Previous   |    Next

doi:10.2204/iodp.proc.346.108.2015

Geochemistry

The geochemistry of sediment and water at Site U1427, within sight of Tottori Prefecture, was completely unknown before drilling. Consequently, after the flying pace at Sites U1422–U1426, the geochemistry group collectively decided that only a few squeezed interstitial water (IW-Sq) samples would be collected in Hole U1427A. By ~26 m CSF-A, interstitial water emerging from the squeezers was obviously yellow, alkalinity had reached 100 mM, and we knew Site U1427 was an extreme example of organic matter decomposition. Being humble guests in Tottori, we felt compelled to collect Rhizon water samples through the upper 10 m and to gather extra volumes of water for future shore-based studies.

Upon completion of Site U1427, the sulfate–methane transition (SMT) was found to lie at ~5 m CSF-A. However, this horizon was not captured in detail with routine sampling in Hole U1427A and was situated within a core gap in Hole U1427B. Sedimentary objectives and time also precluded sampling for the construction of headspace (HS) CH₄ and dissolved hydrogen sulfide (HS^–) profiles across the SMT and NH₄⁺ and metal profiles immediately below the seafloor. A return to Site U1427 several weeks later allowed us to fully document this location.

The seafloor at Site U1427 is at 330 m water depth (see “Operations”). However, the marginal sea has a very shallow thermocline. Thus, despite the shallow water depth, the position of the seafloor at Site U1427 appears to be bathed in Japan Sea Proper Water (JSPW) with a salinity of 34.06 (Senjyu, 1999).

Sample summary

After initial operations at Site U1427, the geochemistry group had collected and analyzed the following samples (Tables T11, T12, T13, T14, T15):

• 2 mudline (ML) samples, representing water poured from inside the liner of Cores 346-U1427A-1H and 346-U1427B-1H.

• 29 interstitial water samples from whole-round squeezing (IW-Sq) of discrete intervals in Hole U1427A.

• 19 interstitial water samples from Rhizons (IW-Rh) of discrete points in Hole U1427B.

• 29 sediment samples from interstitial water squeeze cakes; thus, the distance between samples varies significantly with depth.

• 74 headspace (HS) gas samples from the top of one section of almost every core.

• 6 Vacutainer (VAC) gas samples.

Upon returning to Site U1427 several weeks later, the geochemistry group collected an additional set of samples from Hole U1427C:

• 1 mudline sample, representing water from inside the liner of Core 346-U1427C-1H.

• 20 IW-Rh samples of discrete points in Hole U1427C.

• 12 HS gas samples from the top of each section of Cores 346-U1427C-1H and 2H.

Carbonate and organic carbon

The carbonate, carbon, and nitrogen profiles at Site U1427 (Fig. F27) are similar to those of Unit I at Site U1426 (see “Geochemistry” in the “Site U1426” chapter [Tada et al., 2015g]). Specific depths for subunits at Site U1427 have not yet been defined. For the purposes of this report, and as advised by the Sedimentology Laboratory, we divide the recovered sedimentary record into an upper unit (0–125.5 m CSF-A) and lower unit (125.5–540.30 m CSF-A).

Sediment at Site U1427 contains modest amounts of carbonate (Fig. F27) when compared to material recovered at other sites during Expedition 346. The average carbonate content across all samples analyzed at Site U1427 is 9.4 wt% (Table T11). Only Site U1426 had similarly high values. Moreover, carbonate contents vary significantly with depth. For the upper unit, carbonate contents vary between 2.0 and 25.1 wt%. The suite of samples in the upper 50 m of Hole U1427A shows up to 10 wt% differences in carbonate contents between consecutive samples spaced 4 m apart. For the lower unit, carbonate contents average 9.7 wt%, which is very similar to that for the upper unit. Samples with relatively high carbonate contents (>8 wt%) are consistent with intervals containing moderate to high abundances of calcareous microfossil remains, such as coccoliths and foraminifer tests (see “Biostratigraphy”). It is likely that higher resolution cycles of carbonate exist at Site U1427, although documenting this full variance will need to wait for shore-based analyses.

The total carbon (TC) content of sediment at Site U1427 ranges between 0.9 and 5.1 wt% (Fig. F27). This range represents the variance in mixtures of carbonate carbon (described above) and organic carbon (described below). For example, the high TC value of 5.1 wt% reflects the addition of 3.0 wt% C from carbonate (24 wt% CaCO₃) and 2.1 wt% C from organic matter (5.25 wt% CH₂O) (see “Geochemistry” in the “Methods” chapter [Tada et al., 2015b]). Of the sites drilled so far during Expedition 346, the relative abundance of carbonate carbon and organic carbon are the most balanced at Site U1427.

