Previous   |    Next

doi:10.2204/iodp.proc.346.105.2015

Geochemistry

Site U1424 redrilled Site 794 (Shipboard Scientific Party, 1990). Consequently, a series of geochemistry profiles for solids, interstitial water, and volatile hydrocarbons already existed for this location (Shipboard Scientific Party, 1990; Brumsack and Zuleger, 1992; Brumsack et al., 1992; Murray et al., 1992). These profiles showed that, similar to other sites in this marginal sea, several key processes impact the geochemistry (Shipboard Scientific Party, 1990; Murray et al., 1992). In shallowly buried sediment, the modest degradation of organic carbon drives a series of microbially mediated reactions. Deeper in the sedimentary column, opal recrystallization and basalt reactions are important.

A short transit time and rapid core recovery kept geochemistry operations at Site U1424 on a fast pace. Some instrument maintenance issues also temporarily impeded progress. Given these issues and the existing knowledge, sampling for geochemistry at Site U1424 targeted intervals of maximum scientific interest. To complement data from Site 794 and to provide a backbone for new analyses, one whole-round sample for squeezing (IW-Sq) and one headspace (HS) sample for gas were collected from each full APC core (or from alternate half APC cores) in Hole U1424A. These were at the base of Section 1 and top of Section 2, respectively, in the relevant cores, unless an issue regarding sampling was identified on the catwalk. To supplement conventional water samples, we also took numerous Rhizon (IW-Rh) samples until they could no longer be inserted into the compacted core.

The Rhizon program at Site U1424 was designed to serve four goals. One suite of Rhizons was placed close to samples for squeezing to see whether the chemistry of IW-Rh samples matched that of IW-Sq samples for several dissolved species measured on board the ship. Although standard Rhizon sampling collects only 10 mL or less of water, this is a sufficient volume to analyze for Cl^– (titration), SO₄^2– and Br^– (ion chromatograph), NH₄⁺ (spectrophotometry), and metals (inductively coupled plasma–atomic emission spectroscopy [ICP-AES]). The labor involved in collecting Rhizon samples is much less than that for squeezed samples, even though the postsampling splitting and analyses of samples requires a similar amount of time per sample.

Two more sets of Rhizons were inserted primarily for shore-based analyses. One was to obtain interstitial water for the generation of high-resolution Cl^– and δ¹⁸O profiles. The other was to obtain interstitial water for a study of Fe speciation.

A fourth suite of Rhizons was inserted across the upper 10 m of Hole U1424A to determine where oxidation of ammonium and reduction of metal oxides occur at this site. Numerous box cores and piston cores have been examined in an effort to establish the depth and the importance of various microbial reactions involving N, Fe, and Mn (e.g., Froelich et al., 1979; Bender et al., 1989). This includes anaerobic oxidation of ammonium (“anammox”), which is an important reaction in the global N cycle (Dalsgaard et al., 2005). However, linkages between reactions in the uppermost few meters and those deeper in the sediment column have not been bridged fully, as exemplified by data at Site 794. Like many locations drilled over the last few decades, Site 794 has a dissolved NH₄⁺ profile that appears to reach zero concentration at the seafloor. However, at this site and at almost all drill sites, the first pore water sample is typically a whole-round sample from squeezing taken several meters below the seafloor. Therefore, the precise depth at which NH₄⁺ goes to zero concentration is unknown. A good understanding of microbial reactions in shallowly buried sediment may also help to unravel issues with sediment color banding and magnetic minerals discussed at Sites U1422 and U1423.

At Site U1424, insertion of Rhizons into sediment became very difficult at ~38 m CSF-A. However, reconstructions of past bottom water salinity from interstitial water profiles usually require samples extending to ~50 m or so below the seafloor (e.g., Adkins et al., 2002). Additional samples for squeezing were thus taken from Hole U1424C. A combination of squeezing and sucking seems optimal for achieving desired liquid.

Sample summary

During operations at Site U1424, the geochemistry group collected and analyzed a range of samples. These included (Tables T10, T11, T12) the following:

• 1 “mudline” (ML) sample.

