Fluid Processes in Deep Crustal Fault Zones

Fluid as a C-O-H dominated phase is widespread, but not ubiquitous, in the Earth's crust. The presence or absence of fluid is in large part a function of thermal history, at least up to the onset of melting. Rocks containing relatively low-temperature assemblages that are subject to further heating release fluid and so are commonly saturated, while rocks undergoing cooling resorb fluid into hydrous minerals and so are dry. Fluid may be introduced from external sources during faulting or magmatic activity, and the degree to which it persists depends on the interplay between injection rates and reaction rates. Where fluids do occur in the crust, fluxes are generally low, so that many aspects of fluid chemistry are dictated by saturation with rock-forming minerals. These mineralogical controls on fluid chemistry and activities of volatile species further affect the rheology of the crust by determining whether or not deformation can be fluxed by fluid processes. It is argued that rocks undergoing progressive metamorphism are wet and experience widespread deformation, while rocks that are cooling are strong and deformation is localized into zones, particularly during times of fluid infiltration. The transition between brittle and ductile behavior may therefore reflect changes in the availability of water rather than changes in temperature. Faults themselves are important loci of fluid flow, but it is often difficult to identify the sources of fluid, because geochemical tracers are mainly reset in a rock-dominated environment. Nevertheless, it is unlikely that faults are commonly effective drains of fluid being released by prograde metamorphism, because the very low permeability of such rocks (inferred from evidence for strong overpressuring) means that fluid cannot easily drain into fractures, even where a strong gradient in hydraulic head exists.