Je suis
Citoyen / Grand public
Étudiant / Futur étudiant
Partenaire public
Enseignant / Elève

Do Hydrothermal Systems Control Detachment Fault Development?


IPGP - Campus Jussieu


Séminaires Géosciences Marines

Salle Bleue

par: Andrew McCaig

Leeds University

Large, long-lived black smoker systems on slow spreading ridge are often closely linked with detachment faults, and for the TAG field it has been suggested that a convex-upward fault controlled hydrothermal upflow leading to venting 4.5 km away from the ridge axis [1,2]. The heat source for hydrothermal flow is thought to be gabbroic intrusions into the detachment footwall, 7 kmbsf and not directly beneath the TAG field[1]. Studies of exposed detachment faults suggest two endmember types; hot detachments (e.g. Atlantis Bank, SWIR) characterised by thick, high temperature (800-950 ¡C) gabbroic mylonite zones, and cold detachments characterised by talc-tremolite schists formed at black smoker temperatures and brittle rather than ductile deformation in gabbroic footwalls. Isotopic data suggests high fluid fluxes along cold detachment faults [2] but only limited flow through footwall gabbros. A simple 1-D thermal model shows that gabbros in the footwall of a detachment held at 400 ¡C by hydrothermal fluid will cool rapidly through the brittle-ductile transition (~750 ¡C), preventing the development of extensive mylonites. Hence the presence of vigorous hydrothermal circulation can profoundly affect crustal rheology. The cold detachment model requires that the fault is already the locus of hydrothermal flow at the time the gabbro body is emplaced immediately beneath it. We suggest that initially, gabbro is intruded at intermediate depth in the hangingwall of a steep, low displacement axial valley fault. As the gabbro solidifies, the hydrothermal system mines down into it via a cracking front. The next gabbro is intruded into wet rocks and cannot rise so far. Eventually gabbro is intruded beneath the fault at 7 kmbsf. If magma is trapped below the fault it becomes kinematically favourable to switch to extensional faulting on a long-lived detachment as the main mode of plate separation. Ê[1] deMartin et al. (2007) Geology 35, 711-714. [2] McCaig et al. (2007) Geology 35, 935-938.