Page 159 - Global Tectonics

OCEAN RIDGES 145

and the different rheological properties of the crust In general therefore there is good agreement
compared to the mantle, oceanic crust being more between the theoretical models for the creation of
ductile at high temperatures than the mantle. As out- oceanic crust and observations made on in situ ocean

lined in Section 6.9, the thermal regime beneath a ridge floor and on ophiolites. Certain aspects however are

crest is influenced by the rate at which magma is sup- still problematic. The evolution of a median valley as
plied to the crust, which depends on the spreading rate. accretion occurs, that is, the way in which its fl anks
As a consequence the brittle–ductile transition (at are uplifted and the normal faults ultimately reversed,
approximately 750°C) occurs at a shallower depth in the is poorly understood. This is particularly true for the
crust at a fast-spreading ridge compared to a slow- amagmatic segments of very slow- and ultraslow-
spreading ridge that has a lower rate of magma supply. spreading ridges where mantle material is emplaced

This in turn implies that at a fast-spreading ridge there directly to the sea floor. The details of the formation
is a much greater volume, and hence width, of ductile of the gabbroic layer 3, from a steady state or tran-
lower crust. This ductile crust effectively decouples the sient magma chamber, are also the subject of much
overlying brittle crust from the viscous drag of the con- debate.
vecting mantle beneath, and the tensile stresses pulling
the plates apart are concentrated in a relatively thin and
weak layer that extends by repeated tensile fracture in
a very narrow zone at the ridge axis. On a slow-spread- 6.11 PROPAGATING
ing ridge the brittle layer is thicker and the volume of
are distributed over a larger area and there is more RIFTS AND
ductile crust much smaller. As a result the tensile stresses
viscous drag on the brittle crust. In this situation the
upper brittle layer deforms by steady state attenuation MICROPLATES
or “necking” in the form of a large number of normal
faults creating a median valley.
Chen & Morgan (1990) demonstrated that for crust The direction of spreading at an ocean ridge does not
of normal thickness and appropriate model parameters always remain constant over long periods of time, but
the transition from smooth topography with a buoyant may undergo several small changes. Menard & Atwater
axial high to a median rift valley is quite abrupt, at a full (1968) proposed that spreading in the northeastern
−1

spreading rate of approximately 70 mm a as observed. Pacific had changed direction five times on the basis

The model also predicts that for thicker crust forming of changes in the orientation of major transform faults
at a slow rate of spreading, as for example on the Reyk- (Section 5.9) and magnetic anomaly patterns. Small
janes Ridge immediately south of Iceland, there will be changes in spreading direction have also been proposed
a much larger volume of ductile crust, and smooth as an explanation of the anomalous topography associ-
topography is developed rather than a rift valley. Con- ated with oceanic fracture zones (Section 6.12).
versely, where the crust is thin on a slow-spreading Menard & Atwater (1968) made the assumption that
ridge, for example in the vicinity of fracture zones on the reorientation of a ridge would take place by smooth,
the Mid-Atlantic Ridge, the median valley will be more continuous rotations of individual ridge segments until
pronounced than at a segment center. Such instances of they became orthogonal to the new spreading direction
thicker or thinner crust than normal are also likely to (Fig. 6.19a). The ridge would then lie at an angle to the
be areas of higher or lower than normal upper mantle original magnetic anomaly pattern. Long portions of
temperatures respectively which will enhance the effect ridges affected in this way might be expected to devolve
in each case. The model was extended by Morgan & into shorter lengths, facilitating ridge rotation and cre-
Chen (1993) to incorporate a magma chamber as ating new transform faults (Fig. 6.19b). The change in

observed on the East Pacific Rise. This enhanced model spreading direction is thus envisaged as a gradual, con-
predicts that a steady state magma chamber can only tinuous rotation that produces a fan-like pattern of
−1
exist at spreading rates greater than 50 mm a and that magnetic anomalies that vary in width according to
the depth to the top of the chamber will decrease as position.
spreading rate increases, whilst retaining the essential An alternative model of changes in spreading direc-
features of the Chen & Morgan (1990) model. tion envisions the creation of a new spreading center