The total organic carbon (TOC) contents of samples from Site U1427 range between 0.5 and 2.4 wt% and average 1.2 wt% (Fig. F27). Although lower than observed at Site U1426, these values are still fairly high for marine sediment (Müller and Suess, 1979; Emerson and Hedges, 1988). The TOC contents appear to vary more within the upper unit. However, this variability may simply reflect the higher resolution of sampling over the upper 50 m. We could not address the origin and preservation state of the organic matter at Site U1427 because the Source Rock Analyzer (Rock-Eval) remained inoperable, even though continued efforts were made to fix the system while on location. However, we suspect that a significant portion of the organic carbon derives from the adjacent landmass. A more complete understanding of organic carbon, including variations over short time/depth increments and in the origin, will have to await postcruise investigations.

The total nitrogen (TN) contents of sediment at Site U1427 are relatively low. For the upper unit, TN has a mean value of 0.25 wt%, a maximum value of 0.33 wt% (in the topmost sample at 1.56 m CSF-A), and a minimum value of 0.16 wt% (at 2.60 and 7.75 m CSF-A). Compared to Site U1426, TN contents are slightly less at Site U1427. However, the difference is not greater than the error of this analysis (±0.07, 1σ, as determined by repeat analyses of an internal standard, n = 6). At both Sites U1426 and U1427, downhole trends in TN contents generally follow those of TOC. However, there is only a modest correlation (r² = 0.51) between the two parameters.

The C:N (atomic) ratio of organic matter is an important parameter for understanding provenance and degradation (Waples, 1985; Meyers, 1997). At Site U1427, this ratio is relatively low, with an average of 5.4. This ratio is partly due to the relatively low values of TC and TN and the limitations on such analyses aboard the R/V JOIDES Resolution (see “Geochemistry” in the “Methods” chapter [Tada et al., 2015b]). However, there also appears to be a true decrease in C:N with depth. The average C:N for the upper unit is 6.5, whereas that of the lower unit is 4.5.

Manganese and iron

Dissolved Mn and Fe profiles at Site U1427 (Table T12; Fig. F28) differ somewhat from those at other locations drilled during Expedition 346. As at other sites, some samples from shallow depths have higher concentrations in Mn, Fe, or both. Moreover, IW-Sq and IW-Rh samples from similar depths have different concentrations of Mn, Fe, or both. However, the absolute concentrations and depth intervals of elevated Mn and Fe concentrations are different at Site U1427.

Mn concentration is very low (0.1 µM) in the mudline sample, similar to observations at other sites. Dissolved Mn reaches a maximum of 5 µM at 0.50 m CSF-A (Fig. F29). Deeper than ~2 m CSF-A, Mn concentrations decrease below detection (0.1 µM) in the Rhizon samples. The squeezed samples are also low, remaining <1 µM for the remainder of samples. With either sampling technique, the main finding is that Mn concentrations are very low at Site U1427, especially compared to other sites drilled during Expedition 346.

Low dissolved Mn concentrations at Site U1427 are consistent with previous results and discussions of Mn cycling at other sites during Expedition 346. In shallowly buried sediment, microbes use Mn oxides during consumption of organic carbon, which releases dissolved Mn to interstitial water (e.g., Froelich et al., 1979). Deeper in the sedimentary columns, precipitation of Mn-bearing carbonate phases has been observed, which removes dissolved Mn (e.g., Middelburg et al., 1987). Site U1427 appears different from other sites because the depths of Mn oxide dissolution and Mn carbonate precipitation are much closer to the seafloor and because carbonate precipitation prevents dissolved Mn concentrations from rising to significant concentrations at depth.

Dissolved Fe concentration is below the detection limit (1 µM) in the mudline sample, similar to that observed at previously drilled sites during Expedition 346. For IW-Sq samples, Fe concentrations show high variability between 1 and 30 µM from the seafloor to ~400 m CSF-A (Table T12; Fig. F28). By contrast, Fe concentrations are below the detection limit for all IW-Rh samples (Fig. F29). We are uncertain how to explain the Fe data, especially given previous findings at Sites U1424 and U1425. We expected to find a small maximum in dissolved Fe using Rhizons across the upper 1 m. We may have missed this peak, despite the reasonably close spacing of samples. We anticipated low dissolved Fe concentrations in the IW-Sq samples, perhaps caused by fine-grained magnetite (see “Geochemistry” in the “Site U1425” chapter [Tada et al., 2015f]). However, we did not expect measurable Fe concentrations at 400 m below the seafloor.

