Previous   |   Next

doi:10.2204/iodp.proc.320321.110.2010

Geochemistry

Shipboard geochemical analyses of interstitial water and bulk sediment samples reflect large variations in sediment composition resulting from shifts in carbonate versus opal production. The large-scale redox state and diagenetic processes of the sediment column are related to overall changes in sediment composition. The interstitial water chemistry points to seawater circulation in the basement, whereas reactions with the basement itself appear to exert little influence on the geochemistry of sediments and interstitial waters.

Sediment gases sampling and analysis

A total of 39 headspace gas samples were taken from Hole U1338A at a frequency of one sample per core as part of the routine environmental protection and safety monitoring program. The concentration of methane (C₁) in the samples was <1.7 ppmv, and no hydrocarbon gases higher than C₁ were detected.

Interstitial water sampling and chemistry

A total of 118 interstitial water samples were collected from Holes U1338A and U1338B, 43 using the whole-round squeezing method and 75 by Rhizon sampling (Table T16). Whole-round squeezing was conducted throughout Hole U1338A and in two locations in Hole U1338B to cover poorly recovered intervals. Samples from the two holes were taken to constitute a single depth profile using the core composite depth below seafloor (CCSF-A) scale. Rhizon sampling was conducted at a resolution of four samples per core in the upper 100 m of Hole U1338A immediately after core sections were cut and transferred to the racks in the core laboratory. Rhizon samplers were inserted into the middle of the whole core through holes predrilled in the end cap at the bottom of sections. A diatom-rich interval (Section 321-U1338B-14H-5) was targeted for high-resolution Rhizon sampling (every 10 cm) in Hole U1338B, after being identified in the first hole. Following WRMSL physical property measurements to pin-point the location of the low-density diatom-rich interval, Rhizon samplers were inserted through holes drilled in the core liner at an angle of 55° to ensure only the center of the core was sampled. Chemical constituents were determined according to procedures outlined in "Geochemistry" in the "Methods" chapter. Analyses of squeezed and Rhizon-sampled waters were conducted during the same analytical sessions, and the data from adjacent Rhizon and squeezed samples agree within the measurement uncertainties in virtually all instances (Table T16; Fig. F23).

Chloride ion concentration (not corrected for Br^– contribution) varies slightly with depth and is generally within 555 to 565 mM (Fig. F23). Chloride values increase from ~559 to >565 mM in the upper 30 m of the section, potentially reflecting the more saline bottom waters of the Last Glacial Maximum (e.g., Adkins and Schrag, 2003). However, the precision of the potassium chromate indicator technique is poor compared to the small variation in chloride, and a potentiometric titration technique (Adkins and Schrag, 2003) would be required to reveal more detail. Alkalinity increases slightly downhole from ~2.7 mM at the sediment/water interface to peak slightly above 4 mM at 140 m CSF. Below this depth alkalinity values remain stable at ~4 mM then decrease toward 3 mM at the base of the section. Seawater contamination of the XCB-sampled waters may be responsible for the apparent decrease, as the APC-cored samples from Hole U1338B both have high values. Sulfate concentrations decrease from 28 mM at the top of the section to a minimum of ~23 mM at ~130 m CSF before increasing with depth to ~26 mM near basement.

A very large dissolved manganese peak of 150 μM at 10 m CSF is captured by the high-resolution interstitial water sampling and is remarkably similar to that observed at Site U1337. These peaks are >3 times greater than the highest dissolved manganese concentrations encountered during Expedition 320. Below this peak, manganese values drop to low values of ~2 μM around 150 m CSF before a much smaller second peak of ~13 μM, which is centered around 220 m CSF. Below 300 m CSF, dissolved manganese is low but always detectable. Dissolved Fe displays three distinctive peaks downcore centered around 40, 215, and 300 m CSF. Interestingly these dissolved Fe peaks of ~5 μM broadly correspond to zones of good magnetic remanance (see "Paleomagnetism"). These large variations in dissolved Mn and Fe reflect changes in redox chemistry that also manifest as changes in sediment color. For example, below ~385 m CSF a sharp transition from pale green to yellow sediments occurs (see "Lithostratigraphy") where the dissolved Fe concentration is below detection. Such tandem changes in sediment color and interstitial water chemistry are well documented for the uppermost sediments where the iron hosted in smectites is reduced (Lyle, 1983). To find such a redox front in a reversed position, extending from the basement upward, suggests seawater has circulated through the basement at some time.

The silicic acid (dissolved silicate) content of the interstitial waters is substantially greater than that of bottom waters (e.g., Peng et al., 1993) and increases with depth from ~700 µM in the uppermost sediments to peak at ~1200 ?M at ~120 m CSF. Below this interval, silicic acid concentrations are high and generally stable between 1100 and 1200 µM. This strong enrichment in silicic acid almost certainly originates from the dissolution and diagenesis of biogenic silica. High-resolution Rhizon sampling (every 10 cm) of a diatom-rich interval in Hole U1338B reveals generally high silicic acid contents throughout but slightly higher concentrations in the diatom mat interval itself (Fig. F24). More precise shore-based measurements are required, but the initial data suggest active dissolution of the diatom opal in this interval.

Ca concentrations display very little variation downcore with values increasing from a seawater value (~10.5 mM) to two maxima of 12 mM at ~300 and ~390 m CSF. Mg and K concentrations scatter around seawater values. As the alteration of basaltic basement exchanges K and Mg for Ca (e.g., Gieskes, 1981), the lack of strong gradients in these elements suggests the basement is no longer reactive at this site.

