IODP Proceedings    Volume contents     Search


Scientific results


During Expedition 348, four lithologic units were identified at Site C0002 based on geological and geochemical characteristics of core and cuttings samples (Fig. F4):

  • Unit II (475–512.5 mbsf in Hole C0002M),
  • Unit III (875.5–975.5 mbsf in Hole C0002N),
  • Unit IV (975.5–1665.5 mbsf in Hole C0002N), and
  • Unit V (1665.5–2325.5 mbsf in Hole C0002N and 1965.5–3058.5 mbsf in Hole C0002P).

Lithologic Unit II is only described in core from Hole C0002M and is dominated by fine-grained turbiditic deposits. Silty claystone is the main lithology, with subordinate fine-grained sandstone and sandy siltstone. Similar to the equivalent interval (also designated Unit II) in nearby Holes C0002B, C0002K, and C0002L, this unit is composed of lower Kumano forearc basin sediment and is dominated by the hemipelagic mud of distal turbidites (Expedition 315 Scientists, 2009).

Lithologic Unit III, sampled in Hole C0002N (previously sampled in Holes C0002B, C0002F, and C0002J) (Figs. F4, F6A) is dominated by silty claystone with trace amounts of very fine, loose sand containing common glauconite grains. Like Unit II, this unit is also composed of lower Kumano forearc basin sediment.

Lithologic Unit IV, sampled only in Hole C0002N during Expedition 348 but equivalent to Unit IV sampled and/or logged in Holes C0002A, C00002B, C0002F, C0002H, and C0002J, is dominated by silty claystone, with sandstone as a minor lithology. Sandstone cuttings in this unit are generally very weakly consolidated and occur as disaggregated loose sand. Lithologic Unit IV is divided into five subunits based on sand content (Fig. F6A). Lithologic Unit IV is interpreted as the upper accretionary prism, composed of accreted Shikoku Basin hemipelagic deposits or trench fill sediment.

Lithologic Unit V was sampled in both Holes C0002N and C0002P (Fig. F6) and is dominated by silty claystone. Fine-grained and moderately cemented sandstone is rarely observed. In Hole C0002P, visual estimates of clay content in the silty claystone increase throughout Subunit VA, and the sediment has a finer texture than material sampled nearby in Hole C0002N. The boundary between Subunits VA and VB (2625.5 mbsf) is defined by a fining of grain size and is located at a depth where the dominant lithology changes from silty claystone to fine silty claystone. This unit is interpreted as accreted Shikoku Basin hemipelagic deposits or trench fill sediment of the middle Miocene age trench.

Structural geology

Several key observations were made on cuttings in Holes C0002N and C0002P, along with structural analyses of the limited cores retrieved in Hole C0002P. Structures observed in intact cuttings (Fig. F7) include slickenlined surfaces, scaly fabric, deformation bands, minor faults, and mineral veins. Slickenlines are observed throughout the entire drilled interval, but scaly fabric is increasingly observed below ~2200 mbsf. The other types of structures are scattered throughout the section.

The limited cored interval exhibits steep bedding dips (Fig. F8) ranging from ~45° to 90°, consistent with the observation of steep bedding in borehole resistivity image log data (see “Logging”) throughout the interval from 2100–3000 mbsf. A fault zone, 90 cm thick with 2 mm angular clasts, is present in one of the cores (2204.9–2205.8 mbsf; Fig. F9). In its present position, the brittle fault zone is associated with a normal faulting sense based on kinematic indicators; however, given the very steep bedding dip documented in cores and throughout the hole in log images, it is plausible that this represents an early thrust rotated after its development or a late normal fault developed at or near its present orientation.

The overall character of the deformation throughout the drilled interval (independent particulate flow with limited evidence for cataclastic deformation) suggests that deformation occurred in a relatively shallow environment (~0–4 km in burial depth), consistent with the present-day depth of this interval in the inner wedge but not precluding modest structural exhumation.

Scanning electron microscope observation of cuttings from the upper part of Hole C0002N indicates little evidence for opal diagenesis, implying maximum temperature (T) <60°–80°C at 1225.5 mbsf. In Hole C0002N, the fabric lacks a strongly preferred orientation in clay-rich materials, except along striated microfaults formed by clays. These zones are extremely localized, with a thickness of a few micrometers or less. In Hole C0002P below 2200 mbsf, development of a regularly spaced fabric in sandstones is characterized by thin (<0.1–1 µm) clay-dominated shear planes. Below 2625 mbsf, compaction fabrics in clay-rich materials can be observed; this fabric is cut by very thin shear zones with almost no wall damage zone.


