H. D. Needham Département de Géosciences Marines, IFREMER, Centre de Brest, B.P. 70,
29280 Plouzané Cédex, France
The Mid-Atlantic Ridge from the equator to the Charlie-Gibbs Fracture Zone (FZ) at 52°30'N displays large changes in the configuration of its axis, the distribution of depth along it and the morphological characteristics of individual rift segments and their lateral offsets. For the 15°- 40°N region, multibeam bathymetric data collected from the RV Charcot and during the Sigma (RV Atalante) cruise of the FARA project has removed some previous ambiguities about the location of the ridge axis and furnished precise information on the character of the axial region there. The along-strike variability, not attributable to differences in average spreading rate, raises still unresolved questions and paradoxes concerning the effects of inherited lines of rifting, plate motion geometry, regional thermal regimes of the mantle and shorter wavelength properties of the accretionary process.
The overall segmentation pattern can be described in terms of major fracture zones and of axial provinces (and subprovinces) identified by the obliquity, Y, of the average rift direction with respect to the azimuth of a corridor accomodating the rift-offset sequence. South of the 17°40'N OD (oblique discontinuity), the ridge crest is characterized by ~N-S rift segments. From the oblique Kurchatov FZ (41°30'N) to 51°N, the ridge axis strikes nearly N-S to about 43°30'N and then follows a curved trend except between 46°30' - 48°30'N. In the central area, from close to the Fifteen Twenty FZ to near the Azores triple point near 39°N, there are four main provinces, corresponding to the northward increase in the regional obliquity of the axial zone.
South of the Kane FZ (~24°N), excluding an excursion of the axis recovered by the ~ 40 km long Charcot offset at 21°25'N, the obliquity Y is ~5° for a rift sequence trending 0°-10°N. This is the only region where rift development can take place in a nearly orthogonal ("normal") stress regime, without major offsets and where the maximum length of rift segments could be approached. North of the 24°40'N OD, the modal rift azimuth is ~16°, but Y changes from ~ 10° between the Kane and Atlantis (30°N) FZs (with an apparent subprovince boundary near 26°30'N) to more than ~20° between the Atlantis and the Hayes FZs (33°30'N), where the rift-offset length ratio is less. North of the Hayes FZ, an overall obliquity of about 35°, reflecting the broad staircase-style configuration of the axis, is interrupted by the Oceanographer FZ and a subprovince (Y= ~27°) between ~ 36° and 37°N. Principal province widths vary according to rift and offset lengths but their high degree of internal organization is confirmed by the approximately linear distribution of the shallow centres of individual rift segments (particularly at the scale of subprovinces) within a band of about the same width as the maximum (~20 km) width of an inner floor (cf. South Hayes segment).
The shape of the present ridge-crest suggests that axial provinces are persistent features inherited from lines of early opening. Segmentation patterns would then relate to the obliquity of the province and to the limiting widths of the zones of high heat and mass transfer within which the mechanical properties of the crust inhibit or enable development of rift segments and offsets (e.g. Needham and Carré, 1990). Consequences of regional ridge crest obliquity on fine scale structural features (e.g., Searle and Laughton, 1981) can be observed in high resolution maps and can be experimentally modelled for a given stress field (e.g. Dauteuil and Brun, 1993). On evidence from the East Pacific Rise (EPR) and the Reykjanes ridge, minimum rift-offset development for a given obliquity is associated with a high thermal regime, with the corollary that maximum development corresponds to a colder province. That the segmentation mode can be unstable is illustrated by the apparent suppression of a step-like pattern of ridge offsets along part of the Reykjanes ridge since magnetic anomaly 13 (Vogt and Avery, 1974). Changes in geometry at the scale of a province will be associated with crustal variations to the degree that accretionary processes and the segmentation pattern (e.g., length and development of rift units, offset ages, obliquity) are interdependent. At the segment scale, inside corner areas would lie closer to to the province axis than their outside counterparts.
