Anne Briais CNRS, 18, Ave E. Belin, 31055 Toulouse Cédex, France
Heather Sloan Dept. Geological Sciences, University of Durham, DH1 3LE, UK
Bramley J. Murton Southampton Oceanography Center, Empress Dock, Southampton SO14 3ZH, UK
Lindsay M. Parson Southampton Oceanography Center, Empress Dock, Southampton SO14 3ZH, UK
We analyzed TOBI side-scan sonar images collected during Charles Darwin cruise CD76 in the axial valley of the Mid-Atlantic Ridge (MAR) between 27°N and the Atlantis Transform Fault. Two survey tracks were made along both sides of the axial valley. Mosaics of the two side-scan sonar swaths provide a continuous image of the axial valley and the inner valley walls along seven second-order segments of the MAR. Intra-segment and inter-segment analysis reveals a large degree of variability of the segments and the discontinuities, but no regional trend in the tectonics or volcanics.
Narrow, strongly back-scattering, linear targets are interpreted to be fault scarps and fissures. We use their analysis to characterize the tectonic activity along the segments and in the discontinuities, as well as the evolution of fault scarps and volcanic terrains. Small graben, or small faults and fissures affecting fresh-looking volcanic ridges, are inferred to reflect recent extensional tectonics. Areas of dense fissures, also frequent on the valley floor, are inferred to reflect older terrain highly affected by extension. Within each segment, large-throw faults are often observed near the inside-corner highs, whereas numerous, small-throw faults characterize the outside corner. Since our data image only the valley floor, this observation implies that the asymmetry is generated very early in the tectonic evolution of the valley walls. Dark, mottled and streaky texture near the base of such large-throw fault scarps is interpreted as serpentinite, sediment and talus.
As in previous works on deep-towed side-scan sonar images (e.g., Parson et al., 1993, Smith et al., 1995), we distinguish two types of volcanic morphologies: hummocky volcanoes or volcanic ridges and smooth, flat-topped volcanoes. Individual hummocks are about 100200 m in diameter. Sub-circular, flat-topped volcanoes are usually 50200 m high and 12 km in diameter. A caldera is often observed at their top. High-reflectivity, hummocky mounds with a sharp contrast and well defined contours are interpreted as the site of the most recent volcanic activity. Hummocky ridges seem to have no preferential location along the segments, or even in the discontinuity areas, but they tend to be more voluminous near the segment mid-points. The volcanic ridges are often cut by small normal faults parallel to the valley walls or merging with the inner faults. The overprinting of these faults and fissures by hummocks indicates that coeval volcanic and tectonic activity plays a major role in hummocky mound formation. Although ridges and individual volcanoes vary in size, volcanic constructions do not appear to grade from one end-member to the other. Smooth, flat-topped volcanoes are concentrated near the segment ends for four out of seven segments mapped in this survey. The remaining three segments count a significant number of flat-topped volcanoes distributed along their length (28°15'N-28°42'N, 28°51'N-29°23'N and 29°23'N-29°49'N). Because their construction is usually related to a relatively high eruption rate (Griffiths and Fink, 1993), the abundance of smooth, flat-topped volcanoes is interpreted as an indicator that these three segments are currently, or were recently, undergoing a magmatically active phase. These segments are actually characterized by axial valleys with shallow mid-points, as well as large Mantle Bouguer anomaly lows (Lin et al., 1990), and are interpreted as magmatic segments (Sempéré et al., 1993).
A model for cyclic emplacement of volcanic terrains in the axial valley
We infer that the two different types of volcanic morphologies observed in the axial valley have different origins in the upper mantle and crust, and that they correspond to different phases in the magmatic activity of a segment. Flat-topped volcanoes are commonly associated with large magma effusion rates, while hummocky volcanoes and small ridges are more frequently interpreted as resulting from low-rate magmatic eruptions (Griffiths and Fink, 1993), each corresponding to a small magma pocket in depth (Smith and Cann, 1992). The progressive accumulation of hummocky volcanoes forms the larger axial volcanic ridges (Smith and Cann, 1992). We propose that flat-topped volcanoes are built during phases of high magmatic supply, whereas hummocky volcanoes and volcanic ridges are constructed more progressively, from the extraction of magma in the magma chamber. Phases of high magmatic activity may result from the arrival near the surface of magmatic waves from the mantle. Part of this magma goes to replenish a magma chamber. Part of it is rapidly extruded in the axial valley, forming flat-topped volcanoes. This phase would lead to dynamic emplacement of magma all along the axial valley. During periods of lower magmatic activity, extrusion of magma from the cooling magma chamber is favored by tectonic extension, leading to the construction of hummocky ridges. This phase would correspond to a passive emplacement of volcanics due to tectonic extension, mostly above the cooling magma chamber near the center of the segment. During this phase, the flat-topped volcanoes previously emplaced near the center of the segment are dismantled by tectonic activity overprinted by hummocky constructions. Those emplaced near the segment ends are progressively dissected by faulting. Many of them may be preserved for a long period of time if the tectonic extension is accommodated along a few normal faults in the axial valley walls rather than distributed in the axial valley floor. Such a cyclic evolution of emplacement and dismantlement of volcanics was suggested by Stakes et al. (1984) for the FAMOUS and AMAR segments of the MAR, and was supported by geochemical evidence. Parson et al. (1993) and Murton et al. (1993) also suggested a cyclic evolution of axial volcanic ridges of the Reykjanes Ridge, based on tectonic and volcanic morphology evidence. In our model, the basalts forming the smooth, flat-topped volcanoes should be more primitive than those forming the hummocks. Such a geochemical evolution is suggested by the composition analysis of MAR basalts (Stakes et al., 1984) and by the composition of lavas forming shield volcanoes in Iceland (e.g., Jakobsson et al., 1978).
Griffiths, R.W. & Fink, J.H., J. Fluid Mech. 252, 667-702 (1993).
Jakobsson, S.P., Jonsson, J. & Hido, S., J. Petrol. 19, 669-705 (1978).
Lin, J., Parmentier, G.M., Schouten, H., Sempéré, J.-C. & Zervas, C., Nature 344, 627-632 (1990).
Murton, B.J. & Parson, L.M., Tectonophysics 222, 237-257 (1993).
Parson, L.M., et al., Earth Planet. Sci. Lett. 117, 73-87 (1993).
Sempéré, J.-C., Lin, J., Brown, H.S., Schouten, H. & Purdy, G.M., Mar. Geophys. Res. 15, 153-200 (1993).
Smith, D.K. & Cann, J.R., J. Geophys. Res. 97, 1645-1658 (1992).
Smith, D.K., et al., J. Volcanol. Geotherm. Res. 67, 233-262 (1995).
Stakes, D.S., Shervais, J.W. & Hopson, C.A., J. Geophys. Res. 89, 6995-7028 (1984).
Index of Volume 1 Number 2
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Thursday, June 20, 1996.
© 1996 Cambridge Publications.