Gas hydrate is a solid composed of water and certain gases (methane, CO2, propane, and ethane) that is stable under high pressure, low temperatures, and adequate concentrations of gas (Sloan, 1990; Kvenvolden, 1993). It commonly occurs in pore space of sediment on continental margins and in permafrost regions. In a number of instances, gas hydrate has been observed as a cement in sands and sandstones (Collett, 1993). Hydrate may form as cement, veins, lenses, or isolated nodules. There is considerable current interest in natural gas hydrate because it may be exploited in the near future as a source of methane (e.g., Kvenvolden, 1988; Max and Lowrie, 1996), and because methane may be released from gas hydrate to the ocean and atmosphere during oceanographic change (e.g., Hatzikiriakos and Englezos, 1994; Dickens et al., 1995).
A critical issue is to identify factors that control or limit the distribution of gas hydrate to certain depth zones in a sediment sequence. Pressure and temperature are obvious first-order constraints (e.g., Kvenvolden, 1993; Dickens and Quinby-Hunt, 1994); however, it is increasingly apparent that certain properties of the sediment surrounding gas hydrate must play important roles in controlling hydrate distribution. For example, gas hydrate in Alaskan permafrost regions and at the Middle America Trench [Deep Sea Drilling Program Leg 76] exists in discrete layers apparently related to lithology (von Huene, Aubouin, et al., 1985; Collett, 1993). At the Chile Triple Junction [Ocean Drilling Program (ODP) Leg 141], Cascadia Margin (ODP Leg 146) and Blake Ridge (ODP Leg 164), the base of gas hydrate stability exists at a depth significantly shallower than expected given pressures and temperatures at in situ conditions (e.g., Bangs et al., 1993; Kastner et al., 1995; Ruppel, 1997) and those on the methane hydrate-seawater equilibrium curve (Dickens and Quinby-Hunt, 1994).
Both experimental observations (Brewer et al., 1997) and theoretical considerations (Ruppel, 1997) suggest that the distribution of gas hydrate is largely controlled by lithology and, in particular, the size of pore spaces between sediment grains. The rationale is that the formation of gas hydrate is inhibited in small pore spaces because the activity of water is decreased by capillary action (Handa and Stupin, 1992).
The Blake Ridge, located ~350 km off the coast of South Carolina (Fig. 1), is a large drift deposit where laterally extensive regions contain gas hydrate (Dillon and Paull, 1983). Leg 164 crew recently drilled Sites 994, 995, and 997 on the crest of the Blake Ridge in order to refine our understanding of in situ characteristics of gas hydrate, including sediment effects on hydrate distribution (Paull, Matsumoto, Wallace, et al., 1996). Because the geology and topography on the crest of Blake Ridge is relatively simple, sediment recovered on Leg 164 provides an unprecedented opportunity to analyze basic properties of hydrated sediment and to understand variations in hydrate distribution related to lithological, chemical, and hydrological factors.
A key result of Leg 164 is that gas hydrate distribution on the Blake Ridge is concentrated into a distinct upper and lower zone. Both hydrate zones are well delineated by proxy measurements for hydrate abundance (Fig. 2A, B), especially pore-water chloride concentrations (Egeberg and Dickens, unpubl. data) and pressure core sampler (PCS) gas volumes (Dickens et al., 1997). Although hydrate probably also occurs as diffuse deposits in pore spaces outside of the two zones, the two well-defined zones contain the most abundant hydrate. Both hydrate zones are well within pressure and temperature conditions for methane hydrate stability of gas-liquid systems (Dickens and Quinby-Hunt, 1994; 1997). However, whereas the abundance of hydrate in the lower zone may be explained by methane cycling across the phase boundary between free gas bubbles and gas hydrate (Hyndman and Davis, 1992; Korenaga et al., 1997), the upper zone lacks a satisfactory explanation.
Sediment layers exhibiting preferred hydrate occurrence might be linked to a relative decrease in intergranular capillary forces compared to the bulk sediment column (e.g., Clennell et al., 1995, 1997; Ruppel, 1997). The purpose of this study is to examine this hypothesis by determining if the upper hydrate zone at Site 994 is characterized by a significant change in lithology relative to underlying (and overlying) sediment. Documentation of such a lithological change, particularly one that is associated with an increase in porosity, would provide important field support for theoretical arguments concerning the influence of capillary forces on hydrate accumulation.