In the 15° - 40°N region, it remains to be shown if there are any meaningful correlations between provinces (and subprovinces) and different "vertical" temperature regimes (different mantle domains) and with longitudinal flow gradients (in particular with respect to the Azores). Apart from the anomalous Fifteen Twenty area, close to the North America - South America triple junction, there is some apparent correspondence of province obliquity with geochemical mantle signatures (e.g., Dosso et al, 1993), but this is of ambiguous significance. Paucity of off-axis fracture zones near the Azores may be attributable to thermally induced instability of offset zones. However, their absence south of the Atlantis FZ for crustal ages between magnetic anomalies 13 and 6 compared with their presence in crust of the same age between the Atlantis and Hayes FZs is clearly unrelated to the influence of the Azores, as is the present axial configuration there.
Rift offsets typically lack the strike-slip tectonic signature of well constrained transform faults. FZ A and the 18°30'N FZ are perhaps examples of limiting cases for small-intermediate offset lengths (20-30km). In the northern part of the area, the locations of the 37°50'N and 38°20'N (North Volcano) segments, assigned on the basis of bathymetric, seafloor imagery and magnetics data collected during the Sigma cruise, indicate that the Pico offset to the south (like that of the Thirty Seven Four North offset) is ~45 km long and that the offset between the 37°50' and 38°20' segments (here referred to as the Atalante offset) is 55 km long, nearly as long as the well defined Atlantis FZ at 30°N. Considering their location within the Azores domain, these offsets, reminiscent of some EPR fracture zones, may be "hot", unstable discontinuities. No progressive change with increasing distance from the Azores is evident in the character of the intersegment zones. These can be compared from one province to another, and within provinces, to guide fine-scale studies of crustal structure (spurs, small isolated structural and possibly volcanic features), processes (fissuring, faulting, hydrothermal activity), of how the accomodation of the discontinuity may be effected and of any dependance on the the properties of the rift zones which are offset.
Depth Distribution and Morphology
Principal features of the distribution of depth along the axis include: (i) the short wavelength variations associated with individual segments, (ii) generally increasing regional depth of the shallowest (sill) areas of these segments between the Azores and the Kane fracture zone, (iii) similar but smoother increase in depth of the rift-offset intersection zones, (iv) shallow depths (less than 1000 m) north of the Pico offset, (v) decreasing regional depth between there and just south of the Hayes FZ, (vi) stability of depth distribution between the Kane and Fifteen Twenty FZs, excluding an axial volcano at 20°30'N and the anomalously shallow North Charcot segment at 21°40'N, (vii) association of shallower sill depths with long rift segments, in turn commonly associated with large offsets, (viii) indication of a step like arrangement of depths at the scale of provinces and subprovinces.
The regional morphology of the ridge crest and gross segment morphology can be used to examine possible relationships with crustal accretion variables at different wave-lengths. For example, the organization of rift mountain pairs in the area north of the Hayes FZ, where their E-W distribution reflects rift mountain inside/outside corner area asymmetry, can be compared with that in areas with a smaller offset gradient. Integrated depths of the rift mountains, reflecting crustal level after isostatic recovery, provide a criterion for assessing some along-strike variations in addition to correlations with rift-offset geometry and to variations in focused magmatic activity and crustal thickness within individual segments (Detrick et al., 1995).
In the Azores area, rift mountain depths are represented by the unrifted, less than 1000 m deep segments north of the Pico offset, directly reflecting the immediate influence of the Azores hot-spot. Rift mountains shallower than 1500 m extend to 35°30'N. Between there and just beyond the Hayes FZ, areas shallower than 2000 m are prominent. Further south, rift mountain depth changes relatively little regardless of the depth of the rift segments. Even for deep rifts (e.g. near 21°N) they lie at approximately the same depth as they do next to nearby inner floors.
Dauteuil, O. & Brun, J.P., Nature 361,145-148 (1993).
Detrick, R., Needham, H.D. & Renard, V., J. Geophys. Res. 100, 3767-3787 (1995).
Dosso, L., Bougault, H. & Joron, J.L., Earth Planet. Sci. Lett. 120, 443-462 (1993).
Needham, H.D. & Carré, D., Oceanol. Acta 10, 343-347 (1990).
Searle, R. & Laughton, A., Oceanol. Acta C4, 5-13 (1981).
Vogt, P.R. & Avery, O.E., J. Geophys. Res. 79, 363-389 (1974).
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.