IODP

doi:10.2204/iodp.pr.319.2009

Site summaries

Site C0009

Site C0009 marked the first riser drilling in IODP history. This allowed us to conduct several scientific operations new to IODP, including measurements of in situ pore pressure and stress using the MDT tool and LOTs, real-time mud gas analysis, and laboratory analyses of cuttings. In situ pore pressure, stress magnitude, and permeability are among the most central data for understanding the mechanics of active fault zones and testing the core NanTroSEIZE hypotheses, but they have previously been unavailable in IODP drilling. Analyses of mud gas and cuttings, although common in other riser drilling programs (e.g., Erzinger et al., 2006), were also conducted for the first time in IODP and will be essential for future riser-based drilling. Because there are no standard IODP procedures for shipboard measurements on cuttings, the scientific party developed several techniques for handling, sampling, and measurement of physical properties, rock chemistry, and sedimentological and structural description (see discussion in "Insights from scientific riser operations").

Drilling at Site C0009 achieved all of the primary planned scientific objectives, although some operations were shortened slightly because of schedule constraints. After riserless drilling and installation of casing to 703.9 m drilling depth below seafloor (DSF), riser operations included collection of cuttings (from 1017 m mud depth below seafloor [MSF] to TD) and core (1509.7–1593.9 m core depth below seafloor [CSF]) to document stratigraphy and sediment composition, measure rock physical properties, and identify structures. Because of limited coring operations and limitations associated with cuttings analyses (see "Insights from scientific riser operations"), we defined a single integrated set of lithologic units based on the range of data available from cuttings, wireline logging data, and cores (Fig. F5A, F5B). We ran LOT and MDT experiments that provide direct measurements of in situ stress magnitude, as well as formation pore pressure and permeability. During all phases of riser drilling, we also collected mud gas for geochemical analyses, from 703.9 m drilling depth below rig floor (DRF) to TD (e.g., Wiersberg et al., 2007). A zero-offset and walkaway VSP experiment was conducted using a wireline array of seismometers within the borehole to define the seismic velocity and structure around the borehole and at the underlying plate boundary.

Lithology

Using the combination of data from wireline logs, cuttings, and limited core, we defined four distinct lithologic units composed of mud/mudstone with variable interbeds of sand, silt, and volcanic ash/tuff. The unit boundaries and lithologies are constrained by macroscopic description of cuttings and sediment from drilling mud and of cores, smear slide and thin section observations, bulk X-ray diffraction (XRD) and X-ray fluorescence (XRF) analyses, and wireline logging data including natural GR, caliper, density, photoelectric effect, spontaneous potential, resistivity (including resistivity images), and sonic velocity (Fig. F5A, F5B). Despite some chemical and physical artefacts in the cuttings and uncertainty in their depth of origin, lithologic boundaries defined by analysis of cuttings are generally consistent with boundaries defined by logging data:

  • Unit I (Pleistocene–modern; 0–467 m wireline log matched depth below seafloor [WMSF]) is a mudstone with cyclical sand-rich layers ranging from ~10 to 50 m thick.

  • Unit II (Pleistocene to ~1.5 Ma; 467–791 m WMSF) is a mudstone composed dominantly of silty clay, with some silt and sand interbeds and minor interbeds of volcanic ash.

  • Unit III (Pliocene–Pleistocene: ~3.8 to ~1.5 Ma; 791–1285 m WMSF) is composed of silty clay and poorly lithified silty claystone, with interbeds of silt and fine sand layers. It is distinguished from Unit II by its overall finer grain size, higher wood/lignite content, slightly increased consolidation state, and higher organic content. On the basis of cuttings, we divided Unit III into two subunits, with Subunit IIIB distinguished from Subunit IIIA by increased wood/lignite and glauconite abundance.

  • Unit IV (late Miocene: ~7.7–7.1 to ~5.59 Ma; 1285 WMSF to TD) is a silty claystone with silt interbeds and rare interbeds of fine vitric tuff. The Unit III/IV boundary is marked by changes in several logging data sets, increased lithification and sediment maturity, and several major compositional changes. It is also defined by a ~1.8 Ma age gap and an angular unconformity that can be traced across the Kumano Basin (Fig. F6).