Alkalinity, ammonium, and phosphate

The alkalinity, NH₄⁺, and PO₄^3– profiles at Site U1427 (Table T12; Fig. F30) are exceptional given their concentration changes over depth. Concentrations of all three are relatively low in the mudline sample, being 2.4 mM, 14 µM, and 2.7 µM, respectively. Below the seafloor, concentrations rise very quickly. At ~26 m CSF-A, alkalinity is 99.5 mM, NH₄⁺ is 11.9 mM, and PO₄^3– is 625 µM. The maximum values are 132.3 mM alkalinity at 352.3 m CSF-A, 39.6 mM NH₄⁺ at 446.5 m CSF-A, and 625 µM PO₄^3– at 26.5 m CSF-A.

Over the past four decades of scientific drilling, extreme profiles of alkalinity, NH₄⁺, and PO₄^3– have been generated at several locations. Examples of drill sites where alkalinity exceeds 100 mM in the upper 200 m below the seafloor include DSDP Site 496/568 (Middle America Trench, Legs 67/84 [Harrison et al., 1982; Hesse et al., 1985]) and ODP Sites 685/1230 (Peru margin, Legs 112/201 [Suess, von Huene, et al., 1988; D’Hondt, Jørgensen, Miller, et al., 2003]), 997 (Blake Ridge, Leg 164 [Paull, Matsumoto, Wallace, et al., 1996]), 1019 (California margin, Leg 167 [Lyle, Koizumi, Richter, et al., 1997]), 1084 (Namibia margin, Leg 175 [Wefer, Berger, Richter, et al., 1998; Murray et al., 1998]), and 1251 (Oregon margin, Leg 204 [Tréhu, Bohrmann, Rack, Torres, et al., 2003]). These sites span a wide range of latitude, longitude, bottom water chemistry, and water depth (Site U1427 = 321 m; Sites 685/1230 = 5086 m). There is, however, a common theme: at each of these sites, organic matter has accumulated along a continental margin with very high linear sedimentation rates (>30 cm/k.y.) over the last 1 m.y. or so.

Sediment buried beneath the seafloor at Site U1427 averages ~3.0 wt% CH₂O with a C:N ratio of 6.5 (above). More crucially, because of high terrigenous input from the adjacent continent (“Lithostratigraphy”), the mean sedimentation rate is ~40 cm/k.y. (“Biostratigraphy”). An enormous quantity of organic carbon and nitrogen are entering the sediment column at Site U1427 and fueling fermentation and methanogenesis. As discussed in previous chapters of this volume and elsewhere (e.g., Sayles and Manheim, 1975; Gieskes, 1975; Harrison et al., 1982; Murray et al., 1998), these processes release alkalinity (as HCO₃^–), NH₄⁺, and PO₄^3– to pore space at depth, but much of the PO₄^3– precipitates into phosphate-bearing minerals.

When shown over the entire depth of the cored site (~550 m CSF-A), both the alkalinity and NH₄⁺ profiles at Site U1427 appear concave downward between the seafloor and 100 m CSF-A (Fig. F30). This is true for all “high alkalinity” sites listed above. However, such broad-scale presentation overlooks an essential aspect of the geochemistry, as documented at Site U1427 (Fig. F31). High-resolution Rhizon sampling over the uppermost 10 m shows a significant inflection in the alkalinity profile at ~5 m CSF-A. Shallower than this inflection, alkalinity changes nearly linearly by ~9 mM/m; below the inflection, it changes nearly linearly by ~3 mM/m. This implies an approximate threefold increase in the net flux of alkalinity added to interstitial water of Site U1427 at ~5 m CSF-A. Moreover, the source of this excess alkalinity does not involve solid organic matter because organic matter contains nitrogen, and no inflection in NH₄⁺ is found at ~5 m CSF-A.

Therefore, we suggest that the very high alkalinity and NH₄⁺ concentrations at Site U1427 have a root cause similar to other high alkalinity sites along continental margins: they mostly reflect the decomposition of considerable organic matter over time, especially through fermentation and methanogenesis. This leads to some testable predictions. Previously drilled locations with very high alkalinity and NH₄⁺ concentrations have unusually high Br^– concentrations, yellow water, and considerable biogenic CH₄ (as cited previously). Note, however, that Br^– and color are rarely measured, and CH₄ concentrations are rarely determined at in situ pressure.

