The results of grain-size analyses, including duplicate runs, are listed on Table 1. In a relatively small number of instances, cumulative curves displayed erratic geometries because of low suspension concentrations or sharp deflections in slope that we regarded as spurious instrument behavior. These samples are identified in Table 1, but their results are not included. All of the reliable data are plotted vs. sample depth at each site. Regression plots also show either weight-percentage clay or mean grain size vs. the sediment index properties (water content, porosity, void ratio, and bulk density). Figure 2 and Figure 3 illustrate results for Site 1023. Similar graphics for the other sites are included in Figure 6, Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12, Figure 13, Figure 14, Figure 15, Figure 16, Figure 17, Figure 18, Figure 19, Figure 20, and Figure 21.
Percentages of sand, silt, and clay do not show any consistent trends as a function of depth at any site. Similarly, there seems to be very little change in the texture of hemipelagic mud from Subunit IA to Subunit IB to Unit II. Sampling of sandy and silty turbidite layers during shipboard measurements was not done in a systematic manner, so the specific depths where we plot excursions toward coarser grain sizes are not representative of the actual turbidite distribution in the cores. In addition, no attempt was made during sampling to distinguish between hemipelagic mud and turbidite mud. Sand-sized grains constitute less than 1% of most muds, and the clay content is typically between 60% and 85% (Fig. 2). Most of these silty clays are moderately well sorted, and skewness values are typically between 1.0 and 0.7. In general, mean grain size for the silt + clay fraction of the muds ranges from 1 to 4 µm, whereas mean grain size for the silt + clay fraction of sandy turbidites is typically between 10 and 20 µm.
Combining all of the data from all of the cores shows a clear segregation between the mud and sand/silt lithologies (Fig. 4). Regression plots show considerable sensitivity of index properties to the content of clay-sized particles (Fig. 3). Water content, porosity, and void ratio generally increase with increasing clay content, whereas bulk density tends to decrease. Correlation coefficients for the linear regressions range from 0.03 to 0.79 (Fig. 3, Fig. 7, Fig. 9, Fig. 11, Fig. 13, Fig. 15, Fig. 17, Fig. 19, and Fig. 21). For most of the data populations, these coefficients indicate that the correlations between clay content and index properties are statistically significant at a confidence level of 95%, but it is also clear that other factors are involved. Superimposed upon the effects of grain size are the mechanical changes that occur with depth-dependent compaction. The compaction gradients for muds differ significantly from those of silt and sand turbidites (Shipboard Scientific Party, 1997b, 1997d, 1997a). Most of the borehole successions display systematic effects of compaction within the mud component, but dewatering is most pronounced at Site 1027, where sediment thickness reaches 600 m. In contrast, the initial porosities of turbidites tend to be significantly lower, and there is less dewatering, if any, with depth.
At Sites 1030 and 1031, mud porosities remain high (65%-80%) throughout the relatively thin sediment cover (<45 m). The pore-water profiles at both sites also show clear evidence of upward fluid flow (Shipboard Scientific Party, 1997a). The textural characteristics of these muds appear to be no different than those of hemipelagic deposits at the other sites (Fig. 18 and Fig. 20). As discussed below, their somewhat unusual physical properties probably persist because the overburden at Sites 1030 and 1031 is too thin to collapse the grain fabric inherited from suspension fallout.