Overall, we interpret the stratigraphic succession as a series of forearc basin–filling mudstones with varying sand and silt turbidite abundance underlain by older slope deposits and/or accretionary prism sediment. Unit I is significantly sandier than other units drilled at Sites C0009 and C0002 (20 km seaward). Unit II is characterized by turbidites that are coarser than those in underlying Units III and IV and the units drilled at Site C0002 but markedly thinner and finer grained than those in Unit I above. Unit III includes two subunits (IIIA and IIIB) containing thinly bedded fine-grained turbidites that were deposited above the carbonate compensation depth (CCD) in the early Kumano forearc basin. The lower subunit (IIIB) had an increased supply of terrigenous organic matter (i.e., wood fragments). The composition of detrital grains points to a source from exposed metamorphic rocks of the Shimanto Belt (e.g., Taira et al., 1988; Isozaki and Itaya, 1990) and is consistent for Units I, II, and III.

Unit IV is mudstone containing thin-bedded fine-grained turbidites. Depositional environment relative to the late Miocene CCD is not conclusive based on initial interpretations of calcareous microfossil test preservation (see "Biostratigraphy"); however, deposition may have been at depths close to the CCD. It resembles Unit IV at Site C0002 in terms of sedimentary facies (Expedition 315 Scientists, 2009). Unit IV at Site C0009, however, is less deformed and we observed no clear sedimentary, geochemical, or structural evidence (see below) to conclusively indicate that this unit is composed of highly deformed frontally accreted trench sediment. Thus, Unit IV could be interpreted as either a weakly deformed package of accreted trench sediments, slope deposits, or sediments deposited in the distal reaches of the early Kumano Basin.

Structural geology and geomechanics

At Site C0009, we documented geologic structures in cuttings from 1097.7 to 1512.7 m MSF, and in core from 1510.5 to 1593.9 m CSF. We also analyzed FMI borehole resistivity images from 710 to 1579.9 m WMSF. Although the FMI's limited coverage of the borehole wall and reduced data quality in the deeper borehole (Unit IV) precluded clear identification of borehole breakouts in resistivity images, we were able to use the FMI caliper to measure borehole enlargement associated with breakouts. We also documented drilling-induced tensile fractures (DITF) in some portions of the hole. Both of these data sets are indicators of minimum and maximum horizontal stress orientation.

Vein structures identified in cuttings from the upper part of Unit IV are similar to those observed on previous drilling expeditions (e.g., Site C0002; Ashi et al., 2008) and in onshore exposures, and they are consistent with formation by dewatering or shaking-induced soft-sediment deformation. Faults identified in the cores exhibit two populations (one dipping at 10°–30° and the other at 50°–70°) reflecting both thrust and normal faults. Although most faults do not exhibit a clear sense of offset, we observed some cross-cutting relationships that indicate a complex deformation history in which faults exhibiting normal displacement both cut and are cut by faults having thrust displacement. We did not reorient structures identified in the cores because the cryogenic magnetometer was not on board during the expedition.

FMI images and caliper measurements provided additional structural data for Site C0009. In Subunit IIIA, beds dip gently north 10°–15°, whereas in Subunit IIIB, they dip gently north-northwest. A 3 m thick zone of deformation and increased bedding dip appears to distinguish Subunits IIIA and IIIB. Faults and fractures are also identified in Units III and IV and dip more steeply northwest than bedding planes, with a mean dip of ~60°. Within Unit IV, caliper measurements from the FMI indicate that the hole is significantly enlarged, with one caliper commonly measuring >16–18 inches in diameter and the other caliper typically close to 12 inches. The orientation of the largest caliper measurement (i.e., borehole enlargement) is stable even as the entire tool rotated in discrete 90° clockwise increments while being pulled uphole. The mean value of that orientation, weighted for the depth of borehole sampled in each interval, defines an alignment 58°–238° (northeast–southwest). We interpret this orientation as a series of breakouts, representing the direction of Shmin. This indicates that the maximum stress in the horizontal plane (SHmax) is oriented at 148°–328° (southeast–northwest), which is similar to the orientation of SHmax at IODP Sites C0001, C0004, and C0006 seaward of the megasplay fault but nearly perpendicular to that at Site C0002 located 20 km to the southeast at the seaward edge of the Kumano Basin (Kinoshita, Tobin, Ashi, Kimura, Lallemant, Screaton, Curewitz, Masago, Moe, and the Expedition 314/315/316 Scientists, 2009) (Fig. F7).

