Bulk marine sediment comprises multiple components; therefore, the terrigenous component (eolian and hemipelagic material) needs to be isolated for study. Previous work (e.g., Mehra and Jackson, 1960; Rea and Janecek, 1981; Hovan and Rea, 1992; Olivarez Lyle and Lyle, 2002) has developed a method to chemically extract the nonterrigenous components. The general procedure involves removing carbonate with weak acetic acid, removing metalliferous oxides and hydroxides with a buffered sodium citrate–sodium hydrosulfite mixture, and removing biogenic silica with one of several sediment-dependent steps. In this study, we adopted a specific procedure for extracting the terrigenous component, which is outlined in the "Appendix."
Once extracted, the terrigenous component was homogenized and the grain-size distribution of all 20 samples was examined using a Coulter Multisizer II housed in the Civil and Environmental Engineering Department at Rice University (Houston, Texas). This device measures the electronic resistance of individual particles and divides results into 256 channels over a size range applicable to the material being analyzed, in this case 1–30 µm. Previous work (Poppe et al., 2000) and repeat analyses conducted for this study indicate that the precision of this instrument is within ±10%.
Nominally, 0.1 g of isolated terrigenous material and 15 mL of 0.008-M sodium hexametaphosphate were placed into a clean, labeled sample vial. The solution was sonicated for 5 min and mixed overnight in a hand-action shaker. After running clean electrolyte solution through the Coulter Multisizer II for 3 min to obtain a background count, ~2 mL of stirred solution was slowly added to clean electrolyte solution until a desired sample density was met. Following subtraction of background counts, grain-size measurements were collected by averaging three 3-min sample runs on the Coulter Multisizer II. This method typically allows counting of ~100,000–200,000 individual particles.