Generation of permeability barriers during melt extraction at mid-ocean ridges

Melt focusing at mid-ocean ridges is necessary to explain the narrowness of the zone of crustal accretion and the formation of large but localized on-axis seamounts at slow and ultraslow spreading centers. It has been proposed that melt focusing is facilitated by the presence of a barrier to upward melt migration at the base of the thermal boundary layer (TBL). We assess the development of a melt impermeable boundary by modeling the geochemical evolution and crystallization history of melts as they rise into the TBL of mid-ocean ridges with different spreading rates. A permeability barrier, associated with a crystallization front controlled by the conductive thermal regime, exists for melt trajectories at slow to fast spreading ridges (>= 10 mm/yr half rate). The effective lateral scope of the barrier, where the slope of the barrier exceeds a critical value that allows buoyant melt transport to the axis, generally increases with spreading rate. At all distances from the axis at ultraslow ridges and off-axis at slow spreading ridges, the weak crystallization front may prohibit formation of an efficient barrier and lead to the possibility that some fraction of melt may be incorporated into the lithospheric mantle, allowing refertilization. The protracted crystallization history and potential absence of an effective permeability barrier may explain the dearth of volcanism at ultraslow ridges and calls for a revision of lateral melt focusing scenarios at ultraslow spreading rates.