METHODOLOGY AND STRATIGRAPHY

Sediment samples were taken at about every 50-100 cm of the Pliocene-Pleistocene level of Hole 1082A. The biogenic opal content (see "Appendix") was determined by two methods based on wet chemical extraction techniques:

1. "OLin" data: At the National Sun Yat-Sen University (Taiwan), Lin analyzed 80 samples over a depth interval from 121.86 to 261.96 meters below sea floor (mbsf), following the basic leaching method of Mortlock and Froelich (1989). This method involves a single extraction of silica with an alkaline solution at 85° C for 5 hr and the measurement of the dissolved silicon concentration in the extract by molybdate-blue spectrophotometry. Instead of the 2-N Na2CO3 suggested by Mortlock and Froelich (1989), 0.5-N NaOH was employed as the basic leaching chemical to ensure complete dissolution. The HCl was replaced by a milder 0.5-M glacial acetic acid solution to help prevent the breakdown of silicates, which would result in overextraction of silicate from the sediment (Murray et al., 1995).

2. "OPer" data: At the University of Bremen (Federal Republic of Germany), Pérez analyzed 142 samples from 100.25 to 250.47 mbsf and 104 samples from 298.6 to 399.5 mbsf using the automated leaching method of Müller and Schneider (1993), which is a modification of the manual sequential leaching method of DeMaster (1981). While stirred constantly, the opal is extracted with 1-M NaOH at 85° C in a stainless steel vessel. The extraction time varies according to the sample type (e.g., sediment type, opal concentration, and presence of clay minerals), and the increase in dissolved silica is continuously monitored as a minor portion of the leaching solution is cycled to an autoanalyzer and analyzed for dissolved silicon by molybdate-blue spectrophotometry. The resulting absorbance vs. time plot is then evaluated following the extrapolation procedure of DeMaster (1981). The weak point in the manual technique is that the slope—for example, the increase in dissolved silica extracted with time—and the extrapolated intercept value are based only on a few measurements. The automated technique of Müller and Schneider (1993) reduces the degree of uncertainty in the determination of the intercept.

Sediment samples from Hole 1084A were taken at about every 150 cm throughout the uppermost 476 mbsf, which represents approximately the last 3 m.y. For this site, the biogenic opal content was determined by the modified "OLin" method only.

Overall diatom abundance and species composition were determined based on smear-slide analysis with phase-contrast illumination at a magnification of 400×. A diatom abundance index (DAI) was established using the following convention: 6 (very abundant or dominant) = >50%; 5 (abundant) = 35%-50%; 4 (common) = 20%-35%; 3 (few) = 5%-20%; 2 (rare) = 1%-5%; 1 (trace) = <1%; 0 (barren) = no diatoms in sample (Lange et al., 1999).

After converting depths (mbsf) to meters composite depth (mcd), age models at Sites 1082 and 1084 were derived according to shipboard magnetostratigraphy and nannofossil datums (Shipboard Scientific Party, 1998a, 1998b).