Preliminary biostratigraphy for Holes C0002M, C0002N, and C0002P is based exclusively on the examination of calcareous nannofossils (Fig. F10). There is a general pattern of well- to moderately preserved nannofossils in the upper part of the site (475.09–985.50 mbsf) and moderate to poor preservation below 985.50 mbsf. Assemblages recovered from 475.09 to 506.68 mbsf (Hole C0002M) are Pleistocene in age, whereas between 875.50 and 3055.50 mbsf (Holes C0002N and C0002P), assemblages indicate late Pliocene to late Miocene age.

Calcareous nannofossils were examined in 25 core samples from interval 348-C0002M-1R-1, 9 cm, to 4R-3, 86 cm, and 287 cuttings samples (348-C0002N-3-SMW through 327-SMW and 348-C0002P-9-SMW through 300-SMW) were examined. A further 25 core samples from interval 348-C0002P-1R-CC, 20 cm, to 6R-CC, 10 cm, were examined.

Hole C0002M

Cores taken from Hole C0002M from 475.00 to 512.50 mbsf yield very well preserved and abundant calcareous nannofossils, indicating Pleistocene age for the upper part of the section. Assemblages suggest Biozone NN20 and a maximum age of 1.67 Ma based on the last occurrence of Gephyrocapsa spp. (<3.92 Ma).

Holes C0002N and C0002P

Cuttings and core samples from Holes C0002N and C0002P between 875.50 and 3055.5 mbsf have assemblages ranging from the late Pliocene to late Miocene. Abundance is lower than in Hole C0002M, and assemblages are less well preserved. Assemblages from Holes C0002N and C0002P show an age range from at least 2.06 Ma to a maximum of 10.734 Ma, based mainly on the last occurrence and first occurrence, respectively, of Discoaster brouweri. Samples obtained from between 2955.5 and 3055.5 mbsf did not yield any calcareous nannofossil zonal marker species; therefore, the deepest section of Hole C0002P was not dated.


Remanent magnetization of archive-half sections from Hole C0002P was measured at demagnetization levels of 0, 5, 10, 15, and 20 mT peak fields to identify characteristic directions. Demagnetizations of 10–15 mT successfully removed low-coercivity components, and magnetic directions after demagnetizations indicate stable constant directions (Fig. F11). Declination, inclination, and intensity profiles after demagnetization at 20 mT are shown in Figure F12. The declination profile represents widely scattered directions, indicative of “biscuiting” of cores during RCB coring operations. The inclination profile reveals dominantly positive inclinations, and the magnitude of inclination is not constant downhole. For example, the calculated mean inclinations using the method proposed by Arason and Levi (2010) are 34.22° for 2172.45–2174.955 mbsf, 61.83° for 2194.005–2196.985 mbsf, and 36.50° for 2210.0–2215.0 mbsf. Some sections exhibit steep negative inclinations, which occur in relatively short intervals. Interestingly, the 2205.195–2205.515 mbsf interval, which shows a clear negative inclination, corresponds to the brittle fault zone (see “Structural geology”). This result suggests different timing of magnetization for this interval than that of the intervals above and below. In order to elucidate the timing of lock-in of these magnetizations, careful evaluation referencing structural analysis (e.g., bedding) results are required during postcruise study.


Broadly, geochemical data from Holes C0002N and C0002P support possible important clay-water reactions controlling dissolved cation abundances, the presence of secondary carbonate in faulted sediments, an increase of thermogenic methane with increasing depth, and overall dominance of methane in the gas fraction.

Interstitial water analyses

One 20 cm long whole-round sample for interstitial water was collected from each of the four Hole C0002M cores. Interstitial water samples were obtained with a new 55 mm diameter Manheim squeezer to conduct tests aimed at defining protocols for maximum squeezing pressure to use on future cores (i.e., to avoid possible stress-induced dehydration of smectite group clays; Fitts and Brown, 1999). Because these cores had been stored in core liner for several weeks, the likelihood of contamination by drilling fluid was very high, and only chlorinities were determined. Step-wise increasing pressures (up to 112 MPa) were applied to assess the effect of squeezing on measured chlorinity, and interstitial water was sampled at several different time intervals. Chlorinity values are higher than those observed at the same depth interval (470.5–500.5 mbsf) during Expedition 338 (average = 377.8 mM), likely due to contamination, but the range in chlorinity was still low (397–419 mM) relative to seawater (559 mM). Results from the squeezing experiments show freshening of interstitial water when the same sample was squeezed at higher pressure and for a longer time (Fig. F13). The mechanism that induces the freshening of interstitial water at high pressure is not yet understood, and later shore-based experiments will be conducted.