Bromide

Site U1427 is the first drilling location during Expedition 346 with a Br^– profile worthy of highlighting (Table T12; Fig. F32). The dissolved Br^– concentration is 0.83 mM in the mudline sample, which is the same as inferred for JSPW (Table T10 in the “Methods” chapter [Tada et al., 2015b]). Below the seafloor, Br^– systematically increases with a concave downward profile. Bromide concentrations are ~1.0 mM at 20 m CSF-A, ~2.0 mM at 400 m CSF-A, and 2.18 mM at 522 m CSF-A. Thus, the shape of the Br^– profile resembles those of alkalinity and especially NH₄⁺.

Similar to alkalinity and dissolved NH₄⁺, interstitial water has extreme Br^– concentrations at depth at Site U1427. Indeed, after a cursory shipboard review, afforded by the relatively luxurious pace of operations, after acknowledging that Br^– is not routinely measured on interstitial water and after excluding sites with evaporative brines at depth, we found only two sites with higher dissolved Br^– concentrations: Site 685/1230 on the Peru slope (2.49 mM at 491 meters below seafloor [mbsf]) (Martin et al., 1993) and Site 997 on Blake Ridge (3.3 mM at 500 mbsf) (Egeberg and Barth, 1998). As noted above, and similar to Site U1427, Sites 685/1230 and 997 have very high alkalinity and NH₄⁺ concentrations.

Bromide is usually considered a highly conservative ion in water chemistry, where concentration changes only with variations in salinity. The reason for exceptionally high dissolved Br^– levels in certain pore waters lies in an understanding of marine halogen cycling (Gieskes and Mahn, 2007). Organic matter landing on the seafloor incorporates Br from seawater. During burial of organic matter and its subsequent decomposition, this Br is released to interstitial water as Br^–. Bromide concentrations thus reflect the integrated accumulation of organic matter over long time frames and generally track NH₄⁺ profiles. For locations with high levels of fermentation and methanogenesis at depth, Br^– concentrations typically exceed 1.5 mM (Martin et al., 1993; Egeberg and Barth, 1998; Gieskes and Mahn, 2007). The very high Br^– concentrations at Site U1427 are consistent with other indicators for extreme organic diagenesis at this location. Moreover, the fact that Br^– and NH₄⁺ concentrations are still rising with depth at 550 m CSF-A (Fig. F32) indicates this diagenesis extends well beyond the upper 100 m of sediment, and most likely continues deeper than 550 m CSF-A.

Absorbance/Yellowness

Interstitial water samples at Site U1427 are visibly yellow. This phenomenon has been observed previously at several scientific drill sites, although only formally documented at a few locations (e.g., ODP Site 808, Nankai Trough [You et al., 1993]; Site 685/1230, Peru margin [D’Hondt, Jørgensen, Miller, et al., 2003]; and ODP Site 1244, Hydrate Ridge [Tréhu, Bohrmann, Rack, Torres, et al., 2003]).

A total of 51 interstitial water samples collected at Site U1427 were analyzed for their “yellowness” by measuring absorbance at three wavelengths: 375, 325, and 227 nm (Table T14). Examination of nine randomly selected samples showed that absorbance at 325 nm exceeded 1.0 for several samples. Consequently, 1 mL aliquots of all 51 samples, including the initial 9 analyzed, were placed into a vial and diluted with 2 mL of deionized water for analyses.

Absorbance measured for all three wavelengths shows obvious trends with depth (Fig. F33). The mudline sample has very low absorbance. Below the seafloor, absorbance rises rapidly. At the broad scale, absorbance reaches maxima for all three wavelengths between 100 and 300 m CSF-A and declines at deeper depths. However, a prominent peak in absorbance for 227 nm at ~6 m CSF-A is not seen at other wavelengths. Similar to observations at Sites U1425 and U1426 (see “Geochemistry” in the “Site U1425” chapter and “Geochemistry” in the “Site U1426” chapter [Tada et al., 2015f, 2015g]), the absorbance at longer wavelengths also decreases more rapidly with depth.

The absorbance of interstitial water samples from Site U1427 has been measured in batches with the JWBL standard (see “Geochemistry” in the “Methods” chapter [Tada et al., 2015b]). This is important because it lessens ambiguities in reported measurements between studies, which can arise from different equipment and configurations (e.g., cell length). Interstitial water at Sites 1230 and 1244 have also been determined for absorbance at 325 nm with the JWBL standard (D’Hondt, Jørgensen, Miller, et al., 2003; Tréhu, Bohrmann, Rack, Torres, et al., 2003). The average yellowness of these sites, both characterized by very high CH₄ concentrations, brackets that at Site U1427.

