b = m/V). All variables are defined in the nomenclature table (Table T2). Porous endcaps at the top and bottom of the cylindrical samples facilitated even drainage from the sample ends.
Whole-round core samples were sealed in the core liner during Leg 194 to preserve the natural sample saturation. For each experiment, a core was removed from its liner and a vertically oriented, cylindrical subsample was trimmed and inserted in the fixed-ring consolidation cell. Each vertically oriented subsample had a diameter of 47.9 mm and an initial height of 19 mm. The mass (m) and volume (V) of each sub-sample was measured to calculate bulk density (b = m/V). All variables are defined in the nomenclature table (Table T2). Porous endcaps at the top and bottom of the cylindrical samples facilitated even drainage from the sample ends.
Initial sample porosity (
o) was calculated with equation 1 (Fig. F2; Table T1), assuming a water density (w) of 1000 kg/m3, a grain density (s) of 2600 kg/m3, and 100% water saturation:
o = (s – b)/(s – w). (1)
The initial porosity calculations neglect salt in the pore fluid. Calculations that account for salt mass in the system (Blum, 1997) result in less than 0.7% change in initial porosity.
Ten consolidation experiments were performed; five on samples from Site 1194 and five on samples from Site 1198 (Table T1; Fig. F2). Each sample was placed in the load frame, where contact between the sample, porous disks, and ram were established manually. Vertical stress was then computer-controlled throughout the consolidation experiment. Stress was increased at 0.1-MPa increments to ~0.5 MPa and then by 0.5-MPa increments to ~4 MPa. The samples were then unloaded to 1 MPa, reloaded to 4 MPa, and then loaded by 0.5-MPa stress increments to 6 MPa. Experiments concluded with unloading at 1-MPa intervals. Each stress increment (increase or decrease) occurred at ~0.1 MPa/s and was followed by a stress hold to allow excess pressure dissipation and completion of primary consolidation. After the completion of primary consolidation and onset of secondary compression, another stress increment was applied. Stress holds for the completion of primary consolidation were <15 min for these samples. Analysis of sample height as a function of time (log-time method) (Lambe and Whitman, 1979; Craig, 1992) was used to confirm the completion of primary consolidation, dissipation of excess pore pressure, and onset of secondary compression. Sample height was measured throughout the experiments and sample diameter was constant; this allowed sample volume and porosity to be calculated throughout the entire experiment with equation 2 (nomenclature is defined in Table T2):
v´ + 
v´ = [Vv´v´ + (Vv´ + v´ – Vv´)]/Vv´ + v´ . (2)
The sample void ratio (e = /[1–]) and stress at the end of each stress hold are used to calculate the compression (cc) and expansion (ce) indices of the samples. The compression index in equation 3 characterizes plastic deformation along the linear portion of the e-log(v´) plot, which is interpreted to represent primary (virgin) consolidation (e.g., Craig, 1992):
v´ – ev´ + v´)/log[(v´ + v´)/(v´)]. (3)
The expansion index in equation 4 describes the elastic portion of the e-log(v´) plot characterized by the linear unloading/reloading paths (e.g., Craig, 1992):
v´ – ev´ + v´)/log[(v´ + v´)/(v´)]. (4)
Sample deformation during stress holds provided estimates of the coefficient of consolidation (cv) and of the permeability (k) of the samples at multiple consolidation states during virgin deformation (Table T3). The log-time method was used to estimate cv for primary consolidation (see Lambe and Whitman, 1979; Craig, 1992), and k was calculated using cv with equation 5:
where
e/v´]) for a given effective vertical stress increment (v´),
e = the change in void ratio during the stress hold (Table T2).
A Malvern Master-Sizer Dynamic Laser Light Scattering/Light Diffraction Particle Size Distribution Analyzer was used to determine grain size on seven samples from Sites 1194 and 1198. The particle size analyzer can measure particle sizes from 0.05 to 900 µm.
We evaluated the grain size distribution for each core section in the study (Table T3).
Dry sediment samples were mixed with deionized water to make a sediment-water mixture. The mixture was placed in a sonicator and stirred to establish a uniform concentration of particles in the water. The mixture was then analyzed to evaluate the grain size distribution; d10, d50, and d90 are reported (d10 = the grain size at which 10% of sample is finer, d50 = the grain size at which 50% of sample is finer, and d90 is the grain size at which 90% of sample is finer) (Tables T2, T3). Duplicate and triplicate analyses were performed and standards were analyzed to demonstrate reproducibility of the distributions.
