From their analyses of samples from Sites 417 and 418, Christensen et al. (1980) and Carlson and Herrick (1990) showed that the grain densities of oceanic basalts decrease with increasing porosity and suggested that a decrease in grain density reflects an increase in the relative amounts of alteration products. The grain density of the freshest basalt measured in the suite of samples from this study is 2910 ± 20 kg m^-3, whereas, the average measured grain density of two almost entirely altered samples is 2100 ± 60 kg m^-3. Therefore, we can use grain density as an alteration index, and according to Christensen et al. (1980) and Carlson and Herrick (1990), an increase in relative amounts of alteration products (lower grain densities) should correlate with increasing porosity. The porosities listed in Table 1 correlate inversely with grain densities (or rather, the porosities directly correlate with relative amount of alteration) as shown graphically in Figure 5A.

It has also been suggested that the infilling of pore space by alteration products initially closes or heals the thinnest aspect ratio pores, or cracks, because they require the least amount of material to fill (e.g., Wilkens et al., 1991). If this is the case, then there should be a relationship between the amount of pore space represented by thin cracks and the amount of alteration products represented by grain density in the samples. As can be seen in Figure 5B, there is a direct relationship between the percentage of total porosity with aspect ratios 0.01 and grain density, which is an indicator of the amount of alteration. Samples that are more altered have a smaller percentage of the total pore space present as thin aspect ratio voids, even though the bulk porosity is higher in these samples. This is a second-order effect compared to the normalized, average crack spectrum mentioned above, but it does suggest that the thinner cracks are the first to become effectively sealed by alteration products, assuming that the rocks had similar aspect ratio spectra when they were fresh.

Tosaya (1982) measured the P- and S-wave velocities and densities of clay-bearing sandstones from the Gulf of Mexico and calculated the bulk and shear moduli from the data. Dvorkin and Nur (1996) later extrapolated the bulk and shear moduli values calculated by Tosaya (1982) to predict the moduli of a zero-porosity rock composed of 100% clays (assumed to be smectites) and reported bulk and shear moduli of 21 and 7 GPa, respectively, for the clay (smectite) minerals. The clay moduli values of Dvorkin and Nur (1982) are lower than the basalt grain bulk and shear moduli values reported by Toksöz et al. (1976), 95 and 42 GPa, respectively. The addition of clay minerals into a basaltic rock will thus lower the grain moduli of the rock in proportion to the relative amount of clays added (Hashin and Shtrikman, 1963).

The apparent grain bulk and shear moduli were used as free parameters in the inversions. Assuming that the Kuster-Toksöz model is a valid construct and assuming that the alteration products consist largely of clays, the apparent grain moduli derived from the models should directly correlate with grain density (or inversely with relative amounts of alteration). The best fitting bulk and shear grain moduli of the samples from Hole 990A are plotted vs. grain density in Figure 6. A direct linear relationship is observed between the apparent bulk and shear grain moduli from the models and the grain densities for this suite of samples, suggesting that the elastic moduli of the solid are significantly affected by alteration. Figure 6 thus shows how an increase in the percentage of low density minerals (presumably alteration products such as clays) lowers the apparent bulk and shear moduli of the grains. Using the least-squares fit to the modeling results (Fig. 6), when the grain density lowers from 2970 to 2740 kg m^-3 (a 7.7% reduction), the P-wave velocity lowers from 6.85 to 6.36 km s^-1 (a 7.1% reduction), and the S-wave velocity lowers from 3.85 to 3.66 km s^-1 (a 5.0% reduction).