Geoscience Reference
In-Depth Information
need not be maintained at second-order discontinuities. On intermediate- and
fast-spreading ridges, second-order discontinuities include overlapping spreading
centres, whereas at slow spreading ridges they are represented by oblique offsets
of the ridge axis and deep depressions in the seabed. Third-order discontinuities
at intermediate- and fast-spreading ridges are also associated with overlapping
spreading centres, but with much reduced offset and depth anomalies. Fourth-
order discontinuities have very small offsets in the summit graben or deviations
in the axis; they rarely have a depth anomaly. On slow-spreading ridges these
fine-scale third- and fourth-order discontinuities are short-lived features, marked
by gaps between and within individual volcanoes.
While classification of ridge segments and discontinuities into four orders is
very helpful, it should be appreciated that this is not a physical constraint but a
description - reality is complex and time-dependent.
Detailed studies of the Oceanographer Transform Fault on the Mid-Atlantic
Ridge have shown that its active region is centred on the deepest part of the
valley and is confined to a zone a few hundred metres to a few kilometres wide.
Figure 9.36(a), a seismic-reflection profile across the Vema transform fault at
6
◦
Nonthe Mid-Atlantic Ridge, shows a narrow region of faulting, which marks
the boundary between the African and South American plates. The deformation
in the sediments is confined to a narrow zone less than a kilometre across.
Several detailed seismic-refraction experiments have been shot along and
across large- and small-offset Atlantic transform faults, the results of which indi-
cate that their crustal structures are anomalous. Instead of normal 6-8-km-thick
oceanic crust, the fracture-zone crust is considerably thinner, generally 3-5 km
thick but in places only 2-3 km thick. The thin crust is confined to a region less
than 10 km wide. In addition, lower than normal compressional crustal veloci-
ties are characteristic, and layer-3 velocities are frequently not observed. Figures
9.37(a)-9.37(e) show seismic models crossing transform faults at slow- and very-
slow-spreading ridges along the strike of the median valley. It is clear that, over
some tens of kilometres, the crustal thickness decreases towards the transform
fault, in addition to the very thin crust occurring in the transform fault itself. The
Kane and Oceanographer transform faults have been studied in some detail by
submersibles and ODP drilling has taken place on Atlantis Bank adjacent to the
Atlantis II fracture zone. It is clearly documented that rocks usually associated
with the lower crust and upper mantle (i.e., gabbroic and ultramafic rocks) out-
crop on escarpments at transform faults. This is hard to explain if the crust is
of normal thickness and again suggests that transform-fault crust is thinner than
normal.
A detailed wide-angle seismic-reflection survey has been made over the Blake
Spur fracture zone (a small western Atlantic fracture zone with about 12 km offset
on 140-Ma-old lithosphere). Normal oceanic crust is present up to 10 km from the
fracture zone, but the seismic-velocity structure in a zone 10-20 km wide centred
on the fracture zone is anomalous (Fig. 9.37(f)). Beneath the sediments there is
