As a part of the shipboard safety and pollution program, volatile hydrocarbons (methane, ethane, and propane) were measured in the sediments of Site 1091 from every core in Hole 1091A using the standard ODP headspace sampling techniques. Results are presented in Table T12 and Figure F19. Headspace methane concentrations were generally low (2-22 parts per million by volume [ppmv]) throughout the sedimentary sequence at Site 1091. Ethane, propane, and other higher molecular weight hydrocarbons were not observed.
Shipboard chemical analyses of the interstitial water from Site 1091 followed the procedures for Sites 1088-1090. The results from the shipboard analyses (Table T13; Fig. F20) were obtained from 33 interstitial water samples from Hole 1091A to a depth of 307 mbsf. Interstitial water samples were taken from every core throughout the entire section.
Unlike previous sites, Site 1091 has
relatively low calcium carbonate (CaCO3) content (except in a few narrow
horizons; see Fig. F21; "Solid Phase Analysis", "Lithostratigraphy")
and very abundant diatom opal. This distinct sedimentary composition results in subtle,
but observable, differences from previous sites in the interstitial water profiles of
chemical species determined by shipboard analyses. First, the chlorinity maximum, possibly
associated with the last glacial period, is somewhat more pronounced (with a single high
value of 571 mM) and is located at a deeper depth (50 mbsf) than at Sites 1088-1090. The
salinity maximum is also observed clearly in the Na+ profile at Site 1091
because of the lack of significant gradients in the upper part of the section for the
major cations and anions other than Cl- (Na+ was calculated by
charge balance). There is a slight minimum in Cl- of 565 mM at 127-137 mbsf.
These same features are reproduced in more detail at Site 1093. At present, it is not
clear why the Cl- profiles at Sites 1091 and 1093 are so distinct from those of
previous sites, but there are two important sedimentological characteristics that might
affect the diffusion process. First, the sedimentation rate is fairly high (the average is
~100 m/m.y. in the upper 100 m), perhaps resulting in an accumulation that was rapid
enough to deepen the Cl- maximum. Second, numerous diatom mats are present
between 85 and 230 mbsf (see "Lithostratigraphy")
that may restrict diffusion enough to preserve a lower salinity signal from previous
interglacial periods. Of course, it is inappropriate to speculate further without
additional shore-based measurements of 
18O and diffusional modeling. Also, given the limitations of
shipboard analyses, there is little that can be said about the decrease in Cl-
below 250 mbsf, except that it presumably results from the influence of lower salinity
waters deeper in the sediment.
The low CaCO3 content of Site 1091 sediments produces Ca+2, Mg+2, and Sr+2 profiles that show little or no evidence of carbonate diagenesis. Ca+2 concentrations are essentially constant at near-bottom-water concentrations (~10.5 mM) down to ~90 mbsf and then decrease slightly to concentrations of ~7 mM in a broad minimum from 200 to 250 mbsf. The decrease in Ca+2, which is not accompanied by any apparent change in Mg+2 concentrations, may reflect authigenic calcite precipitation resulting from the increase in alkalinity. Mg+2 and Sr+2 concentrations are essentially constant throughout the section within the analytical uncertainty of these determinations.
K+ concentrations are also constant throughout the section at near-bottom-water concentrations of ~10.4 mM, suggesting little net interaction of K+ with clay minerals or basement (which is at ~1200 mbsf at Site 1091). Li+ does show some evidence for small decreases from bottom-water concentrations in the uppermost 150 m, possibly connected to authigenic calcite precipitation. Below 250 mbsf, Li+ increases steeply as a result of a deep source, perhaps a higher concentration of volcanic detritus in deeper sediments or interaction with basement. Because basement is at ~1200 mbsf at Site 1091, the steep increase in Li+ is more likely related to the composition of deeper sediments, as indicated by the increased mud content and significantly lower sedimentation rates at the bottom of the section (see "Lithostratigraphy", "Chronostratigraphy").
