Eighty-five interstitial water samples were squeezed from selected 10- to 50-cm-long whole-round samples for chemical and isotopic analyses at Site 1173. Sample depths ranged from 1.4 to 687.5 mbsf. In Cores 190-1173A-1H and 2H, samples were collected from every section; in Core 3H, samples were collected from three sections; in Cores 4H and 5H, two samples were collected per core; and from Cores 6H to 73X, one sample was collected per core (except for Cores 62X and 69X, which were not sampled). Elemental concentrations are reported in Table T15 and plotted as a function of depth in Figure F26. Bottom water was sampled with the WSTP. All samples were analyzed for concentrations of 10 major and minor dissolved anions and cations that sensitively reflect microbially mediated or inorganic water-rock (sediment and oceanic basement) reactions. The former include alkalinity, sulfate, ammonium, and phosphate, and the latter are Cl, Ca, Mg, Na, K, and Si. Salinity and pH were also measured. The microbially mediated reactions are most intense in the top 150 m of the sediment section, where organic carbon is most abundant (0.85-0.27 wt%) (see "Organic Geochemistry") and porosities are highest (60%-70%) (see "Physical Properties"). Each of the dissolved species controlled by inorganic water-rock reactions shows the most dramatic changes in concentration values between ~345 mbsf, which corresponds to the boundary between lithostratigraphic Unit II (upper Shikoku Basin facies) and Unit III (lower Shikoku Basin facies), and ~390 mbsf, which corresponds to the décollement-equivalent horizon (~390-420 mbsf), based on preliminary interpretations. A distinct minimum in alkalinity is also observed at the latter depth.
Cl concentrations were all determined in duplicate, resulting in a relative analytical uncertainty of 0.1%. Concentrations increase from near bottom-water concentrations of 555 to 572 mM (a 3.1% increase) at a depth of 80 mbsf. This trend is consistent with the diffusion of lower chlorinity interglacial water into the sediments. Cl concentrations then smoothly decrease to 563 mM at a depth of 337 mbsf, where the gradient changes sharply. Between 337 and 620 mbsf, concentrations drop to 496 mM, corresponding to ~9% dilution relative to seawater Cl. The gradient between these depths is not smooth. There are significant minima centered at ~377 and 499 mbsf. Below 620 mbsf, concentrations monotonically increase to 518 mM. The source of the impressive broad (~350 m) low-Cl zone in lithostratigraphic Unit III cannot be resolved aboard ship. Distinguishing between an in situ source from hydrous mineral dehydration reactions or fluid transported from a deeper-seated source situated arcward will be possible with shore-based mineralogical, chemical, and isotopic analyses of the sediments and interstitial waters. It is interesting to note that the low-Cl zone is found in the same lithostratigraphic unit as at ODP Site 808 (Leg 131), but the dilution at Site 808 is larger, ~20%.
Na concentrations increase from a near-seawater value to a broad maximum, an 8% increase, between ~100 and 300 mbsf. This increase is due to a combination of processes: diffusion of low salinity interglacial water, ion exchange driven by ammonium production, and ash alteration. Below the Unit II/III boundary, concentrations generally decrease toward basement with a concentration of 355 mM at 680 mbsf. The decrease in Na in Unit III is charge balanced by an increase in Ca. This Ca for Na exchange is consistent with albitization of the basement. It is also interesting to note that similar to Cl, there is a local minimum just above the approximate stratigraphic level of the décollement, but that the molar decrease in Cl is approximately twice that of Na, indicating that the decrease is not simply due to dilution but also is the result of fluid/rock reaction.
K concentrations are ~10% higher than bottom-water concentrations from the sediment-water interface to 300 mbsf. The difference is caused by K expulsion from clay mineral ion exchange sites, partial dissolution of volcanic ash, and slightly elevated concentrations during the glacial ocean. Similar to Si (see "Silica"), K concentrations decrease sharply between 300 and 360 mbsf, across the Unit II/III boundary, and increase slightly near the approximate stratigraphic level of the décollement zone. Below this, concentrations remain constant to basement at ~2 mM, indicating that >80% of the original K in the interstitial waters was incorporated into solid phases between 60° and 110°C. The relative invariance of both Si and K concentrations below 360 mbsf suggests that they are involved in the same diagenetic reaction of authigenic silicate formation. The most likely candidates are K zeolites. The sharp boundary between the two K concentration depth intervals suggests high rates of reaction above and below this sharp boundary. At temperatures below ~150°C, K is depleted from the aqueous phase, as seen here, whereas, at elevated temperatures, as at hydrothermal vents, K is released into the aqueous phase.
