Gas bubbles rise through sediments because they are buoyant, however, they become trapped when their diameter exceeds that of the sediment pore throats. Gas accumulations, formed when bubbles coalesce, may grow to occupy voids larger in size than the sediment pore spaces. Within these accumulations gas pressure rises above hydrostatic pressure until such time as migration is initiated either by capillary flow or fracture flow. Capillary flow is controlled manly by the size of sediment pore throats: the height of the gas column required to initiate capillary flow increases by a factor of ten (approximately) for each decrease in grain size: from coarse sand to fine sand, to silt, to silty clay. The height of the gas column also doubles (very approximately) with an increase in depth of 500m. Fracture migration is initiated when the buoyancy of the gas column exceeds the confining pressure. This is a function of the overburden pressure. As a first approximation the critical gas column height is about one thousandth of the depth below seabed, although grain size and compaction (particularly the overconsolidation of clays) are also influential.
In shallow marine sediments fracture flow will occur before capillary flow in all sediments except coarse sands. Increases in water depth cause an increase in the gas column height because gas density increases as pressure increases. However, as this effect is very small, the precautions required for operations in shallow gas areas on the continental shelf apply equally to deep water sites.
Gas (particularly methane) is widespread beneath the seabed in both shallow and deep water environments (Hovland and Judd, 1988; etc.). Accumulations of gas are important not only as potentially exploitable resources, but also as hazards to drilling (Prince, 1990), the integrity of foundations (Holrnes et al., 1997), and marine slope instability (Cochonat et al., 1996). Gas seeps and gas trapped in seabed sediments are also useful guides to petroleum prospectivity (Brekke et al., 1997). In all these cases the presence of gas demonstrates that migration has occurred from a source sediment, be it a deep thermogenic source, or a shallower biogenic (bacterial) source Gas tends to migrate towards the seabed and will (eventually) escape at the seabed unless trapped somewhere en route. Ample evidence of the presence of gas accumulations, and of gas migration, has been presented elsewhere (e.g Judd and Hovland, 1992; Heggland, 1997).
The purpose of this paper is to investigate the mechanisms of gas migration, and factors which may enable or Impede it. It is recognised that gases such as methane may migration in solution. For secondary migration, the migration of free gas (gas bubbles), considered here, is considered to be a more efficient gas transport mechanisms than diffusion (Krooss, 1989). Previous work on this subject has focussed on migration mechanisms at considerable depths (> lkm) below the surface (seabed), and is primarily m e d at investigating the migration of petroleum from source to reservoir, and from reservoir through the cap rock or seal (e.g. Gurevich et al., 1993;
