r boundary is, however, intact and occurs within Section 320-U1333C-14H-4. As is the case in Hole U1333B, Chron C13n is incompletely recovered, although the combined data from Holes U1333B and U1333C provide a complete record. Chron C15r is lost in a gap between Cores 320-U1333C-14H and 15H; therefore, the upper portion of Magnetozone N22 is correlated with Chron C15n and the lower portion with Chron C16n.1n. The cryptochron recognized within Chron C18n.1n in Hole U1333B is also recognized in Hole U1333C, where it spans ~30 cm. At the base of Hole U1333C, correlation with the GPTS is more difficult because of coring gaps and infrequent reversals. The Chron C18r/C19n boundary and the upper portion of Chron C19n fall within a break between Cores 320-U1333C-18H and 19H. The lowest reversal in Hole U1333C and for Site U1333 is the Chron C20n/C20r boundary, which occurs at 161.5 m CSF. Three XCB cores were collected below the last APC core (320-U1333C-21H), but these recovered only a small amount of material within core catchers and so were not measured.

Top of page   |   Previous   |   Next