Alongaxis Segmentation Of Oceanic Ridges

Many early investigations of ocean ridges were essentially two-dimensional in that they were based on quite widely spaced profiles oriented perpendicular to their strike. More recently "swath"-mapping systems have

volcan®

Dikes volcan®

Dikes

e D6

e D6

Fig. 6.11 Interpretive models of magma chambers beneath a fast (a) and slow (b) spreading ridge (modified from Sinton & Detrick, 1992, by permission of the American Geophysical Union. Copyright © 1992 American Geophysical Union).

Distance (km)

Fig. 6.11 Interpretive models of magma chambers beneath a fast (a) and slow (b) spreading ridge (modified from Sinton & Detrick, 1992, by permission of the American Geophysical Union. Copyright © 1992 American Geophysical Union).

been employed which provide complete areal coverage of oceanic features. These systems have been used to reveal variations in the structure of ocean ridges along strike. A review of these developments was provided by Macdonald et al. (1988).

Studies of the East Pacific Rise have shown that it is segmented along its strike by nontransform ridge axis discontinuities such as propagating rifts (Section 6.11) and overlapping spreading centers (OSC), which occur at local depth maxima, and by smooth variations in the

depth of the ridge axis. These features may migrate up or down the ridge axis with time.

OSCs (MacDonald & Fox, 1983) are nonrigid discontinuities where the spreading center of a ridge is offset by a distance of 0.5-10 km, with the two ridge portions overlapping each other by about three times the offset. It has been proposed that OSCs originate on fast-spreading ridges where lateral offsets are less than 15 km, and true transform faults fail to develop because the lithosphere is too thin and weak. The OSC geometry is obviously unstable, and its development has been deduced from the behavior of slits in a solid wax film floating on molten wax, which appears to represent a reasonable analogue (Fig. 6.12a). Tension applied orthogonal to the slits (spreading centers) causes their lateral propagation (Fig. 6.12b) until they overlap (Fig. 6.12c), and the enclosed zone is subjected to shear and rotational deformation. The OSCs continue to advance until one tip links with the other OSC (Fig. 6.12d). A single spreading center then develops as one OSC becomes inactive and is moved away as spreading continues (Fig. 6.12e).

Fast-spreading ridges are segmented at several different scales (Fig. 6.13). First order segmentation is defined by fracture zones (Section 4.2) and propagating rifts (Section 6.11), which divide the ridge at intervals of 300-500 km by large axial depth anomalies. Second order segmentation at intervals of 50-300 km is caused by nonrigid transform faults (which affect crust that is still thin and hot) and large offset (3-10 km) OSCs that cause axial depth anomalies of hundreds of meters. Third order segmentation at intervals of 30-100 km is defined by small offset (0.5-3 km) OSCs, where depth anomalies are only a few tens of meters. Finally, fourth order segmentation at intervals of 10-50 km is caused by very small lateral offsets (<0.5 km) of the axial rift and small deviations from axial linearity of the ridge axis (DEVALS). These are rarely associated with depth anomalies and may be represented by gaps in the volcanic activity within the central rift or by geochemical variation. Clearly fourth order segmentation is on the same along-axis length scale as the intervals between pure melt pockets in the melt lens documented by Singh et al. (1998) (Section 6.6).

Third and fourth order segmentations appear to be short-lived, as their effects can only be traced for a few kilometers in the spreading direction. Second order segmentations, however, create off axis scars on the spreading crust consisting of cuspate ridges and elongate basins that cause differential relief of several hundred

/

Two knife cuts in frozen wax

V-

film, spreading initialized

V

(a)

/

Propagation of spreading

centers along strike

a

(b) /

Spreading centers overlap and curve towards each other, encircling a zone of shear and rotational deformation. OSC geometry established

Spreading centers overlap and curve towards each other, encircling a zone of shear and rotational deformation. OSC geometry established

/

Progressive shear and

rotational deformation

y

continues until one

OSC links with

F-

the other

(d)

1

A continuous spreading

//il

center is established,

w

abandoned OSC and

1/

overlap zone rafted

1

away

(e)

1

Fig. 6.12 Possible evolutionary sequence in the development of an overlapping spreading center (redrawn from MacDonald & Fox, 1983, with permission from Nature 302,55-8. Copyright © 1983 Macmillan Publishers Ltd).

meters. The scars do not follow small circle routes about the spreading pole, but form V-shaped wakes at 60-80° to the ridge. This indicates that the OSCs responsible for the segmentation migrate along the ridge at velocities of up to several hundred millimeters per year. Figure 6.14 summarizes the three general cases for the evolution of such ridge-axis discontinuities in terms of the movement of magma pulses.