Biostratigraphy

We established an age model for the drilled sequence at Site C0009 from calcareous nannofossil biostratigraphy, primarily utilizing cuttings samples (704–1610 m MSF) and augmented with core samples from 1509.7 to 1593.9 m CSF (Figs. F5A, F6). We analyzed 72 cuttings samples spaced at 5–30 m intervals, with the closest spacing near key lithologic and zonal boundaries. Calcareous nannofossils are generally abundant throughout the section, and preservation of fossils is moderate in the upper part but there is a trend toward poorer preservation below ~1290 m MSF within Unit IV. Except for a few samples from the top sand-rich interval in lithologic Unit II, most samples yielded abundant nannofossils and the major age-diagnostic taxa appear reasonably continuous throughout their ranges. Therefore, most of the important Neogene and Quaternary datums (summarized by Raffi et al., 2006) and hence the nannofossil zones of Martini (1971) were recognized. However, frequent downhole contaminations of species prevent the use of some important nannofossil datums based on first (or first common) occurrences.

Our results indicate that the sampled sequences range in age from late Miocene (>7.1 Ma) to Pleistocene (<0.43 Ma). Unit II and the upper half of Subunit IIIA are early Pleistocene, and the rest of Unit III is Pliocene (Fig. F6A). Unit IV is late Miocene, and we documented an unconformity spanning ~1.8 m.y. (5.6–3.8 Ma) between the late Pliocene (Subunit IIIB) and the late Miocene (Unit IV). The upper Miocene sequence continues to the base of the drilled section, with probable ages younger than 7.88 Ma.

Geochemistry

During riser drilling at Site C0009, we conducted scientific mud gas monitoring (e.g., Erzinger et al., 2006). The principal formation gases extracted from the returning drilling mud were hydrocarbons, mainly methane, with up to 14 vol% CH4 detected during initial drilling of the 12¼ inch hole and up to 3 vol% detected during hole opening to 17 inches. We also detected traces of ethane (up to 16 parts per million by volume [ppmv]) and propane (up to 3 ppmv). During drilling of the 12¼ inch hole, methane concentrations were relatively low above ~800 m MSF. Methane concentrations were generally highest within Subunit IIIB from 1050 to 1220 m MSF. Below 1280 m MSF (approximately corresponding to the Unit III/IV boundary), methane concentrations decreased abruptly to ~4–5 vol%. During hole opening to 17 inches, a better depth resolution was achieved by modifying the gas extraction set-up; both data sets exhibit similar downhole trends.

The distribution of methane correlates directly with the presence of wood and lignite in cuttings, implying that they are the primary source of hydrocarbons (Figs. F5A, F8). The molecular composition of hydrocarbons clearly suggests a microbial source (the ratio CH4/C2H6 was consistently >500 and typically ~1000), although this ratio might be somewhat biased by the greater solubility of methane in the drilling mud relative to ethane and propane. This interpretation is consistent with an estimated temperature of ~50°C at the base of the borehole, which is too low for thermogenic hydrocarbon generation. Based on the correlation between stratigraphic observations and mud gas distribution combined with the overall fine-grained nature of the sediments in Unit III, gas migration through permeable strata seems unlikely to play a significant role in the observed gas distribution. This interpretation is also consistent with the absence of gases indicating a contribution from greater source depth (i.e., helium or heavy hydrocarbons). Taken together with observations from physical property measurements (see "Physical properties") these observations indicate a moderate gas saturation (Sw = ~10%) that was probably generated in place.

The total organic carbon (TOC) content of drill cuttings samples ranges from 0.93 to 8.7 wt%, with the highest values in Subunit IIIB between 1080 and 1240 m MSF. These values are consistent with the high abundance of wood/lignite observed in this unit and the occurrence of methane, as noted above. TOC and total nitrogen (TN) exhibit a very similar depth distribution. The TOC/TN ratio ranges from 13.8 to 74.5 with an average of 29.8; values are generally highest in Unit III and decrease in Unit IV. Marine organic matter typically exhibits TOC/TN ratios in the range of 6–9, whereas the ratio for terrestrial organic matter is typically >10. This suggests that the source of organic matter in much of the borehole is terrestrial, which is in agreement with the wood and coal fragments observed in drill cuttings.

Physical properties

We obtained a wide range of physical property data from wireline logs and measurements on cores and cuttings (Fig. F5). These data included bulk density and porosity, P- and S-wave velocity (VP and VS, respectively), magnetic susceptibility, electrical resistivity, and thermal conductivity (cores only). In general, bulk density gradually increases downsection and porosity (computed from bulk density and estimated from resistivity logs) decreases, which is consistent with a trend of increasing compaction with depth. P- and S-wave velocities increase with depth overall, but the P-wave velocity profile is marked by excursions to lower values within Unit III.

