Previous   |    Next

doi:10.2204/iodp.proc.322.209.2014

Methods

Sampling

We tested 27 whole-round samples of mud and mudstone from Sites C0011 and C00012 in the Shikoku Basin (Table T1). Their distributions by depth and lithostratigraphic unit are shown in Figure F2, as are bedding dips from intervals in close proximity. Six samples from Site C0018 were also tested, four from a mud-rich slope-basin facies and two from a sand-rich facies. The whole-round samples were cut onboard the D/V Chikyu within several hours of recovery, capped and taped, sealed with wet sponges in aluminum vacuum bags, and maintained at ~4°C during shipment and storage to prevent moisture loss. Immediately prior to laboratory testing, the samples were extruded from the core liners and a trimming jig was used to shape cylindrical specimens.

Moisture and density

After trimming each cylinder of mudstone for a CRSC test, we retained two or three pieces of trimmings to measure water content by oven drying to constant mass at 105°C (ASTM, 2006). The value of water content is equal to the difference in weight before and after oven drying divided by the oven-dried weight (Blum, 1997; Expedition 316 Scientists, 2009). The calculated values of void ratio (e) include a correction for salt in the pore water. Those values can be compared to shipboard data from nearby sample intervals to assess sample disturbance and/or loss of moisture during shipment and storage. In some instances, shore-based e values are greater than shipboard e values; the higher values can be attributed to drilling/coring disturbance (e.g., dilation of microfractures) or subtle differences in the mudstone’s mineral composition and/or texture between the adjacent intervals. Deformation of the core was particularly acute during Expedition 322 with use of the rotary core barrel system.

Consolidation tests

CRSC tests were conducted at the University of Missouri (USA) in the oedometer system described by Guo et al. (2011), following the protocols and configurations specified by the American Society for Testing and Materials (ASTM, 2006). According to these procedures, each cylinder of sediment is placed in a stainless steel specimen ring (diameter = 4.14 cm) to maintain a condition of zero lateral strain. The consolidation cell and lines are then evacuated of air using a vacuum pump. The specimen is backpressured to ~200–400 kPa using de-aired synthetic seawater (1.75 g NaCl in 500 mL distilled water) for 24 h to ensure saturation and to dissolve any air remaining in the system lines. During backpressuring, the axial load actuator maintains height with a strain of 0.2%. Constant-rate-of-strain loading is applied using a computer-controlled load frame, with the sample base undrained and sample top open to the backpressure. The sample height, the applied vertical total stress, and the basal pore pressure are constantly monitored. The maximum axial load of our loading frame is 44 kN; for specimens with a diameter of 4.14 cm, that load corresponds to a maximum vertical total stress of 33 MPa. With the backpressure ranging from 200 to 400 kPa, the maximum vertical effective stress ranges from approximately 32 to 33 MPa. For each test, the rate of displacement (or strain rate) is specified in order to keep the anticipated ratio of basal excess pore pressure to total axial stress below 0.10. Displacements are measured by a linear variable differential transformer (LVDT) mounted at the top of the consolidation cell, and the displacements are corrected after ASTM (2006) to account for the compliance of the testing system. The resulting strain rates are documented in Table T2.

As documented previously by Guo et al. (2011), the following equations (ASTM, 2006; Long et al., 2008) were used to compute values of axial strain (ε), base excess pressure (Δu), average effective axial stress (σ′_v), hydraulic conductivity (K), intrinsic permeability (k), the coefficient of volume compressibility (m_v), and the coefficient of consolidation (c_v):

and

All of the symbols, definitions, dimensions, and units of measure for these and other equations are defined in Table T3. During a CRSC test, hydraulic conductivity values are not reliable until the axial strain distribution reaches steady state (ASTM, 2006). Conversion of each value of vertical hydraulic conductivity under laboratory conditions to vertical intrinsic permeability (Table T2) requires knowledge of permeant properties; for that, we assumed a fluid viscosity of 0.001 Pa·s for water at 20°C and a fluid density of 1027 kg/m³. Calculations of in situ vertical hydraulic conductivity (Table T2) similarly requires knowledge of temperature-dependent changes in permeant properties within the formation; we assumed interstitial water salinity of 35‰ and followed the equations of El-Dessouky and Ettouny (2002) and Fofonoff (1985), as constrained by measurements of borehole temperature (Expedition 333 Scientists, 2012a, 2012b). After each test was finished, sample disturbance was evaluated by comparing the Δe value to e_i (Lunne et al., 1997). Most of the samples rate “poor” to “very poor” using this criterion (Table T1).

For each set of test results, the compression index (C_c) refers to the slope of the virgin portion of the e versus log(P) curve (Fig. F3). The compression index is calculated using the following equation:

Consolidation test results can also be used to define a relation between intrinsic permeability and porosity in the form: log k = log k_o + a × n, where a is a constant (e.g., Neuzil, 1994). The in situ intrinsic permeability is determined by numerically fitting a regression curve through the middle of each band of test results and extrapolating along that line to the value of in situ porosity or void ratio (Long et al., 2008). The value of in situ void ratio is picked from each consolidation-test curve at the point of intersection with maximum past effective normal stress (P′_c) (Fig. F3). That point is also referred to as “preconsolidation stress” (e.g., Dugan and Daigle, 2011). To solve for each specimen’s value of P′_c, we employed both the Casagrande (1936) method and the strain energy density (SED) method (Becker et al., 1987). In the case of one-dimensional consolidation,

We also compare each P′c value to the calculated value of in situ hydrostatic vertical effective stress (σ′_vh). The gradient of hydrostatic vertical effective stress at each site was calculated by subtracting the value of hydrostatic pore pressure from the overburden pressure (total normal stress). The overburden pressure was computed by integrating the shipboard values of bulk density over the depth range of drilling (Expedition 322 Scientists, 2010a, 2010b; Expedition 333 Scientists, 2012a, 2012b, 2012c). Under conditions of monotonic and uniaxial loading, a specimen’s P′_c value should be equal to σ′_vh at each equivalent sampling depth (e.g., Holtz and Kovacs, 1981), but there are many reasons why that expectation might not hold. To quantify such departures, the overconsolidation ratio (OCR) is defined as P′_c/σ′_vh. An OCR of unity indicates normal consolidation. Values of OCR < 1 indicate underconsolidation, a condition that is often caused by fluid pressures greater than hydrostatic. Values of OCR > 1 indicate overconsolidation. True overconsolidation usually results from erosional unroofing of overburden, whereas an apparent state of overconsolidation can result from lateral tectonic stresses greater than vertical effective stress or from strengthening of sediment fabric by cementation.

Top of page   |    Previous   |    Next