Earth Science分享 Learning


[转载]Plate Tectonics: Geological Aspects (J. TARNEY)

已有 2823 次阅读 2018-3-27 17:09 |个人分类:地质学基础|系统分类:科研笔记| 地质, 地幔, 矿物 |文章来源:转载


Plate Tectonics: Geological Aspects



Knowledge of mantle petrology and the constitution of the deeper mantle is important in trying to understand several aspects of plate tectonics. For instance, is there whole-mantle convection or two-layer convection? What are mantle plumes? What are superplumes? Does the subducting slab penetrate into the lower mantle? What happens to the slab at depth? Is the sub-continental mantle different from the oceanic mantle? First, some basic facts.

Principal Internal Subdivisions of the Earth


Upper MantleMoho-400620.10
Transition Zone400-100010000.17
Lower Mantle1000-29002450.41
Outer Core2900-51541770.30
Inner Core5154-6371120.02


Note that the crust makes up quite a small proportion of the total Earth. The main problems that have occupied geologists over the years are: What is the nature of the crust-mantle boundary (the MOHO). What is the nature of the low velocity zone? Is the lithosphere diferent in composition from the asthenosphere? What happens in the transition zone? What is the nature of the deep mantle? 

Fig. 1. The Earth in proportion. The thickness of crust is exaggerated. Note that the convecting upper mantle is quite shallow; but it is now known that at least some subducting slabs penetrate the 700 km discontinuity, and could well descend to the bottom of the lower mantle. When heated by the hot liquid core they could eventually rise as mantle plumes.

The Continental Crust

Though we know quite a lot about the upper crust, there is still quite a lot of uncertainty about the lower crust. Is there a real Conrad discontinuity separating the lower from the upper crust? Is the lower crust made up of dry granulite-facies rocks. Is it more mafic than the upper crust, perhaps as a result of intrusion of mafic magmas into the lower crust (called "underplating"). Or is it more mafic as a result of extraction of silicic granitic magmas from the deep crust?

There a number of regions where we think tectonic activity has brought segments of the lower crust up for inspection. Notable examples are Kapuskasing and Pikwitonei in Canada (Precambrian crust), the Lewisian of NW Scotland (also Precambrian), Calabria in S. Italy and the Ivrea Zone in the Alps (both Phanerozoic).

Fig. 2. Diagram (based on field and geophysical studies) to show how deep crust can be thrust up to high crustal levels. Kapuskasing structure, Ontario, Canada). The Lewisian of NW Scotland can be interpreted similarly, the high-grade lower crustal granulites being thrust over amphibolites >2.5 Gyr ago.



Fig. 3. Estimates as to the extent to which we have sections thru two segments of Archaean crust and one segment of ‘recent’ Alpine crust. There is uncertainty as to whether actually we can directly sample the lowest crust. The Alpine sequence is in a series of thrust slices, but some rocks in region of the Ivrea zone have been down to greater depths and have re-bounded again to the surface. Perhaps the more important question is what causes high-P rocks to exhume?

The Moho and the Lower Crust

In the early 1960s there was considerable discussion between petrologists and geophysicists as to the nature of the MOHO. The P-wave velocity of most regions of the uppermost mantle beneath both continents and oceans lies in the range 8.2 ±0.2 km/sec. This in itself would restrict the composition of the mantle below the Moho to some combination of the following minerals (which have the appropriate properties):

Olivine Pyroxenes Garnet (minor spinel, hornblende, phlogopite)

The two principal rock types carrying these minerals are: PERIDOTITE (olivine + pyroxenes) and ECLOGITE (pyroxene + garnet), which are of ultramafic and mafic composition respectively.

The nature of the lower crust is less certain. Exhumed Precambrian high-grade granulite-facies rocks (as in Figs. 2 & 3), which have equilibrated at depths of 25-30 km, have an intermediate (dioritic) bulk composition. But deep crustal xenoliths brought up in volcanic breccia pipes tend to be more mafic (gabbroic) in composition. So has the continental crust in part been underplated subsequently by mafic magma? Also, recent deep seismic investigations have revealed strong horizontal reflections in the deep (mainly post-Archaean) crust - what is the cause of these reflections? Do they represent mafic intercalations, differences in fluid content, crustal viscosity, or the bottoming out of shear zones (cf. Kusznir & Matthews, 1988; Meissner & Kusznir, 1987; Warner, 1990; Reston, 1990a). Comparison of crustal reflection profiles across the various tectonic zones of Europe (Wever et al., 1987; Sadowiak et al., 1991) has identified several different types of deep crustal structure: (a) abundant lamellae above the Moho as in BIRPS SWAT 4, (b) bands of reflectors as in BIRPS WINCH 3, (c) hyperbola-like diffractions as in BIRPS SWAT 6-9, (d) "crocodile" diverging reflectors observed in old collision zones, but not so far in the UK, (e) "ramp and flat" stuctures as in BIRPS SWAT 4 2/3, and (f) "fishbone" pattern observed across the Brabant massif. These features (cf. Meissner, 1989) are thought to represent different types of crustal structure. However it is felt that lower crustal viscosity (Meissner & Kusznir, 1987) is a more important factor controlling development of reflectors than is composition, and the current view is that good lower crust reflectors might characterise mature crust, but that this pattern could be destroyed by either compression/collision ("crocodiles", etc.), igneous intrusions or significant extension.

Nonetheless it is commonly assumed that the lower crust is gabbroic in composition, either igneous (gabbro), or metamorphic amphibolite (wet) or granulite (dry).

The MOHO as a phase transition

Fig. 4 shows various suggestions (made at one time or other) for how the MOHO beneath oceans and continents could be a phase transition (change of mineralogy, but not a major change in composition). Serpentine is the hydrated variety of peridotite (with ca 12% water, thus lower density). Eclogite is the high-pressure form of basalt or gabbro. But are these models realistic? The serpentine-peridotite model is now discounted.


The MOHO and the gabbro-eclogite transformation: Basalt transforms to eclogite at high pressures according to the equation:

Olivine + pyroxene + plagioclase jadeitic pyroxene + garnet


For this to be capable of explaining the MOHO it must be a relatively sharp transition, of no more than a few km. Green and Ringwood (1967) studied this experimentally to 30 kilobars (= ca 100 km) using a quartz tholeiite and an alkali basalt (Fig 5a, b).

Fig. 5a. Experimental studies of Green & Ringwood (1967) on quartz tholeiite basalt showed that transformation to eclogite occurred over a considerable depth interval. Note that eclogite has a lot more quartz than the equivalent basalt. But mantle eclogites have no quartz. Where does the silica go?


With both compositions the transformation was found to be gradual. The disappearance of the low density phase (plagioclase) and its replacement by the high density phases (garnet and jadeite) occurred over a pressure range of ca 10 kb (= ca 25 km). Thus the MOHO cannot be a phase transformation and must be a compositional transition.

The MOHO as a chemical transition


Possible scenarios for the oceans and continents are shown in Fig. 6.


In oceanic regions the form of the MOHO is known from dredging at fracture zones where peridotites (often serpentinised) have been recovered by dredging and drilling along with cumulate gabbros and pillow basalts. Actual sections of ocean floor are preserved in ophiolite complexes.

The nature of the sub-continental MOHO is less certain, partly due to lack of knowledge of the lower crust.


Composition of the Upper Mantle

Our petrological knowledge of the upper mantle composition comes from several sources:

(1) Nodules brought up in volcanic pipes;

(2) Large sections of mantle found in obducted ophiolites;

(3) Slices of mantle thrust up in mountain belts such as the Alps (Ivrea-Verbano Zone)


and the Caledonides (especially Norway); and


(4) Modelling "backwards" from erupted basalt compositions.

(1) Olivine rich nodules are quite common in erupted alkali basalts worldwide, and are almost all of spinel lherzolite composition (olivine, orthopyroxene, Cr-diopside, spinel). In kimberlite (diamond) pipes there is a greater diversity in that both garnet peridotite and eclogite xenoliths occur (the former dominant). Some nodules contain diopside-rich, phlogopite-rich or amphibole-rich veins. It is thought that these nodules are representative of the sub- continental mantle. This material seems to be rather refractory (could not yield much basalt on melting), but at the same time can be quite enriched in incompatible trace elements such as Sr, Ba, K, Rb and the light rare- earths.

(2) The peridotitic material in ophiolite complexes (obducted ocean floor) is mainly HARZBURGITE (ol+opyx) or DUNITE (ol), often cut by pyroxenite (opyx) veins and sometimes having chromite segregations (podiform chromites). There is a consensus that this mantle is the refractory residue left after basalt extraction at mid-ocean ridges. But are the pyroxenite veins the result of silica-rich fluids coming off the subduction zone? (see in Fig. 5a all the free quartz present in eclogite in subduction zones) - note that many ophiolites are thought to be fragments of "back-arc" spreading centres.

(3) In the Ivrea Zone and the Lanzo-Seisia Zone of the Italian Alps, the peridotite slices are overlain by layered gabbros. They are mainly LHERZOLITE (ol-opyx-cpyx -spinel, or -hornblende or -phlogopite). Most carry veins composed of orthopyroxene, orthopyroxene-spinel, diopside-orthopyroxene, phlogopite-diopside, or hornblende. Different segments different veins, suggesting a complex make-up of the sub-continental mantle. Some of the rock types are similar to those brought up in volcanic breccia pipes.

(4) Considerable progress has been made in understanding the compositions of basaltic rocks in recent years, and interpreting them in terms of melting models. So it is possible to "model back" to the primary mantle from which the basalt was derived, and estimate its composition. We now know that there are several distinct types of mantle, that have been kept separated for many hundreds of millions of years. They have distinct trace element and isotopic characteristics. (You may come across references to them as DMM, HIMU, EM1, EM2 and PREMA), but there is uncertainty as to where they are located.

Upper Mantle Mineralogy


The variation in upper mantle mineral assemblages with temperature and pressure can be determined experimentally. But there is uncertainty about composition. Because a lot of observed mantle material is not primary, but has had a liquid (basalt) fraction removed by partial melting, Ringwood coined the term 'PYROLITE' for primitive fertile mantle - in effect dunite with basalt put back in!


PYROLITE = 3 parts DUNITE + 1 part BASALT

Mantle mineralogy varies mainly on account of the nature of the aluminous phase, which is P-T dependent, i.e.

Olivine (MgFe)2SiO4 + Orthopyroxene (MgFe)2Si2O6 as major phases, plus:

Plagioclase CaAl2Si2O8 or
Spinel (MgFe)Al2O4 or
Clinopyroxene (NaCa)2(AlSi)2O6 or
Garnet Mg3Al2Si3O12 or
Hornblende or

Phase diagram (Fig. 7), though rather complicated, shows that plagioclase peridotite can exist only at very shallow depths where the geothermal gradient is high; spinel- and pyroxene peridotites have a larger stability field in the upper mantle; but garnet peridotite will occur at deeper levels (hence common as nodules in kimberlite pipes).


Fig. 7. Summary of phase relations in pyrolite (after Green & Ringwood) appropriate to upper mantle conditions. Wet solidus for small amount of hornblende breakdown. Note that only oceanic geotherm intersects this wet solidus.

Methods for investigating deep mantle mineralogy

Direct sampling of the deeper mantle is obviously impossible. Observed seismic velocity-depth functions however constrain the densities of likely mantle phases. Moreover, possible phase transformations in the transition zone of the mantle are very difficult to verify experimentally. For instance, in the 400-900 km depth region pressures are in the range 130-340 kilobars and temperatures 1500-3000°C ... beyond the range of most experimental equipment until recently. Now with diamond anvil apparatus, laser heating and on-line X-ray determinations it is possible to reach into this range, at least momentarily.

Indirect methods have also proved reasonably successful. Fortunately high pressure phases tend to crystallise in structures which are already known (isomorphs). For instance we can compare silicates with germanates because germanates form a series of crystal structures closely parallel to those of silicates, but the transformations occur at lower pressures.

Thermodynamics requires that the high pressure polymorph be denser, which limits possible structures. Once structure is known the bond lengths between cations and anions enable densities to be calculated.

The Radius Ratio (Rcation/Ranion) determines type of crystal structure. At high pressures effective radii contract differentially, thus altering the radius ratio. Thus a new high pressure phase may appear when radius ratios exceed certain critical values. Large ions (e.g. Oxygen 1.40Å) contract more under pressure than small ions. Oxygen is more polarizable than smaller cations:

ElementPolarizabilityRadius (Å)


3.1 × 10-24cm31.4


0.04 × 10-24cm30.26


0.12 × 10-24cm30.72


Pressure thus increases covalent component of chemical bond.

Phase transitions in the deeper mantle

Refinement of seismic wave data has shown number of discontinuities (Fig. 8):

These zones are:

(1) LOW VELOCITY ZONE: from below lithosphere to about 200-250 km. Not always 

present. Asthenosphere has S-wave attenuation . . small amount of liquid, perhaps
ca 1% melting?

(2) MINOR DISCONTINUITY around 350 km.



(5) Between 900 and 2700 km no major discontinuities, but some smaller ones. In general

increase in seismic velocities and density explained by self-compression of homogenous material.

The following explanations have been proposed to explain these discontinuities (Fig. 9):

350 km. Pyroxene forms a complex solid solution with pre-existing garnet in which one-quarter of silicon atoms are octahedrally co-ordinated, leading to 10% increase in density of pyroxene component:

Mg3(MgSi)Si3O12 & Ca3(CaSi)Si3O12

400 km. Olivine transforms to beta-Mg2SiO4 which has SPINEL structure and is 8% denser than olivine.

