The main objectives for Site 1254 with respect to the organic geochemistry are (1) characterizing the hydrologic system along the décollement and the upper conduit observed during Leg 170 at Site 1040; (2) determining the chemistry and composition of pore fluids and gases from the décollement to compare with profiles measured during Leg 170 and to evaluate possible spatial heterogeneity and any possible changes through time for diagenetic and hydrologic modeling; (3) constraining pathways of fluid return to the surface and to evaluate the effects of this flow system on element fluxes; and (4) determining the horizon for long-term fluid flow monitoring.
At the prism Site 1254, volatile hydrocarbon gases (primarily methane, ethane, and propane) were sampled by headspace and vacutainer techniques and analyzed by gas chromatography (see "Organic Geochemistry" in the "Explanatory Notes" chapter). A major focus was to exactly determine the depths of the two main flow conduits discovered at Site 1040 during Leg 170 (Kimura, Silver, Blum, et al., 1997). The décollement conduit was of special interest for the deployment of the OsmoSampler and for determining where the base of the décollement is (see "Operations" and "Inorganic Geochemistry"). Therefore, samples were taken at a high frequency (i.e., generally three vacutainers and four headspace samples per core) and analyzed as soon as possible to allow for real-time monitoring. Vacutainer results are summarized in Table T13 and shown in Figure F45, whereas headspace data are compiled in Table T14 and shown in Figure F46.
The hydrocarbon concentrations in the vacutainers are significantly higher (up to 99.9 vol%) than those obtained by headspace technique, but both show similar hydrocarbon ratios for equivalent depth intervals as well as similar concentration profiles (Figs. F45, F46). The difference between the two data sets is readily explained by the different sampling techniques causing a loss of hydrocarbons, especially the more mobile methane, before the headspace sediment is sealed in the glass vial (Stein et al., 1995). Hence, scattering in the headspace data points occurs more frequently. Vacutainer results were found to show much less scatter. Besides the hydrocarbon concentrations, the fractions of O2, N2, and CO2 in the gas samples are reported in Table T13 to estimate possible contamination by air during the sampling procedure. This approach is applicable to samples in which methane is the major gas component in the sediments because of production by methanotrophy (e.g., 150-356 mbsf). For the vacutainer plots (Fig. F45), data points highly contaminated by air were omitted (see Table T13). Therefore, the results of the vacutainer samples are considered more reliable at Site 1254 and are preferred for the discussion.
Gas voids were observed throughout the cored sediment section and were sampled with a plastic syringe. The gas in the voids consisted mainly of methane but also showed considerable amounts of higher alkanes up to the homologous pentane. Methane concentrations were very high (7 × 105 to 9 × 105 ppmv in the vacutainers) throughout the cored sediment intervals above and within the décollement and drastically dropped down to 4 × 104 ppmv directly below the décollement at 363 mbsf (Fig. F45). Ethane concentrations were significantly lower (~200-800 ppmv) and showed a peak concentration of 1500 ppmv at 360 mbsf. Based on the data from Site 1040 (Kimura, Silver, Blum, et al., 1997), propane was chosen as an indicator for flow conduits because it was absent above the prism fault as well as below the décollement. Its thermogenic origin (80°-120şC) (Claypool and Kvenvolden, 1983) implies that propane is produced in far deeper sediment layers and transported upward through the flow conduit system including the décollement. Propane concentrations show at least two significant peaks, at 216 mbsf and in the décollement at 355 mbsf, with maximum concentrations of 326 and 370 ppmv, respectively. A smaller peak was detected in the décollement zone at 322 mbsf. At these depths also, higher hydrocarbons (i.e., butane, isobutane, and isopentane) were detected in considerable amounts. The three propane concentration peaks correlate with structural and lithologic observations (see "Structural Geology"). The uppermost peak at ~220 mbsf corresponds to a highly fractured interval that is interpreted as a fault zone, whereas the two lower peaks are connected to the décollement fault zone. The lithology shows a brecciated sandy interval at 345-356 mbsf, near the 370-ppmv propane peak at the base of the décollement, whereas the smaller peak at ~325 mbsf is near a fractured foliated interval just above the décollement (at 320-330 mbsf). These lithologies may act as flow conduits. Compared to the Site 1040 data, propane data at Site 1254 document that the uppermost flow conduit is observed ~20 m deeper. The fluid conduit in the décollement is observed at almost the same depth at Site 1040; the same is true for the lithologic boundary with the underthrust section (see "Structural Geology"). Overall, the results verify and better define the vertical structure of the flow conduits as determined during Leg 170.
CH4/C2H6 ratios (Table T13) show values between 1500 and 2000 in the whole prism section cored (150-360 mbsf), with two minima of 1088 and 586 at the flow conduits, indicating migrating thermogenic hydrocarbons. Below 363 mbsf (i.e., in the underthrust sediments), the methane-to-ethane ratio rapidly increases to ~10,000, indicating methanogenesis as the major source for hydrocarbons.
The depth distributions of inorganic carbon (represented as calcium carbonate), organic carbon, nitrogen, and sulfur in the solid phase from the cores in Hole 1254A are reported in Table T15. The combined concentration-depth data from Sites 1254 and 1040 are plotted for CaCO3, total organic carbon (TOC), and total sulfur in Figure F47.
Calcium carbonate concentrations are generally low (<10 wt%), whereas the TOC content fluctuates (typically ~1.0 wt%) and is slightly elevated (up to 1.5-2.0 wt%) at the base of the décollement zone (>330 mbsf). Total sulfur concentrations vary between 0 and 1.5 wt%, similar to the amounts measured in Subunits P1B and U1A during Leg 170 (Kimura, Silver, Blum, et al., 1997). In particular, the higher sulfur concentrations (~1.0-1.5 wt%) are observed at the thrust fault (~220 mbsf) and in the décollement zone (350-360 mbsf), although these observations must be regarded as tentative, given the poor accuracy of the sulfur data.
To characterize the type of organic matter in the sediments, total organic carbon/total nitrogen (TOC/TN) values and hydrogen index (HI) values from Rock-Eval pyrolysis have been used. Unfortunately, the S3 channel of the Rock-Eval did not record any data and, hence, no oxygen index (OI) or S2/S3 ratio could be calculated (Table T16).
Most of the TOC/TN values range from 4 to 8 (Table T15), which indicate a predominantly marine origin of the organic material (Bordovskiy, 1965; Emerson and Hedges, 1988). In contrast, the generally low HI values (50-120 mg HC/g TOC) (Table T16) suggest a significant input of terrigenous material (Emeis and Kvenvolden, 1986). Whereas the data from Site 1040 (Kimura, Silver, Blum, et al., 1997) also suggested this composition of organic matter, the HI values were significantly higher, suggesting marine-dominated TOC. The elevated OI values (not determined at Site 1254) indicated a significant input of terrigenous organic matter. At Site 1254, the low content of pyrolizable carbon as well as the combination of temperatures of maximum hydrocarbon generation (Tmax) <440°C and low production indexes (PI < 0.2) reflect organic matter that is immature for oil production. In addition, the PI depth profile shows a distinct peak value of 0.31 in Section 205-1254A-14R-3, which might be interpreted as a sign for migrated hydrocarbons. Although a similar pattern was not reported for Site 1040, the coincidence with the depth of the fluid conduit in the décollement is noteworthy. Furthermore, during whole-round preparation from Cores 205-1254A-13R, 14R, and 15R for the inorganic geochemical analyses, a weak but distinct smell of aromatic compounds (i.e., of oil/gasoline) was noticeable.