Marcia Maia1, Susanna Sichel2,Anne Briais3, Daniele Brunelli4,5, Nicolas Ferreira1, Marco Ligi5, Thomas Campos6 , Bérengère Mougel7 and Christophe Hémond1
1 Laboratoire Domaines Océaniques, CNRS-Université de Bretagne Occidentale, IUEM, Rue Dumont d’Urville, 29280 PLouzané, France
2 LAGEMAR, Universidade Federal Fluminense, Av. Litorânea, Niteroi, Brazil
3 GET, Observatoire Midi-Pyrenées, Av. Edouard Bélin, 35000, Toulouse, France
4 Dipartimento di Scienze Chimiche e Geologiche, Università di Modena e Reggio Emilia, Via Campi 183, 41125, Modena, Italy
5 ISMAR - CNR - Geologia Marina, Via Gobetti, 101, 40129, Bologna, Italy
6 Departamento de Geologia, Universidade Federal do Rio Grande do Norte, Natal, Brazil
7 Institut de Physique du Globe de Paris, Paris, France
Large portions of slow-spreading ridges have mantle-derived peridotites emplaced either on, or at shallow levels below the sea floor. Mantle and deep rock exposure in such contexts is linked to extension through low-angle detachment faults at oceanic core complexesor, along transform faults, to transtension due to small changes in spreading geometry.
At the St. Paul transform, in the Equatorial Atlantic, a large body of ultramafic rocks forms the archipelago of St. Peter & St. Paul. These islets, emplaced near the axis of the Mid-Atlantic ridge on the northern transform boundary of the St. Paul transform system, have intrigued geologists since Darwin’s time. They are made of variably serpentinized and mylonitized peridotites and the continuous uplift rate of 1.5 mm/yr reveals that they are presently under tectonic deformation. The existence of an abnormally cold upper mantle or cold lithosphere in the Equatorial Atlantic was, until now, the preferred explanation for the origin of these ultramafics.
The COLMEIA cruise, held in the Equatorial Atlantic in January/February 2013 in the area of the St. Paul transform system, is part of a joint effort between France and Brazil for the study of the Mid-Atlantic ridge near the St. Peter St. Paul’s islets. The scientific objective of the cruise was to study in detail the temporal evolution of the complex Saint Paul transform plate boundary, and the origin of the St. Peter -St. Paul ultramafic massif.
During the cruise we acquired multibeam echosounder bathymetry, backscattering, water column acoustic data, gravity, magnetics and seismics. Thirty-one successful dredges returned a wide variety of rocks, including gabbros and peridotites. Fifteen CTD stations with nephelometric profiles were cast and a hydrothermal plume signal was found, with a source probably located in the Mid-Atlantic Ridge segment south of the St. Paul transform system. No hydrothermal plume signal was found inside the transform system. Five autonomous hydrophones were moored in the SOFAR channel around the study area in order to monitor the seismic activity and whale presence and were recovered mid-2014. Of the five instruments, three recorded acoustic signals for one year. The preliminary analysis of these recordings shows a very high seismicity in the area near the islets.
Both the bathymetry data and the rocks recovered by dredging suggest that the image of a regional amagmatic Mid-Atlantic ridge is a simplistic view of the processes active in the St. Paul system. The ridge segments are short and narrow, with deep axial valleys. Axial depths are below 4000 m on average, and reach 5400 m in some nodal basins. There is no evidence for a clearly defined neo-volcanic ridge on the axial valley floors, but a few round volcanoes were observed in the axial valley of the central segment. The pattern of off-axis abyssal hills is highly variable from one segment to another. The northern segment displays a long sequence of magmatic abyssal hills. The central segment shows both hummocky ridges probably of magmatic origin, but also ridges where peridotites have been dredged. The southern segment shows few short, symmetric ridges made of peridotite and gabbros. Both the central and the southern segments display asymmetric core complexes nucleating at segments ends. This variety of off-axis morphologies suggest that accretionary processes along the intra-transform segments are unstable and highly variable in space and time. Thus, significant variations in the spreading style were recognized, with a more magmatic northern segment and comparatively less magmatic central and southern segments. However, the existence of long-lived core complexes at the western flanks of these later segments suggests that, instead of an amagmatic regime with mantle exhumation, we are observing a reduced melt extraction regime probably controlled by a cold, thick lithosphere where magma is retained in the crust to create large gabbro bodies.
Another striking result is the evidence that the origin of the St. Peter & St. Paul archipelago is linked to compressive stresses along the transform fault. The islets are the summit of a large push-up ridge formed by deformed mantle located in the center of a positive flower structure, where large portions of mylonitized mantle are uplifted. The transpressive stress field can be explained by the propagation of the northern Mid-Atlantic Ridge (MAR) segment into the transform domain, which induced the migration and segmentation of the transform fault creating a series of restraining step-overs. A counterclockwise change in plate motion at ~11 Ma initially generated extensive stresses in the transform domain, forming a flexural transverse ridge. Shortly after the plate reorganization, the MAR segment started to propagate southwards, adapting to the new spreading direction. Enhanced melt supply at the ridge axis, possibly due to the Sierra Leone thermal anomaly induced the robust response of this segment.
