Mid-Atlantic Ridge

Our knowledge of the distribution of hydrothermal systems along the slow-spreading Mid-Atlantic Ridge (full spreading rate 2-4 cm/year) is still in its infancy and, with only four deep-water active hydrothermal fields known, of which to date only two---TAG and Snake Pit---have been studied in any detail, a generalized model of controls on their occurrence cannot yet be derived.

The TAG hydrothermal field at 26 08N on the Mid-Atlantic Ridge is one of the largest sea floor deposits and has been intensively studied over the past few years. It is located at the mid-point of a 40 km long ridge segment at the base of the eastern median valley wall. The hydrothermal field extends over an area of at least 5 km x 5 km and consists of presently active low and high temperature zones, as well as a number of relict deposits [ Rona et al., 1993a,b]. High temperature activity is confined to a mound that lies on crust at least 100,000 years old. It is distinctly circular in plan view with a diameter of about 200 m, and is mineralogically zoned. The venting fluids have a wide range of temperatures (up to 363 C) and two distinct chemistries, which can be related through processes of conductive cooling, mixing with entrained seawater, and precipitation and dissolution of various mineral phases within the mound [ Tivey et al., 1994]. Geochronological studies indicate that hydrothermal activity on the mound has been intermittent over at least the past 20,000 years with a periodicity of 5-6,000 years, and that the current activity began about 50 years ago after a hiatus of about 4,000 years [ Lalou et al., 1990, 1993]. Ages of about 102,000 years from one of the relict zones and about 125,000 years from the low temperature field higher on the eastern median valley wall indicate that hydrothermal processes have been active since the formation of the underlying crust [ Lalou et al., 1988, 1993]. However, the mechanism by which upflow of hydrothermal fluids can be focused episodically in one area over such long periods of time is unclear. It has been suggested that listric faults associated with the median valley wall provide the pathways for fluids being heated by a source at the zero-age neovolcanic axis [e.g. Thompson et al., 1985]. Alternatively, discrete volcanic centers may act as the heat source for localized activity and exert some structural control. This is supported by observations of very recent volcanics on the volcanic dome associated with the presently active mound [ Zonenshain et al., 1989]. More recently, based on observations of east-west faults high on the eastern wall in the vicinity of the low temperature field, Karson and Rona [1990] suggested that the intersection of these transfer faults with ridge-parallel faults may concentrate hydrothermal activity. However, no direct evidence exists for the extension of east-west faults from the low temperature field to the presently active TAG mound.

The Snake Pit hydrothermal field at about 23 22N covers an area of 150 m x 300 m along the shallowest portion of an intensely fissured neovolcanic ridge. It consists of three mounds aligned east-west and elongated in a direction parallel to the ridge axis. The eastern one is the most active with black smoker chimneys; the central one exhibits only diffuse, low temperature flow and relict chimneys; and the western one is highly tectonized but with black smoker fluids emanating from ``beehive'' structures [ Thompson et al., 1988; Fouquet et al., 1993]. Geochronological studies suggest two major episodes of activity at Snake Pit, the first one between 2,000-4,000 years ago, and the current one beginning about 80 years ago [ Lalou et al., 1990, 1993]. Fouquet et al. [1993] have proposed a model for the formation of Snake Pit that involves an initial hydrothermal event related to faulting and fissuring at the summit of the neovolcanic ridge which produced the elongate mounds. A later tectonic event related to graben formation faulted the western mound, and the hydrothermal circulation was rejuvenated after a volcanic episode, with the upflow zones more focused at the western and eastern mounds.

The other two Mid-Atlantic Ridge sites---Broken Spur and Lucky Strike---were cursorily investigated by submersible in 1993, with further studies being carried out in 1994. At Broken Spur, which is located at about 29 N in the axial graben of the neovolcanic ridge in a similar setting to Snake Pit, three discrete black smoker sites consisting of structures similar to those observed at Snake Pit have been identified [ Murton et al., 1993]. In contrast, the setting of the Lucky Strike hydrothermal field at 37 N is more similar to Axial Seamount on the Juan de Fuca Ridge than to the other Mid-Atlantic Ridge sites. It is located at a depth of 1600 m in a depression between three cones that make up the summit of the Lucky Strike Seamount. Active vent sites are dispersed over an area at least 700 m long and 300 m wide, although the bounds of the field are not yet determined [ Langmuir et al., 1993]. The dispersed nature of the distribution of hydrothermal activity at this site contrasts with the more localized distribution seen at Snake Pit and TAG, although the presence of both active chimneys and extensive relict sulfide deposits attest to episodicity in venting similar to the other Atlantic sites.

In spite of the differences in volcanic and tectonic setting of Mid-Atlantic Ridge sites, they all share two characteristics in common: the volume of the hydrothermal deposit at each site is on average larger than observed at most other mid-ocean ridge hydrothermal sites, and there is evidence for episodicity in hydrothermal activity over long time periods. This suggests there are fundamental differences in the persistence or availability of heat sources and the maintenance of effective fluid flow conduits at slow-spreading ridges compared with fast-spreading ridges.