500-550 km. Calcium silicate CaSiO3 component of garnet transforms to extremely dense PEROVSKITE structure. Also beta-Mg2SiO4 transforms to gamma-Mg2SiO4 with 2% increase in density.

650 km. The spinel structure disproportionates to MgSiO3 with PEROVSKITE structure and MgO with a ROCK SALT structure, i.e. Mg2SiO4 > MgSiO3 + MgO. Additionally the MgSiO3.Al2O3 component transforms to an ILMENITE structure and any sodium present will transform to a high pressure form of NaAlSiO4 having CALCIUM FERRITEstructure.

Lower Mantle. Below 700 km no more major transformations are possible - the minerals are as close-packed as they can get. There is thus then a slow progressive increase in density to the mantle-core boundary.

Fig. 9. Density changes with depth in the mantle, and the changes in mineral structure that have been proposed to explain them


A point of interest is whether this sharp increase in density at 650-700 km acts as a barrier to mantle convection. If the slab cannot penetrate this boundary, does it pile up above 700km? Are there two convecting zones in the mantle: one above, one below the 700km discontinuity? Does this also coicide with a chemical boundary? Is there any chemical interchange across the boundary layer?


Fate of the subducted slab: Ringwood 1991 Model

One of the problems of plate tectonics is the fate of the subducting slab. This can be traced, from seismic evidence, to descend to about 650 km; but the evidence is somewhat conflicting regarding the extent to which it penetrates the dense 650 km discontinuity. (See references by Jordan and Hilst). Because the phase changes with depth are now known in some detail, both for ultramafic mantle material and for subducted basaltic ocean crust, it is possible to calculate their modal compositions with depth. For instance, the modal composition of pyrolite with depth is shown in Fig. 10:


Fig. 11 shows the same calculations for basaltic ocean crust. Note that the plate which is subducting is not uniform mantle pyrolite but, because of melting at the ridge axis, it has segregated into a basaltic ocean crust (ca 5 km thick), residual harzburgite (from which the basalts were extracted) underlain by ordinary pyrolite. Knowing the mineral proportions and the densities of the minerals in each of the main rock types, undepleted pyrolite, depleted harzburgite, and basaltic ocean crust, it is then possible to calculate the density changes in each of these rock types with depth. The thermally equilibrated densities for these three rock types with depth are shown in Fig. 12.

Fig. 12. Densities (g/cc) of thermally equilibrated basaltic ocean crust, subducted harzburgite lithosphere compared with undepleted pyrolite mantle to depths of 800 km. Note that the ocean crust is mostly more dense and the harzburgite is less dense than pyrolite down to 650km depth, but then their positions are reversed.



Fig. 13. Density differences between basalt - pyrolite and harzburgite - pyrolite as the subducted ocean crust sinks. The basaltic slab becomes less dense than mantle pyrolite in the depth range 650 - 750 km.


The important point is that the subducted plate sinks because the basaltic component of the slab (now eclogite) is ca. 0.2 - 0.1 g/cc more dense than the enclosing host pyrolite to depths of 650 km, and exerts a strong 'slab-pull' force at subduction zones. The harzburgite part of the plate may also be slightly more dense initially because it is cold, but is inherently less dense once it has thermally equilibrated with the surrounding mantle pyrolite. However, because of the phase changes in pyrolite at the 670 km discontinuity, the basaltic crust suddenly becomes 0.2 g/cc less dense than the pyrolite in the depth range 650-750 km, whereas the harzburgite component of the slab becomes very slightly more dense. These effects are very clearly shown in Fig. 13. Ringwood (1991) argues that these changes then have the effect of trapping subducted basaltic ocean crust at the 670 km discontinuity, as shown in Fig. 14.

Fig. 14. The effect of density differences is that basaltic ocean crust becomes trapped at the 670km discontinuity.


Ringwood has suggested that the slab piles up at the base of the upper mantle, as shown in Fig. 14. By the end of the Archaean (2500 my ago) he envisages that the mantle structure around the 650km discontinuity would be as shown in Fig. 15. This layer is source for diamond-bearing kimberlite magmas according to Ringwood et al. (1992).


Fig. 15. Likely mantle structure at the end of the Archaean as a result of subducted mafic ocean crust piling up at the 650 km discontinuity (after Ringwood).
Fig. 17. Comparison between oceanic and continental lithosphere.


Assuming constant spreading rates (present day) it can be calculated that, throughout Earth history, the amount of ocean crust which may have accumulated at the 650 km discontinuity would be at least 100 km thick. However because the harzburgite is inherently less dense and potentially more buoyant than the surrounding mantle, then when it heats up it may begin to ascend as blobs or diapirs, as shown in Fig. 16 (below).




Fig. 16. Models of mantle differentiation involving storage of ocean crust at the 650 km discontinuity.


This has interesting consequences as a mechanism to generate mantle plumes and 'hotspots'. Most plumes need to be generated at a discontinuity, either the 650 km one or at the core-mantle boundary. The mantle model (Fig. 16) shows that these plumes rise and penetrate the lithosphere to become the source of hotspot ocean islands. If these rising diapirs cannot penetrate the lithosphere, they may just add to the base of the lithsophere, and melts may penetrate it and metasomatise and chemically alter it. The ocean lithosphere is young (almost all less than 200 my) whereas the sub-continental lithosphere is older, cooler, thicker and more complex, as shown in both Figs. 16 & 17.

Because of the different thermal regimes, and the influences of plumes, it is likely that there have been differences in the make-up of the lithosphere during Earth history. Fig. 18 (below) shows the likely structure of the modern mature Phanerozoic ocean lithosphere (left), which is regarded as being less depleted with increasing depth. Ocean plateaus (centre) have a very thick ocean crust, with (implicitly) a much more depleted harzburgitic mantle underlying it. In the Archaean (right) one suggestion is that the high mantle temperatures would have led to very high degrees of melting, to produce high-Mg komatiitic lavas, and leaving an extremely depleted pure-olivine dunitic residue. This oceanic structure is much more like that of modern oceanic plateaus, so was there plate tectonics in the Archaean or plateau tectonics (see later PlateLect-F)?


Models of Mantle Differentiation

This widely accepted model implies that mantle (or at least the upper mantle) is continually differentiating to form continental crust by a two-stage process. The crust formed is permanent and is not recycled back into the mantle.

(1) Primitive pyrolite mantle rises at mid-ocean ridges, melts to form basaltic ocean crust overlying refractory harzburgite plate.

(2) Plate sinks back into the mantle at subduction zones. Hydrated altered ocean crust dehydrates and causes melting of the basaltic ocean crust and of the overlying mantle wedge to yield andesitic magmas.

(3) Andesitic magmas fractionate en route to the surface to produce more siliceous magmas. Hence sialic crust accretes laterally at continental margins, is of low density and is indestructible.


Fig. 19. Simple "box model" of mantle evolution, showing how melting at spreading ridges produces ocean crust, which is then altered by hydrothermal activity and then subducted. Part of this subducted crust is then melted to form continental crust, and the residues then subducted to become part of the reservoir of the depleted (DMM) mantle. Small degree melts migrate upwards to enrich the sub-continental mantle and provide the source for alkali basalts. Sediment subduction may modify the sub-continental lithosphere. (after Tarney et al. 1980)


A consequence of course is that if the continental crust has been extracted from the convecting mantle, the convecting mantle must have become progressively depleted in lithophile elements. This is now known as the 'DM' mantle reservoir. This is the reservoir that supplies depleted mid-ocean ridge basalt ("MORB"). The real story is a little more complicated, as may be deduced form Fig. 19. Sediments may be subducted and contaminate the lithosphere under continental margins as well as the material stored at the 650 km discontinuity. Small degree melts permeate upwards and vein both the sub-continental and sub-oceanic lithosphere, but because the former is older, we generally observe more complex effects under the continents.

Mantle Convection

The diagrams below show some conceptual models of how the mantle may be convecting, and possible relationships between the upper and lower mantle (after Allègre et al.). There is still a very intense debate on whether the lower mantle is involved in manconvection.


Fig. 20. Box models for crust-mantle evolution. On the left continental growth occurs through igneous contributions from both the upper and lower mantle. On the right the continental crust has mainly been extracted from the upper mantle, which is therefore "depleted" relative to lower mantle.


Slab Penetration into Lower Mantle?

Fig. 22. Cartoon showing how subducting slabs may either lay themselves out along the 650 km discontinuity (a), or penetrate the discontinuity to enter the lower mantle as in (b). The latter gives active back-arc spreading (see later lecture).


The most recent analysis of the fate of the oceanic crust as it subducts into the mantle beneath the West Pacific island arcs (van der Hilst & Seno, 1993), suggests that whereas that subducting beneath the Izo-Bonin arc and Shikoku Basin, south of Japan may be deflected and "laid-out" along the 650 km discontinuity in the transition zone (Fig. 22(a)), that further south beneath the Mariana Arc may penetrate into the lower mantle (Fig. 22(b)):

UPDATES (1994/6)

Ringwood's Megalith Model:

The essence of the Ringwood megalith model is that while ocean lithosphere subduction is initially thermally driven because the downgoing slab is cold [the basaltic (eclogite) layer is 5% more dense than the surrounding mantle, and compensates for the depleted harzburgite which is 2% lighter – see Fig. 13], the compositional buoyancy difference between the two becomes significant at 700km depth, resulting in the mechanical separation of basaltic from harzburgite/dunite components (see Fig. 15). However, recent modelling by Gaherty & Hager (1994), using a range of viscosity contrasts for eclogite vs harzburgite, shows that the two are unlikely to separate. The slab buckles and folds as it reaches the 700 km discontinuity, but there is no obvious separation of eclogitic and harzburgitic components. The compositional buoyancy differences are subordinate to the overall thermal buoyancy.

Nature of the Lower Mantle:

While it is generally known that the convecting Upper Mantle (above the 670km discontinuity) is chemically depleted in lithophile elements because of the progressive growth and extraction of the continental crust from it, it has been commonly thought that the Lower Mantle is largely undepleted. However, Kerr et al. (1995) have proposed that the Lower Mantle is also depeleted, in part because of the return of subducting slab material right through the 670km discontinuity into the lower mantle: see also van der Hilst & Seno (1993). This implies that there is much more interchange between Upper and Lower mantle than was first thought. The Lower Mantle feeds into the upper mantle in the form of large hot plumes (see later lecture). Figure 23 below shows how cool subducted material may go right into lower mantle, or get stuck termporarily at the 670km discontinuity and then 'drip' into the lower mantle:



Fig. 23 (after Kerr et al. 1995) shows 2-layer convection of the mantle, as subducting plates lodge at the 670km discontinuity, or get convected back into the upper mantle; but with periodic interchange between the two as cold plates avalanche down into the lower mantle, and deep mantle plumes are displaced and rise to form major oceanic plateaus and continental large igneous provinces ("LIPs"). There may be composition differences between upper mantle and lower mantle as a result of such processes through Earth history.

Larson and Kincaid (1996) then go on to argue that the breakup of major continents, as occurred with the Gondwana supercontinent in the Mesozoic (ca. 130Ma), leads to more rapid subduction of old cold ocean crust. These cold slabs then avalanche down and penetrate the 670km thermal boundary layer into the lower mantle. One effect is to raise the 670km TBL; another is to displace material from the deep lower mantle (D") which appears as major mantle plumes during the mid-Cretaceous magnetic superchron (120 Ma - 80 Ma). See later notes on mantle plumes.


REFERENCES: Mantle Mineralogy

(These references are probably more than you require at this stage, but as some aspects are followed up in more detail later in your courses, they may be useful to you.)


ALLèGRE, C.J. 1982. Chemical geodynamics. Tectonophysics 81, 109-132.

ALLèGRE, C.J. & TURCOTTE, D.L. 1986. Implications of a two-component marble-cake mantle. Nature 323, 123-127.

BLUNDELL, D.J. 1990. Seismic images of continental lithosphere. Journal of the Geological Society, London 147, 895-913.

GAHERTY, J.B. & HAGER, B.H. 1994. Compositional vs. thermal buoyancy and the evolution of subducted lithosphere. Geophysics Research Letters 21, 141-144.

IRIFUNE, T. & RINGWOOD, A.E. 1987. Phase transformations in a harzburgite composition to 26 GPa: implications for dynamical behaviour of the subducting slab. Earth and Planetary Science Letters 86, 365-376.

IRIFUNE, T. & RINGWOOD, A.E. 1993. Phase transformations in subducted ocean crust and buoyancy relationships at depths of 600-800 km in the mantle. Earth and Planetary Science Letters 117, 101-110.

JORDAN, T.H. 1975. The continental tectosphere. Review of Geophysics and Space Physics 13, 1-12.

JORDAN, T.H. 1978. Composition and development of the continental tectosphere. Nature 274, 544-548.

JORDAN, T.H. 1981. Continents as a chemical boundary layer. Philosophical Transactions of the Royal Society, Lond. A301, 359-373.

KEARY, P. & VINE, F.J. 1991. Global Tectonics. Blackwell Scientific Publ., 302pp.

KERR, A.C., SAUNDERS, A.D., TARNEY, J., BERRY, N.H & HARDS, V.L. 1995. Depleted mantle-plume geochemical signatures: no paradox for plume theories. Geology 23, 843-846.

KUSZNIR, N. and MATTHEWS, D.H. 1988. Deep seismic reflections and the deformational mechanics of the continental lithosphere. Journal of Petrology Special Lithosphere Issue, pp. 63-87.

LARSON, R.L. & KINCAID, C. 1996. Onset of mid-Cretaceous volcanism by elevation of the 670 km thermal boundary layer. Geology 24, 551-554.