Slow - spreading ridge

Fig. 6.13 Summary of the hierarchy of segmentation on fast- and slow-spreading ridges. S,, S2, S3, and S4 - first to fourth order ridge segments. D,, D2, D3, and D4, - first to fourth order discontinuities (redrawn from Macdonald et al., 1991, Science 253,986-94, with permission from the AAAS).

The different scales and hence "orders" of ridge segmentation were first recognized on the fast-spreading East Pacific Rise. Segmentation also exists on the slow-spreading Mid-Atlantic Ridge but takes on somewhat different forms, presumably because the ridge crest is cooler and hence more brittle (Sempere et al., 1990; Gente et al., 1995) (Fig. 6.13). First order segmentation is defined by transform faults, but overlapping spreading centers are absent and second order segments are bounded by oblique offsets of the ridge axis associated with deep depressions in the sea floor. Third and fourth order segmentation is in the form of geochemical variations and breaks in volcanic activity in the inner valley floor. The latter generate discrete linear volcanic ridges 2-20 km long and 1-4 km wide (Smith & Cann, 1993). Again first and second order segmentation is long-lived and third and fourth order segmentation is short-lived. Segmentation on the ultraslow-spreading Gakkel Ridge, which does not even exhibit transform faulting, is in the form of volcanic and tectonic, or magmatic and amag-matic, segments (Michael et al., 2003) (Section 6.9).

The first order segment boundaries, transform faults, are marked by pronounced bathymetric depressions (Section 6.12). They are often underlain by thinner crust than normal and anomalously low sub-Moho seismic velocities that may be due to partial serpentini-zation of the mantle as a result of seawater percolating down through the fractured crust. This thinning of the crust in the vicinity of fracture zones is particularly marked on the slow-spreading Mid-Atlantic Ridge (White et al., 1984; Detrick et al., 1993b). By contrast, the central portions of segments are elevated, have crust of normal thickness, and thinner lithosphere. This implies that the supply of magma from the mantle is focused at discrete points along the ridge axis at segment centers. These regions of thicker crust and enhanced

Fig. 6.14 Three possible general cases for the evolution of ridge-axis discontinuities. Arrows along axis refer to direction of propagation of magmatic pulses. t,, t2,... ,tn refer to a time sequence. Cases 2 and 3 apply to second and third order discontinuities (after Macdonald et al., 1988, with permission from Nature 335,217-25. Copyright © 1988 Macmillan Publishers Ltd).

Fig. 6.14 Three possible general cases for the evolution of ridge-axis discontinuities. Arrows along axis refer to direction of propagation of magmatic pulses. t,, t2,... ,tn refer to a time sequence. Cases 2 and 3 apply to second and third order discontinuities (after Macdonald et al., 1988, with permission from Nature 335,217-25. Copyright © 1988 Macmillan Publishers Ltd).

magma supply are characterized by negative mantle Bouguer anomalies (MBAs) and the areas of thinned crust between them by positive MBAs (Lin et al., 1990). These latter areas include second order discontinuities in addition to transform faults. Evidence for the focusing of magma supply at segment centers is less obvious on the fast-spreading East Pacific Rise, but discontinuities and variations in the size and width of the magma chamber correlate with segmentation (Toomey et al., 1990). These observations suggest that magma is emplaced at segment centers and migrates laterally along the ridge axis towards the segment ends. Increasingly it has been recognized that on cooler slow-spreading crust this can mean that segment ends are starved of magma and that parts of the crustal section consist of serpentinized mantle.

The extension of oceanic crust at ridge crests can occur either by the intrusion of magma or by exten-sional faulting. If ridge crests in the vicinity of transform faults are deprived of magma, amagmatic extension becomes more important. Perhaps the most spectacular expression of this is the occurrence of major low-angle detachment faults (Ranero & Reston, 1999; MacLeod et al., 2002) (Section 7.3) on the inside corners of slow-spreading ridge-transform intersections that give rise to large corrugated and striated domes of ser-pentinized peridotite and gabbro (Plate 6.1 between pp. 244 and 245). These corrugated domes are exposed fault planes that deform the upper mantle and lower crust of oceanic lithosphere. The corrugations parallel the spreading direction and indicate the direction of motion on the fault. The offset on these faults is typically at least 10-15 km (Cann et al., 1997). These exposures are thought to result from processes involving extension, detachment faulting, and crustal flexure that are similar to those that form metamorphic core complexes in zones of continental extension (Sections 7.3, 7.6.2, 7.7.3). For this reason the zones of exhumed peridotite at ridge-transform intersections are referred to as oceanic core complexes. Examples include the Atlantis Massif at the Mid-Atlantic Ridge-Atlantis transform intersection (Plate 6.1 between pp. 244 and 245) (Black-man et al., 1998; Schroeder & John, 2004; Karson et al., 2006) and along the Southeast Indian Ridge south of Australia (Baines et al., 2003; Okino et al., 2004).