The yellowness of marine water has been ascribed to dissolved organic carbon in general (Briucaud et al., 1981) and fulvic acid more specifically (Krom and Sholkovitz, 1977). At Sites 808 and 1230, the absorbance of water at 325 nm correlates fairly well with independent measurements of dissolved organic carbon (You et al., 1993; Smith, 2005). Considering the degree of yellowness and that Site 808 also has high amounts of CH₄, considerable fermentation and methanogenesis may necessarily occur within the sediment column of Site U1427.

Volatile hydrocarbons

Similar to Site U1426, significant gas expansion occurred during recovery of sediment at Site U1427. Indeed, visible gas cracks and gas voids began appearing inside of core liners within 15 m CSF-A of the seafloor. In theory, these gas release features should first appear when gas concentrations at in situ pressure and temperature conditions surpass those of solubility for interstitial water at shipboard pressure and temperature conditions. At even relatively low in situ gas concentrations, recovery of sediment cores from depth results in gas escaping from interstitial water (Paull et al., 2000).

Methane is the dominant hydrocarbon gas (>99.9%) in HS samples from Site U1427 (Table T13), and the HS CH₄ profile (Fig. F34) appears to exhibit a pattern observed at many other locations (Paull et al., 2000). Methane concentrations in headspace are very low (<50 ppmv) in the first two samples from the upper 3.3 m CSF-A. CH₄ concentration in the next sample, at 12.8 m CSF-A, is 44,000 ppmv. Below this apparent maximum, CH₄ values rapidly decrease to 10,000 ppmv at ~32 m CSF-A and then gradually decrease further to the bottom of the cored hole.

Methane profiles such as found at Site U1427 are mostly an artifact of degassing (Paull et al., 2000). At sites with very high in situ CH₄ concentrations, headspace CH₄ values are low above a shallow SMT. Below the SMT, CH₄ concentrations rise abruptly with depth and shortly cross the CH₄ solubility curve at 1 atm pressure. Within several meters after this crossover depth, gas begins escaping from interstitial water. Once gas voids start occurring, headspace values decrease significantly such that they approximate those on the solubility curve at 1 atm. However, values on this curve decrease with sedimentary depth because core temperatures, even after recovery, are warmer, and the solubility of CH₄ decreases with increasing temperature (Rettich et al., 1981).

Headspace ethane concentrations range between 0 and 9 ppmv in Hole U1427A and generally increase with depth. As a consequence, the downhole CH₄ to C₂H₆ (C1:C2) ratio exceeds 1000 shallower than 260 m CSF-A but gradually decreases, reaching 250 at the bottom of the site. Similar to the CH₄ concentration profile, the C1:C2 profile partly reflects the effect of degassing (Paull et al., 2000) and the different gas solubility of CH₄ and C₂H₆ (Rettich et al., 1981). In general, the C1:C2 ratio increases with depth simply because of differential degassing (Paull et al., 2000). However, a thermogenic origin for some of the CH₄ and other gases at depth cannot be excluded. Small amounts of propane and propene (<26 ppmv) occur in the headspace samples.

Several gas samples were taken from gas voids (Table T15). For these VAC samples, CH₄ concentrations range from 800,000 to 900,000 ppmv, whereas C₂H₆ concentrations range between 50 and 80 ppmv. Importantly, higher molecular weight hydrocarbon gases are absent, presumably because they are diluted by the extremely high CH₄ concentrations. VAC gas samples have been shown to more accurately represent the relative abundance of gases at in situ conditions (Paull et al., 2000).

Overall, the gas concentration profiles at Site U1427 are very similar to those found at other scientific drill sites, notably the same ones with very high interstitial water alkalinity and NH₄⁺ concentrations, such as Sites 997, 1230, and 1251 (cited previously). The singular unusual aspect of the gas profiles at Site U1427 is the rapid increase in CH₄ between samples at 3.3 and 12.8 m CSF-A. We contend this is an artifact of sample resolution and that analyses of detailed headspace gas sampling would show a steep CH₄ concentration gradient rising between 5 and 12 m CSF-A.

All drilling locations discussed in the above paragraphs, with the exception of Site 1084, have gas hydrate within sediment at depth. However, at 330 m water depth, the gas hydrate stability zone is very thin at Site U1427, and in situ CH₄ concentrations probably do not cross the gas hydrate solubility curve.