Dissolved Si concentrations increase from ~600 to 1250 µM at 304 mbsf. There is a small decrease to 1065 mM at 336.9 mbsf and then a sharp decrease within the distance of one core to 230 mM at 346.5 m. There is no concentration trend from this depth to the bottom of the hole. The sharp discontinuity in silica concentrations at this depth range, which corresponds to the lithostratigraphic boundary between Units II and III, indicates that (1) distinct water-sediment reactions control the silica concentrations above and below this depth interval and (2) that both reaction rates are greater than the diffusive transport rate. Biogenic silica, mostly diatom and sponge spicule dissolution, controls the high dissolved silica values above the boundary, as indicated by the diatom and sponge spicule abundances in lithostratigraphic Units I and II and absence in Unit III (see "Biostratigraphy"). An as-yet-unidentified authigenic silicate reaction(s), possibly zeolite formation, controls Si concentrations below the boundary.
Mg concentrations generally decrease with depth, whereas Ca concentrations generally increase. There are, however, significant changes in their concentration gradients that indicate the occurrence of a variety of distinct reactions. Mg and Ca concentrations decrease from near seawater values to a local minimum of ~46 mM for Mg and 4 mM for Ca at 23 mbsf. This decrease may be due to the formation of dolomite. Dolomite formation and other carbonate reactions are driven by high alkalinity. Both then increase, Mg to 54 mM at 48 mbsf and Ca to 7.6 mM at 70 mbsf. The Mg increase is due to ion exchange driven by increasing ammonium concentrations, whereas Ca is involved in carbonate reactions. Mg then smoothly decreases to a second local minimum just below the Unit II/III lithostratigraphic boundary, consistent with the uptake of Mg into a silicate phase (see "Silica"). There is a small increase in Mg with a maximum of 10.5 mM at 459 mbsf and then a smooth decrease toward basement reaching a concentration of 1.7 mM at 678 mbsf, consistent with a low-Mg basement fluid.
Ca concentrations increase smoothly from the Unit I/II boundary to a small local maximum at ~400 mbsf (coincident with a local Cl minimum). Concentrations then smoothly increase with depth reaching a concentration of 87 mM at 678 mbsf, indicating a significant source of Ca from the basement.
Microbially mediated reactions and diffusion control sulfate concentrations in interstitial waters. Sulfate concentrations decrease nearly linearly with depth in Core 190-1173A-1H. By a depth of 5.9 mbsf, sulfate is undetectable. This is consistent with shallow sulfate reduction being dominated by reduction by methane produced at greater depth. Sulfate concentrations remain below detection until a depth of 238 mbsf. Between this depth and 634 mbsf, there is a smooth increase to 10 mM. In the last three cores, concentrations appear to decrease, with the deepest sample having a concentration of ~8 mM. Unlike Si and Cl, there is no discontinuity in the sulfate concentration gradient at 340 mbsf. The reappearance of dissolved sulfate below 200 mbsf is most likely a relic, a result of the rapid burial of the sulfate-containing Shikoku Basin sediments by the overlying turbidites. The sulfate profile below ~600 mbsf, where temperatures are ~90°C, is controlled by slow microbial activity (see "Microbiology"). Below this, inorganic reactions in the sediment and oceanic basement (i.e., anhydrite precipitation) occur as evidenced by the reversal in the sulfate gradient at the base of the section.
Ammonium concentrations monotonically increase from 0.7 mM at 1.4 mbsf to a maximum of 6.2 mM between 24 and 40 mbsf. Ammonium production is microbially mediated by organic matter fermentation, as is phosphate production. An active ion exchanger, ammonium at high concentration expels other ions (particularly Mg and K) from clay mineral ion exchange sites, as indicated, for example, by the increase in Mg concentrations in the same depth interval. Ammonium concentrations then decrease to 0.2 mM at 678 mbsf with a decreasing gradient with depth. The change in gradient suggests that more than one sink for ammonium exists. We propose that an as-yet-unidentified microbially mediated reaction influences the concentration-depth profile below ~100 mbsf, and at greater depths, where temperatures are too high for microbial activity, it diffuses into oceanic basement. As with sulfate, there is no gradient discontinuity at 340 mbsf. Phosphate concentrations peak at a shallower depth than alkalinity or ammonium. The very low concentration of phosphate below 100 mbsf is controlled by the solubility of apatite, the major sink for phosphate.
Similar to ammonium, alkalinity has a concentration maximum in the upper part of the hole that is driven by the microbial fermentation of organic matter and sulfate reduction. However, the maximum is found at greater depth. At 10 mbsf, the alkalinity concentration is 10 mM, rising to a maximum of 54 mM at 60.5 mbsf.