An unequivocal case for high Nusselt number hydrothermal convection in sediment-buried igneous oceanic crust

New observations of seafloor heat flow, precisely located along seismic reflection profiles crossing a buried ridge on the eastern flank of the Juan de Fuca Ridge, show a nearly exact inverse correlation between heat flow and sediment thickness, such that the basement-sediment contact appears isothermal to within 10 K, despite a factor of three local variation in sediment thickness. We have used these observations with numerical models to infer hydrothermal heat-transport properties of the upper oceanic crust at this 3.5 Ma site. Model results show that, while fluid circulation is stimulated by the effects of basement topography even at sub-critical Rayleigh number conditions, the creation of a nearly isothermal basement surface requires very high heat-transport efficiency. Lower limits for the Nusselt number ( Nu ≥ 25), for the Rayleigh number ( Ra ≥ 4000), and for the permeability (κ ≥ 10−11 m2), are provided by assuming that high permeability is distributed throughout the uppermost 600 m of relatively low-velocity igneous crust at this site. Relatively high permeability can also be inferred by considering the calculated fluid pressure regime in light of what is known about the relationship between fluid seepage through the sediment section and the underlying basement topography from geochemical data: Local super-hydro-static pressure and fluid discharge above buried basement ridges can occur only if basement permeability is higher than 10−13 m2. Unfortunately, no constraint on the actual distribution of high permeability below the top of the igneous crust is provided by the thermal regime based on the heat-flow and seismic observations. Equally uniform upper basement temperatures can be produced by fluid flow in a thinner layer (of thickness h) having a correspondingly higher Nusselt number and permeability. Only the products of Nu × h, and κ × h2 are constrained. Bulk permeabilities (averaged over intervals a few hundred meters thick) measured in boreholes that have penetrated the upper oceanic crust are typically less than 10−13 m2. The much higher formation-scale permeability we infer may be a consequence of the relative youth of the crust at this site, although it is more likely that the interconnected fractures and extrusive volcanic unit contacts and voids that contribute most to the bulk formation permeability are relatively infrequent and not representatively sampled by drilling.