Gas hydrate is an icelike mineral that contains methane or other low molecular weight gases in a lattice of water molecules. Methane hydrates are stable under the temperature and pressure conditions generally found in the Arctic and near the seafloor at water depths >300 m. They are quite common beneath the slope of both active and passive continental margins where methane originates from the decomposition of organic matter by biogenic and/or thermogenic processes. International interest in this material has increased considerably in the past several years because of increasing recognition that the large volumes of gas stored in these structures represent a significant fraction of the global methane budget (see review by Kvenvolden and Lorenson, 2001) and may, therefore, be a potential energy resource for the future (e.g., Milvov and Sassen, 2002). Several authors have also suggested that sudden widespread dissociation of subseafloor gas hydrates in response to changing environmental conditions may have had a significant effect on past climate (e.g., Revelle, 1983; Nisbet, 1990; Paull et al., 1991; Katz et al., 1999; Dickens, 2001). These effects remain speculative, as the volume of gas stored in the gas hydrate reservoir and its behavior during changing environmental conditions are currently poorly constrained.

In order to evaluate the economic potential of hydrates, their role as a natural hazard, and their impact on climate, we need to know the following:

• How are hydrates and underlying free gas distributed vertically and horizontally in the sediment?
• What is the gas hydrate inventory on Hydrate Ridge?
• What controls their distribution (i.e., lithologic controls on fluid migration and on hydrate nucleation and growth)?
• What are the effects of this distribution on the mechanical properties of the seafloor?
• How can hydrate and gas distribution be regionally mapped using remote sensing geophysical techniques?
• How does hydrate respond to changes in pressure and temperature resulting from tectonic and oceanographic perturbation?
• How can we use the isotopic record as a proxy for past tectonic and climate changes?
• Does the hydrate system harbor a rich biosphere? This question is of broad interest, particularly given the recent recognition that the biosphere extends deeper into the earth and that it has a larger impact on the geologic record than previously thought.

These questions were the focus of Ocean Drilling Program (ODP) Leg 204, which was dedicated to understanding the biogeochemical factors controlling the distribution and concentration of gas hydrates in an accretionary margin setting. Coring was guided by a three-dimensional (3-D) seismic site survey (Trehu and Bangs, 2001; Trehu et al., 2002) and by logging-while-drilling (LWD) data acquired at the beginning of the leg. These two data sets provided a "road map" to guide coring and sampling, enabling us to anticipate the depths at which gas hydrates should be expected and better focus special sampling tools. Accurate quantification of in situ hydrate and gas concentrations is difficult because of hydrate dissociation and gas loss during core retrieval (Paull and Ussler, 2001). A major focus of Leg 204 was, therefore, to acquire samples under pressure using the ODP pressure core sampler (PCS) system and the recently developed Hydrate Autoclave Coring Equipment (HYACE) system, which includes a laboratory transfer chamber for maintaining pressure while making physical property measurements. Extensive use was made of infrared (IR) cameras on the catwalk to rapidly identify potential hydrate-bearing samples and preserve them for careful study. Extra attention was also given to making high-resolution measurements of the chemistry of interstitial waters, resulting in a large number of interstitial water samples from this cruise. We also had frequent deployments of tools to measure in situ temperature and pressure, especially in zones where LWD data indicated rapid changes in the physical properties of the sediments, and conducted an extensive suite of downhole and two-ship seismic experiments.