Five whole-round samples (10–41.5 cm in length) were collected from Hole C0002P (Cores 348-C0002P-2R through 6R; 2176.28 to 2211.31 mbsf). Perfluorocarbon tracer data indicate that only Section 348-C0002P-2R-3 had appreciable contamination (~5%) by drilling fluid. Samples were processed to obtain pore water using the ground rock interstitial normative determination (GRIND) method as was used during Expedition 338 (Strasser, Dugan, Kanagawa, Moore, Toczko, Maeda, and the Expedition 338 Scientists, 2014). Salinity and chlorinity alternate between high and low values downhole (Fig. F14), a pattern that is paralleled by several other major and minor ions, including Br, NH4+, Na+, K+, Mg2+, Ca2+, Li, Mn, Ba, Sr, and to a lesser extent Rb and Cs. Although the possibility of localized brines in the formation cannot be excluded, it is unlikely that such large fluctuations in concentrations could occur in samples taken only 10 m apart. It is more likely that the variations in concentrations are related to interaction between interstitial water and rock during core retrieval and sample processing with the GRIND method, perhaps related to variations in clay mineralogy among samples. Two sections (348-C0002P-3R-2 and 5R-2) yielded ion concentrations within interstitial water ranges observed at shallower depths in Holes C0002B and C0002J, with chlorinities of 428 and 387 mM, respectively. If correct, the very low values observed in Hole C0002P require substantial freshening relative to typical seawater values (556 mM), and at these depths and in situ temperatures the only likely potential source is dehydration of clay-bound waters or opaline silica. Further analysis is necessary to evaluate the significance of the GRIND data.

Carbonate, total organic carbon, total nitrogen, and carbon/nitrogen ratios

Carbonates (as CaCO3), total organic carbon (TOC), and total nitrogen (TN) were analyzed on 15 samples from the four cores from Hole C0002M. CaCO3 ranges from 4.26 to 13.67 wt%, and the median (6.24 wt%) is higher than the two values determined in the same interval from Hole C0002L during Expedition 338 (6.1 and 1.4 wt%). TOC is low (0.46–0.82 wt%) and decreases slightly with depth. TOC and TN are similar to those reported from Expedition 338; the low TOC/TN ratio values in Hole C0002M indicate a marine origin for the organic matter.

In Holes C0002N and C0002P, CaCO3, TOC, and TN were determined from the 1–4 and >4 mm cuttings size fractions (Fig. F15). Excluding intervals contaminated by artificial cement derived from the area around the casing shoe, CaCO3 varies from 0.46 to 9.6 wt% (median = 3.2 wt%). Carbonate abundance generally decreases downhole. A prominent maximum is observed around 1920.5 mbsf, where CaCO3 is as high as 7.94 wt%. A broader maximum at 2620.5 mbsf shows CaCO3 values as high as 5.80 wt%. Cuttings from these depth intervals contain fragments of carbonate veins and sediment with veins, suggesting the local increases in carbonate are associated with fault or fracture zones. The abundant carbonate observed in the fault zone in Core 348-C0002P-5R, however, was not evident in the cuttings from the same interval, likely because the signal from the thin fault zone was diluted by mixing of cuttings over a ~5–10 m interval. Increases in CaO and MnO observed in the X-ray diffraction (XRD) data at both carbonate maxima are consistent with the observed increase in CaCO3 and possibly the presence of Mn-bearing carbonate.

In Hole C0002M core samples, TOC is low (0.46–0.82 wt%) and decreases slightly with depth. TOC and TN are similar to Expedition 338. The low TOC/TN ratio values indicate a marine origin for the organic matter. In Holes C0002N and C0002P, TOC values range from 0.47 to 2.07 wt% (median = 0.9 wt%) and gradually decrease with depth (Fig. F15). TOC values at the bottom of Hole C0002N (2040–2320 mbsf) do not overlap those at the beginning of Hole C0002P, even though data for CaCO3 and TN do match in the overlapping interval. Data from core samples in Hole C0002P are in closer agreement with those of cuttings from Hole C0002P than with cuttings from Hole C0002N. The relatively high TOC/TN ratios (7.58–35.58; median = 17.53) suggest most of the organic matter is of terrestrial origin. In contrast, TOC/TN in Hole C0002P core samples tend toward a marine origin except for two samples from Sections 348-C0002P-2R-4 and 6R-4. Thus, the high TOC/TN ratios in cuttings might reflect contamination by drilling fluid.