Most physical property measurements exhibit changes at the major lithologic unit boundaries. For example, the slope of the compaction trend increases slightly across the Unit III/IV boundary at 1285 m WMSF; we interpret this to reflect a change in either lithology or compaction history (i.e., secondary consolidation associated with the increased age of the sediment in Unit IV or possible lateral compression). In addition to the change in the slope of the porosity-depth trend, we observe a decrease in measured density (and concomitant increase in computed porosity) at the Unit III/IV boundary. We suggest this is an artifact of the higher clay content in Unit IV relative to overlying units; postexpedition XRD analyses of the clay-sized fraction will address this by quantifying the abundance of hydrous clay (i.e., smectite minerals).

On the basis of P-wave, S-wave, and resistivity logs, we identified four distinct zones of increased gas content in Unit III. Specifically, we observe intervals characterized by low VP, low VP/VS ratio (and thus low Poisson's ratio), and increased resistivity that correlate clearly with depths of increased methane measured by mud gas analyses (Fig. F8). A preliminary calculation suggests a gas saturation of ~10%. We also observe a significant change in caliper at the Unit III/IV boundary. Above this, the hole remains in gauge, whereas below the boundary we observe significant enlargement with one caliper measuring up to 16–18 inches (see "Structural geology and geomechanics"). This enlargement, taken together with the decrease in Stoneley wave velocity in this interval, may be related to increased deformation in Unit IV.

Moisture and density (MAD) density and porosity data from cores are in good agreement with the wireline logging data. In contrast, measurements conducted on cuttings appear to significantly overestimate porosity (by as much as ~12%) and underestimate bulk density. This discrepancy may be related to the small size of the cuttings that led to swelling during washing or exposure to drilling mud in the borehole or water retention on the surfaces of small particles after washing. Although the absolute values of porosity measured on cuttings are most probably well in excess of true formation porosity, downhole trends may reflect real relative variations in formation bulk density and porosity. This and several other issues encountered in analysis of cuttings are discussed in more detail in "Insights from scientific riser operations."

Downhole measurements

Stress, pore pressure, and permeability

Measurement of in situ pore pressure and stress using the MDT wireline logging tool assembly was one of the new scientific operations for IODP conducted during Expedition 319. This wireline tool measures borehole pressure, formation pore pressure at the bed and meter scale, and least principal stress magnitude. We ran the MDT tool successfully 12 times at Site C0009. This included nine single probe tests to measure in situ pore pressure and fluid mobility and three dual packer tests in 1 m thick isolated intervals: one to measure formation permeability from a drawdown and recovery cycle and two to measure in situ stress (two of the single probe tests were conducted at the same depth stations as dual packer tests). The single probe test uses a circular probe sealed against the borehole wall to extract fluid and reduce formation pore pressure in a small volume and then records the subsequent pressure recovery. The dual packer test isolates an interval of the formation (configured for 1 m at Site C0009) to either draw down the pressure (again to estimate in situ pressure and permeability over a larger scale than the single probe test) or to increase the borehole pressure to create a hydraulic fracture and measure the least principal stress magnitude.

The nine in situ pore pressure measurements indicate that formation pore pressure is hydrostatic or elevated by only a few percent of the hydrostatic value to depths of at least 1463.7 m WMSF (the depth of the deepest reliable measurement) (Fig. F9A). The slight apparent overpressure could be due to pressurization of the formation by weighted drilling mud, an incorrect estimation of the hydrostatic pressure (e.g., if the integrated seawater density from mean sea level to the measurement point was slightly underestimated), or the presence of real but slight overpressure. The permeablities we measured with the single probe tests range from ~10–16 to 10–14 m2 (Fig. F9B). Overall, the variation in permeability values is consistent with lithologic characteristics inferred from GR logs; the higher permeabilities correspond to zones of lower GR (i.e., sandier intervals). Analysis of the dual packer drawdown-recovery test at 1539.69 m WMSF in the clay-rich and fine-grained Unit IV yielded a permeability value of 1.3 x 10–17 m2. It is important to recognize that the MDT tool is generally designed for use in formations with permeabilities > ~10–15 m2. In low-permeability formations, both pore pressure and permeability estimates should be viewed with caution because the pressure recovery time may be considerably longer than the tool deployment time.