• 40 interstitial water samples from whole-round squeezing (IW-Sq). Of these, 23 were taken from Hole U1424C, primarily to collect water for shore-based analyses of δ¹⁸O.

• 86 interstitial water samples from Rhizons (IW-Rh).

• 29 sediment samples, mostly from interstitial water squeeze cakes but some from physical property samples, to evaluate differences between dark and light sediment horizons.

• 16 headspace (HS) gas samples.

We assume that in situ interstitial water over short depth increments has similar chemistry (although the Fe data at Site U1424, below, may challenge this basic tenet). This is generally true for IW-Sq and IW-Rh samples from similar depth intervals. However, for some species over some depth intervals, the different sampling approaches lead to distinct chemistry. This is a recurring theme, one that we aim to understand with future work during Expedition 346 and on shore. At this juncture, it is important to note four items. First, all IW-Sq samples collected from Hole U1424A were examined and split into aliquots before any IW-Rh samples from this hole. Second, most IW-Sq samples were analyzed in batches distinct from most IW-Rh samples. Third, the number of interstitial water samples over a given depth increment is often many fewer than IW-Rh samples. Fourth, deionized water and artificial seawater, when either slowly pushed through a squeezer or slowly pulled through a Rhizon, show no anomalous concentrations of any species analyzed in the shipboard laboratory. In summary, it is difficult to ascertain the root cause of apparent deviations in the concentrations of several species because, once removed from sediment, the IW-Sq and IW-Rh samples were treated differently. However, differences are not caused by contamination.

Carbonate and organic carbon

Similar to previous sites drilled in the Japan Basin during Leg 127 (Tamaki, Pisciotto, Allan, et al., 1990) and Expedition 346 (Sites U1422 and U1423), sediment recovered at Site U1424 has low amounts of carbonate (Fig. F25). Most samples have <1.0 wt%. However, some samples have between 1 and 6 wt%, and two samples, at 4.94 and 141.25 m CSF-A, have ~32 and 44 wt%, respectively. The deeper of these samples is associated with a layer rich in foraminifers, whereas the shallower sample is associated with a layer rich in diatoms and radiolarians but where carbonate has accumulated inside the radiolarian shells (see “Lithostratigraphy” and “Biostratigraphy”). In general, the variance in carbonate content with respect to depth is similar to that obtained at Site 794 (Shipboard Scientific Party, 1990).

Regarding total organic carbon (TOC), Subunit IA (0–42.9 m CSF-A, see “Lithostratigraphy”) has the highest values. TOC contents range from 0.3 to 1.4 wt% on material from 5 cm thick intervals used for interstitial water squeezing and from 0.1 to 3.9 wt% across 2 cm thick intervals chosen to represent light (low TOC) and dark (high TOC) horizons (see “Physical properties”). As established by previous work (Föllmi et al., 1992; Tada et al., 1992), TOC varies significantly at the centimeter- to meter-scale in sediment from the Japan Basin, particularly in younger strata correlative with Subunit IA. The TOC contents of samples from Subunit IB (42.9–64 m CSF-A) and Unit II (~64–158.8 m CSF-A) are generally lower (mean = 0.38) and less variable (standard deviation = 0.28) than those of overlying sediment, although only a small number of samples were examined. A similar result was found at ODP Site 795 (Tamaki, Pisciotto, Allan, et al., 1990).

Sediment total nitrogen (TN) contents at Site U1424 vary much less than TOC contents. The mean value of TN is 0.23 wt%. However, the maximum TN content (0.47 wt% at ~3.41 m CSF-A) is only double this amount, whereas the minimum TN content (0.13 wt% at 122.25 and 141.25 m CSF-A) is less than half this amount. Consequently, the C/N ratio of organic carbon appears to change significantly, from <0.6 to >10 on a molar basis. The lower C/N values are probably incorrect because of errors in total carbon (TC) and TN measurements at low concentrations. A similar observation was made previously for sediment in the Japan Basin and may relate to temporal differences in the type of organic carbon, the preservation of organic carbon, or both (Shipboard Scientific Party, 1990). Nitrogen may also adsorb onto clay (Shipboard Scientific Party, 1990).

