Permeability of the continental crust: Implications of geothermal data andmetamorphic systems

In the upper crust, where hydraulic gradients are typically <1 MPa km(-1), advective heat transport is often effective for permeabilities k greater th an or equal to 10(-16) m(2) and advective mass (solute) transport for k gre ater than or equal to 10(-20) m(2). Regional-scale analyses of coupled grou ndwater flow and heat transport in the upper crust typically infer permeabi lities in the range of 10(-17) to 10(-14) m(2), so that heat advection is s ometimes significant and solute advection should nearly always be significa nt. Analyses of metamorphic systems suggest that a geochemically significan t level of permeability can exist to the base of the crust. In active metam orphic systems in the mid to lower crust, where vertical hydraulic gradient s are likely >10 MPa km(-1), the mean permeabilities required to accommodat e the estimated metamorphic fluid fluxes decrease from similar to 10(-16) m (2) to similar to 10(-18) m(2) between 5- and 12-km depth. Below similar to 12 km, which broadly corresponds to the brittle-plastic transition, mean k is effectively independent of depth at similar to 10(-18.5+/-1) m(2). Cons ideration of the permeability values inferred from thermal modeling and met amorphic fluxes suggests a quasi-exponential decay of permeability with dep th of log k approximate to -3.2 log z - 14, where k is in meters squared an d z is in kilometers. At mid to lower crustal depths this curve lies just b elow the threshold value for significant advection of heat. Such conditions may represent an optimum for metamorphism, allowing the maximum transport of fluid and solute mass that is possible without advective cooling.