MEISSNER, R. 1989. Rupture, creep, lamellae and crocodiles: happenings in the continental crust. Terra Nova 1, 17-28.

MEISSNER, R. and KUSZNIR, N. 1987. Crustal viscosity and the reflectivity of the lower crust. Annales Geophysicae 5B, 365-373.

MEISSNER, R., MATTHEWS, D.H. and WEVER, T. 1986. The Moho in and around Britain. Annales Geophysicae 4B, 659-666.

MENZIES, M.A. 1990. Archaean, Proterozoic, and Phanerozoic lithospheres. In: M.A. Menzies (Editor) Continental Mantle, Clarendon Press, Oxford, pp.67-86.

RINGWOOD, A.E. 1974. The petrological evolution of island arc systems. Journal of the Geological Society, London 130, 183-204.

RINGWOOD, A.E. 1975. Composition and Petrology of the Earth's Mantle. McGraw-Hill, New York.

RINGWOOD. A.E. 1982. Phase transformations and differentiation in subducted lithosphere: implications for mantle dynamics, basalt petrogenesis, and crustal evolution. Journal of Geology 90, 611-643.

RINGWOOD, A.E. 1985. Mantle dynamics and basalt petrogenesis. Tectonophysics 112, 17-34.

RINGWOOD, A.E. 1986. Dynamics of subducted lithosphere and implications for basalt petrogenesis. Terra Cognita 6, 67-77.

RINGWOOD, A.E. 1991. Phase transitions and their bearing on the constitution and dynamics of the mantle. Geochimica et Cosmochimica Acta 55, 2083-2110.

RINGWOOD, A.E. & IRIFUNE, T. 1988. Nature of the 650-km seismic discontinuity: implications for mantle dynamics and differentiation. Nature 331, 131- 136.

RINGWOOD, A.E., KESSON, S.E., HIBBERSON, W. & WARE, N. 1992. Origin of kimberlite and related magmas. Earth and Planetary Science Letters 113, 521-538.

SADOWIAK, P., WEVER, T. and MEISSNER, R. 1991. Deep seismic reflectivity patterns in specific tectonic units of Western and Central Europe. Geophysics Journal International 105, 45-54.

TARNEY, J., WOOD, D.A., SAUNDERS, A.D., CANN, J.R. & VARET, J. 1980. Nature of mantle heterogeneity in the North Atlantic: evidence from deep sea drilling. Phil. Trans. Roy. Soc. London A297, 179-202.

van der HILST, R. & SENO, T. 1993. Effects of relative plate motion on the deep structure and penetration depth of slabs below the Izu-Bonin and Mariana island arcs. Earth and Planetary Science Letters 120, 395-407.

WEVER, T., TRAPPE, H. and MEISSNER, R. 1987. Possible relations between crustal reflectivity, crustal age, heat flow and viscosity of the continents. Annales Geophysicae 5B, 255-266.

WARNER, M.R. 1990. Basalts, water or shear zones in the lower continental crust? Tectonophysics 173, 163-173.

WYLLIE, P.J. 1971. The Dynamic Earth. Wiley, London


Germanates as high pressure models of silicates

Trying to elucidate the petrological nature of the deep mantle is not easy because it is difficult to re-create such high-pressure - high-temperature conditions in the laboratory. At least for any length of time: Laser heating and momentary shock treatment can do it for a short time, but as silicate reactions usually take a long time to reach equilibrium condition, this leads to huge uncertainties in P-T parameters. However, in the early years, mineral chemistry principles were used to predict high pressure behaviour.

Use of germanates to model high pressure silicates was first suggested by Goldschmidt in 1931. Si and Ge are tetravalent and in same group in the Periodic Table.

RADII:Si4+ 0.26AGe4+ 0.40A


Silicates and germanates usually isostructural and there is a wide range of germanate structures. So, if it is possible to synthesize a germanate structure at moderate pressures it is likely that an equivalent silicate structure will exist at higher pressures. If a germanate displays a phase transformation at a given pressure, the corresponding silicate often displays the same transformation but at a much higher pressure. This is because the critical radius ratio RGe/ROxygen for transformation to a new phase is attained at much lower pressures with Ge. Some germanates (e.g. GeO2) can crystallise at zero pressure while the equivalent silicate needs 100 kb pressure.

Many transformations in germanates involve change from 4- to 6-fold co-ordination with oxygen. Compare:

NaAlSi2O8>NaAlSi2O6 + SiO228kb (silicate)


>Jadeiite + Rutile str
NaAlGe2O8>NaAlGe2O6 + GeO25kb (germanate)

2CoSiO3>2Co2SiO+ SiO2100kb (silicate)


>Spinel + Rutile str
2CoGeO3>2Co2GeO4 + GeO210kb (germanate)



These lines of reasoning allowed predictions to be made as to what types of structure might exist at depth in the Earth.


That's All Folks!





Magma production at the Earth's mid-ocean ridge system far exceeds that in any other tectonic environment, and this has been so since the early Precambrian. It is the dominant way in which internal heat is dissipated. The structure of a mid-ocean ridge is shown below:

Note how the lithosphere thickens as it moves away from the ridge. Because the Earth's magnetic field oscillates between north and south at intervals of a few hundred thousand (or the odd million) years the basalts erupted then take on the current magnetisation, and so give rise to the seafloor magnetic lineations (patterns shown above) that can be used to date the ocean floor. Melting of pyrolite mantle extracts basaltic liquids to form the ocean crust, leaving a residue of harzburgite (ol+opyx) forming the underlying lithosphere.

The ocean lithosphere suffers extensive hydrothermal alteration at the ridge (see below), but the rocks eventually finish up subducting back into the mantle:

It is because these fluids are released in the Benioff Zone as the slab is subducted that magmas are able to be generated in the mantle wedge above the subduction zone. It is fluid, not friction, which is responsible for active margin magmatism. But it is ridge processes which make it all possible. So we need to look at these.

Why does melting occur? Melting temperatures of most silicate minerals increase with increasing pressures. So temperatures of solid mantle material at depth may be higher than the melting point of mantle near the earth's surface. As hot deep mantle rises beneath spreading ridges it will, as pressure falls, rise above its solidus, and begin melting.

The simplified situation is as follows:

As the uprising mantle crosses the geotherm it begins to melt, and as the solidus temperature of mantle falls with decreasing pressure, the temperature of the melt increases relative to this solidus, thus effectively giving higher degress of melting with decompression, as shown. The amount of melt generated will be limited by the latent heat of fusion (which is high for silicates), and as the melting range of mantle peridotite lies between ca. 1100°C and ca. 1700°C, it is likely that most ridge basalts are partial (rather than complete) melts of mantle. The magma may enter a chamber in the ocean crust and begin crystallising, giving the following P-T path:

There is the possibility of superheat (i.e. temperature above the liquidus) if the magma can rise quickly, but it is apparent that most magmas are erupted or emplaced without superheat (a possible exception are ultramafic lavas called komatiites).

Because we haven't yet been able to drill very far down into oceanic crust, the only way we can begin to understand what happens to the basaltic magma as it rises up at the ridge is to look at ophiolite complexes. There are many of these in the Alpine belt, although we are not always sure that these mafic slivers represent true ocean basin crust or whether some (or all) may represent marginal basin crust or the roots of island arcs.

Nonetheless, by putting together information from a number of ophiolite complexes, particularly Troodos on Cyprus, we come up with the following idealised section:

Not every ophiolite has all these components complete, and it is not always for tectonic reasons. Often the gabbro is missing, or the sheeted dykes, and in some cases the dykes may intrude the harzburgite. Of course sheeted dykes can only be formed if there is a continuously extending magma chamber (try doing it without!). So if sheeted dykes are missing it may mean that there has not been such a magma chamber. In fact there is a lot of debate on this issue. Some geophysical studies indicate a possible continuous magma chamber beneath the East Pacific Rise. However, the EPR is a smooth fast-spreading ridge, and maybe there is enough thermal input to keep a continuous magma chamber going. On the other hand in the slow-spreading Atlantic with its central rift valley and irregular topography, there is no direct evidence for a continuous magma chamber. Some workers, including those at Leicester, suggest that with slow-spreading ridges, each eruption may be a distinct event, and that any magma chamber is only short-lived. Some sections of the Atlantic ridge, like the FAMOUS area (south of the Azores) have numerous small volcanic cones, and this is now being recognised all over the Atlantic.

A consequence is that that there may be a variety of magma chamber profiles, with those from fast-spreading ridges having fat "onion" shapes, those from rather slower-spreading ridges having "leek" shapes. Very slow spreading ridges (e.g. SW Indian Ridge) may just have dykes feeding lavas which directly overly peridotite. There are ophiolites with this profile, where the dykes cut harzburgite tectonite and gabbro is only locally developed. Even with the type Troodos ophiolite, which has a moderatley thick gabbro section, geochemical studies have shown that the gabbros are in fact a compound of a number of small bodies.

Transform Fault Effects

It has long been known that the ocean crust is much thinner in the vicinity of oceanic transform faults. Also that a greater variety of rock types can be drilled or dredged in the vicinity of transforms, and that there is usually a significant topographic difference between the two sides of a transform fault (esp. the larger ones).

The latter effect arises because the ocean crust sinks as much as 3 km over the first 50 m.y. of its existence. So the greater the age difference of adjacent bits of ocean crust across a transform, then the greater the height of the transform wall. Obviously if the wall is 1 km high, then a large amount of rubble will fall down onto the lower plate, and deeper parts will become exposed. Moreover as the transform fault moves, the movement can deform the basalts into hornblende schists.

The thinner crust arises from the cold-wall effect, i.e. that the mantle rising up adjacent to the transform fault are actually in contact with older, and therefore cooler, oceanic crust on the other side. Cooler conditions give less melt and therefore thinner crust. Thinner crust also means there is more likelihood of mantle being exposed in the transform wall, again increasing the variety of rock types.



There has been a very great deal of interest worldwide in the metamorphism and hydrothermal alteration of oceanic crust. After all there are few geological situations where you have a large red-hot magma chamber below and a 3 km column of ocean water on top trying to dowse it. There are a number of important questions that could be asked:

(a) how extensive is the metamorphism, and how far distant from the ridge do the metamorphic effects extend?

(b) if there is extensive hydrothermal activity, does this lead to equally extensive mineral deposits which could be mined?

(c) would the metamorphism affect the magnetic anomaly patterns that are so useful for dating ocean crust?

(d) is the ocean crust so hydrated that this represents an important source of fluids at subduction zones?

(e) does the hydrothermal interchange influence the chemical budget of the oceans?


There is no doubt that ophiolite complexes (obducted bits of ocean crust in mountain belts) are usually >90% altered, and there was debate during the 60's and 70's whether this was a result of metamorphism in the mountain belt during orogensis, or resulted from ocean floor metamorphism. The latter is now the favoured explanation. Many of the samples of ocean crust recovered from the ocean floor by drillling or dredging are altered.



Cann (1979) recognises 5 different mineral assemblage facies in oceanic basalts recovered by dredging, drilling etc. The rocks characteristically preserve igneous textures.

(1) Brownstone Facies

Low temperature ocean floor weathering or cool hydrothermal alteration. Products usually have yellowish brown tint due to oxidising conditions (bluish grey if reducing). Mineral assemblages not in equilibrium; just replace specific primary phases.

Olivine replaced by Celadonite (K-rich dioctahedral Fe-illite) under more extreme alteration. This fills vesicles and replaces glass. Under reducing conditions this is a Saponite (Mg-rich trioctahedral smectite). Pyrite common. Thus basalt has clay alteraton products.

Plagioclase remains fresh, though under extreme alteration may be partly replaced by K-feldspar (Adularia).

Glass: Where basalt glass is common, Palagonite (orange coloured disordered illite) occurs, often associated with a low temperature zeolite (Phillipsite) and Calcite.

Fig. 8. Diagram from Cann (1979) tries to indicate how the minerals in a basalt affected by hydrothermal activity contribute to the secondary phases. At low temperatures it is mainly the basalt glass and olivine which are unstable and contribute to brownstone facies minerals, but plagioclase and then augite and iron oxide become progressively involved at higher grades until the whole rock recrystallises.


(2) Zeolite facies ( Temperature above 50-100°C.)

Here Phillipsite is replaced by higher temperature zeolites - Analcite and Natrolite. Distinct zones of zeolites occur on Iceland.

Mafic minerals replaced by Saponite or Saponite-Chlorite mixed layer minerals, coarser grained than in Brownstone Facies. Plagioclase may also be partly replaced by saponite, but augite stays fresh. Upper limit of facies (250-300°C) marked by disappearance of zeolites and saponite and incoming of albite and chlorite.

(3) Greenschist Facies

Albite ± chlorite ± actinolite ± epidote ± sphene. Degree of alteration variable, primary assemblages may be completely replaced. Augite is commonly a relic, veins are common, often quartz-bearing. Assemblages may or may not be equilibrium ones. Upper limit of facies marked by disappearance of albite, chlorite and actinolite and the appearance of green aluminous hornblende associated with more calcic plagioclase (An20-30).

(4) Amphibolite Facies

Hornblende+Ca-plagioclase + titanomagnetite±epidote. This assemblage is most commonly developed in coarser grained rocks - dykes and gabbros - obviously of deeper origin. Degree of metamorphism variable. Some primary hornblende occurs in gabbros or diorites, but it is clear that amphibolite facies metamorphic assemblages are superimposed on this. So metamorphism closely follows magmatic activity.

The results can be summarised in the following table. Note that it is not just the mafic ocean crust (basalt or dolerite) that is altered. The mantle itself is often brought up along faults, transforms and fracture zones, and this is frequently altered to serpentine (ca. 13% water) at temperatures below 450°C. But other hydrous minerals such as talc, tremolite and chlorite are possible at higher temperatures. There are also 3 different varieties of serpentine: antigorite, chrysotile (the glossy variety) and lizardite.