In summary, and despite only analyzing a small total number of samples, we were able to document the carbon chemistry of sediment sampled at Site U1424 and demonstrate its similarity to Site 794 (Shipboard Scientific Party, 1990). Total carbon content is mainly controlled by organic carbon, perhaps from mixed sources, and high amounts of carbonate occur in sporadic samples.

Manganese and iron

The dissolved Mn and Fe profiles (Table T11; Fig. F26) at Site U1424 show peaks in the upper 20 m below the seafloor and some variability with greater depth. Interstitial water obtained from squeezers and Rhizons at similar depth generally have similar Mn and Fe concentrations, although at a fine scale, some discrepancies may exist. For example, dissolved Mn concentrations at similar depth appear to be consistently higher in IW-Rh samples relative to IW-Sq samples. However, most IW-Rh samples were analyzed for Mn concentrations after different storage and in a different batch than the IW-Sq samples. In any case, the Rhizon samples lead to dissolved Mn and Fe profiles with a depth resolution similar to those in many piston core studies.

The Mn profile at Site U1424 is relatively smooth and fairly similar in shape to those at Sites U1422 and U1423. Mn concentrations continuously increase from less than the detection limit (~0.5 µM) at the seafloor to ~113 µM at 8.5 m CSF-A. Below this depth, dissolved Mn decreases rapidly to 70 µM at 17.8 m CSF-A and then gradually decreases to ~40 µM at 60 m CSF-A. Manganese concentrations are relatively constant (50–70 µM) from 57 to 160 m CSF-A.

The liberation of dissolved Mn at shallow burial depths (~11 m CSF-A) is captured well by Rhizon sampling (Fig. F27). Presumably, this increase in interstitial water Mn concentration derives from Mn oxide reduction of organic matter (Froelich et al., 1979). As at Sites U1422 and U1423, the subsequent decrease in dissolved Mn below 11 m CSF-A and increase in dissolved Mn below 60 m CSF-A are likely caused by the formation and dissolution of a Mn-carbonate phase, such as rhodochrosite (Middelburg et al., 1987). The argument for such a process at Site U1424 and other sites in the Japan Basin is compelling, especially when compared with alkalinity profiles. For example, between 20 and 60 m CSF-A, dissolved Mn concentrations never approach the detection limit as they do at Sites U1423 and U1422. At deeper depths, Mn concentrations decrease to ~50–70 µM at Site U1424 with an alkalinity of ~10–12 mM (Fig. F28), but values were ~20–30 µM at Site U1422 with an alkalinity of ~15–20 mM. Thus, below the shallow subseafloor peak, the dissolved Mn profile appears coupled to alkalinity, which mostly represents HCO₃^– at depth.

Fe concentration is below detection limit (~0.9 µM) at the seafloor but reaches a maximum of ~47 µM at 1.8 m CSF-A. Values decline to 3 µM at 6.6 m CSF-A and then rise to a second relative maximum of 23 µM at 10.4 m CSF-A. Dissolved Fe concentrations return to below detection limit at 17 m CSF-A and remain at low concentrations for the rest of the hole, except for small increases centered at ~50, 80, and 120 m CSF-A.

The double spikes in dissolved Fe at shallower depths (<20 m CSF-A) are captured well by the high-resolution Rhizon sampling (Fig. F27), and they may have a causal relationship to shallow Fe peaks identified at Sites U1422 and U1423 (see “Geochemistry” in the “Site U1422” chapter and “Geochemistry” in the “Site U1423” chapter [Tada et al., 2015c, 2015d]). However, the detail of the interstitial water Fe profile at Site U1424 also allows comparisons with sedimentary profiles of parameters related to Fe content and mineralogy. In particular, highs and lows in the dissolved Fe profile clearly correspond to highs and lows in magnetic susceptibility (Fig. F27).

