INTRODUCTION

Clathrate hydrates of gas ("gas hydrates") are ice-like structures where cages of water molecules are stabilized by gas molecules (Sloan, 1990). Gas hydrates principally composed of CH₄ and water occur naturally in the pore space of marine sediment where appropriate high pressure and low-temperature conditions exist, and there is an adequate supply of methane (Kvenvolden, 1993).

The amounts of gas trapped in oceanic sediment gas hydrates (7.5 to 15 10¹⁸ g C as CH₄; Kvenvolden, 1988, 1993; MacDonald, 1990; Gornitz and Fung, 1994) comprise the Earth's largest fossil fuel reservoir. However, such global estimates must be regarded as speculative. Bottom-simulating reflectors (BSR's) have been interpreted as a phase boundary between overlying CH₄ hydrate and underlying free CH₄ gas for nearly three decades, but the amount and distribution of CH₄ in these two zones has been a long-standing and contentious issue (e.g., Shipley et al., 1979; Katzman et al., 1994; Wood et al., 1994; Holbrook et al., 1996; Paull, Matsumoto, Wallace, et al., 1996; Hovland et al., 1997).

On a local scale, measurements of pore-water Cl^- and the stable isotopes of oxygen (O) and hydrogen (H) on samples recovered from depth represent a potentially powerful method for quantifying gas hydrate amounts in marine sediment (Hesse, 1990; Hesse and Harrison, 1981). The use of these species for quantification of gas hydrate amounts is based on the fact that salts are excluded during hydrate formation, and the cage-building water molecules are enriched in the heavy isotopes O and H. The aqueous phase from which the hydrates form gets progressively enriched in salts and depleted in the heavy isotopes of O and H (Hesse, 1990). Hence, in closed systems the amounts of hydrate formed may be estimated from the Cl^- concentration and isotope signatures (H and O) of the surrounding pore water (Ussler and Paull, 1995).

However, hydrates decompose rapidly when the pressure is released during sampling. Thus, pore water extracted from hydrate-bearing sediments are usually mixtures of interstitial water and meltwater from the hydrates. Although quantification of gas hydrate amount from pore-water Cl^- and stable isotope measurements is conceptually simple, the method is not straightforward because in situ pore-water chemistry prior to gas hydrate dissociation is unknown. Estimates of gas hydrate amount made by this method, therefore, are entirely dependent on assumed in situ concentrations of Cl^-. Previous investigators (Kvenvolden and Kastner, 1990; Froelich et al., 1995; Kastner et al., 1995; Yuan et al., 1996) have assumed in situ pore-water Cl^- concentrations similar to that of seawater. As acknowledged by these authors, such an assumption neglects advection and diffusion of ions and results in overestimation of gas hydrate amount (Ussler and Paull, 1995). Until recently (Egeberg and Dickens, 1999), a paucity of information concerning fluid flow, in situ pore-water chemistry, and gas hydrate amount precluded alternative and better assumptions.

The aim of this work is to utilize the special conditions (upward migration of saline pore water) over the Blake Ridge Diapir to estimate the composition of the interstitial water (Cl^-, ²H), and to use these parameters to determine the amount of hydrates, based on analysis of pore water squeezed from whole-round cores and analysis of decomposed samples of hydrate.