Sulfate, sulfide, and barium

Almost universally, when examined in detail, drill sites with extreme alkalinity, NH₄⁺, and CH₄ concentrations at depth are characterized by a near-linear dissolved SO₄^2– profile and a maximum in dissolved HS^– in the upper 0–20 m of sediment. We now return to the prominent alkalinity inflection at ~5 m CSF-A and present the interstitial water SO₄^2– (Fig. F35) and HS^– (Fig. F31) profiles at Site U1427 (Table T12).

Sulfate concentration is 28.5 mM in the mudline sample, which is slightly higher than the 28.2 mM inferred for JSPW (Table T10 in the “Methods” chapter [Tada et al., 2015b]). Below the seafloor, dissolved SO₄^2– decreases to 0.0 mM at ~5 m CSF-A. This drop is nearly linear (r² = 0.998). Beneath 5 m CSF-A, concentrations remain at 0.0 mM. Sulfide concentration is below the detection limit (~5 µM) at the seafloor, but starting at ~0.4 m CSF-A, HS^– concentrations rapidly increase, reaching a maximum value of 4498 µM at ~5 m CSF-A. Deeper than ~5 m CSF-A, HS^– concentrations decrease to below the detection limit by ~17 m CSF-A.

The linearity of the SO₄^2– profile strongly suggests that net SO₄^2– removal from sediment at Site U1247 occurs over a thin horizon at 5 m CSF-A rather than over a significant thickness beneath the seafloor. The maximum in dissolved HS^– strongly indicates production of HS^– over the same horizon. The most likely explanation is that a significant upward flux of CH₄ reacts with downward diffusing SO₄^2– via anaerobic oxidation of methane (AOM) (Reeburgh, 1976; Borowski et al., 1996):

As clear from this equation and as discussed previously (see “Geochemistry” in the “Site U1422” chapter [Tada et al., 2015c]), removal of SO₄^2– across a thin horizon by AOM should generate significant HCO₃^– and HS^–, which must necessarily cause a sharp inflection in alkalinity. The total impact of AOM upon shallow carbon chemistry can be determined through flux calculations that include the loss of SO₄^2– and CH₄, the change in alkalinity, and authigenic carbonate precipitation (below). Such calculations have been made for piston cores northeast of Site U1427 and show that AOM consumes almost all SO₄^2– in shallow sediment (Snyder et al., 2007). We therefore maintain that methane generation and consumption most likely underlies much of the geochemistry at Site U1427.

A similar theme has emerged in our presentation of dissolved SO₄^2– and Ba profiles at sites within the marginal sea (see “Geochemistry” in the “Site U1422” through “Site U1426” chapters [Tada et al., 2015c, 2015d, 2015e, 2015f, 2015g]). Sulfate reduction through oxidation of solid organic carbon, AOM, or some combination of both decreases SO₄^2– concentrations to low values. Over time and with sediment burial, barite grains pass into the low-SO₄^2– fluids and dissolve (von Breymann et al., 1992; Dickens, 2001). Consequently, interstitial water Ba concentrations are high at depth. The SO₄^2– and Ba profiles at Site U1427 (Fig. F36) bring a twist to this discussion.

Dissolved Ba concentration is below the detection limit (1 µM) in the mudline sample. Below this depth, Ba concentrations remain very low until ~2 m CSF-A and then begin to rise. As found at previous sites, the increase in dissolved Ba begins above the SMT, and Ba concentrations are higher in IW-Sq samples compared to IW-Rh samples. These aspects have been discussed in previous site chapters. The new consideration is the magnitude of the rise in dissolved Ba concentrations. Within 5 m below the SMT, dissolved Ba concentrations are <80 µM, at least for IW-Rh samples. Indeed, interstitial water Ba values do not rise higher than 250 µM for any sample across the entire sediment column.

Dissolved Ba concentrations below the SMT are much lower at Site U1427 than at other sites drilled during Expedition 346 (von Breymann et al., 1992). We offer two possible explanations. First, the very high sedimentation rate at Site U1427 adds a significant downward burial flux of dissolved Ba below the SMT. This would counter the upward diffusion of dissolved Ba and diminish the expected recycling of Ba across the SMT (Dickens, 2001). Second, the amount of Ba entering the sediment column as barite may be much less at Site U1427 than at sites drilled in deeper water. Much of the barite incorporated into deep-sea sediment forms within the water column in association with sinking organic matter (Bishop, 1988). However, the actual process appears to occur over a depth range, such that barite contents increase significantly with greater water depth (Ganeshram et al., 2003; van Beek et al., 2007; Dehairs et al., 2008).