This intriguing result suggests at least two different explanations for the interstitial water Fe profile. First, peaks and troughs in dissolved Fe reflect the dissolution and precipitation of multiple Fe-bearing phases, presumably caused by microbial consumption of organic material. However, the shallower Fe maximum is even shallower than the Mn maximum. The canonical sequence of reactions that oxidize organic matter should reduce Mn oxides at shallower depths than Fe oxides (e.g., Froelich et al., 1979). The Fe diffusion gradients with respect to depth would also be very large. Alternatively, the measured “dissolved Fe” represents the amount of very small magnetite particles that can be passed from a given sample through either the pores of Rhizons (150–200 nm) or the pores of filters (450 nm) attached to the squeezers. Magnetotactic bacteria produce single-domain (35–120 nm) magnetite in shallow-marine sediment (e.g., Kopp and Kirschvink, 2008), and it is possible that we are extracting this component during the collection of interstitial water. Future work on the magnetite within the sediment or Rhizons with different pore sizes may resolve this issue.

The complete removal of Fe from ~18 to 38 m CSF-A suggests the formation of an Fe-bearing mineral. Iron sulfides may precipitate in this depth interval as HS^– is released to interstitial water during sulfate reduction. As recognized at Sites U1422 and U1423 and further complicated by the relationship between dissolved Fe and magnetic susceptibility, the Fe story within sediment of the Japan Basin remains complex. Several Fe-bearing solids, potentially including oxides, sulfides, carbonates, and clay, can precipitate and dissolve at depth, which would influence the Fe profile. Furthermore, we are now uncertain whether we are truly measuring dissolved Fe. Shore-based studies focused on authigenic minerals, including the possible presence of bacterial magnetite, are required to fully understand variations in the abundance of dissolved Fe and sedimentary Fe with respect to depth.

Alkalinity, ammonium, and phosphate

Alkalinity, ammonium, and phosphate (Table T11; Fig. F28) profiles at Site U1424 show trends broadly similar to those observed at Sites U1422 and U1423, but with lower concentrations for all three species. For example, the alkalinity maximum at Site U1422 is 22 mM at 84 m CSF-A.

Our interpretations of the alkalinity, NH₄⁺, and PO₄^3– profiles at Site U1424 are the same as for previous locations. Microbial decomposition of solid organic matter releases HCO₃^–, NH₄⁺, and PO₄^3– to interstitial water, whereas precipitation of authigenic minerals removes some of the HCO₃^– and PO₄^3– at depth. The lower concentrations of alkalinity, NH₄⁺, and PO₄^3– at Site U1424 relative to Sites U1422 and U1423, therefore, indicate less organic decomposition over time. The general lack of CH₄ at Site U1424 (below) further supports this interpretation. In terms of paleoceanographic objectives, this may suggest a smaller flux of organic carbon to the seafloor at Site U1424 during the late Quaternary relative to the other locations.

As for Mn and Fe, the alkalinity, NH₄⁺, and PO₄^3– profiles demonstrate the ability of Rhizons to enhance the depth resolution of interstitial water profiles (Figs. F28, F29). Alkalinity and NH₄⁺ concentrations determined on samples from Rhizons and squeezers are similar at Site U1424. Rhizon and squeeze samples also reveal the same PO₄^3– trend, although PO₄^3– concentrations are noticeably higher in IW-Rh samples relative to IW-Sq samples at concentrations higher than ~20 µM.