Li concentrations decrease from ~26 µM at the surface to a minimum of ~3 µM around 250 m CSF before increasing sharply with depth to seawater values at the base of the section. The interstitial water Sr profile is a mirror image to that of Li except the decrease from the peak of 400 µM at 200 m CSF is punctuated by a sharp drop of >100 µM between ~260 and 290 m CSF. Li and Sr profiles indicate seawater circulation in the basement, as their values tend toward seawater values near the basement. The source of Sr to the pore waters is most likely the dissolution and recrystallization of biogenic calcite (e.g., Baker et al., 1982). High-resolution (every 10 cm) Rhizon sampling reveals elevated Sr concentrations in a diatom-rich interval of Hole U1338B (Fig. F24), suggesting a link between opal and biogenic carbonate diagenesis.

The dissolved Ba concentration is low in all samples, as expected given the high dissolved sulfate concentrations throughout the sediments. B concentrations in the interstitial waters generally range between 440 and 380 µM.

Bulk sediment geochemistry

Major and minor elements

At Site U1338, bulk sediment samples at a frequency of one per core were analyzed for Si, Al, Fe, Mn, Mg, Ca, Na, K, Ti, P, Ba, Cu, Cr, Sc, Sr, V, Y, and Zr (Table T17).

The Al, Mg, Fe, and Ba contents of the bulk sediment all display a similar downcore pattern with two distinctive peaks centered around ~50 and ~225 m CCSF-A (Fig. F25). Mg and Fe also increase toward the base of the section, and Fe peaks broadly correspond to the dissolved Fe peaks observed in interstitial waters. The Ca and silicon oxide contents of the bulk sediment mirror each other precisely (Fig. F25), demonstrating the important role of opal versus biogenic calcite production or preservation in determining the composition of the sediments at this site. Generally the silicon oxide content is higher and more variable above 270 m CCSF-A and vice versa for the calcium (carbonate) content.

Carbonate leachates from Sites U1336–U1338

The trace element composition of the bulk carbonate fraction was determined for selected samples from Sites U1336–U1338 (Table T18; Fig. F26). Carbonate was selectively dissolved in 1 M acetic acid adjusted to pH 5 with sodium acetate (Tessier et al., 1979; Lyle et al., 1984; Kryc et al., 2003) following a wash in pure water buffered to pH 10 with ammonium hydroxide. Further details and evaluation of this methodology are given in "Geochemistry" in the "Methods" chapter.

The Mg/Ca ratio of the bulk carbonate generally ranges between 2 and 6 mM/M, with Site U1336 samples exhibiting the lowest values and Site U1337 samples the highest. Mn/Ca ratios are generally high, ranging between 2 and 6 mM/M at Sites U1337 and U1338. The Mn/Ca ratio of the bulk carbonates measured for Site U1336 are all below 2 mM/M, which may be related to the much lower dissolved Mn content of interstitial waters at Site U1336 compared to Sites U1337 and U1338. The Sr/Ca ratio of Site U1337 and U1338 bulk carbonates varies between 2 and 3 mM/M, which is consistent with Sr/Ca ratios measured on modern coccolith samples (e.g., Stoll et al., 2002). The Sr/Ca ratio of the bulk carbonates from Site U1336 is generally lower than that at Sites U1337 and U1338, ranging between 1 and 2 mM/M. This lower Sr/Ca ratio possibly reflects a higher degree of carbonate recrystallization at Site U1336 compared to Sites U1337 and U1338. The Fe content of the bulk carbonate leachates was below detection in most samples from Sites U1337 and U1338 but was detectable in all samples measured from below 100 m CSF at Site U1336. The Li content of the bulk carbonates investigated was below detection (<4.8 μM) for all samples. This finding negates the possibility of carbonates providing the large sedimentary Li sink inferred from the interstitial water profiles.

Sedimentary inorganic and organic carbon

Calcium carbonate and inorganic carbon concentrations were determined on 245 and 62 sediment samples from Holes U1338A and U1338B, respectively (Table T19; Fig. F27). Hole U1338B samples were used to fill the stratigraphic gaps in Hole U1338A sampling because of the low recovery and core disturbance associated with XCB coring below Core 321-U1338A-27X. Calcium carbonate concentrations range between 26 and 88 wt% with substantial variability in the upper 273.31 m CCSF-A, corresponding to the alternation between calcite and opal in the upper two lithologic units (see "Lithostratigraphy"). Below 273.31 m CCSF-A (lithologic Unit III), calcium carbonate contents become generally high and stable between 66 and 91 wt% compared with the upper part of the stratigraphic column, with the exception of distinctive decreases at 295.66–297.00, 327.95–350.96, and 390.95–424.42 m CCSF-A.

TOC was determined for 60 sediment samples from Holes U1338A and U1338B (Table T19; Fig. F27). In the upper ~230 m CCSF-A, TOC content is generally high and variable, between 0.09 and 0.46 wt%, whereas below ~230 m CCSF-A TOC content is <0.09 wt%. Downhole TOC variability is most likely related to lithologic changes with higher TOC being found in the more biosiliceous intervals. To test this, TOC measurements were performed on two samples targeted specifically for their high biosiliceous content from 327.95 and 350.96 m CCSF-A. The results reveal high TOC values of 0.50 and 0.20 wt% and indicate that higher TOC values are found in the biosiliceous-rich intervals.

Top of page   |   Previous   |   Next