Calcium, magnesium, and strontium

Calcium concentration is 9.9 mM at the mudline and decreases to 3.6 mM at 4.3 m CSF-A, where there is an inflection point in the profile and the rate of decrease is reduced (Table T12; Fig. F37, F38). This lower rate of decrease is maintained until a minimum of 1.06 mM is reached at 31.75 m CSF-A, where at this deeper, second inflection point, concentrations begin generally increasing with depth to 5 mM at 540.3 m CSF-A. The increased sample resolution of the upper 10 m CSF-A shows a linear decreasing trend in values from 10 mM, with an inflection point deeper than 4 m CSF-A (Fig. F38).

At the mudline, strontium concentration is 88 µM and decreases fairly linearly to 32 µM at 31.75 m CSF-A (Fig. F37, F38). At deeper depths, concentrations generally begin to increase to a maximum of 138 µM at 540.3 m CSF-A.

Magnesium concentration is 50 mM at the seafloor and remains constant until 10 m CSF-A (Fig. F38), where a maximum of ~65 mM is observed between 40 and 70 m CSF-A. Deeper than 70 m CSF-A, concentrations begin to decline, reaching a minimum of 16.2 mM at 540.3 m CSF-A (Fig. F37).

The decline in calcium and strontium concentrations from 0 to 31.75 m CSF-A suggests that calcite formation is removing both of these elements from interstitial water (Snyder et al., 2007). The relatively constant magnesium concentrations over the same depth interval exclude the possibility of dolomite formation and lends support to the calcite hypothesis. Calcium and strontium both show similar downcore peaks and troughs in their concentration profiles; these elements were analyzed in different analytical runs with different dilutions and standards, indicating that these features are likely real and indicative of changes in authigenic carbonate phases.

Chlorinity and sodium

At Site U1427, Cl^– concentrations with depth yield a profile (Table T12; Fig. F39) somewhat similar to those at other locations drilled during Expedition 346. The mudline sample has a Cl^– concentration of 545 mM, which is the same as the 545 mM expected for JSPW (Table T10 in the “Methods” chapter [Tada et al., 2015b]). Immediately below the seafloor, Cl^– concentrations begin to decrease. The values show considerable scatter when examined in detail (Fig. F39), which probably reflects the relatively low precision of the shipboard automated Cl^– titration system. Nonetheless, Cl^– concentrations clearly decrease to a minimum of between 510 and 520 mM between ~10 and 40 m CSF-A. Deeper than 40 m CSF-A, Cl^– concentrations vary between ~520 and 540 mM.

The decrease in dissolved Cl^– within the upper 50 m of the sediment column is now a robust finding at every site drilled so far during Expedition 346. As discussed previously in this volume, less saline “deep water” probably existed within this basin during the recent past. The significance of the result at Site U1427 concerns the magnitude of the Cl^– minimum and water depth. This location appears to have the lowest Cl^– values, perhaps because it was at ~200 m water depth and within the thermocline during the LGM. A fresher mixed layer in the marginal sea during the LGM with a nominal salinity (S) of 31 would explain the Cl^– minima at Site U1427 and foraminifer assemblages and stable isotopes in piston cores (Oba et al., 1991). The Cl^– concentrations of deeper fluids at Site U1427 may represent the long-term average salinity of the thermocline in the marginal sea. Sodium concentrations range from 450 to 470 mM from the mudline to ~3 m CSF-A. From 3 to 60 m CSF-A, concentrations range from 430 to 450 mM. Deeper than 60 m CSF-A, concentrations increase gradually with depth to the bottom of the hole, ranging from 475 to 500 mM.

Potassium

Potassium concentration is 10.4 mM at the mudline and quickly increases to ~12 mM at 0.5 m CSF-A (Table T12; Fig. F40). Potassium increases to a maximum of 16 mM at 70 m CSF-A before decreasing slightly to between 13 and 14 mM for the remainder of the sampled depths.

Boron and lithium

The dissolved boron profile at Site U1427 (Table T12; Fig. F41) exhibits a different pattern than those found at previously drilled sites during Expedition 346. B concentration is 423 µM in the mudline sample, which is similar to the ~420 µM expected for JSPW (Table T10 in the “Methods” chapter [Tada et al., 2015b]). Concentrations then rapidly increase with depth to a distinct peak of ~760 µM at ~40 m CSF-A. This rise is consistent across both IW-Sq and IW-Rh samples, at least over the uppermost 10 m where overlap occurs, and has a concave downward shape. Deeper than 40 m CSF-A, B concentrations become less coherent and more scattered, varying between 430 and 710 µM.