Summary of Mineral Assemblages in Altered Crust

BrownstoneCeladonite +Phillipsite + Palagonite + Saponite ?


Saponite + mixed layers + analcite + natrolite?


Chlorite + Albite
+ Actinolite +
Epidote + sphene
AmphiboliteHornblende +
Plagioclase +

Iron Oxide
+ Olivine
+ Enstatite
GabbroAugite + 
Plagioclase +
Hypersthene + 
Iron Oxide +
+ Diopside




General Comments:

Greenschist and amphibolite facies metamorphism of the ocean floor differs from these facies in normal regional metamorphism in that:

(a) The thermal gradient is very high: can be several-100°C per km compared with 30 - 50°C/km in regional metamorphism.

(b) No garnet is developed in mafic rocks (pressures are not high enough).

(c) The rocks lack deformation textures (except in samples recovered from transform fracture zones)

(d) Very variable degree of recrystallization, because lower grade metamorphic assemblages are superimposed on earlier higher grade ones. This is because hydrothermal activity continues under cooler conditions as crust progressively moves away from ridge. (In regional metamorphism it is more common for the rocks to equilibrate at one set of P-T conditions)

Despite ca. 20 years of drilling, the deepest drill holes in the ocean floor (several hundred metres) have still only penetrated brownstone- and zeolite-facies rocks. No greenschist facies or amphibolite facies rocks penetrated. To see what happens deeper down, we really need to examine ophiolite complexes.

Ocean floor metamorphism - Sarmiento Ophiolite, Chile


The Sarmiento ophiolite (Saunders et al. 1979) is one of a series of discontinuous mafic lenses that represent the mafic floor of an extensional back-arc basin that was closed and uplifted in mid-Cretaceous times in the southern Patagonian Andes. Excellent vertical exposures have enabled the distribution of metamorphic zones resulting from the hydrothermal metamorphism to be established (Stern & Elthon, 1982?).

Lithological sequence at Sarmiento consists of:

  • 2 km pillow lavas

  • 300 m sheeted dykes

  • 1 km gabbros with plagiogranite

Four main metamorphic equilibrium mineral facies can be recognised:

(1) Zeolite Facies. Zeolites, palagonitized glass ± smectites ± calcite ± quartz ± pyrite ± sphene ± albite.

(2) Greenschist facies. Chlorite, epidote, Na-plagioclase, sphene, ± quartz ± calcite ± biotite ± pyrite.

(3) Lower Actinolite Facies. Low-Al (2-5% Al2O3) fibrous green amphiboles, Ca-plagioclase, sphene ± biotite ± calcite.

(4) Upper Actinolite Facies. Higher-Al (5-8% Al2O3) brown-green amphiboles, Ca-plag. (>An50), titanomagnetite, ± ilmenite ± biotite.

These are arranged vertically in the complex, with metamorphic grade increasing downwards. However, intensity of metamorphism also varies and is at a maximum within the sheeted dyke unit. Moreover, the lower temperature facies may be superimposed on the higher grade ones. The histograms show the intensity of metamorphism in each of the components of the complex for each of the four metamorphic grades.

Significant Points:

(1) The intensity of metamorphism (recrystallisation) is greatest in the sheeted dykes. This is because the vertical dyke margins permit easy access of circulating fluids, coupled with the fact that higher temperatures speed reaction rates. The high water-rock ratios in the sheeted dyke zone mean that the rocks are strongly leached. Any chemical elements not required by the newly forming minerals (hornblendes and chlorite) are removed upwards by the fluids. This means elements like Rb, U, Th, K, some Sr, Ba and chalcophile elements like Zn, Cu and Pb are removed from the deep dyke rocks. Some of the former group will be re-absorbed by the zeolites and clays in the uppermost 'Brownstone' part of the section, others are dissolved in seawater, whereas the chalcophile elements form valuable mounds of sulphide on the seafloor (black smokers). Overall, the hydrothermal activity achieves a major amount of vertical chemical redistribution within the ocean crust. The mobile elements are moved to the top of the ocean crust. These are the same elements that become "mobile" when the ocean crust is subducted and arc magmas are generated.

(2) As thermal gradients fall (i.e. crust moves away from ridge axis) circulating water permits lower grade assemblages to form - superimposed on high grade ones. But this secondary metamorphism is of lower intensity because circulation channels become blocked with the growth of secondary minerals. The sedimentary cap that progressively builds up on top of the basalts will eventually block circulation. There is a good example of this in the "Megaleg"

Chemical fluxes in oceanic crust - the 'MEGALEG'

The Deep Sea Drilling Project Legs 51-53 drilled two deep 200 m holes in Cretaceous (110 m.y.) oceanic crust in the western Atlantic near Bermuda. The holes were only 450m apart, but one hole, Hole 417A, drilled some of the most altered basalts found on the ocean floor, whereas those in Hole 417D were relatively fresh. All alteration was at 'Brownstone' facies. Basalts in both holes petrographically similar. The differences just reflect the relative access by circulating fluids.


Compare compositions (water free):


417A hyaloclastite417A avge417D avge















Note significantly higher K2O and lower CaO in altered rocks, particularly the most fragmentary hyaloclastite 417A, which is the most permeable of rock types.

Interpretation: Hole 417A is on basement 'high' which remained uncovered by sediment for a considerable time thus permitting long-term circulation of warm (30°C) water. Hole 417D was a located in a topographic depression which became quickly filled with sediment which blocked extensive water circulation.

The results were first interpreted to suggest that the ocean crust may be a tremendous 'sink' for K2O and Rb transferred from sea water through hydrothermal circulation involving major volumes of seawater. However, how much of the Rb, K2O, etc. is derived through leaching from more altered rocks (e.g. in sheeted dyke unit) at deeper levels?

To see what other chemical changes occur during alteration of oceanic crust it is best to look at the "type" ophiolite, the Troodos Complex on Cyprus. This has been intensively studied through field investigations, mining operations and by scientific drilling. Troodos formed at 91 Ma, but hydrothermal activity continued for a further 40 Ma after crust formation (Gallahan & Duncan 1994).

Chemical Changes in Oceanic Crust - Troodos Ophiolite





1. Strontium Isotopic Composition

Studies by Spooner et al. (1977) of zeolite- to amphibolite-facies altered basalts on Cyprus show that 87Sr/86Sr ratios are increased relative to fresh basalts and gabbros.



0.80760 ± 3

Altered Basalt


Fresh Gabbros

0.70338 ± 10

Cretaceous Seawater


Similar results have been obtained on altered ocean basalts. High 87Sr/86Sr in altered mineral products can be leached away in dilute acid so that unaltered minerals yield the original magmatic Sr isotopic ratios.

Spooner suggested that interchange of seawater Sr with ocean crust Sr occurs during hydrothermal circulation and may buffer Sr isotope composition of seawater:


Fresh Ocean Crust Av.0.703

Seawater Average


Continental Rocks

> 0.712

There is a considerable variation in 87Sr/86Sr in seawater with time that can be linked to varying plate activity. This is discussed in more detail below.

2. Oxygen and Hydrogen Isotopic Compositions

Spooner et al. (1977) showed that oxygen isotope ratio values in Troodos and other Mediterranean ophiolites were higher ( d18O = ca. 9) than in fresh 'mantle-derived' basalts (d18O = 6) and were consistent with alteration by seawater at high temperatures of ca. 350°C. The interesting point about this is that because oxygen is the most abundant element in any rock, it is necessary to exchange almost all the oxygen in the rock to significantly change the isotopic ratio. In other words, water-rock ratios are high, or large volumes of seawater interact with ocean crust at spreading centres. The implication is that if oxygen can be exchanged on this scale then many other elements can be changed too.

Heaton and Sheppard (1977) showed that the isotopic composition of hydrogen in water in equilibrium with chlorite and amphibole from altered dykes from Cyprus was indistinguishable from that of seawater.

Comment: Altered oceanic crust (now with higher 87Sr, 18O and 2H contents) which is subducted at Benioff Zones may modify the isotopic composition of island arc magmas from "normal" mantle values. The hydrous fluids driven off as the subducting slab heats up as it goes down subduction zones will be enriched in the heavy isotope of these elements. So it is not surprising that island arc magmas differ in their isotopic ratios from other mantle-derived igneous rocks.


Sulphide Ore Deposits in Oceanic Crust

Sulphide deposits have been found on East Pacific Rise. They occur at positions of discharge of hydrothermal systems ("black smokers"). On Cyprus ore bodies are 500 m x 350 m x 50 m, and consist of pyrite and chalcopyrite with accessory marcasite, sphalerite and galena. Chemical and isotopic data suggest that the sulphide deposits mostly formed on seafloor:

(a) Fluid inclusions in ore material have composition of seawater.

(b) Ore material has 87Sr/86Sr = 0.7075.

(c) Hydrogen isotope composition same as seawater.

Calculations suggest ore bodies may have formed in 100,000 yrs. Fluid inclusion studies suggest that the temperature of the plume of rising hydrothermal fluid was 300 - 350°C. Spooner suggests that contained sulphur has isotopic composition of seawater sulphate. So ocean crust ore sulphide may be largely of reduced seawater sulphate origin. The following diagrams illustrate some of the processes of convective seawater circulation and the respective mineral zones in the formation of hydrothermal mounds on the ocean floor (l.h. side) and other submarine envirnments (r.h. side of 1st picture):


These diagrams illustrate the importance of fault-control on the location of the discharge zones of hydrothermal activity, of permeability in allowing the hydrothermal solutions to circulate, of P-T-pH-Eh in controlling which minerals are stable and thus which elements are leached and which are deposited.

Comment: Large amounts of sulphide in adddition to chloride and hydroxyl are added to ocean crust as a result of hydrothermal activity. A lot of pre-concentration of potential ore metals already occurs in the ocean crust. So what happens when ocean crust goes down subduction zones?

At subduction zones chlorine- and sulphide-rich fluids are released during dehydration. Could this give us a possible explanation for porphyry copper deposits that occur commonly at continental margins like the Andes? Spooner has stressed that water is needed as transport medium, chloride for metal complexing and sulphur for fixing the metals as solid phases. All these are present in ocean crust as it is subducted.


Metal Deposits on Ocean Floor

Sulphide ores common in ophiolites. Could we locate them on ocean floor? Many hydrothermal discharge zones have now been found at ocean ridges (mostly in the East Pacific, but also now in the Atlantic) by submersible investigations. These are potentially important ore reserves in terms of total volume, but individual deposits are too small to mine economically, even by remote techniques. Discharge areas may be located by trace element profiles in seawater near the ocean bottom. So we can find them, but to exploit them is another matter.


REFERENCES: Oceanic Crust

CANN, J.R. 1974. A model for ocean crust structure developed. Geophysical Journal of the Royal Astronomical Society 39, 169-187.

CANN, J.R. 1979. Metamorphism in the ocean crust. In: TALWANI, M. et al. (eds) Deep Drilling Results in the Atlantic Ocean: Ocean Crust. Maurice Ewing Series 2, 230-238. American Geophysical Union

CARLSON, R.L. & JOHNSON, H.P. 1994. On modelling the thermal evolution of the oceanic upper mantle: An assessment of the cooling plate model. Journal of Geophysical Research 99, 3201-3214.

ELTHON, D. & STERN, C.R. 1978. Metamorphic petrology of the Sarmiento ophiolite complex, Chile. Geology 6, 464-468.

FLOWER, M.J. 1991. Magmatic processes in oceanic ridge and intraplate settings. In: FLOYD, P.A. (ed) Oceanic Basalts. Blackie, Glasgow, pp.116-147.

GALLAHAN, W.E. & DUNCAN, R.A. 1994. Spatial and temporal variability on crystallisation of celladonites within the Troodos ophiolite, Cyprus: Implications for low-temperature alteration of the oceanic crust. Journal of Geophysical Research 99, 3147-3161.

LIN, J. & PARMENTIER, E.M. 1989. Mechanisms of lithosphere extension at mid-ocean ridges. Geophysical Journal International 96, 1-22.

MARSH, B.D. 1989. Magma chambers. Annual Reviews of Earth and Planetary Sciences 17, 439-474.

NICOLAS, A. 1989. Structures of ophiolites and dynamics of ocean lithosphere. Kluwer, Amsterdam, 367pp.

NICOLAS, A., FREYDIER, Cl., GODARD, M. & VAUCHEZ, A. 1993. Magma chambers at oceanic ridges: How large? Geology 21, 53-56.

RONA, P.A. 1985. Hydrothermal mineralization at slow spreading centers: Red Sea, Atlantic Ocean and Indian Ocean. Marine Mining 5, 117-145.

RONA, P.A. 1986. Mineral deposits from sea-floor hot springs. Scientific American XX, 66-74.

SAWKINS, F.J. 1976. Metal deposits related to intracontinental hotspot and rifting environments. Journal of Geology 84, 427-430.

SAUNDERS, A.D. & TARNEY, J. 1984. Geochemical characteristics of basaltic volcanism within back-arc basins. In KOKELAAR, B.P. & HOWELLS, M.F. (eds) Marginal Basin Geology. Special Publication of the Geological Society, London 16, pp.59-76.

SAUNDERS, A.D. & TARNEY, J. 1991. Back-arc basins. In: FLOYD, P.A. (ed) Oceanic Basalts. Blackie, Glasgow, pp.219-263.

SAUNDERS, A.D., TARNEY, J., STERN, C.R. & DALZIEL, I.W.D. 1979. Geochemistry of Mesozoic marginal basin floor igneous rocks from southern Chile. Geological Society of America Bulletin 90, 237-258.