We conducted hydraulic fracturing (HF) tests near the top and bottom of the 12¼ inch open-hole section at 874.30 and 1462.3 m WMSF using the MDT dual packer tool. The deeper test did not yield reliable results. For the shallower test, we observed a repeatable and clear instantaneous shut-in pressure of 34.8 MPa, which we interpret as the least principal stress (S3) (Fig. F9A). This value is smaller than the vertical stress (Sv). If the principal stresses are horizontal and vertical, then this implies S3 = Shmin; the measured value corresponds to an effective stress ratio (Shmin/Sv) of 0.82. As part of standard riser drilling operations, we also conducted a LOT at the base of the 20 inch casing (708.6 m DSF). The leak-off pressures are 30.22–30.25 MPa, and as was the case for the MDT HF test, are less than Sv, also indicating that S3 = Shmin. However, the value of S3 obtained from this test is considerably lower than from the MDT HF test, with Shmin/Sv = 0.44. Although there is considerable uncertainty in determining the S3 from this type of test (e.g., Zoback, 2007), the LOT was repeatable and appears reliable.

Vertical seismic profiling

We conducted a walkaway VSP experiment in Hole C0009A using an array of 16 seismometers within the Vertical Seismic Imager (VSI) wireline tool set inside the 13 inch casing between 2989 and 3217 m DRF. The objective of the walkaway VSP experiment was to image the structure around the megasplay fault and master décollement below the borehole and evaluate seismic anisotropy of the basin sediment and accretionary prism around the borehole. The shooting vessel (Kairei) shot a single 53.4 km long offset transect in the dip direction of the subducting plate and a circular shooting offset of 3.5 km radius around the borehole. We obtained 880 shot records in the dip line and 275 shots in the circular line.

From the walkaway VSP records, we were able to identify direct wave arrivals, seismic phases associated with multiples in the water column, refractions from the accretionary prism, and reflections from interfaces below the seismometers within the accretionary prism as well as from the splay fault and probably from the plate boundary. The dense seismic array in the hole was sensitive enough that seismic waves traveling upward and downward were coherent and clearly distinguishable. We also observed subordinate seismic waves generated by direct waves hitting the structure of the borehole and traveling along the casing at ~6 km/s. These phases were prominent in the vertical component at frequencies higher than 20 Hz and make it difficult to see high-frequency phases from the formation. Therefore, deeper and weaker seismic phases appear clearer in horizontal component records because of smaller effects from such casing related phases.

Following the walkaway VSP experiment, we conducted a zero-offset VSP experiment using air guns deployed at 60 m offset from Chikyu and obtained by moving the seismic array upward from the depth of the walkaway VSP experiment in ~122 m intervals to obtain coverage around the entire length of the borehole. As was the case for the walkaway VSP, zero-offset VSP records also exhibited the subordinate seismic phases traveling vertically in the casing. We also observed acoustic waves propagating inside the casing, and those probably reflected back at the bottom of the casing (at the casing shoe or float collar). In horizontal component records, there are clear phases propagating downward with apparent velocities in good agreement with velocities from wireline sonic logs and with the velocities from precruise seismic processing. We picked these phases as P-waves propagating in the formation.

Cuttings–Core-Log-Seismic integration

We combined the zero-offset VSP (check shot) with wireline sonic velocities to derive a velocity-depth function at Site C0009 and used this velocity function to correlate the wireline, cuttings, and core data with time-based seismic data (Fig. F5). The velocities from the sonic log and check shot are lower than those used for 3-D seismic processing prior to this expedition. This is primarily due to a zone of low velocity and VP/VS ratio from ~1030 to 1200 m WMSF in the wireline sonic data that corresponds to a zone of increased free gas in Unit III (Figs. F5A, F8).

We identified several prominent seismic surfaces that can be traced regionally within the Kumano Basin (Unconformities UC1–UC4; Figs. F4, F6). The uppermost of these surfaces, Unconformity UC1, is a prominent reflector within Unit II at ~600 mbsf (Fig. F6A). Unconformity UC2 is located at ~750 m WMSF, 40 m above the interpreted Unit II/III boundary (Figs. F5A, F6A). Unconformity UC3 is a prominent positive polarity reflector and marks an angular unconformity at the base of the low-velocity zone in Subunit IIIB (Figs. F5A, F6A), and we interpret that it is caused by the strong increase in impedance at the base of the gas zone. Unconformity UC4 lies at the Unit III/IV boundary and is marked by a decrease in density and an increase in velocity; the net result is a positive increase in impedance (Fig. F5A).