At the coarse depth scale (Fig. F41), dissolved Li increases gradually downhole from a mudline value of 25 µM to a value of 38 µM at 70 m CSF-A. Concentrations then rise more rapidly, such that Li reaches ~680 µM at ~540 m CSF-A. This is the highest Li concentration measured in interstitial water thus far during Expedition 346.

However, the full Li profile at Site U1427 (Fig. F41) shows the power of supplementing conventional IW-Sq samples with detailed IW-Rh samples where the sediment characteristics allow (Fig. F42). Over the upper 15 m, Li concentrations decrease from 25 µM at the seafloor to ~15 µm at 5 m CSF-A. Li then rises back to ~27 µM at 18 m CSF-A. The gradual coarse-scale rise in Li with depth has a significant trough precisely at the SMT. Similar decreases in Li across the SMT have been found at other high-alkalinity sites (D’Hondt, Jørgensen, Miller, et al., 2003; Tréhu, Bohrmann, Rack, Torres, et al., 2003).

Silica

Comparing water samples analyzed with dual measurements, dissolved Si concentrations (by inductively coupled plasma–atomic emission spectroscopy) and dissolved H₄SiO₄ concentrations (by spectrophotometry) agree well (r² = 0.97). Typically, the Si values are slightly less than the H₄SiO₄ values, but this is not considered significant, and we discuss the H₄SiO₄ specifically.

The dissolved H₄SiO₄ profile at Site U1427 (Table T12; Fig. F41) resembles those at other sites drilled in the marginal basin (Murray et al., 1992), with one main exception. The mudline sample has a concentration of 145 µM. Below this bottom water value, there is a characteristic rapid increase in H₄SiO₄ concentrations to 652 µM at ~1.6 m CSF-A. However, at Site U1427, H₄SiO₄ concentrations change much less with depth compared to other locations. Dissolved concentration is only ~1250 µM at ~540 m CSF-A.

Similar to previously discussed sites, the H₄SiO₄ profile has a fairly straightforward explanation (e.g., Kastner et al., 1977). Within the uppermost meter or so below the seafloor, H₄SiO₄ concentrations increase significantly. This could be caused by temperature, as noted for K⁺ (Sayles and Manheim, 1975). More likely, it is caused by rapid dissolution of siliceous microfossils (Schink et al., 1974). The increase in dissolved H₄SiO₄ with depth relates to higher solubility at greater temperature and changes in mineralogy (“Lithostratigraphy”). In any case, the rise in H₄SiO₄ concentrations is slower at Site U1427 because the geothermal gradient is much more gradual than at previously drilled sites (“Downhole measurements”).

Rhizon commentary

A zone of exceptional interest—the SMT—was hypothesized to exist between 3.0 and 7.5 m CSF-A, based on alkalinity and SO₄^2– measurements of routine IW-Sq samples in Hole U1427A. Multiple IW-Rh samples were spread across the depth horizon in Hole U1427B. Other than the unfortunate coincidence of the SMT with a gap between Cores 346-U1427B-1H and 2H, analyses of the Rhizon samples captured the expected kink in alkalinity and inflection in SO₄^2– (Fig. F31).

Preliminary conclusions

In their now classic review of interstitial water chemistry, Sayles and Manheim (1975) stated: “The most complex pattern of interstitial water diagenesis is found in rapidly deposited, organic-rich, sediments containing a large proportion of terrigenous silts and clays in addition to variable amounts of biogenic carbonate. These sediments are typical of continental margin areas …” Site U1427 exemplifies this type of location. Following early literature (Gieskes, 1975; Sayles and Manheim, 1975), the geochemistry at Site U1427 is explained by the rapid burial of organic carbon driving a series of microbial reactions, including oxic respiration, metal oxide reduction, and sulfate reduction (fermentation and methanogenesis).

Here, we diverge from most previous work to make two summary comments. First, at the broad scale, the overall pattern of diagenesis and interstitial water chemistry at Site U1427 is remarkably predictable. Second, this chemistry cannot be understood or appreciated without a return flux of CH₄ that drives AOM at an SMT. We suggest that basic aspects of geochemistry at Site U1427 are representative of many continental margins.

Top of page   |    Previous   |    Next