Figure 6.15 illustrates the along axis variation of oceanic crust for slow- and fast-spreading ridges as envisioned by Cannat et al. (1995) and Sinton & Detrick (1992) respectively. Indeed Cannat et al. (1995) suggested that serpentinized rocks may be much more common in the oceanic crust than previously assumed, even in areas distant from fracture zones. They dredged in the region of the North Atlantic Ridge at 22-24 °N over areas of positive gravity anomalies, indicative of relatively thin crust, and over areas with a normal gravity field. Over the former, which comprised some 23% of the area surveyed, they encountered serpentinite with very few of the basaltic rocks which normally characterize oceanic layer 2. They suggested that as magmatic centers grow, migrate along the ridge axis and decline, the normal oceanic crust would similarly migrate and would enclose those regions of serpentinitic crust that originate where magma was absent. This work is important as it implies that serpentinized peridotite is more common in slow-spreading oceans than previously recognized. There are wide ranging implications. Perido-tite is much more reactive with seawater than basalt and on weathering would release magnesium, nickel, chromium and noble metals. Sepentinite also contains far more water than altered basalt, which could account for much of the water supplied to the mantle in subduction zones (Section 9.8), although at the present day the only examples of oceanic crust formed at slow-spreading rates entering subduction zones are the Caribbean and Scotia arcs.

Segmentation of ocean ridges appears to be controlled by the distribution of partial melts beneath them (Toomey et al., 1990; Gente et al., 1995; Singh et al., 1998), which feed magma chambers at discrete locations along them and create local depth anomalies. The ridge model of Sinton & Detrick (1992), described above, precludes extensive mixing within the small axial magma chamber along the ridge, and could explain the observed geochemical segmentation. With time the magma may migrate away from its sources, creating a gradual increase in depth of the axis as the pressure within it gradually wanes. This phenomenon may explain the noncoincidence of magma chamber and rise culmination noted by Mutter et al. (1988). The brittle shell overlying the magma stretches and cracks and magma intrudes so that eruptions follow the path of magma migration. After eruption the removal of supporting magma gives rise to the formation of an axial summit graben. Evidence for the pulselike, episodic spreading of ridges, in which sea floor spreading occurs by fracturing, dike injection and copious volcanism, has been provided by seismological studies and direct observation (e.g. Dziak & Fox, 1999;

Magmatic segment center Transform fault Magmatic segment center km

Asthenosphere t'-TryV,

Segment center

Slow-spreading ridge

Segment center

Fast-spreading ridge

Volcanics/basalt

Dikes

Gabbro

Ultramafic/peridotite

Lower crust

Melt lens

Crystal mush zone (partially molten)

Transition zone (largely solid)

Fig. 6.15 Along axis sections illustrating the variation in crustal structure between segment centers and segment ends on slow- and fast-spreading ridges, as envisioned by Cannat et al. (1995) and Sinton and Detrick (1992) respectively (redrawn from Cannat et al., 1995, with permission from the Geological Society of America, and modified from Sinton and Detrick, 1992, by permission of the American Geophysical Union. Copyright © 1992 American Geophysical Union).

Tolstoy et al., 2001) and studies of ophiolites (Harper, 1978). Ridge axis discontinuities consequently occur where the magmatic pulses finally come to rest. The variable bathymetry and associated geophysical and geochemical differences imply that adjacent segments of ridge have distinct, different sources in the mantle. First to third order segmentation is caused by the variable depth associated with magma migration; fourth order effects are caused by the geochemical differences in magma supply.

Was this article helpful?

0 0
How To Have A Perfect Boating Experience

How To Have A Perfect Boating Experience

Lets start by identifying what exactly certain boats are. Sometimes the terminology can get lost on beginners, so well look at some of the most common boats and what theyre called. These boats are exactly what the name implies. They are meant to be used for fishing. Most fishing boats are powered by outboard motors, and many also have a trolling motor mounted on the bow. Bass boats can be made of aluminium or fibreglass.

Get My Free Ebook


Post a comment