RESULTS AND DISCUSSION

Hydrocarbons >C₁ were found in all sediments analyzed from Site 1109 (Table T1). The shallowest sample (1.4 mbsf) contained small quantities of ethane and propane (<30 pmol/g). Low levels of isobutylene + 1-butene, 3-methylpentane, and hexane were also detected. The concentrations of ethane, propane, and butane remained relatively low (<100 pmol/g) throughout Site 1109 and had no consistent trend with depth. In contrast, the longer-chain components increased in concentration with depth. The concentration of hexane increased from 15 pmol/g at the surface to 716 pmol/g at 660.7 mbsf with a concomitant increase in both 2-methyl- and 3-methylpentane (Table T1; Fig. F3). The highest concentrations of these compounds were present at the base of the hole where low levels of trans-2-butene and isopentane were present for the first time. Shipboard headspace methane concentrations had a typical downhole profile with initially low methane concentrations (~2-4 ppmv) increasing (1000-10,000 ppmv) once sulfate had been removed from the pore water at ~100 mbsf (Taylor, Huchon, Klaus, et al., 1999) (Fig. 3A). Methane concentrations remained relatively constant until ~640 mbsf, after which concentrations decreased to 3-6 ppmv at the bottom of the hole. The high C₁/C₂ ratio throughout the hole (>1000) indicated that the methane was of biogenic origin (Taylor, Huchon, Klaus, et al., 1999). Unusually, the entire sedimentary sequence was sampled at this site, down to an older (50 Ma) basement (massive dolerite) (Monteleone et al., this volume), and methane concentrations decrease towards this basement. Decreases in deep methane concentrations have been observed at other ODP sites; however, these decreases were coincident with increases in pore water sulfate (Mather and Parkes, 2000; Cragg et al., 1990), indicating methane oxidation via anaerobic sulfate reduction (e.g., Hoehler et al., 1994). In contrast, at this site there was no sulfate increase and therefore some other mechanism must be responsible for the decrease in methane.

All sediments analyzed from Site 1115 contained hydrocarbons >C₁ (Table T2). Low concentrations (<20 pmol/g) of ethane, propane, 2-methyl- and 3-methylpropane, and hexane were present in the surface sample (1.5 mbsf). The concentrations of ethane, propane, butane, and isobutylene + 1-butene remained low (<60 pmol/g) throughout Site 1115, and as with Site 1109, had no consistent trend with depth. 2-Methylpentane and 3-methylpentane concentrations had a broad subsurface peak maximizing at 554.8 mbsf. Hexane was also elevated at this depth but continued at similar concentrations until 736.2 mbsf (Table T2; Fig. F4). Small amounts of cis-2-butene, 1,3-butadiene, and pentane were detected in the lower part of the core. 2,2-Dimethylbutane occurred for the first time at the base of the hole at a concentration of 102 pmol/g. As found at Site 1109, shipboard headspace methane began to increase in concentration below 210 mbsf coincident with the removal of pore water sulfate (Fig. F4A) (Taylor, Huchon, Klaus, et al., 1999). Concentrations remained between ~1000 and 10,000 ppmv throughout the rest of the hole except for a narrow zone around ~550 mbsf, coinciding with the regional unconformity and hiatus resulting from the emergence of the forearc sequence (Taylor, Huchon, Klaus, et al., 1999). The C₁/C₂ ratio indicates a biological source for methane at this site (Taylor, Huchon, Klaus, et al., 1999).

Low molecular weight hydrocarbons (C₁-C₇) have been detected in a number of Holocene sediments at part-per-billion concentrations (e.g., Hunt, 1975; Hunt and Whelan, 1979; Hunt et al., 1980; Rowe and Muehlenbachs, 1999; Simoneit et al., 1988; Whelan et al., 1980) and also in numerous ODP sediments (e.g., Berner and Faber, 1993; Froelich et al., 1995; Ingle, Suyehiro, von Breymann, et al., 1990; Kastens, Mascle, Auroux, et al., 1987; Mountain, Miller, Blum, et al., 1994; Westbrook, Carson, Musgrave, et al., 1994; Whelan, 1979; Whiticar and Suess, 1990; Whiticar et al., 1994). The thermal generation of LMWH is generally thought to be a high-temperature, ~>50°C process (Hunt, 1996), although the exact formation mechanism for these components in thermally immature sediments is still unknown (Mango, 1997). As the thermal gradients at both Site 1109 and 1115 are low, with extrapolated sediment temperature reaching only ~25° and ~22°C at the base of Site 1109 and 1115, respectively (Taylor, Huchon, Klaus, et al., 1999), this would appear to exclude thermogenic origin of these LMWH. However, low-temperature thermal generation of nonmethane components is also considered to be possible at temperatures below 62° and even down to 20°C (Rowe and Muehlenbachs, 1999). Hence, at these sites, both biological and thermogenic mechanisms may be responsible for LMWH generation.

Low-temperature bacterial production of a range of low molecular weight branched hydrocarbons including 2- and 3-methylpentane from farnesol, ß-carotene, and -2-pinene has been demonstrated in laboratory experiments (Hunt et al., 1980). As sterile controls showed no formation of LMWH, it was concluded that their production was a result of low-temperature bacterial processes. Therefore, it seems reasonable to assume the same is true at these sites. The reason for the much higher concentrations of 2-methyl- and 3-methylpentane at depth at these sites, however, is unclear. It may, however, reflect the more terrigenous nature of organic matter at depth, which was deposited during continental rifting. These sites became increasingly marine during subsequent seafloor spreading where terrigenous input would decline. Terrigenous plant material would contain potentially more isoprene biopolymers than marine-derived organic matter (Harwood and Russell, 1984). Hence, there would be greater potential for generation of branched-chain LMWH within the deeper terrestrial sections. An additional aspect may be that even the small increases in temperature at these sites may enhance bacterial degradation of sedimentary organic carbon, as shown by Wellsbury et al. (1997), thus contributing to the formation of branched LMWH compounds at depth (Figs. F3, F4).

The elevated LMWH concentrations at both sites appear to coincide with a minimum in methane (Figs. F3, F4). This is surprising and may indicate a different origin for LMWH to methane, as has been previously suggested (Rowe and Muehlenbachs, 1999) or by preferential consumption of methane by bacteria compared to the other LMWH components. Further investigation of the LMWH distributions in ODP sediments may help resolve this issue.