Each lithologic unit is imaged as a distinct seismic facies. Lithologic Unit I is characterized by laterally continuous high-amplitude seismic reflections (Fig. F6A). These reflections are largely parallel; however, in the upper part of the section, they converge and onlap to the south-southeast (apparent direction in the plane of the seismic cross section shown in Fig. F6B) as Unit I thins. We interpret that the well-stratified and laterally continuous reflections are caused by abundant 10–50 m thick sand cycles as recorded in GR data that were deposited in deep water as turbidite deposits. Unit II is characterized by lower frequency reflections and slightly reduced amplitude relative to Unit I (Fig. F6A). The reflectors are generally not continuous; many intersect each other when traced laterally. The low amplitude in Unit II is broadly consistent with an interpretation of silty clay with few silt or sand interbeds. There are several bright reflections within Subunit IIIA, but the unit is generally less reflective than overlying units (Fig. F6A). The upper part of Subunit IIIB is relatively transparent and corresponds to a zone with uniformly low velocity abundant wood fragments identified in cuttings and elevated mud gas methane concentrations.

Unconformity UC3 lies near the base of Unit III and is interpreted as an angular unconformity. It truncates underlying reflectors and marks a transition from a transparent section (containing free gas) above to high-amplitude tilted reflectors in a statified basin below. GR values are slightly lower above than below Unconformity UC3, and we interpret this to record a lithologic change. Unconformity UC4 lies at the Unit III/IV boundary and separates reflective strata above from less reflective but more chaotic and discontinuous reflectors below. Reflections within Unit III onlap Unconformity UC4 with an apparent north-northwest–south-southeast direction.

Site C0010

Operations at Site C0010 included drilling with MWD/LWD across the megasplay fault to a TD of 555 m LSF, casing the borehole (with casing screens at the fault), conducting an observatory dummy run to test strainmeter and seismometer deployment procedures, and installation of a simple pore pressure and temperature monitoring system (smart plug). Although the smart plug is relatively simple, it marks the first observatory placement in NanTroSEIZE. All of the planned science objectives for Site C0010 were achieved; although casing operations were adjusted to fit hole conditions after drilling to TD (560 m DSF) with casing to 500 m DSF instead of the planned 525 mbsf outlined in the prospectus. MWD/LWD data (gamma ray and resistivity, including resistivity at bit images) were collected allowing (1) definition of major lithologic unit boundaries and of the shallow megasplay fault zone and (2) determination of the preferred placement of the screened joints interval within the fracture zone interval for the temporary observatory. Through comparison with previously drilled Site C0004 (Kinoshita, Tobin, Ashi, Kimura, Lallemant, Screaton, Curewitz, Masago, Moe, and the Expedition 314/315/316 Scientists, 2009) these data also provide insights into along-strike differences in the architecture of the megasplay fault and hanging wall.

After drilling the hole and in preparation for a future permanent observatory installation, a dummy sensor run was carried out to evaluate reentry operations during instrument deployment. After casing was completed and the borehole cemented, two dummy run tests were conducted, including adjustments for the effects of the Kuroshio Current, which reached speeds of 4.5 kt during the experiment. After the dummy run reentry simulations were completed, the smart plug was installed for short-term (1–2 y) data collection and storage to monitor temperature and pore pressure within the megasplay fault zone.

Lithology

We defined three logging units at Site C0010 on the basis of LWD/MWD measurements (GR and bit resistivity) (Fig. F10) and guided by previous results at nearby Site C0004 (Expedition 314 Scientists, 2009; Expedition 316 Scientists, 2009) (Fig. F11). Although in detail the sites are not identical (e.g., see "Log-Seismic integration"), the observations at Site C0004 provide valuable constraints on lithologic variations at Site C0010 (Kinoshita et al., 2008; Kimura et al., 2008). In particular, data from previous drilling at Site C0004 show that lithologic changes are best defined based on GR measurements.

We define three distinct lithologic packages at Site C0010. From top to bottom, these are slope deposits (Unit I, 0–182.5 m LSF), thrust wedge (Unit II, 182.5–407 m LSF), and overridden slope deposits (Unit III, 407 m LSF to TD). Logging Unit I is divided into two subunits. Subunit IA (0–161.5 m LSF) is characterized by GR and bit resistivity patterns similar to logging Unit I at Site C0004 (Figs. F10, F11) and we interpret it as hemipelagic slope sediments composed primarily of mud with minor distal turbidite interbeds. We interpret Subunit IB (161.5–182.8 m LSF) as slope sediments composed of material reworked from the underlying thrust wedge.