SINTON, J.M. & DETRICK, R.S. 1992. Mid-ocean ridge magma chambers. Journal of Geophysical Research 97, 197-216.

SPOONER, E.T.C., CHAPMAN, H.J. & SMEWING, J.D. 1977. Strontium isotopic contamination and oxidation during ocean floor hydrothermal metamorphism of the ophiolitic rocks of the Troodos Complex, Cyprus. Geochimica et Cosmochimica Acta 41, 873-890.

THOMPSON, G. 1991. Metamorphic and hydrothermal processes: basalt - seawater interactions. In: FLOYD, P.A. (ed) Oceanic Basalts. Blackie, Glasgow, pp.148-173.

WHITE, R.S. 1991. Structure of ocean crust from geophysical measurements. In: FLOYD, P.A. (ed) Oceanic Basalts. Blackie, Glasgow, pp.30-48.


You may wonder what the strontium isotopic composition of seawater has to do with plate tectonics? Surprisingly the variation in the Sr isotopic composition of ocean seawater with time it is turning out to be an excellent monitor of past plate tectonic activity. But we are only just beginning to understand why. For instance the present day seawater Sr isotopic composition (expressed as 87Sr/86Sr) is 0.709, and because ocean water is globally well mixed, all modern shells and limestones that incorporate seawater Sr have this ratio. By measuring the Sr isotopic ratios in limestones or shells from older geological formations it is possible to log the Sr isotopic variations in seawater with time. The early work by Peterman et al. (1970) and others produced the following curve:

It can be seen that there is a progressive increase from the late Jurassic to the present day. However there is also a decrease from the early Carboniferous to the late Jurassic. So there must be some geological process or processes which produce a decrease in 87Sr/86Sr as well as those which produce an increase. So what are these processes?

Rubidium-87 (87Rb) is radioactive and decays to 87Sr, so that the ratio 87Sr/86Sr must increase in the Earth with time (86Sr is unradiogenic so stays constant). However the rate at which the 87Sr/86Sr ratio will increase depends on the elemental Rb/Sr ratio of the rock. The Earth's mantle is low in Rb relative to Sr, so mantle derived rocks, which have a very low Rb/Sr ratio, tend to have low 87Sr/86Sr (the ratio has increased from only 0.699 to 0.703 over the last 4500 m.y.!). However, crustal rocks such as granites and shales are rich in Rb and have a high Rb/Sr ratio so, given time, become enriched in 87Sr and have a high 87Sr/86Sr. So if seawater interchanges chemically with crustal rocks its ratio of 87Sr/86Sr will increase, wheareas if it interchanges with mantle rocks this ratio will be pulled down again. Clearly to change the ratios in a reservoir the size of an ocean must indicate (at least) two very big process, one pushing the ratio up and the other dragging it down. What are they?


Significance of Increasing Sr Isotope Ratios

Clearly, enhanced erosion of high 87Sr/86Sr continental material will drive the ratio up. As feldspars and micas weather and breakdown to clays their radiogenic Sr is released. This is very soluble in river water and finishes up in the sea. But enhancement can only come if the proportion of mountain belts is increased. And this will be brought about by continental collision at the end of the Wilson Cycle. So does the sharply increasing curve since the late Jurassic then monitor the development of our recent mountain belts such as the Alps, Himalayas and Andes? There have been a number of attempts to model this most recent 100 m.y. growth, since the seawater variation curve for this period is very well known (after Richter et al. 1992):

The Sr flux into the oceans can be estimated from the large rivers, e.g.,



RiverSr Flux (mol/yr)87Sr/86Sr
Amazon2.2 × 1090.7109


0.2 × 1090.7183

Himalayan-Rivers total

7.7 × 1090.7127

Global Total

3.3 × 10100.711


The growth is most rapid between 40 m.y. to the present, and particularly in the period 20 - 15 m.y. Richter et al. (1992) find that they can model this rapid growth mainly as a consequence of uplift of the Himalayan-Tibetan plateau, following the collision of India with Asia. See the high values for the Sr flux in the Himalayan Rivers (Indus, Ganges, Irrawady, Yangtze, Mekong, etc.). So can we detect mountain belt formation in the geological record by rapid increases in the Sr isotopic ratio of seawater?

This poses an interesting problem as to why, if mountain belts are repeatedly formed throughout history, the Sr ratio does not just go on up and up. After all, if the seawater ratio can rise by 0.002 in the last 40 m.y., why has it risen by only 0.010 in the last 4500 m.y. Some equally powerful process must be bringing it down again.

The curve for the whole of the Phanerozoic (Cambrian to present) was established by Burke et al. (1982) by very careful work on limestones of various ages, taking care to avoid rocks that experienced later diagenetic effects:

It can be seen that we actually have to go back to the Cambrian before the seawater Sr ratio was as high as it is at the present day. In fact there has been a general decline for a period of almost 400 m.y. between the Cambrian and the late Jurassic, and superimposed on this is at least five sharp falls.

Significance of Decreasing Sr Isotope Ratios

The main reason for the decrease must be hydrothermal exchange of seawater with hot basalt at mid-ocean spreading centres. As we have seen from ophiolites, the breakdown of feldspars with low "mantle" Sr ratios releases Sr into the seawater. At the same time, zeolites and clays growing in the low-temperature altered basalts take Sr out of seawater with a high "continental" component. Because ocean crust is eventually subducted at trenches, the net effect is to remove some of this "continental" Sr component located in the hyrothermally-altered ocean crust deep into the Earth's mantle.

So can we correlate these periods of rapidly falling Sr isotope composition with periods of faster spreading? Or with the breakup of major continents like Gondwanaland? Note that if there is enhanced ridge activity, this probably means that the uprising mantle is hotter and less dense, and that most likely it will displace ocean water volume, like Iceland, and flood the continental shelves. Of course the more the continents are flooded, the less active will be the rivers, and the lower their content of "continental" Sr. But, in contrast, once there is a major continental collision, like that following the closure of Tethys and the formation of the Alps and the Himalayas, then ridge activity must stop, or slow, for a while until new plate configurations are established. So is this why the curve oscillates around. It is advisable to try to model it. This is what Richter et al. (1992) have tried to do.


Modelling the Phanerozoic Curve

Richter et al. (1992) used the following numbers:


Total Sr in Oceans:

1.25 × 1017 (mol)0.7092

River flux:

3.3 × 1010 (mol/yr)0.711

Hydrothermal flux:

0.82 × 1010(mol/yr)0.7030

Diagenetic flux:

0.3 × 1010 (mol/yr)0.7084


The latter flux is that due to carbonate diagenesis, but is not an important controlling factor.

They then tried to model the Sr isotopic variation throughout the Phanerozoic by making various assumptions about plate motions, collisions, spreading rates, etc. Obviously things get more uncertain the further back in time, and because there are parts of the Earth where the geology is not very well known. The result is shown below:

The model accounts reasonably well for the large-scale structure of the sewater Sr isotopic curve, but fails to reproduce several of the local maxima and minima, especially in the period 300 - 100 m.y. However the "highs" in the Cambrian, Devonian and present day do correlate with extensive mountain building.

The interesting point about such graphs is that they tell us that there may be something missing from our current plate tectonic models. For instance, Richter et al. assume that all mountain building is due to continent collision. But there was no major continent collision as such involved in the uprise of the Andes, which is also a major contributor to seawater Sr. The Andean uplift may have resulted in part from the docking and attempted subduction of major ocean plateaus. This would add a new dimension to the story because the formation of hot ocean plateaus would substantially enhance the ridge hydrothermal Sr flux, to be followed later by an enhanced "continental" flux as mountain belts were formed as these hot thick plateaus tried to subduct.

For further discussion of long-term changes in geological features, and possible implications, see Moores (1993).


BURKE, W.H., DENISON, R.E., HETHERINGTON, E.A., KOEPNICK, R.B., NELSON, H.F. & OTTO, J.B. 1982. Variation of seawater 87Sr/86Sr throughout Phanerozoic time. Geology 10, 516-519.

HESS, J., BENDER, J.L. & SCHILLING, J.-G. 1986. Evolution of the ratio of strontium 87 to strontium 86 in seawater from the Cretaceous to the Present. Science 231, 979-984.

MOORES, E.M. 1993. Neoproterozoic oceanic crustal thinning, emergence of continents, and origin of the Phanerozoic ecosystem: A model. Geology 21, 5-8.

PALMER, M.R. & EDMOND, J.M. 1989. The strontium isotope budget of the modern ocean. Earth and Planetary Science Letters 92, 11-26.

PETERMAN, Z.E., HEDGE, C.E. & TOURTELOT, H.A. 1970. isotopic composition of strontium in seawater throughout Phanerozoic time. Geochimica et Cosmochimica Acta 34, 104-120.

RICHTER, F.M., ROWLEY, D.B. & DePAOLO, D.J. 1992. Sr isotope evolution of seawater: the role of tectonics. Earth and Planetary Science Letters 109, 11-23.

VEIZER, J. 1985. Strontium isotopes in seawater through time. Annual Reviews of Earth and Planetary Sciences 17, 141-156.



As the concept of sea floor spreading gained acceptance in the late 60's, the consequences for geology gradually began to dawn. One of the first to recognise how plate tectonics could be applied to the geological record was J. Tuzo Wilson. If continents rift apart to form ocean basins, other oceans must close. This may be repeated throughout Earth history. Example: the IAPETUS ocean between England & Scotland in the Lower Palaeozoic, closed in the Caledonian; later opening of the Atlantic, almost in the same place. The cycle is known as the Wilson Cycle:

(1) Rifting of continents by mantle diapirism

(2) Continental drift, seafloor spreading & formation of ocean basins

(3) Progressive closure of ocean basins by subduction of ocean lithosphere

(4) Continental collision and final closure of ocean basin

The two diagrams below (Figs 1 & 2) illustrate some simple (if old) concepts of continental rifting (e.g. the Gondwana continent) at the start of the Wilson Cycle. Uprising plume causes doming of crust with magma chamber developing underneath. As extension continues, an ocean basin forms, and thick sedimentary sequences develop at continental margins as rivers dump sediments in deep water. However in reality may be a bit more complex . . .


CONTINENTAL RIFTING: rrr and RRR triple junctions

Four main stages can be recognised in the tectonic development of a typical rifted passive margin:


(1) The RIFT VALLEY stage involves early graben formation prior to continental splitting. This stage may be associated with domal uplift caused by uprise of hot upper mantle material - but this uplift is not ubiquitous and may be connected with underlying mantle hotspots. Example: African Rift Valley.

(2) The YOUTHFUL stage, lasting about 50 my after the onsett of seafloor spreading, while the thermal effects are still dominant. This stage is characterised by rapid regional subsidence of the outer shelf and slope, but some graben formation may persist. Example: Red Sea.

(3) The MATURE stage during which more subdued regional subsidence may continue. Example: most of the present Atlantic continental margins.

(4) The FRACTURE stage when subduction starts and terminates the history of the continental margin.

Fig. 3. The continent of Africa is thought to have been split by a series of rift valleys in various states of development. Those in East Africa are still in thick crust. Those in West Africa are associated with thick oil-bearing sediments. In the Red Sea area the rifting has gone so far as to form a narrow ocean. In the south-east Madagascar has been completely separated from Africa by rifting.

There are many examples of Stage 1. East African Rift Valley is the classic example. But also the Midland Valley of Scotland, the Rhine Graben, the Oslo Graben. These rifts have never got beyond stage 1. Commonly the volcanism associated with these rifts is highly alkaline and undersaturated in silica.

What initiates rifting? There has been considerable discussion on this over the years. Some have ascribed rifting to up-doming of the crust over a hot-spot; certainly parts of the E African rift system are very elevated, compared with other sectors, suggesting that the doming reflects an underlying hot low-density mantle plume. In other cases, geophysical models suggest the asthenospheric mantle is rising to high levels beneath the rift. However it is also apparent that rifting can take place without extensive uplift; in such cases it may be the convective processes in the underlying asthenosphere which are causing the extension. To rift a continent apart it needs the rifts associated with various possible thermal domes to link together. Morgan (1981, 1983) has suggested that as continents drift slowly over hotspots the hotspots weaken the plate - like a blowtorch impinging on the base - and these weakened zones become the sites of continental rifting.

Burke & Whiteman (1973), following the doming hypothesis, suggested that in these domal regions, three rifts would develop, forming an 'rrr' triple junction. Although it is possible that all three rifts might develop into an ocean ('RRR'), it is more likely that two of these rifts would develop into an ocean ('RRr'), leaving the third rift as a 'failed arm'. They demonstrated / speculated that on many continents it was possible to recognise these RRr junctions. The 'failed arm' rift would eventually subside as the thermal anomaly decayed and become the site of a major depositional basin, or a major river channel and delta. The Benue Trough in Nigeria is regarded as an example of such a failed arm following the opening of the S. Atlantic. When oceans eventually close it is possible to recognise these failed arms as depositional basins oriented perpendicular to the collision mountain belt (most basins tend to be aligned parallel to mountain belts). These are termed 'aulacogens'.

Fig. 4. A. Doming by a mantle plume associated with volcanicity. B. Rifting (rrr junction) is initiated. C. Further development results in two of the rifts developing into an ocean, the third is a failed arm (aulacogen). D. Less likely is that all three arms develop into oceans. E. A common situation is that the failed arm develops into a major river system feeding the continental margin. F. Expansion of oceans on a finite earth is not possible: there must be plate subduction, somewhere, sometime. G. Closure of oceans results in island arc development above the subduction zone. H. Continued closure results in collision with major fold and thrust belts. But often the failed arm (aulocogen) is still preserved.


