Journal of Conference Abstracts

MARFLUX/ATJ. Hydrothermal Activity Along the Mid-Atlantic Ridge Axis, in the Area of the Azores Triple Junction. Mn, CH₄, He Plumes from Dynamic Hydrocasts

Henri Bougault IFREMER, Centre de Brest, B.P. 70, 29280 Plouzané Cédex, France

henri.bougault@ifremer.fr

Joël Radford Knoery IFREMER, Centre de Brest, B.P. 70, 29280 Plouzané Cédex, France

Jean Luc Charlou IFREMER, Centre de Brest, B.P. 70, 29280 Plouzané Cédex, France

H. David Needham IFREMER, Centre de Brest, B.P. 70, 29280 Plouzané Cédex, France

Pierre Appriou Dept. Chimie, Fac. Sciences UBO, 6 Av. Le Gorgeu, 29200 Brest, France

Martine Aballéa Dept. Chimie, Fac. Sciences UBO, Av. Le Gorgeu, 29200 Brest, France

Philippe Jean-Baptiste Lab. Géochim. Isotopique, CEA CEN Saclay, LMCE Bat. 709, 91191 GIF s/Yvette, France

Chris R. German Southampton Oceanography Center, Empress Dock, European Way, Southampton, SO14 3ZH, UK

Miguel Miranda Centro de Geofisica, Univ. de Lisboa, Rua da Escola Politecnica 58, 1200 Lisboa, Portugal

Charles H. Langmuir Lamont Doherty Earth Observatory, Columbia University, Palisades, NY 10964, USA

The Mid-Atlantic Ridge (MAR) in the area of the Azores Triple Junction (ATJ)

From 36°N to 40°N, the MAR displays the influence of the Azores Hot Spot. The axial depth varies from about 3000 m away from the ATJ to 1200 m at the ATJ (38°40'N) with a minimum of 800 m at the axial volcano of 38°20'N. In addition to this main trend, the ridge segments show quite different characteristics. The Lucky Strike, Menez Gwen and 38°20'N segments are controlled by an extremely large magmatic budget: the depth variation between the central part and the ends of these segments is extreme (i.e.: from 1700 m to 2800 m for Lucky Strike). All segments display a classic rift valley or a graben, with one exception: the South ATJ segment does not display an axial rift but instead is inflated and shows a reverse V shape cross-section. The Famous segment represents the typical morphology of the MAR segments, whereas the Amar segment is tectonically more complex. The ridge segments are delineated by conventional transform faults or by more complex offsets of different types. The main purpose of the MARFLUX/ATJ project (MAR hydrothermal fluxes at the ATJ) was to describe the variation in hydrothermal activity along these various segments presenting these extreme differences in their characteristics. Hydrothermal activity is evaluated by using Mn, CH₄ and He hydrothermal tracers in the water column along the ridge axis.

The Dynamic Hydrocasts

The normal and traditional way to sample seawater is to collect samples vertically. This operation is usually conducted either by attaching bottles along a cable at pre-selected depths or by using a CTD-rosette. This punctual way of sampling, at selected locations and selected depths, provides detailed vertical profiles but is not convenient to describe the variation in hydrothermal activity at the segment scale (20 to 80 km). The Dynamic Hydrocast was developed to overcome this difficulty both to describe hydrothermal plumes of different tracers and to locate potential hydrothermal sites. Instead of sampling along a vertical line, Dynamic Hydrocasts sample horizontally or along a line parallel to the bottom. The equipment is composed of four modules attached at selected depth along a non conductive towing cable. Each module is composed of ten bottles (made of a piston and a cylinder) and a modified seabird unit. In order not to sample "punctually", each bottle is filled up proportionally to the distance covered thanks to a propeller acting a pump. When a bottle is filled, the system goes automatically to fill the next one. The distance over which each bottle is filled can be adjusted by the propeller. The seabird unit records temperature, salinity, depth and the time during which each bottle is filled. Each measurement made from the Dynamic Hydrocast samples represents an average value over the distance where the bottle was filled. At the end, we obtain an array of data along a vertical cross-section of seawater made up of forty "average" concentrations distributed along four parallel lines of ten samples each. This data set allows us to obtain a picture of a vertical cross-section of a plume for the various elements or compounds investigated. Figure 1 is an example of a vertical cross section of the hydrothermal plume along the Amar segment.

Chemical hydrothermal anomalies everywhere in the water column

From the results recorded by Dynamic Hydrocasts, hydrothermal anomalies are detected everywhere between 36°N to 38°N along a cumulative distance of 108 miles. The hydrothermal plume is present everywhere within the rift valley of a ridge segment with minimum values in the plume larger than ten times the seawater background (i.e. Fig. 1). It is likely that the rift valley prevents the hydrothermal inputs into seawater from being dispersed in the open ocean and that currents along the axial valley homogenise the plume. Long-lived tracers (i.e. He) display a smaller source signal/segment background than short-lived tracers (i.e. Mn). The unexpected results of this continuous record along axis consist in both the pervasive character of hydrothermal plumes from 36°N to 38°N and the amplitude of the signals, particularly in the Famous and Amar segments. From the ATJ northwards, discrete vertical hydrocasts reveal weak hydrothermal signals, if any, except in the area of 39°40'N.

Fig. 1: Hydrothermal plume. The three figures represent a vertical cross section of ³He, and of methane and manganese iso-concentrations along the axis of the Amar segment of the MAR. Between 1700 and 2300 m depth, chemical signals are detected everywhere along the segment, with values at least 10 times larger than the seawater background. The three chemical tracers have different properties and consequently do not display identical cross-sections of the plume. All chemical tracers focus on a potential hydrothermal area, the Rainbow site at 27 km, near 36°25'N.

Hydrothermal sites in different geological settings

Two hydrothermal sites have already been recognised in the area of interest: Lucky Strike (Langmuir et al., in press) and Menez Gwen (Fouquet et al., 1995). Both sites are located on the central topographic highs of segments which are characterised by intense magmatic production. These segments are almost like huge volcanoes centred on the middle of the segments and cross-cut by a graben. The plumes recorded by the dynamic hydrocasts show that there are potential hydrothermal areas in other settings: the rift valley walls and the ends of segments, transforms or more complex offsets. The south-west wall and the southern east-west offset of the Famous segment display two distinct large plumes at different depths. A potential site (Rainbow: German et al., 1996) has been clearly identified in the Amar segment: the Dynamic Hydrocast along the Amar segment shows consistent plume maxima for He, CH₄ and Mn at this location (Fig. 1, at 27 km). These plume maxima, several hundreds of meters above the floor of the valley, are associated with a non-typical axial structure resulting from a complex tectonic offset. It is thought that the south-west end of Famous and the Rainbow site on Amar are two examples of complex offsets controlled by tectonics which favours seawater circulation into the crust.

From previous studies of hydrothermal activity along the MAR and from this continuous record by the Dynamic Hydrocasts along 108 miles of the ridge axis during the MARFLUX/ATJ project, it appears that hydrothermal activity along the MAR would be as common as along the EPR but that it develops in more diverse settings. We think that different types of Ocean / Lithosphere exchanges are still to be discovered along the slow spreading ridges as demonstrated recently at 14°45'N (Krasnov et al., 1995).

References

Fouquet, Y. et al., Nature 377 (1995).

German, C.R. et al., Earth Planet. Sci. Lett. 93-104 (1996).

Krasnov, S.G. et al., In Hydrothermal Vents and Processes, The Geological Society, London, 43-64 (1995).

Langmuir, C. et al., Earth Planet. Sci. Lett. (in press)