Logging Unit II (182.5–407 m LSF) is a thrust wedge comprising the hanging wall of the megasplay fault and is correlated with logging Unit II and lithologic Units II and III at Site C0004. The thrust wedge at Site C0010 has higher GR than at Site C0004, which may indicate higher clay content (Fig. F11). Logging Unit III (407 m LSF to TD) is composed of slope sediments overridden by the thrust wedge and correlates with logging Unit III and lithologic Unit IV at Site C0004. On the basis of the LWD data and coring results from Site C0004, we interpret Unit III as hemipelagic muds with minor turbidite interbeds and rare volcanic ash layers.

Structural geology and geomechanics

We measured the attitudes of faults, bedding, and breakouts from LWD resistivity image data. As a result of ship heave during logging operations, the resistivity image data exhibit variable quality. In most of the bedding data, we identified eastward dips of ~45°–60°. However, there is considerable scatter in both dip magnitudes and bedding orientations, and because of limited data quality there are a limited number of observations (N = 11). Faults dip to the west and south, with most dips ranging from ~40° to 80°. We observe the highest concentration of faults near the base of the thrust wedge. Notably, the two logging runs through the lowermost ~60 m of the thrust wedge (2900–2970 m LRF) are markedly different. The first logging run (during drilling) exhibits faults with a wide range of dips, generally to the south and west. The second run, which was conducted after reaming in an open hole, exhibits steeply dipping faults (>45° and mostly >60°), and shallowly dipping faults like those noted in the first logging run are not observed.

Borehole breakouts show that SHmax trends northwest–southeast, similar to other sites on the outer slope along the NanTroSEIZE transect (Sites C0001, C0004, and C0006) (Kinoshita et al., 2008; Tobin et al., 2009) (Fig. F7). A sharp discontinuity in stress orientation occurs across the base of the thrust wedge and is consistent with a fault discontinuity (Barton and Zoback, 1994). In contrast, the trend of breakouts at Site C0004 changes gradually downhole and is continuous across the base of the wedge (Kinoshita et al., 2008). The enlargement of breakouts during the interval between the two logging runs indicates that the breakouts grew with time in this environment, in contrast to observations from more lithified rocks (Zoback, 2007).

Physical properties

Physical properties from logging data collected at Site C0010 include GR, bit resistivity, shallow-, medium-, and deep-button resistivity, and ring resistivity (Fig. F10). In addition, we estimated porosity from resistivity. However, this approach is limited by the fact that there are no data from cores, cuttings, or logging to calibrate the transform from resistivity to porosity (e.g., Kinoshita et al., 2008; Conin et al., 2008).

Ring resistivity exhibits an overall increase downhole, ranging between 0.7 and 0.9 Ωm in the slope sediments of Unit I (0–182.5 m LSF). Resistivity in the stratified overridden slope deposits of Unit III (407–554 m LSF) below the thrust wedge (Unit II) follows the same trend, with values between 0.8 and 1.2 Ωm. Resistivity values in both Units I and III are similar to those for slope deposits at Site C0004 (Fig. F11). The thrust wedge (Unit II) is characterized by significantly higher overall resistivity than the slope deposits above and below. In the lower portion of Unit II (260–407 m LSF) we observe considerable scatter in resistivity, with fluctuations from 1.5 to 2.5 Ωm over distances of ~10–20 m.

The overall higher resistivity and the fluctuations in resistivity in the thrust wedge correlate tightly with increased overall GR values and excursions to lower GR values, respectively (Figs. F10, F11). The overall higher resistivity values may reflect increased compaction within the thrust wedge relative to the overlying and underlying slope sediments, increased clay content leading to increased tortuosity, or a combination of the two. We suggest that the large fluctuations superimposed on the overall trend reflect either interbedded coarser grained zones or more intensely faulted and fractured intervals.

We estimated porosity from resistivity using parameters for Archie's law derived at nearby Sites C0001 and C0004 that also penetrated slope sediments and the underlying wedge (Kinoshita et al., 2008; Kimura et al., 2008; Conin et al., 2008), where both logging and core data were available. For the slope apron (Unit I) and underthrust (Unit III) sediments at Site C0010, an exponential porosity-depth trend (Athy, 1930) fits the computed porosities well. In contrast, the estimated porosity of the thrust wedge is markedly lower because of its considerably higher resistivity, although as noted above the higher resistivity may be partly due to increased clay content. Preliminary calculations suggest that lithologic effects are unlikely to fully explain the magnitude of resistivity increase within the thrust wedge, indicating that it may reflect increased compaction state related to its burial history or increased in situ stress in the wedge relative to the slope sediments.

