Gas hydrate, fluid flow and free gas: Formation of the bottom-simulating reflector

Gas hydrate in continental margins is commonly indicated by a prominent bottom-simulating seismic reflector (BSR) that occurs a few hundred metres below the seabed. The BSR marks the boundary between sediments containing gas hydrate above and free gas below. Most of the reflection amplitude is caused by the underlying free gas. Gas hydrate can occur without a BSR, however, and the controls on its formation are not well understood. Here we describe two complementary mechanisms for free gas accumulation beneath the gas hydrate stability zone (GHSZ). The first is the well-recognised hydrate recycling mechanism that generates gas from dissociating hydrate when the base of the GHSZ moves upward relative to hydrate-bearing sediment. The second is a recently identified mechanism in which the relationship between the advection and diffusion of dissolved gas with the local solubility curve allows the liquid phase to become saturated in a thick layer beneath the GHSZ when hydrate is present near its base. This mechanism for gas production (called the solubility-curvature mechanism) is possible in systems where the influence of diffusion becomes important relative to the influence of advection and where the gas–water solubility decreases to a minimum several hundred metres below the GHSZ.We investigate a number of areas in which gas hydrate occurs to determine where gas formation is dominated by the solubility-curvature mechanism and where it is dominated by hydrate recycling. We show that the former is dominant in areas with low rates of upward fluid flow (such as old, rifted continental margins), low rates of seafloor uplift, and high geothermal gradient and/or pressure. Conversely, free-gas formation is dominated by hydrate recycling where there are rapid rates of upward fluid flow and seabed uplift (such as in subduction zone accretionary wedges).Using these two mechanisms to investigate the formation of free gas beneath gas hydrate in continental margins, we are able to resolve a number of problems. These include why sub-BSR free-gas zones are generally thin in convergent margins and thick in passive margins, and why areas such as the Gulf of Mexico and Mackenzie Delta do not have regional BSRs. Finally, we use our understanding of how BSRs are formed to show that regional gas hydrate is unlikely to exist in substantial quantities in passive continental margins outside areas with an observable BSR.