The high-resolution NH₄⁺ profile in shallow sediment (Fig. F29) is of particular interest. The nine Rhizons taken within 2 m of the mudline provide an unusual look at processes occurring between the seafloor and deeper sediment. Dissolved Mn and alkalinity both increase across this depth interval (Fig. F27). By contrast, the NH₄⁺ profile has a kink at 40 cm. Shallower than this depth, concentrations are ~10 µM and essentially those of bottom water, as determined by measuring the mudline sample. Below this depth, concentrations increase rapidly. The Mn and alkalinity profiles clearly show that core disturbance and seawater mixing do not cause the inflection in the NH₄⁺ profile. Instead, the kink in the NH₄⁺ concentration gradient, which happens to mark the depth separating overlying red sediment from underlying green sediment (see “Lithostratigraphy”), likely indicates a change in NH₄⁺ flux. Two processes could induce such a change. One explanation would be NO₃^– diffusing downhole from the seafloor to 40 cm, where it begins oxidizing organic matter and producing NH₄⁺ (e.g., Froelich et al., 1979). Alternatively, NH₄⁺ generated at depth from the degradation of organic material by other electron acceptors, such as dissolved SO₄^2–, diffuses upward and reacts. The postulated reaction could involve O₂ (nitrification) or NO₂^– (anammox). Onshore flux calculations regarding pore water at Site U1424 should reveal more about such processes.

Volatile hydrocarbons

Methane is the only hydrocarbon gas in the headspace samples at Site U1424. No ethane or heavier hydrocarbons were detected. Compared to CH₄ concentrations at Sites U1422 and U1423, values at Site U1424 are extremely low, consistently less than ~5 ppmv through the bottom of the hole (Fig. F30). Nonetheless, there is a CH₄ profile at Site U1424 (Fig. F31). The CH₄ concentration is zero in the shallowest sample at 1.5 m CSF-A. Deeper, values increase to a maximum of ~5 ppmv at 46 m CSF-A and then decrease to 0 ppmv at 131 m CSF-A (Table T12).

With such low CH₄ concentrations, a prominent sulfate–methane transition (SMT) should not exist at Site U1424. Instead, low CH₄ concentrations might coincide with significant concentrations (>10 mM) of dissolved SO₄^2– (D’Hondt et al., 2004). Indeed, this is the case (Fig. F31).

Sulfate and barium

Sulfate concentrations demonstrate a classic concave downward profile (Table T11; Fig. F31) indicative of the continuous use of SO₄^2– to oxidize organic material through the entire sediment column that was sampled. Unlike as at Sites U1422 and U1423, SO₄^2– does not reach zero concentration. The general absence of CH₄ at Site U1424 further supports SO₄^2– being the final electron acceptor within the upper 160 m of sediment at this location. This conclusion was also reached at Site 794, which has a similar SO₄^2– profile.

The presence of SO₄^2– throughout the entire sampled sequence prohibits the dissolution of barite. Consequently, Ba concentrations in interstitial water remain low throughout the sediment at Site U1424 (Fig. F32). This profile is contrary to Sites U1422 and U1423, where Ba concentrations increase dramatically at depths where barite passes below the SMT.

By focusing on small-scale variations in dissolved Ba, a disparity between IW-Sq and IW-Rh samples becomes apparent. Ba concentrations measured on the Rhizon samples remain low and essentially equivalent to those of bottom water (<2.5 µM) for the upper 40 m below the seafloor. By contrast, Ba concentrations measured on the squeezed samples increase to 10–20 µM immediately below the seafloor and remain at this level throughout the upper 40 m. Unlike for some other species, the discrepancy is consistent across numerous samples, independent of other possible factors, such as batch analysis or processing time. Either Ba is being removed from water during sampling by Rhizons, Ba is being added to water during sampling by squeezing, or both. Significant dilution of Rhizon water samples with deionized water or seawater is not a valid explanation because several other species examined have very similar dissolved concentrations across IW-Sq and IW-Rh samples. It is possible that small amounts of barite precipitate during Rhizon sampling, as this would not impact concentrations of other species significantly. Alternatively, and somewhat analogous to one explanation for the dissolved Fe data, very fine grains of barite may enter the interstitial water obtained through squeezing. We note that barite grains <0.45 µm in diameter, the pore size of the filter, can be found in marine sediment (Griffith and Paytan, 2012).

