Introduction

Gas-hydrate samples have been recovered at about 16 areas worldwide (Booth et al., 1996). However, gas hydrate is known to occur at about 50 locations on continental margins (Kvenvolden, 1993) and is certainly far more widespread so it may represent a potentially enormous energy resource (Kvenvolden, 1988). But adverse effects related to the presence of hydrate do occur. Gas hydrate appears to have caused slope instabilities along continental margins (Booth et al., 1994; Dillon et al., 1998; Mienert et al., 1998; Paull & Dillon, (Chapter 12; Twichell & Cooper, 2000) and it has also been responsible for drilling accidents (Yakushev and Collett, 1992). Uncontrolled release of methane could affect global climate (Chapter 11), because methane is 15-20 times more effective as a "greenhouse gas" than an equivalent concentration of carbon dioxide. Clearly, a knowledge of gas-hydrate properties is necessary to safely explore the possibility of energy recovery and to understand its past and future impact on the geosphere.

Gas hydrate exists in the natural environment at pressure and temperature conditions that make it difficult to study in situ. For that reason, the U.S. Geological Survey's Woods Hole Field Center developed a laboratory system (GHASTLI - Gas Hydrate And Sediment Test Laboratory Instrument) to simulate natural conditions within the gas-hydrate-stability region. Using this system, gas hydrate can be formed in reconstituted sediment (Winters et al., 2000) and field samples containing gas hydrate can be preserved (Winters et al., 1999) while physical properties are measured.

GHASTLI determines a number of properties of the host sediment prior to gas-hydrate formation, after hydrate has formed, and after dissociation. Currently, four main types of data are measured: (1) acoustic properties (both compression-wave and shear-wave), (2) shear strength and related engineering

U.S. Geological Survey 384 Woods Hole Road Woods Hole, MA 02543 USA

Institute for Geophysics, Univ. of Texas 4412 Spicewood Springs Road, Bid. 600 Austin, TX 78759 USA

properties under triaxial test conditions, (3) permeability, and (4) electrical resistivity. These results are used to model and predict the relation between the natural environment, sediment fabric, and the existence of gas hydrate.

Acoustic properties can dramatically change as gas hydrate forms, and this relationship allows remote identification of hydrate. Initial velocities derived from seismic data have been used to quantify the presence of gas hydrate (Lee et al., 1993). This acoustic relationship then has been used as a means of mapping the extent of hydrate and may be a means of estimating resource potential. Also, acoustics have been used to model gas-hydrate interaction with host sediment (Dvorkin and Nur, 1993; Pecher et al., 1999). Currently, a number of theories exist that relate the effect of natural gas-hydrate saturation to acoustic behavior, but testing of these theories has not been possible without a laboratory system capable of simulating in situ conditions. Berge et al. (1999) showed that velocity behavior changes occur when pore-volume saturation exceeded about 35% using sand and R11 (CC13F) refrigerant as the hydrate former. It is not yet known if hydrate formed with R11 exhibits behavior similar to natural gas hydrate.

Sediment strength properties can be affected by the presence of gas hydrate, however, measurements of the effects of hydrate on sediment strength are scarce. There is currently concern that offshore drilling in the Ormen Lange gas field in the North Sea could dissociate existing gas hydrate and cause seafloor instabilities (Offshore, 2000). The nearby Storegga slide, which caused tsunamis to impact both Scotland and Norway about 7,000-8,000 ka, is thought to have been influenced by gas-hydrate dissociation. Concerns exist that gashydrate dissociation related to mass wasting may release methane, a potent "greenhouse gas", to the atmosphere (Chapter 11). A better understanding of strength properties can be used in stability calculations of: (1) natural slopes (Kayen and Lee, 1991), (2) structures placed on land or the seafloor (pipelines and drilling platforms), and (3) wells drilled to recover natural gas and oil (Briaud and Chaouch, 1997).

Sediment permeability is reduced if significant amounts of gas hydrate occupy intergranular pore space. The changes that are produced are important for slope-stability calculations (Kayen and Lee, 1991) and in determining if gas hydrate can effectively form a seal that traps bubble-phase gas beneath the gas-hydrate-stability zone (GHSZ). The presence of gas beneath the GHSZ is a major contributor to the formation of a bottom-simulating reflector (BSR) present on seismic-profile records.

Because electrical resistivity of gas hydrate is much higher than naturally-existing pore water, measurement of that parameter can be used to indicate when gas hydrate has formed in the laboratory and can be used as a comparison to well-logging measurements. Well logging has provided much information about the presence and properties of gas hydrate in the natural state, for example, Collett et al., 1984, 1999; Collett & Ladd, 2000; and Miyairi et al, 1999. Gas-hydrate saturation has also been calculated from well-log information (Lee and Collett, 1999).

Figure 1: GHASTLI system showing the syringe pumps that control pressure and the main pressure vessel, located just to the right of center in the photo, that contains the sediment test specimen.

Gas hydrate exists naturally in different materials and under varying environmental conditions. Understanding the factors responsible for hydrate formation, preservation, and concentration may enable prediction of locations where methane from gas hydrate can be economically and safely recovered. GHASTLI provides a means for making measurements related to the above topics under conditions that can be closely controlled. Sediment grain size, specimen bulk density, effective stress, and pore-fluid characteristics can be varied to assess their impact on gas-hydrate formation and behavior.

Was this article helpful?

0 0

Post a comment