Hydrothermal systems, thermal boundary layers and detachment faults in slow-spread ocean crust

We have known for the last 50 years that Europe and America have been moving apart at about 2cm/yr by processes of seafloor spreading that generate new oceanic crust at the submarine mid-Atlantic Ridge. This is one of the fundamental processes of Plate Tectonics, and has shaped the planet that we live on. Yet because we cannot use standard remote sensing techniques using electromagnetic radiation to study the seafloor, in many ways we know more about the surface of Mars than we do about the floor of the Atlantic! Over the last 12 years improved sonar surveys of the mid Atlantic Ridge have revealed a new mode of seafloor spreading where a significant part of the plate divergence is taken up by slip on long-lived, convex upward detachment faults, rather than mainly by magmatic intrusion. Up to half of the Atlantic seafloor may have formed in this way. These detachment faults are associated with large hydrothermal systems producing black smokers venting 400 C fluids on the seafloor. On fast (10-15 cm/yr) spreading ridges such as the East Pacific Rise, black smoker systems are small, short-lived, and located in zones of active volcanism, and are supplied with heat by shallow (1-2 km) magma chambers that are there more or less all the time. These systems have been modelled extensively, and a key element is the existence of a thin conductive boundary layer between molten magma and the hydrothermal fluid. On the mid-Atlantic ridge, black smoker systems are more widely spaced, larger, and longer lived, and often are located a few km away from the zone of active volcanism. These systems may in some cases be controlled by fluid flow up detachment faults, with heat supplied by episodic magma chambers as deep as 7km below seafloor, and much less numerical modelling work has been done on them. We have identified a fossil thermal boundary layer in a detachment fault sampled by drilling. In this proposal we plan to investigate this boundary layer more thoroughly, as well as the complex interrelationships between faulting, magmatism and hydrothermal circulation at slow spreading ridges. We will address this problem by building thermal and hydrothermal numerical models to predict both the asymmetric thermal structure produced by detachment faulting and the hydrothermal circulation patterns associated with permeable fault zones and localised magmatism. The hydrothermal models have to be very sophisticated because of the complicated properties of water, which changes density and viscosity very rapidly in the temperature range of black smoker systems. Hence we will work with experienced modellers in Paris to achieve our aims. We will test these models using data on cooling rates of rocks from IODP core in the footwall of an exposed detachment fault in the Atlantic - these cooling rates are calculated by comparing the compositions of natural minerals with experimental data on diffusion rates of trace elements. The aim of our models is not to replicate nature precisely (there are too many unknowns to do that) - but to test the range of parameter values that generate acceptable results. For example, the model must generate vents with the temperature measured on the seafloor and the heat output estimated from geochemical data - what are the minimum values of fault zone thickness and permeability that allow this to happen? These values can then be compared with physical models of permeability based on fracture densities and seismicity distributions. Because it is hard to observe subsurface geology or fluid flow directly, modelling is often the only way of determining whether hypotheses are realistic. At the end of this project we will have a better understanding of one of the most important but least accessible parts of the Earth System - the formation of new lithosphere at ridge crests, and the complex interactions between the ocean and the crust that occur as a result of this process.