Calcium, magnesium, and strontium

The Ca, Mg, and Sr profiles at Site U1424 (Table T11; Fig. F33) are similar to those for the upper 160 m at Site 794 (Shipboard Scientific Party, 1990). Over the depth drilled, dissolved Ca and Sr increase slightly, whereas dissolved Mg decreases slightly. These profiles exhibit minimal change in concentration gradients with respect to depth, being primarily dominated by alteration of ash and basalt below the base of drilling, as well as diffusion (Murray et al., 1992).

The Ca profile shows a slight increase in concentration downhole throughout the sediment column (Fig. F33), suggesting a minimal amount of net calcium carbonate formation or dissolution. As noted above, the relatively low level of organic matter decomposition at this location maintains alkalinity <14 mM throughout the drilled sediment column (Fig. F28). Apparently, this concentration is insufficient for precipitation of calcium carbonate phases. However, it is worth noting that the solubility of calcite is ~2 orders of magnitude greater than that of rhodochrosite. The latter phase may very well exist in sediment at Site U1424.

The Sr profile is problematic because the IW-Rh samples show a high degree of scatter (Fig. F33). We presently lack an explanation for these data but can exclude addition of Sr as water passes through the Rhizons.

Chlorinity and sodium

Cl^– and Na concentrations of the mudline sample (543 and 571 mM, respectively) are similar to those of inferred present-day Japan Sea Proper Water (JSPW) (545 and 468 mM) (Table T10 in the “Methods” chapter [Tada et al., 2015b]). Below the seafloor, from 5 to 60 m CSF-A, Cl^– concentrations are significantly lower than present-day bottom water values (Table T11; Fig. F34). This occurs at all sites drilled so far during Expedition 346 and is both interesting and important. As discussed in “Geochemistry” in the “Site U1423” chapter (Tada et al., 2015d), we suggest the interval with low Cl^– concentrations represents a nonsteady-state transient because the deep water of the marginal sea was significantly fresher during the Last Glacial Maximum. At Sites U1422 and U1423, Na concentrations were distinctly lower than bottom water values when Cl concentrations were low at shallow subseafloor depths. At Site U1424, the Na profile shows considerable scatter and lacks evidence for fresher interstitial water (Fig. F34).

Potassium

The first-order trend of the dissolved K profile (Table T11; Fig. F34) is an overall decrease with depth. This decrease in K concentrations continues to deeper depths at Site 794, reaching ~1 mM at 540 meters below seafloor (mbsf) and the contact with basalt (Shipboard Scientific Party, 1990). The profile probably results from the alteration of numerous ash layers within the sediment column or basalt (Murray et al., 1992), as both can incorporate potassium.

K concentration of the mudline sample is 10.3 mM, which is very close to the 10.2 mM predicted for JSPW (Table T10 in the “Methods” chapter [Tada et al., 2015b]). Therefore, a curious feature of the K profile is enrichment in shallow sediment. Concentrations increase from the above bottom water value to 12–13 mM just below the seafloor. This jump in K concentration also occurs at Sites U1422 and U1423, although it is captured in more detail at Site U1424. The increase in K may relate to the dissolution of a K-bearing phase or formation of an authigenic mineral that can exchange K⁺ with other cations such as NH₄⁺, Li⁺, or Na⁺.

Lithium and boron

Lithium increases from ~26 µM in the mudline sample to 68 µM in the deepest sample with an inflection at ~20 m CSF-A (Table T11; Fig. F35). Analyses of interstitial water samples from Site 794 show that this trend continues until ~350 m CSF-A. At this depth, which corresponds to the opal-A/opal-CT transition, Li concentrations decrease (Murray et al., 1992). Over this greater depth scale, the Li profile mimics the Si profile. This was originally interpreted as reflecting Li release by dissolution of diatoms and subsequent Li removal by alteration of ash, basalt, or authigenic clay formation (Shipboard Scientific Party, 1990; Murray et al., 1992). The higher resolution sampling across the upper 160 m shows that in the shallow sediment column Li profile is concave upward, whereas the Si profile is concave downward. Trends and concentrations of Li from IW-Sq and IW-Rh samples agree well.

