Mid Ocean Ridge Electromagnetics
|Deployment of a seafloor EM instrument from the R/V Melville. On deck are Wayne Crawford, Kerry Key and Lisl Lewis. (Rob. L. Evans)|
Crustal Resistivity and Hydrothermal Circulation
The process of new crustal formation at a midocean ridge involves a complex series of interactions between newly delivered melt from the mantle, and seawater which penetrates deep into the crust through cracks and faults, removing heat from the crust. Both the distribution of melt within an axial magma chamber and the spatial pattern of hydrothermal convection remain poorly constrained, but both interconnected melt and hot hydrothermal fluids are electrically conductive and so can potentially be identified using EM methods.
In order to measure conductivity within the crust, artificial sources of electromagnetic fields are required. This kind of experiment, known colloquially as controlled-source EM (CSEM), provides the geophysical backbone of my research.
My PhD thesis involved collection and analysis of a CSEM data set from the East Pacific Rise (EPR). Those data, the first collected by what has now become a reliable and productive transmitter, were able to place good constraints on the porosity structure of the upper oceanic crust at the ridge (Evans et al., 1991), were able to bound the lateral extent of an axial melt body containing connected partial melt (Evans et al., 1994) and, when combined with seismic velocity data, were used to constrain the large-scale bulk permeability of the crust (Evans, 1994), one of the few available constraints on crustal permeability at a ridge crest.
I have also carried out a different kind of CSEM experiment, known as the magnetometric resistivity (MMR) technique. This method, pioneered by Nigel Edwards, is relatively straightforward, as the transmitter is simply a vertical conducting cable extending from the seasurface to the seafloor. The magnetic field generated by the source is measured by remote seafloor magnetometers. The first experiment of this type I completed was in the overlapping spreading center between the Cleft and Vance segments of the Juan de Fuca ridge. Here, crustal conductivities were significantly high, high enough to require raised temperatures beneath the ridge axes, but with a sharp fall-off away from the ridge (Evans et al., 1998). This result was the first to suggest that EM techniques might be able to identify the patterns of hydrothermal circulation beneath ridges.
In 2000, I completed a much larger experiment in collaboration with Spahr Webb, this time with 10 seafloor magnetometers and over 200 transmission stations. Magnetometers were placed on the ridge crest in areas of known hydrothermal activity, in axial sites devoid of venting, and further off-axis to a distance of approximately 4km. Data collected at sites off-axis show higher seafloor resistivities than at axial sites. This response is the opposite to that expected from porosity controlled resistivity structure, with a thicker high-porosity extrusive layer 2A (as required seismically) off-axis. An obvious explanation for the reduced axial resistivities is that the upper-most few hundred meters of crust are much hotter beneath the ridge crest than a few kilometers off-axis, lowering the pore-fluid resistivity (Evans et al., 2002)
In the Atlantic, the end result of hydrothermal circulation is often a large sulfide mound. An example of this is the TAG mound, and I have carried out several surveys looking at the structure of TAG. EM work to this end involved numerical modeling studies of potential experiments (Evans & Everett, 1994), and the development of time domain EM equipment suitable for small scale studies which was eventually deployed during an Alvin dive program. The data collected in that experiment identified the sulfide feature as a conductive body with significant variability in structure related to variations in sulfide thickness and type (Cairns, Evans & Edwards, 1996). I followed this experiment at TAG with another, this time using seafloor gravity measurements made inside the Japanese Shinkai 6500 submersible. In two dives, sufficient data were collected to bound the mass of sulfide in the mound (about 3 million tonnes - a result later confirmed by ODP drilling) and to rule out the presence of a large sulfide rich root extending deep beneath the mound (Evans, 1996)
Magnetotelluric Methods: The MELT Experiment
The MELT experiment used a combination of seismic and MT measurements to
understand the style of melt delivery from the mantle to the fast spreading southern East Pacific Rise.
Results of the MELT experiment show asymmetry in structure across the ridge. The western side of the ridge appears to contain a small percentage of interconnected melt, broadly consistent with seismic results, but there is little evidence for melt more than a few kilometres to the east of the ridge (Evans et al., 1999; Baba et al., in press).
The MELT experiment has also provided insights into
the evolution of oceanic plates. In a paper published in Nature (Evans et al., 2005) we show how seismic and electrical structure appears to be controlled by compositional effects more than by thermal effects with a relatively flat boundary in structure at a depth of ~60km which we interpret to be the boundary between dry oceanic mantle overlying hydrated mantle.