Headspace gas and mud-gas results

Drilling mud-gas monitoring took place continuously while drilling Holes C0002N (838–2330 mbsf) and C0002P (1954–3058 mbsf). The drilling mud-gas was analyzed for alkanes (methane, ethane, propane, etc.) and nonhydrocarbon (222Rn, He, H2, Xe, N2, O2, etc.) gases by GeoServices and by using the scientific drilling mud-gas monitoring system onboard the Chikyu (SciGas system). In Hole C0002N, measurements from the SciGas system yielded lower hydrocarbon gas concentrations compared to the data from GeoServices. Before starting Hole C0002P, the SciGas system was improved, leading to relatively higher gas concentrations and better detection of relative changes in the gas composition in that hole. Headspace gas samples from Hole C0002M and C0002P cores were dominated by methane, which was as high as 7,354 and 23,455 ppm, respectively. Concentrations found in headspace gas samples were up to 2 orders of magnitude higher than the drilling mud-gas samples from the same interval, indicating underestimation of formation hydrocarbon abundances by the real-time mud-gas monitoring.

Total mud-gas concentrations were dominated by methane, with maximum concentrations of ~8% at around 1305 mbsf (see Fig. F71 in the “Site C0002” chapter [Tobin et al., 2015]). Downhole gas concentrations steadily declined to <0.2% and rose again at ~2184 mbsf. Overall, ethane and propane were only present in minor concentrations, and higher homologs (i.e., n-butane, i-butane, n-pentane, and i-pentane) typically remained <0.01%. Starting at 2200 mbsf, ethane and propane steadily increase with depth. Over the same interval, the Bernard parameter (C1/[C2 + C3]) showed an overall decline with depth and when combined with the methane carbon isotope ratio, indicated the onset of a thermogenic regime at ~1700 mbsf (see Fig. F80 in the “Site C0002” chapter [Tobin et al., 2015]). A clear thermogenic signature was reached at ~2325 mbsf.

Nonhydrocarbon gases were dominated by atmospheric components (N2 + O2 = 98.3%), which is consistent with an overall N2/Ar ratio <100. Hydrogen was detected at values up to 0.78% at ~3043 mbsf, and 222Rn values typically remained below 450 Bq/m3.

Physical properties

Shipboard physical property measurements, including moisture and density (MAD), electrical conductivity, P-wave velocity, natural gamma radiation, and magnetic susceptibility, were performed on cuttings samples from 870.5 to 3058.5 mbsf and on core samples from 2163 and 2218.5 mbsf.

MAD measurements were conducted on seawater-washed cuttings (“bulk cuttings”) in two size fractions, >4 and 1–4 mm from 870.5 to 3058.5 mbsf and handpicked intact cuttings from the >4 mm size fractions from 1222.5 to 3058.5 mbsf (Fig. F16). The bulk cuttings show grain density of 2.68–2.72 g/cm3, bulk density of 1.9–2.0 g/cm3, and porosity of 50%–35%. The intact cuttings show almost the same grain density (2.66–2.70 g/cm3) but have higher bulk density (2.05–2.41 g/cm3) and lower porosity (37%–18%) than the bulk cuttings. The grain density agreement suggests that the measurements on both bulk cuttings and intact cuttings are of good quality, and the differences in porosity and density are real. It is likely that the values from the bulk cuttings are affected strongly by artifacts of the drilling process. Thus, the bulk density and porosity data on intact cuttings probably better represent true formation properties. Combined with MAD measurements on handpicked intact cuttings and discrete core samples from previous expeditions, porosity generally decreases from ~60% to ~20% from the seafloor to 3000 mbsf at Site C0002.

Electrical conductivity and P-wave velocity of discrete samples, which were prepared from both cuttings and core samples from 1745.5–3058.5 mbsf, range between 0.15 and 0.9 S/m and 1.7 and 4.5 km/s, respectively. The electrical resistivity (a reciprocal of conductivity) of discrete samples is generally higher than the LWD resistivity data, but the overall depth trends are similar. The P-wave velocity of discrete samples is lower than the LWD P-wave velocity (see “Logging”) between 2200 and 2600 mbsf, whereas the two are in close agreement deeper than 2600 mbsf.