Development of Continental Rifts

Early ideas on the development of rifts are conceptualised in the diagram shown in Fig. 5. This is based on the African rift system, where there is significant rift magmatism. There is notable extension, shown by the widening of the diagram block by at least 50 km. At the same time there is uplift or ascent of the more ductile mantle, especially the asthenosphere. The crust, and particularly the upper crust, is assumed to act in a brittle fashion.

Fig. 5a. Progressive formation of a rift valley through extension of the lithosphere and continental crust (by about 50 km). Note that uprise and decompression of the underlying asthenosphere results in magma formation. The crust responds by brittle fracture. Early rift sediments are downfaulted into the developing rift (graben). Erosion takes place on the sides of the rift valley.

The first stage assumes that graben-like faults begin to form in the brittle crust.

The second stage shows simultaneous necking of the lithosphere with uprise of an asthenosphere diapir. The decompression associated with the latter causes melting of the mantle to give alkaline basaltic magmas. Pre-existing sediments are downfaulted into the graben.

The third stage is accompanied by significant extension and by more uprise of the asthenosphere. The latter causes doming of the crust (which is evident along the E. African rift system, but is variably devloped. New sediments are deposited within the graben as a result of erosion of the uplifting sides of the graben. So there are both pre-rift and syn-rift sediments within the developing rift valley, but sediments on the flanks are progressively erodied away. Note the complex normal-faulting within the rift valley itself.

The fourth stage (Fig. 5b – below) shows the actually rifting-apart of the continent, so the asthenosphere rises towards the surface, causing decompression and extensice melting. New basaltic oceanic crust is formed.

Finally, sea-floor spreading takes over as the ocean basin widens. The rift sedimentary sequence is buried beneath younger marine sediments.

Note: on this diagram the sediments at the continental margin are shown as not very thick. This is because the model is based on the East African Rift System, which does not have a great deal of subsidence associated with rifting. However, other rifted continental margin sequences are very different, with thick sedimentary sequences.



Continental Shelf Sediments

The real situation at passive continental margins is shown in Fig. 6 (below). This is typical of a number of crustal cross-sections across the continental shelf of the eastern Atlantic seaboard of North America, projected down to 30 km -- based largely on gravity and magnetic evidence, plus some seismic profiles -- and some extrapolation from land geology based on deep drill holes.

The critical point is the huge thicknesses of Mesozoic and Tertiary sediments, here shown as almost 15 km, but in other cross-sections this can be even thicker. Note that at the bottom of this pile are volcanics and volcanogenic sediments, and evaporites, which most likely are shallow water. Also, massive carbonate reef structures, which must also be shallow water, but also must indicate progressive subsidence .. .. slow enough that shallow water sedimentation can keep pace with it.

In many sections of the continental shelf off this eastern seaboard of the USA there is a major coast-parallel magnetic structure, possibly a major intrusion. But its age is unknown.

Fig. 6. Profile of deep structure of continental shelf off Atlantic coast of eastern North America -- ?typical of passive continental margins. (Based on gravity, magnetics and seismic data) Critical points regarding this profile are (a) the large thickness of post-rift sediments of Mesozoic-Tertiary age, up to 15 km, and (b) that most of these sediments are shallow-water type. Note: volcanics and evaporites and reef (or carbonate banks)


Rift Terminology

Continental Rift: elongate tectonic depression with which the entire lithosphere has been modified in extension

Rift System: Tectonically interconnected series of rifts

Modern Rift: A rift that is teconically or magmatically active

Paleorift: A dead or dormant rift

Failed Arm: Branch of a triple junction not developed into an ocean basin

Aulacogen: Paleorift in ancient platform that has been reactivated by compressional deformation

Active Rifting: Rifting in response to thermal upwelling of the asthenosphere

Passive Rifting: Rifting in response to remote stress field



Rifts and Mineralisation

Rifting structures are often good sites for mineralisation. This arises for three reasons:

(1) They can be the sites of thick clastic sedimentation. These sediments hold vast amounts of inter-granular salt water (brines). The brines may be in contact with reducing sediments, such as carbonaceous shales, also a ready supply of sulphur/sulphate. As the sediments compact, these brines are expelled and can move laterally for large distances until they move up the rift faults. Having been buried deep the brines get hot, and can be very corrosive. So en route they can dissolve considerable amounts of metals. However, when they rise up the rift faults and cool, these metals will be precipitated out. This can be enhanced because oxidising meteoric water (groundwater) may also penetrate down these faults, so metals wil be precipitated out when the two meet.

(2) Rift structures are also thermally anomalous hot zones. This is because they are frequently underlain by igneous intrusions -- granite (or perhaps in some cases gabbro) plutons. This magmatic heat drives the hydrothermal systems. Importantly, these hydrothermal systems can last for many millions of years, so the hot fluids in these hydrothermal systems can leach away at the rocks within the rift system and precipitate the leached metals nearer the surface. Because the rift structures remain topographically low structures for many tens of millions of years, these metals concentrations can be preserved, without being eroded, for long periods.

(3) The rift zones may be the sites of diverse rocks, particularly basaltic lavas, which can release their metals on hydrothermal alteration. However, because the rift faults can extend very deep (well into the upper mantle in some cases), there may also be a component of deep fluids and metals in the hydrothermal system.




The references below will lead you to some of the discussion on rifting and the Wilson Cycle:

BAKER, B.H., MOHR, P. & WILLIAMS, L.A.J. 1972. Geology of the eastern rift system of Africa. Geological Society of America Special Paper 136, 1-67.

BOSWORTH, W. 1985. Geometry of propagating continental rifts. Nature 316, 625-627.

BOSWORTH, W. 1987. Off-axis volcanism in the Gregory rift, East Africa: implications for models of continental rifting. Geology 15, 397-400.

BOTT, M.H.P 1995. Mechanisms ofrifting: Geodynamic modeling of continental rift systems. In: K.H. Olsen (ed.) Continental rifts: evolution, structure, tectonics. Developments in Geotectonics25, 27-43. Elsevier, Amsterdam

BRAILE, L.W., KELLER, G.R., WENDLANDT, R.F., MORGAN, P. & KHAN, M.A. 1995. The East African Rift system. In: K.H. Olsen (ed.) Continental rifts: evolution, structure, tectonics. Developments in Geotectonics, 25, Elsevier, Amsterdam

BURKE, K. & DEWEY, J.F. 1973. Plume generated triple junctions: key indicators in applying plate tectonics to old rock. Journal of Geology 81, 406-433.

BURKE, K. & WHITEMAN, A.J. 1973. Uplift, rifting and break-up of Africa. In TARLING, D.H. & RUNCORN, S.K. (eds) Implications of continental drift to the earth sciences. Academic Press, London. 735-755.

DEWEY, J.F. & BURKE, K. 1974. Hotspots and continental break-up: implications for collisional orogeny. Geology 2, 57-60.

DUNCAN, C.C. & TURCOTTE, D.L. 1994. On the breakup and coalescence of continents. Geology 22, 103-106.

GURNIS, M. 1988. Large-scale mantle convection and the aggregation and dispersal of continents. Nature 332, 695-699.

MORGAN, W.J. 1981. Hotspot tracks and the opening of the Atlantic and Indian Oceans. In Emiliani, C. (ed) The Sea. Volume 7, 443-487. Wiley, New York.

MORGAN, W.J. 1983. Hotspot tracks and the early rifting of the Atlantic. Tectonophysics 94, 123-139.

MURPHY, J.B. & NANCE, R.D. 1992. Mountain belts and the supercontinent cycle. Scientific American 266, 84-91.

OLSEN, K.H. & MORGAN, P. 1995. Introduction: Progress in understanding continental rifts. In: K.H. Olsen (ed.) Continental rifts: evolution, structure, tectonics. Developments in Geotectonics, 25, 3-26. Elsevier, Amsterdam

SPOHN, T. & SCHUBERT, G. 1982. Convective thinning of the lithosphere: a mechanism for the initiation of continental rifting. Journal of Geophysical Research 87, 4669-4681.

WHITE, R.S. & McKENZIE, D.P. 1989. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research 94, 7685-7730.

WILSON, J.T. 1966. Did the Atlantic close and then re-open? Nature 211, 676-681.




Fig. 1. Simplified relationships at a continental margin. There can be more than 10 km of shallow-water sediments at the margin – implying slow subsidence. How?

Passive continental margins are those associated with continental rifting and the subsequent formation of ocean basins. They differ from active continental margins which are associated with subduction. The continental shelves around the Atlantic are typical passive margins: however there are some quite large differences in the morphology of continental margins around the Atlantic: the reasons for which are not fully understood (but see White et al. 1987; White & McKenzie 1989). There is of course considerable interest in continental margins because of their potential as major oil reservoirs. Hence much has been learned in the last few years.

One aspect of continental margins that has always been puzzling is the existence of very thick – but relatively shallow-water – sedimentary sequences. There can be as much as 15 km of Mesozoic and later sediments at some continental margins bordering the N. Atlantic. How can these very thick sequences be reconciled with gradual but progressive subsidence? Over the years various ideas (summarised in Bott 1979, 1982) have been put forward:

Gravity Loading Hypothesis: This attributes subsidence to sediment load (effectively replacing seawater with denser sediment), and is based on isostacy.

The amount of subsidence depends on relative densities of seawater (1.03), sediment (2.15 – 2.55) and the underlying mantle (3.3). If the sea is filled with sediment then in theory a sediment thickness of over twice the initial depth can develop. In fact a total thickness of 14 km can form near the base of the initial slope. If the lithosphere is treated as elastic the downwarping can extend about 150 km beyond the local sediment load. See Fig. 2 below.

Problem: This mechanism is not easily reconciled with substantial sequences of shallow water sediment. It can only work if the sediments were deposited in deep water initially. If initial water depth is less than 200 m, then sediment loading effect is negligible.

Fig. 2. Gravity loading hypothesis. This depends on replacing low density water by higher density sediment . . .

Thermal hypothesis: This assumes that continental lithosphere near the embryo margin is heated at time of continental rifting - this reduces density of lithosphere permitting isostatic uplift. Subsequently, as the ocean widens, lithosphere cools with time-scale of ca. 50 my and will subside to original position. However if erosion occurred during uplift stage, real subsidence can occur, enhanced by sediment loading.

Fig. 3. Thermal hypothesis of Sleep. This was the first to recognise that heating up the mantle (by a plume or whatever) could produce substantial crustal uplift (and erosion), followed by thermal subsidence. Compare the models by McKenzie and Wernicke later . .

Problem: Even with an extreme initial elevation of about 2 km, the amount of subsidence, even with sediment loading, is not much more than 2 km. So not able to explain thick sequences of over 5 km.

A modification of this thermal model assumes that the thermal event transforms the base of the crust to denser granulite facies mineral assemblages, which may also be invaded by basic magma. If this causes an increase in density of 0.2, it can be calculated that the maximum depth of sediment permitted would only be about 3 or 4 km. Thus insufficient to account for large sediment thicknesses.

Fig. 4. Modification of thermal hypothesis according to Falvey (who argues that heating will cause dense granulite to form).

Problem: such models predict a gap of many m.y. between onset of spreading and the first marine sedimentation - which is not observed.

Crustal Thinning hypothesis: The continental crust and the lithosphere have an upper brittle zone, 20 km thick, overlying a much weaker layer which deforms by ductile flow. Thus crust may thin by progressive creep of middle and lower crustal material towards the sub-oceanic upper mantle. It is argued that this may give rise to jerky subsidence.

Fig. 5. After the initial rifting the lower crust deforms by plastic flow. Could the lower continental crust flow UNDER oceanic crust in the manner shown?

An alternative hypothesis suggests that extreme thinning of the continental crust can occur in a rift valley setting by plastic necking. Then, as the ocean basin forms the passive continental margin will gradually subside.

Fig. 6. Necking of continental crust?






Problem: a typical rift zone is about 50 km wide, thus transition zone at a continental margin would be only 25 km wide. Observed continental margin sequences are however much wider than this.


Normal-fault based mechanisms:


Early hypotheses assumed that graben formation required a wedge of crust about 60 km wide to sink isostatically between inward-dipping normal faults. As the upper crust forms graben by wedge subsidence the ductile lower crust compensates by plastic flow.

Fig. 7. Can normal faulting lead to displacement of ductile mantle by flow?


Problem: Calculations suggested that a subsidence of ca 5 km could occur for an initial 20 km wide trough. Not really enough. But getting nearer.


Faulting near continent-ocean contact:

This mechanism permits limited subsidence as normal faulting accompanies downdrag of the cooling ocean lithosphere. The oceanic lithosphere subsides on a time scale of about 50 my, so consistent with shallow water sediments. However note that the zone of subsidence is too narrow.

Fig. 8. Does normal faulting occur at continental margins in the manner shown in B ?


None of the above mechanisms, either alone or together, seem capable of explaining the observed thick sedimentary sequences at continental margins that are formed at the start of the Wilson Cycle. New ideas were clearly required. These began to develop in the late 1970's as we began to understand more about the thermal behaviour of the lithosphere and about the nature of listric faults.

Continental Lithosphere: The mantle forming the plates is more rigid than the underlying asthenosphere. But this rigid mechanical boundary layer (MBL) varies in thickness. It is thin at the ridges, but thickens to 60 or even 100+ km in old oceanic lithosphere. It may be much thicker under the continents, but it is also older - in fact the lithosphere under the continents is usually as old as the continent above. So it may be cool, and may have experienced enrichment by small degree mantle melts, the components of which may be stored in hydrous minerals.