Log-Seismic integration

We used the time-depth data acquired at Site C0004 (Expedition 314 Scientists, 2009) to correlate the logs to the seismic data at Site C0010. The Unit I/II boundary marking the top of the thrust wedge is imaged with a weak positive polarity seismic reflection that is coincident with a sharp increase in resistivity (Figs. F10, F11). The Unit II/III boundary between the base of the thrust wedge and overridden slope sediment is recorded by a prominent negative polarity reflection. At this boundary, the resistivity drops markedly. We interpret that velocity and density also decrease at this boundary, resulting in the negative impedance contrast.

At Site C0010, GR values in the thrust wedge increase gradually from 65 API at the top to 120 API at the thrust wedge center and decrease to 80 API at the base (Fig. F10). Several fluctuations in GR values with depth are superimposed on this overall trend, with values ranging from 60 to 80 API units; resistivity cycles within the thrust wedge parallel decreases in GR values. We observe greater seismic reflection amplitudes within the thrust wedge at Site C0010 than at Site C0004 (Fig. F11; cf. Fig. F4B). We conclude that the fluctuations in composition or physical properties within the thrust wedge at Site C0010 recorded in GR and resistivity drive differences in velocity and density (impedance) that generate seismic reflections (see "Lithology" and "Physical properties"). These results suggest significant variation in composition or other properties in the thrust wedge over a scale of only a few kilometers. A striking feature of seismic profiles across the thrust wedge is that the negative polarity reflector is weak at the tip of the thrust wedge and increases where it is more deeply buried (Fig. F4B). The most likely explanation for this is that there is increasing consolidation in the thrust wedge relative to the underlying material.

Observatory

Sensor dummy run test

As part of our operations at Site C0010, we conducted two sensor dummy run tests to simulate future installation of long-term observatory instruments, with the goal of documenting vibration and shock associated with running the instrument package through the water column and reentering the borehole. During the first dummy run, two seismometers, an accelerometer-tiltmeter, a strainmeter, and nine miniature temperature loggers (MTLs) were attached to the sensor tree (Fig. F12A, F12B). In addition, two full joints and one pup joint of tubing were attached below the strainmeter to replicate the planned future installation procedure. Because of the high current in the area of Site C0010 (~4–5 kt surface current), we ran the instrument carrier to ~500 m DRF in a low-current area (LCA) and then drifted toward the drill site. Unfortunately, while drifting to the site for reentry, one seismometer was dropped from the instrument carrier, and the strainmeter was detached from the bottom of the carrier because the current caused strong vibrations of the drill pipe. After visual confirmation that these components were lost (by ROV at ~1650 meters below sea level [mbsl]), we retrieved the instrument carrier without reaching the seabed. However, acceleration, tilt, and temperature data in the water column were collected during this dummy run.

After the first run, we repaired the instrument carrier and conducted a second dummy run using the accelerometer-tiltmeter attached to the instrument carrier to evaluate shock, acceleration, and vibration during reentry. This run also included a dummy strainmeter, with similar dimensions and mass to a real strainmeter, and two full joints of 3½ inch tubing attached below the instrument carrier. The bottom of the sensor tree was reentered into the wellhead three times to 5–7 m DSF. The reentry operations were generally very smooth; during the third reentry, the tubing below the sensor tree hit against the reentry cone two times. After the sensor tree was recovered, the accelerometer-tiltmeter was checked. Unfortunately, the recording stopped because of damage sustained by vibration in the water column and no acceleration and tilt data were obtained for the reentry test.

Smart plug installation

After the observatory dummy runs were completed, we suspended Hole C0010A by installing a smart plug instrument package below a mechanically set retrievable casing packer (Fig. F12C, F12D). The retrievable packer was set inside casing above two screened casing joints; the smart plug and screen placement in the casing were configured to continuously monitor pore pressure and temperature in an isolated interval of formation including the splay fault and also monitor hydrostatic pressure as a reference (Fig. F12C). The smart plug contains two high-precision pressure transducers with period counters and four temperature sensors (one as part of each pressure gauge for compensation, one platinum chip thermistor, and one stand-alone MTL) in a shock-proof housing (Fig. F12D). The self-contained instrument has a recording lifetime of ~7 y.

The smart plug is designed to thread onto the casing packer. Based on difficulties with the dummy run caused by the strong Kuroshio Current, we further secured the smart plug to the crossover below the packer by tack-welding prior to running it into the water column (Fig. F12D). The smart plug safely entered the hole and the packer was set at 364 m DSF. Retrieval of the bridge plug and instrument package is anticipated for 2010 or 2011 when a more sophisticated long-term monitoring system will be deployed at Site C0010 (Fig. F13).