Oceanic Fracture Zones

Transform faults in the oceans are well defined, in the absence of sedimentary cover, by fracture zones. These are long, linear, bathymetric depressions that normally follow arcs of small circles on the Earth's surface perpendicular to the offset ridge (Bonatti & Crane, 1984). The apparent relative simplicity of oceanic fracture zones is no doubt due in part to the fact that they are commonly studied from the sea surface several kilometers above the ocean floor. Direct observations of a fracture zone on the Mid-Atlantic Ridge (Choukroune et al., 1978) have shown that it consists of a complex swarm of faults occupying a zone 300-1000 m in width. Searle (1983) suggests that these multi-fault zones are

The Leaky Transform Zone

Fig. 6.23 The tectonic elements of the Juan Fernandez microplate, together with magnetic anomalies numbered according to the timescale of Fig. 4.8. TR, transform; PT, paleo-transform; FZ, fracture zone; WIPF, WOPF, EIPF, EOPF: western/eastern, inner/outerpseudofaults (redrawn from Larson et al., 1992, with permission from Nature 356,571-6. Copyright © 1992 Macmillan Publishers Ltd).

Fig. 6.23 The tectonic elements of the Juan Fernandez microplate, together with magnetic anomalies numbered according to the timescale of Fig. 4.8. TR, transform; PT, paleo-transform; FZ, fracture zone; WIPF, WOPF, EIPF, EOPF: western/eastern, inner/outerpseudofaults (redrawn from Larson et al., 1992, with permission from Nature 356,571-6. Copyright © 1992 Macmillan Publishers Ltd).

wider and more common on fast-spreading ridges such as the East Pacific Rise.

Fracture zones mark both the active transform segment and its fossilized trace. It has been suggested (Collette, 1979) that the fractures result from thermal contraction in the direction of the ridge axis. The internal stresses caused by contraction are much larger than the breaking strength of the rocks, and it is possible that fracture zones develop along the resulting lines of weakness.

Dredging of fracture zones has recovered both normal oceanic crustal rocks and rocks which show much greater metamorphism and shearing. Very commonly large blocks of serpentinite lie at the bases of the fracture zones. Bonatti & Honnorez (1976) and Fox et al. (1976) have examined specimens recovered from the thick crustal sections exposed in the large equatorial Atlantic fracture zones, which were found to consist of ultramafic, gabbroic and basaltic rock types and their metamorphosed and tectonized equivalents. Serpentinite intrusion appears to be quite common within fracture zones, accompanied by alkali basalt volcanism, hydrothermal activity, and metallogenesis. Investigations of the Vema Fraction Zone (Auzende et al., 1989) have indicated a sequence similar to normal oceanic layering. St Peter and St Paul rocks, in the equatorial Atlantic, which lie on a ridge associated with the St Paul Fracture Zone, are composed of mantle peridotite.

In the North Atlantic, fracture zone crust is very heterogeneous in thickness and internal structure (Detrick et al., 1993b). It is often thin (<3 km) with low

Fig. 6.24 Differential topography resulting from transform faulting of a ridge axis.

Fig. 6.26 Development of a leaky transform fault because of a change in the pole of rotation.

Fig. 6.25 Different types of basement morphology across fracture zones (redrawn from Bonatti, 1978, with permission from Elsevier).

Fig. 6.26 Development of a leaky transform fault because of a change in the pole of rotation.

Fig. 6.25 Different types of basement morphology across fracture zones (redrawn from Bonatti, 1978, with permission from Elsevier).

seismic velocities, and layer 3 is absent. The crustal thinning may extend several tens of kilometers from the fracture zone. Geologically, this structure may represent a thin, intensely fractured and hydrothermally altered basaltic layer underlain by serpentinized ultra-mafic rocks. The apparent thickness variations may reflect different extents of serpentinization. The thin mafic crust is thought to be a result of reduced magma supply at ridge offsets as noted in Section 6.7.

Ocean fracture zones must bring oceanic crust of different ages into juxtaposition. The depth of the sea floor is dependent upon its age (Section 6.4), and so it would be expected that a scarp would develop across the fracture zone from the younger, higher crust to the lower, older crust (Menard & Atwater, 1969; DeLong et al., 1977) (Figs 6.24, 6.25b). The rate of subsidence of oceanic lithosphere is inversely dependent upon the square root of its age (DeLong et al., 1977), so the higher, younger crust subsides more rapidly than the lower, older side. The combination of contraction in the vertical plane and horizontally perpendicular to the direction of the ridge axis would result in a small component of dip-slip motion along the fracture zone away from the active transform fault. DeLong et al. (1977) have suggested that this small amount of dip-slip motion could give rise to fracture zone seismicity and deformation of rocks within the floor and walls.

Transverse ridges are often found in association with major fracture zones and can provide vertical relief of over 6 km. These run parallel to the fractures (Bonatti, 1978) on one or both margins. They are frequently anomalous in that their elevation may be greater than that of the crest of the spreading ridge (Fig. 6.25c,d). Consequently, the age-depth relationship of normal oceanic lithosphere (Section 6.4) does not apply and depths differ from "normal" crust of the same age. The ridges do not originate from volcanic activity within the fracture zone, nor by hotspot activity (Section 5.5), but appear to result from the tectonic uplift of blocks of crust and upper mantle. Transverse ridges, therefore, cannot be explained by normal processes of lithosphere accretion. Bonatti (1978) considers that the most reasonable mechanism for this uplift is compressional and tensional horizontal stresses across the fracture zone that originate from small changes in the direction of spreading, so that transform movement is no longer exactly orthogonal to the ridge. Several small changes in spreading direction can give rise to episodic compression and extension affecting different parts of the fracture zone. This has caused, for example, the emergence of parts of transverse ridges as islands, such as St Peter and St Paul rocks, and their subsequent subsidence (Bonatti & Crane, 1984).

Lowrie et al. (1986) have noted that, in some fracture zones, the scarp height may be preserved even after 100 Ma. They accept that some parts of fracture zones are weak, characterized by active volcanism, and maintain the theoretical depths predicted for cooling lithosphere. Other parts, however, appear to be welded together and lock in their initial differential bathymetry. The differential cooling stresses would then cause flexure of the lithosphere on both sides of the fracture zone. Future work will reveal if there is any systematic pattern in the distribution of strong and weak portions of fracture zones.

There are certain oceanic transform faults in which the direction of the fault plane does not correspond exactly to the direction of spreading on either side so that there is a component of extension across the fault. When this occurs the fault may adjust its trajectory so as to become approximately parallel to the spreading direction by devolving into a series of fault segments joined by small lengths of spreading center (Fig. 6.26). A fault system in which new crust originates is termed a leaky transform fault (Thompson & Melson, 1972; Taylor et al., 1994). An alternative mechanism for leaky transform fault development occurs when there is a small shift in the position of the pole of rotation about which the fault describes a small circle. The fault would then adjust to the new small circle direction by becoming leaky.

Continental rifts and rifted margins

0 0

Post a comment