The B profile shows quite a bit of scatter in both IW-Sq and IW-Rh samples until 65 m CSF-A. Deeper than this depth, B concentrations become fairly constant at ~400 µM.

Silica

Overall, dissolved Si exhibits a concave downward profile (Table T11; Fig. F35), presumably explained by biogenic Si dissolution throughout most of the sampled sediment column. Because the solubility of opal increases with higher temperature and because the mass of silica in sediment is much greater than that in surrounding pore water, dissolved Si concentrations should increase with depth. Interestingly, however, the dissolved Si profile suggests the situation is more complicated at Site U1424.

The IW-Rh samples show anomalies over the upper 20 m below the seafloor, where Si concentrations vary significantly over short depth. These lows and highs in dissolved Si may indicate zones of biogenic silica dissolution or precipitation. More obvious is a ~20 m thick zone of anomalously low Si concentrations centered around ~56 m CSF-A, which is apparent in squeezed samples from both Holes U1424A and U1424C. Silicon may be entering an authigenic phase at this horizon, although a Si sink is not identified. We note the depth interval corresponds to an interval of slower sedimentation rates (see “Stratigraphic correlation and sedimentation rates”).

Additional Rhizon commentary

As noted in previous paragraphs and shown in various figures, the water samples collected from Rhizons have fairly similar chemistry to water samples collected from squeezers. However, there appear to be subtle differences, although the cause of such variance remains uncertain with the samples, processing, and analyses at Site U1424. A suite of closely spaced IW-Sq and IW-Rh samples, analyzed for range of dissolved species, should prove illuminating.

The geochemistry program at Site U1424 uncovered two issues regarding Rhizon samples not acknowledged in previous chapters or elsewhere. The total number of Rhizons collected over several hours in Hole U1424A (87) overwhelmed the geochemistry laboratory. The current Laboratory Information Management System (LIMS) data system for sample documentation was not built with Rhizon sampling in mind. Basically, the ability to collect very large numbers of samples needs to be tempered with the reality of labeling and processing such samples. However, we predict that Rhizon sampling will become more prevalent in shipboard geochemical studies, and the added benefits they bring to expeditions far outweigh these temporary logistical issues.

We also experienced difficulties inserting Rhizons into the sediment of Core 346-U1424A-5H and deeper. A similar depth limitation to Rhizon sampling was encountered during IODP Expedition 302 (Dickens et al., 2007) and may be related to compaction of clay-rich sediment.

Preliminary conclusions

The geochemical analyses performed on samples from Site 794 provided a framework for Expedition 346 geochemists to build upon. We chose to supplement the previous work, mostly with higher resolution sampling focused on shallower depths, where degradation of organic material influences interstitial water chemistry prominently. This sampling included extensive use of Rhizons for the dual purposes of making comparisons of interstitial water chemistry between sampling techniques and of documenting thin sediment horizons marked by changes in concentration gradients of dissolved species.

Where comparable in terms of depth and analyses, the geochemistry results at Site U1424 are similar to those at Site 794. At this location, organic carbon and biogenic silica enter the sediment column and drive a series of reactions, some involving other sedimentary components. Alteration of underlying basalt also impacts the interstitial water profile. The primary, large-scale difference in geochemistry between Site U1424 and other drill sites within the Japan Basin is that the long-term accumulation of organic matter is significantly lower.

The Rhizon sampling program added a new dimension to geochemistry at this location and may at future drill sites in general. The dissolved Fe profile shows major peaks that clearly correspond in depth to sediment intervals with relatively high magnetic susceptibility. However, we remain uncertain whether Fe, before the addition of nitric acid and ICP-AES analysis, was dissolved in water or instead occurred as very small particles of magnetite. The dissolved NH₄⁺ profile drops to zero concentration at ~0.40 m CSF-A and where the color of sediment changes from red to green. These changes in the NH₄⁺ gradient may be controlled by nitrification or anammox but ultimately link the shallow nitrogen cycle discussed in piston core studies to the deep nitrogen pool documented in drill cores.

Top of page   |    Previous   |    Next