BOTT, M.H.P. 1979. Subsidence mechanisms at passive continental margins. American Association of Petroleum Geologists Memoir 29, 8-19?

BOTT, M.H.P. 1982. The mechanism of continental splitting. Tectonophysics 81, 301-309.

KUSZNIR, N.J. & ZIEGLER, P.A. 1992. The mechanics of continental extension and sedimentary basin formation: a simple-shear/pure-shear flexural cantilever model. Tectonophysics 215, 117-131.

WHITE, R.S., SPENCE, G.D., FOWLER, S.R., McKENZIE, D.P., WESTBROOK, G.K. & BOWEN, A.N. 1987. Magmatism at rifted continental margins. Nature 330, 439-444.

WHITE, R.S. & McKENZIE, D.P. 1989. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research 94, 7685-7729.




There is no doubt that when ocean basins open there is considerable subsidence of the continental shelves over a wide area, and not just over the immediate rifted margin. This is well exempified by the South Atlantic at ca. 127 Ma, just as the first oceanic crust formed:

Fig. 9. A very large area in the south Atlantic was submerged following break-up at 127m.y. Why? Both Chile and Argentina have modest on-land oil reserves in Patagonia to the west of the Falkland Plateau. DSDP site 330 drilled oily sediments in 1974. Why did Argentina go to war over the Falklands?


Drilling at the eastern spur of the submerged Falkland Plateau revealed that it was continental (granite gneisses) and that there was a dry caliche surface (Mediterranean climate) just before opening of the Atlantic, but that there had been at least 2 km subsidence since then. Initial sediments very oily, deposited under anoxic conditions in a basin with restricted circulation. So the initial rift stage was the one that favoured oil accumulation. Why? It is important to understand the mechanism of development of these basins.

Modern Ideas

It became apparent from COCORP-type deep reflection seismic profiling that many (if not the majority) of steeply dipping normal faults are actually curved (concave-upward) and become shallow-dipping and sub-horizontal at depth. These are now known as listric faults. As the lithosphere is stretched during continental extension, the ductile deeper crust thins by pure shear, while the upper crust is broken up and pulled apart by listric faults which 'bottom out' in the ductile layer. At the surface of course these have the appearance of graben. This is the essence of McKenzie-type and other recent models of basin formation. As the sub-continental (i.e. mantle) lithosphere is thinned by stretching it is of course partly replaced by hotter asthenosphere. This will gradually cool on a time scale of the order of 50 - 100 m.y., and as it cools it becomes denser and the shallow basin above gradually subsides and is progressively filled with shallow-water sediment. The amount of subsidence will depend on the initial amount of stretching. This can usually be estimated and is known as the stretching factor, or "beta factor". The parameter b is defined quite simpy as b/a where awas the initial width and b is the stretched width. A b factor of 1.2 will give ca. 3 km subsidence. With complete rifting (to form ocean crust and an ocean basin) then b approaches infinity.

Note that during the development of sedimentary basins, subsidence occurs in two stages:

(1) as a result of tectonic stretching – on a short time scale, ca. 10 my, and

(2) as a result of thermal subsidence – long time scale, ca. 50 – 100+ my.

Considerable information is now available on North Sea basins as a result of drilling operations and syntheses of the large amount of seismic data (see, e.g. Badley et al. 1988; Gibbs 1984; Sclater & Christie 1980) so their subsidence history is well known. The northern Viking Graben suffered two episodes of rifting – in the Permo-Triassic and in the Middle Jurassic – during which the basin was progressively widened. Stretching factors in the Permo-Triassic were quite small (b = 1.1 – 1.3), whereas in the Late Jurassic were much larger in the northern N. Sea (b = >1.6). Each rifting episode was followed by more substantial thermal subsidence. In the central part of the Viking Graben almost 10 km of sediment has accumulated since the onset of the first rifting episode. As the second rifting phase ended 140 my ago at least 90% of the subsidence resulting from thermal relaxation must have occurred by now. Note that whereas normal faults during the rifting phase tend to be listric, those accompanying thermal subsidence are planar.

An important secondary factor in such models is that the sediments initially deposited in such basins will be 'cooked' slightly as a consequence of the increased heat from the underlying asthenosphere – vital in maturation and migration of petroleum. But sedimentary basins are not only important as oil reservoirs: the expulsion of heated fluids from such basins can leach metals too, thus if suitable host rocks exist valuable mineral deposits can be formed. A number of important mineral deposits are attributed to this mechanism.

Further development of lithosphere stretching models have been proposed by Wernicke, by Lister et al., Coward and others (see references below).

The important difference is in the recognition of low-angle detachments (superficially like thrusts, but with movement sense as in normal fault), first proposed for the Basin & Range province in the western USA. These may bottom out in the lower crust or the upper mantle. The main effect is to introduce asymmetry compared with the pure shear uniform-stretching McKenzie-type model, so that basins associated with the thermal subsidence phase may be offset from the thin-skinned basins associated with the initial rifting. Magmatic effects (melting resulting from the uprising asthenosphere) may be offset from the main sedimentary basins. Because of the asymmetry, the continental margins on the two sides of an opening ocean may have very different profiles. Many other complications may ensue. Consult the references below if you want the full story!

At least 3 types of continental margin have now been recognised:

(1) volcanic,
(2) non-volcanic and
(3) rift-transform.

(1) Volcanic margins tend to be narrow and have a thick igneous crust between continental and normal ocean crust. A thick zone (3 – 5 km) of seaward-dipping volcanic reflectors is typical. Suggestions of convective circulation in uprising asthenosphere to explain volcanism, or that the underlying asthenosphere was hotter than usual. Examples: Voring Plateau, western Rockall Bank, East Greenland. See White et al. (1987 & 1988).

White & McKenzie (1989) have developed these models further to quantitatively relate the volume of volcanics produced at continental margins to the temperature of the underlying mantle. If the temperature is 100°C above normal the volume of magma will be doubled. Also they have developed a relationship between the degree of stretching and the temperature of the mantle to predict whether the rifted margin will rise above sealevel or subside below it. When rifting occurs above hotspot plumes there is usually an accompanying large volume of magma.

(2) Lithospheric deformation on non-volcanic margins is dominated by block faulting and many listric faults. Stretching over a broad zone (100–300 km). May be sediment starved (Red Sea, Galicia Bank, Goban Spur– Irish Sea) or heavily sedimented (e.g. eastern USA margin).

(3) Rift-transform margins evolve in environments where there was a significant component of strike-slip shear as well as extensional strain deformation during opening (e.g. region between W. Africa and Brazil; Falklands Plateau; also Gulf of California).

These different types of margin may have very different petroleum potential. Need to know more about them to aid in locating future supplies. Note that the important petroleum reservoirs in the North Sea are in 'failed-rifts' – where the North Atlantic tried (unsuccessfully) to open quite a long time before it eventually succeeded!

There is a rapidly growing literature on models for continental rifting and basin formation: try to read some of those below, and especially note the diagrams. In any case they may prove useful to you next year.

Another problem of concern is why do we get basaltic magmatism associated with some basins and not with others. Latin and White (1990) have tried to argue that magmatism is more likely with uniform pure shear stretching (McKenzie model) than the asymmetric simple stretching model of Wernicke. This is because asthenosphere uprise is more focussed in pure shear model:

Fig. 13. Comparison of thermal conseqences of McKenzie’s pure shear model and Wernicke’s pure shear model of extensional sedimentary basins. It is argued with the simple shear model it is very difficult to produce sufficient decompression to allow magma formation.


This then has very different thermal consequences:

Fig. 14. With pure shear the temperature of the uprising asthenosphere an exceed the solidus of the mantle and allow melting.Fig. 15. With simple shear the temperature of the uprising asthenosphere never reaches the solidus - so no melting occurs.

Basin Inversion

Basins that have formed by rifting and thermal subsidence don't always remain basins. and may suffer later uplift and erosion. This is known as basin inversion. This happened to many of the Permo-Triassic basins in Western Europe (see Ziegler 1982) and is particularly evident in the NW part of the British Isles and adjacent continental margin. Could this be due to tectonic compression before all the thermal subsidence took place, with the excess sediment being removed by erosion? It is apparent that most of NW Britain was blanketed by Mesozoic sediment that has been removed (viz. Chalk in calderas on Arran) since the early Tertiary, and deposited in basins to the east. Some offshore basins with b factors near 2.0 have a short fall in the expected sediment thickness of ca. 4 km. So something has caused epeirogenic uplift in the early Tertiary over most of NW Britain.

Unfortunately, there is no evidence of enough tectonic compression (Roberts 1989) to account for this uplift by crustal thickening. So what else? Brodie & White (1994) have suggested instead that it may result from magmatic underplating by basalt. They calculate that 5 km of basalt (density 2.8) underplated into the lower crust above the Moho would initially cause 600m uplift. Additionally, with the "amplification" effect of erosion this may increase to ca. 2.5 km. Of course in this general region we know that the Iceland plume was initiated ca. 60 Ma ago (early Tertiary), and one 'rrr' arm extended down through Western Scotland to Lundy. A lot of basalt lavas were erupted. But was much more magma underplated? We know from their geochemistry that many of these basalt magmas have suffered crustal contamination. Are they just a small representative of much more that was ponded in the lower crust? See later lecture on plumes.

The interesting point is that many sedimentation features – basin development, basin inversion, epeirogenic uplift enhancing erosion – may all have their origin in mantle thermal processes. Hence it is important to understand the mantle!

REFERENCES: Sedimentary Basins & Continental Margins

BADLEY, M.E., PRICE, J.D., RAMBECH DAHL, C. & AGDESTEIN, T. 1988. The structural evolution of the northern Viking Graben and its bearing upon extensional modes of basin formation. Journal of the Geological Society, London 145, 455-472.

BARR, D. 1987. Lithospheric stretching, detached normal faulting and footwall uplift. In COWARD, M.P., DEWEY, J.F. & HANCOCK, P.L. (eds) Continental Extensional Tectonics. Geological Society of London, Special Publication 28, 75-94.

BARTON, P. & WOOD, R. 1984. Tectonic evolution of the North Sea Basin: crustal stretching and subsidence. Geophysical Journal of the Royal Astronomical Society 79 987-1022.

BRODIE, J. & WHITE, N. 1994. Sedimentary basin inversion caused by igneous underplating: Northwest European continental shelf. Geology 22, 147-150.

BUCK, W.R. 1991. Mode of continental lithospheric extension. Journal of Geophysical Research 96, 20161-20178.

BUCK, W.R., MARTINEZ, F. STECKLER, M.S. & COCHRAN, J.R. 1988. Thermal consequences of lithospheric extension: Pure and simple. Tectonics 7, 213-234.

COCHRAN, J.R. 1983. Effects of finite rifting times on the development of sedimentary basins. Earth and Planetary Science Letters 66, 289-302.

COOPER, M.A. & WILLIAMS, G.D. 1989. Inversion Tectonics. Geological Society of London, Special Publication 44, 000pp.

COWARD, M.P. 1986. Heterogeneous stretching, simple shear and basin development. Earth and Planetary Science Letters 80, 325-336.

GIBBS, A.D. 1984. Structural evolution of extensional basin margins. Journal of the Geological Society, London 141, 609-620.

HELLINGER, S.J. & SCLATER, J.G. 1983. Some comments on two-layer extensional models for the evolution of sedimentary basins. Journal of Geophysical Research 88, 8251-8269.

HOUSEMAN, G. & ENGLAND, P. 1986. A dynamical model of lithosphere stretching and sedimentary basin formation. Journal of Geophysical Research 91, 719-729.

KENT, P., BOTT, M.H.P., MCKENZIE, D.P. & WILLIAMS, C.A. (eds) 1982. Evolution of sedimentary basins. Philosophical Transactions of the Royal Society, London A305, .

JARVIS, G.T. 1984. An extensional model of graben subsidence - the first stage of basin evolution. Sedimentary Geology 40, 13-31.

KLEMPERER, S. 1988. Crustal thinning and the nature of extension in the northern North Sea from deep seismic reflection profiling. Tectonics 7, 803-821.

LATIN, D. & WHITE, N. 1990. Generating melt during lithospheric extension: Pure shear vs. simple shear. Geology 18, 327-331.

LE PICHON, X., ANGELIER, J. & SIBUET, J.C. 1982. Plate boundaries and extensional tectonics. Tectonophysics 81, 239-256.

LEEDER, M.R. 1983. Lithospheric stretching and North Sea Jurassic clastic source areas. Nature 304, 510-514.

LISTER, G.S., ETHERIDGE, M.A. & SYMONDS, P.A. 1986. Detachment faulting and the evolution of passive continental margins. Geology 14, 246-250.

LISTER, G.S., ETHERIDGE, M.A. & SYMONDS, P.A. 1989. Detachment models for the formation of passive continental margins. Tectonics 10, 1038-1064.

McKENZIE, D.P. 1978. Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters 40, 25-32.

MUTTER, J.C., BUCK, W.R. & ZEHNDER, G.M. 1988. Convective partial melting I: a model for the formation of thick basaltic sequences during the initiation of spreading. Journal of Geophysical Research 93, 1031-1048.

RESTON, T.J. 1990. Mantle shear zones and the evolution of the northern North Sea Basin. Geology 18, 272-275.

ROBERTS, D.G. 1989. Basin inversion in and around the British Isles. Geological Society of London, Special Publication 44, 131-150.

ROWLEY, D.B. & SAHAGIAN, D. 1986. Depth-dependent stretching: a different approach. Geology 14, 32-35.

SAWYER, D.S., SWIFT, B.A., SCLATER, J.G. & TOKSOZ, M.N. 1982. Extensional model for the subsidence of the northern US Atlantic continental margin. Geology 10, 134-140.