The redox characteristics of Site 1091 sediments can be classified as reducing based on the observation of dissolved H2S (by scent, although often faint) in Cores 177-1091A-3H through 31H. However, sulfate concentrations decrease only modestly from near-bottom-water concentrations of ~28 mM at 4 mbsf to ~22 mM around 300 mbsf. Numerous observations of pyrite in smear slides throughout the section (see "Lithostratigraphy"), as well as the moderately high total organic carbon (TOC) and total sulfur (TS) contents (see Fig. F21; "Solid Phase Analysis"), suggest that sulfate concentrations may not reflect the long-term redox character of these sediments (see below). Furthermore, downhole sulfate reduction rates must be relatively slow, considering the observed maximum alkalinity and ammonium concentrations of ~10 and 1 mM, respectively, at ~150 mbsf.
Both Sites 1089 and 1091 have similar TOC content (~0.5 wt%), but the fact that sulfate concentrations decrease to near zero by 50 mbsf at Site 1089 and only decrease by about 20%-25% by 300 mbsf at Site 1091 reflects a fundamental difference in the nature of organic-matter degradation. The difference in diffusion and burial rates between Sites 1091 and 1089 could conceivably only explain part of the difference in sulfate profiles. One possible explanation stems from the observed downhole variations in alkalinity, ammonium, and phosphate (Fig. F20) at Site 1091. Alkalinity and ammonium increase gradually to maxima around 150 mbsf, whereas the maximum in phosphate is observed between 20 and 30 mbsf. The phosphate maximum probably reflects more labile bulk organic matter that is oxidized in the upper 30 to 40 m of sediment. During periods of very rapid sediment accumulation (glacials), this upper zone could become intensely reducing and may be the source of the significant pyrite formation seen lower in the section. Downhole, alkalinity and ammonium increase as a result of degradation of a more refractory organic fraction with very low phosphate content. This low-phosphate organic fraction is likely opal-intrinsic, implying that it would only become available for oxidation upon the slow dissolution of the opal.
The presence of dissolved Mn+2 and Fe+2 concurrent with dissolved H2S is curious and not altogether consistent with the scenario of more intensely reducing conditions during glacial periods (because reactive Mn and Fe would have been removed from the surface sediments if such conditions existed). One possible explanation for this problem is that mildly reducing conditions at depth could have worked slowly on the mud-rich sediments sampled at the bottom of the section (and presumably present deeper below the sampled interval) to reduce relic Mn and Fe. Downhole, sulfate reduction rates are slow and insufficient to precipitate all of the dissolved Fe+2. The scent of H2S was faint in many samples, and was not detected at all in Cores 177-1090A-32H and 33H. Furthermore, the increases in Mn+2 and Fe+2 below 200 mbsf (Fig. F20) are consistent with the hypothesis that the deeper mud-rich sediments are at least part of the source of the Mn+2 and Fe+2 observed higher in the section.
The shipboard solid phase analysis at Site 1091 consisted of measurements of inorganic carbon, total carbon, total nitrogen (TN), and TS (for methods see "Geochemistry" in the "Explanatory Notes" chapter). The results of Hole 1091A are presented in Table T14 and Figure F21. CaCO3 contents in Hole 1091A range from 0.2 to 58.9 wt%, with an average value of 5.7 wt%. With our coarse sampling resolution, the percentage of CaCO3 appears to be very low (nearly zero) in the upper Pleistocene sedimentary sequence from 12 to 78 mbsf. TOC contents vary between 0.17 and 0.91 wt%, with an average value of 0.60 wt%. TOC contents at Site 1091 are slightly higher than those of Sites 1089 and 1090, which are located between the Subtropical Convergence and the Subantarctic Front. TN contents are generally low (0.07-0.15 wt%). Although the TS content of most samples is less than 0.5 wt%, one high value (1.33 wt%) was observed in Sample 177-1091A-28H-2, 69-70 cm (256 mbsf). TOC/TN values vary between 1.7 and 7.9, indicating a predominance of marine organic material (Fig. F22). Pyrolysis analyses were not performed because of the low organic carbon in these sediments.