SCLATER, J.G. & CHRISTIE, P.A.F. 1980. Continental stretching and explanation of the post mid-Cretaceous subsidence of the central North Sea Basin. Journal of Geophysical Research 85, 3711-3739.

VOORHOEVE, H. & HOUSEMAN, G. 1988. The thermal evolution of lithosphere extending on a low-angle detachment zone. Basin Research 1, 1-9.

WERNICKE, B. 1981. Low-angle normal faults in the Basin and Range province: nappe tectonics in an extending orogen. Nature 291, 645-648.

WERNICKE, B. 1985. Uniform-sense normal simple shear of the continental lithosphere. Canadian Journal of Earth Sciences 22, 108-125.

WERNICKE, B. & BURCHFIEL, B.C. 1982. Modes of extensional tectonics. Journal of Structural Geology 4, 105-115.

WHITE, N. 1989. Nature of lithospheric extension in he North Sea. Geology 17, 111-114.

WHITE, N. & McKENZIE, D. 1988. Formation of "steer's head" geometry of sedimentary basins by differential stretching of the crust and mantle. Geology 16, 250-253.

WHITE, R.S. 1987. When continents rift. Nature 327, 191.

WHITE, R.S., SPENCE, G.D., FOWLER, S.R., McKENZIE, D.P., WESTBROOK, G.K. & BOWEN, A.N. 1987. Magmatism at rifted continental margins. Nature 330, 439-444.

WHITE, R.S. 1988. A hot-spot model for early Tertiary volcanism in the N. Atlantic. In MORTON, A.C. & PARSON, L.M. (eds) Early Tertiary Volcanism and the Opening of the NE Atlantic. Geological Society of London, Special Publication 39, 3-13.

WHITE, R.S. & McKENZIE, D.P. 1989. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research 94, 7685-7729.

WOOD, R. & BARTON, P. 1983. Crustal thinning and subsidence in the North Sea. Nature 304, 561.

ZIEGLER, P.A. 1982. Geological atlas of Western and Central Europe. Shell International, The Hague, 130pp. [JT has slides of relevant maps]



Subduction Zones are where cool lithospheric plates sink back into the mantle. It takes about 50 my for the ocean lithosphere that formed in the hot (>1000°C) environment at mid-ocean ridges to cool to an equilibrium state and sink to its maximum depth below sea-level. Although there is no universal agreement on the balance of forces that drives plate tectonics, the "slab-pull" force is thought to be an important one. For instance the Pacific Plate is the fastest moving plate (ca. 10 cm/yr), and this is the plate that supplies most of the Earth's subducting lithosphere, and thus where the overall slab-pull force will be the larger. The normal argument is that the cool ocean crust will more easily convert to dense eclogite which, as we have seen in Lecture 1, is much more dense than pyrolite.

What is most surprising is the great variation in geological features associated with subduction. There is a huge difference between the East Pacific and the West Pacific. Not only that, but there are differences along the Andean margin, and also quite major differences as we go back in time. But it is important to understand subduction because this is where the continental crust grew progressively with time.

Subduction is where tectonics, structural geology, sedimentation, igneous petrology, metamorphism, geochemistry, geophysics and applied geology all interact. Typical "textbook" features of a mature continental margin subduction zone are shown below. The cartoon shows sediment being scraped off the downgoing plate to form an accretionary wedge, and that a forearc basin is forming on top of the wedge as it is dragged down (and is presumably fed by volcanic debris from the arc). However, the cartoon avoids the issue of how and where the volcanic magmas come from. To what extent does the basaltic subducted slab contribute to arc magmas? Is it just the fluids carried down in altered oceanic crust that migrate into the mantle wedge overlying the subduction zone and cause melting? Ot what extent do sediments carried down the subduction zone then contribute to arc magmas? Why are arc volcanoes nearly always situated about 110 km above the Benioff Zone? What happens to material taken down the subduction zone?




Marginal basins are a common feature of the Western Pacific. Examples (north to south) are the Sea of Japan, the West Philippine Basin, the Parace Vela & Shikoku Basins, the Mariana Trough, the Woodlark Basin, the Fiji and Lau Basins. By contrast marginal basins are rarer in the Eastern Pacific. The two examples in the Atlantic are the Caribbean and the Scotia Sea.

Marginal basins are small oceanic basins, usually adjacent or "marginal" to a continent, which are separated from larger oceans by an island arc. Some marginal basins at continental margins may be imperfectly developed and represented by thinned crust, often associated with basic volcanism. Karig (1971, 1974) divided marginal basins into:

(1) Active marginal basins with high heat flow.

(2) Inactive marginal basins with high heat flow.

(3) Inactive marginal basins with normal heat flow.

The first two are thought to have formed by back-arc spreading, either still active (1), or recently active (2). The third may represent basins formed by even older back-arc spreading, or normal ocean crust that has been "trapped" behind a recently developed oceanic island arc.



The commonly held model of an arc - back-arc system has the following components:

(1) Subduction Zone

(2) Fore-arc region with accretionary sedimentary prism

(3) Frontal Arc

(4) Active Arc

(5) Marginal Basin with spreading centre

(6) Remnant Arc

(7) Inactive Marginal Basin

Although the extensive fore-arc region of many island arcs was thought to be composed of off-scraped sediments, drilling has not substantiated this. It appears that - at least at intraoceanic arcs - abyssal sediments on the downgoing plate are largely subducted.

That the back-arc region is a zone of asthenospheric upwelling is supported by seismic evidence which suggests a low-Q (seismic attenuation) zone behind the arc, compatible with a small amount of melt in the back-arc region:

Magnetic anomalies in back-arc basins are not so well developed, nor have such symmetrical linear patterns, as those in the normal ocean basins. There have been difficulties in identifying the anomalies. It has been suggested by Lawver & Hawkins (1978) that spreading may be more diffuse and not constrained to one central well-defined spreading centre. Good dateable magnetic anomaly patterns were first described from the Scotia Sea back-arc basin (IA Hill). Spreading in some basins may be asymmetric, with accretion favoured on the active arc side.


Models for Back-arc Spreading (see Karig, 1974)

Active Diapirism: One of the earliest models, based on the Mariana Arc System, is that of an uprising diapir splitting the arc. The diapir is initiated either as a result of frictional heating at the subduction zone, or more likely through fluids released from the dehydrating subducting slab. The rising diapir then splits the arc in two and the two halves are progressively separated by seafloor spreading:

Passive Diapirism: This results from regional extensional stresses in the the lithosphere across the arc system. In effect the downgoing slab, although acting like a conveyor belt, also has a vertical component that causes "roll-back". The arc and forearc then stays with the subduction zone, as a result of a supposed trench suction force:

Stepwise Migration: Here it is assumed that the subducting slab is snapped off near the hinge, presumably because something on the downgoing slab is too light to go down, and so a new subduction is initiated oceanwards. The arc stays near the hinge and the asthenosphere wells up behind it:

Convection-driven: This model proposed by Toksoz & Bird (1978), and requires that subsidiary convection cells are driven by the downward drag of the downgoing slab. Calculations suggest that spreading would occur about 10 my after the start of subduction. This might explain why back-arc spreading is more common in oceanic regions ™ the lithosphere is thinner and thus more easily disrupted than under continents:

Uprising Harzburgite Diapir: This model (Oxburgh & Parmentier 1978) depends on the fact that refractory lithosphere (which has lost its basalt component at mid-ocean ridges) is less dense and inherently more buoyant than normal fertile mantle. Thus it would rise if heated to same temperature as surrounding mantle. Such diapirs could in theory be derived from subducting lithosphere, although it is doubtful that subducting lithosphere could be heated within 10 my; more likely it takes 1000 - 2000 my according to megalith concepts of Ringwood (1982):

Old and Young Lithosphere: Molnar & Atwater (1978) have argued that it depends on the dip of the subducting slab whether extension occurs in the back arc region. In the W. Pacific it is old (Jurassic), cold and dense lithosphere that is subducting - with very steep dip and strong vertical component. Thus extensional conditions in back-arc region. In the E. Pacific, on the other hand, the lithosphere subducting beneath the Andes is young (Tertiary), warm and less dense, and subducts at a shallow angle. Thus convergence is more compressive than extensional. Uyeda & Kanamori (1979) have characterised these two extreme types of subduction as Mariana and Chilean type respectively. See also Dewey (1981)

Other models: Various researchers have since commented on the possible causes of back-arc spreading, including assessments of dependence on absolute and relative plate motions. Consult some of references listed below. Experimental laboratory studies have been carried out by Kincaid & Olsen (1987), observing the effects of continued subduction where the subducting slab 'hits' the 650 km discontinuity. The results show that steep subduction does produce a significant roll-back effect on the hinge, which will generate extensional conditions in the back-arc region. Note that with subduction rates of about 7 cm/yr it would take about 10 my before newly subducted ocean lithosphere would 'hit' the 650 km discontinuity and begin to initiate 'roll-back' of the hinge, and thus extensional conditions.



The Mariana Arc is perhaps the type intra-oceanic arc system, and the most extensively studied through marine geophysical studies, dredging and drilling (particularly Legs 58, 59 and 60 of DSDP in late 1970's). From west to east it consists of the following features:

(1) West Philippine Basin

(2) Kyushu-Palau Ridge (a remnant arc)

(3) Shikoku & Parece-Vela Basins

(4) West Mariana Ridge (a remnant arc)

(5) Mariana Trough

(6) Active Mariana Arc

(7) Mariana Fore-arc (made of old arc)

(8) Mariana Trench (up to 11 km deep)

(9) The subducting Pacific Plate (Jurassic age)


West Philippine Basin: This may be 'trapped' in origin and not strictly formed by back-arc spreading. It appears to pre-date the Kyushu-Palau Ridge. Magnetic anomalies suggest active spreading in the early Tertiary (62-40 Ma) with the NW-SE trending Central Basin Fault as the spreading centre. The Oki-Daito Ridge in the northern West Philippine Sea is aligned parallel to this feature and has been regarded as an old remnant arc: however drilled samples from the Oki-Daito Ridge are alkaline basalts, not island arc basalts. Drilled samples from the W. Philippine Basin are fairly typical MORB.

The Philippine Basin is slowly subducting to the west beneath Taiwan, etc. The subduction rate is much less than that of the Pacific Plate beneath the Marianas.


Kyushu-Palau Ridge: This is over 2000 km long and rises 2 km above the adjacent basin floors. Consists of vesicular lava flows, dykes and sills, interbedded with volcaniclastic breccias lying below Middle Oligocene oozes. Lavas all belong to Island Arc Tholeiite (IAT) Series, typical of the most primitive island arcs. Now an inactive Remnant Arc that was active between about 42 and 32 my ago.


Parece-Vela and Shikoku Basins: Magnetic anomaly patterns indicate back-arc spreading between 30 and 17 my in Parece-Vela and between 26 and 15 my in the Shikoku Basin in north. Basaltic sills common in sediments near basement, indicating high rates of sedimentation near near ridge axis. Basalts are vesicular. Similar to MORB.


West Mariana Ridge: Shallower and younger than the Kyushu-Palau Ridge. Drilling penetrated about 1000 m of volcaniclastic material composed of basalts, basaltic andesites, rare andesites and plagioclase phenocrysts. Their character is calc-alkaline, with much higher contents of Ba and Sr than those of K-P Ridge. Arc was active 17-8 my ago. So now a Remnant Arc. Arc built up when spreading in P-V / Shikoku Basins ceased.


Mariana Trough: This is 1500 km long, 250 km wide. Rough topography, high heat flow. Magnetic lineations poorly developed, but suggest back arc spreading from about 6 my ago - i.e. when activity on West Mariana Ridge ceased. Near the West Mariana Ridge metabasalts, gabbros and anorthositic cumulates were drilled - deeper part of a rifted-apart arc? Basalts in Mariana Trough are MORB-like, but have some arc characteristics. Vesicular. Spreading still in progress. Further north, on Iwo-Jima Ridge, there is an incipient back-arc basin just beginning to form - the Bonin Trough.


Mariana Active Arc: This consists of numerous small islands and seamounts, on the eastern edge of the extensive Fore-arc region. Lavas are mainly basalts, basaltic andesites and andesites.


Mariana Fore-arc: The forearc region shows a history of continual subsidence. The basement is Eocene in age (similar to Kyushu-Palau Ridge) and consists of two distinct lava types:


(1) Island Arc Tholeiites (very similar in character to those of Kyushu-Palau Ridge). These magmas can normally be easily distinguished from calc-alkaline basalts from more mature arc systems.

(2) Boninites, or high-magnesian andesites. These are unusual lavas, combining high Si with high Mg, Ni and Cr. They are thought to have formed by wet-melting of rather refractory lithosphere.

(3) Dacites also occur on Guam.


Drilling and dredging in the trench area of the fore-arc has recovered mainly volcanic materials. No scraped-off sediments from the oceanic plate - with the implication that all sediment is being subducted, and that the fore-arc itself is suffering tectonic erosion as a result of the rasping action of the downgoing slab.




Combining evidence from magnetic anomalies, drilling, dredging and geochronology, the geologic history of the arc system can be pieced together. In the period immediately preceding the development of the arc, the plate configuration in the eastern Indian Ocean and western Pacific was dominated by the rapid movement of India northward. There were some major N - S oriented transform faults at this time, so about 60 Ma ago the plate tectonic configuration probably looked like this:


India was just about to collide into Asia to form the Himalayas, Australia had just begun to separate from Antarctica, and note the very large ridge offsets on the N-S transforms. The critical point at this time was that slab-pull associated with the rapidly-moving Indian Plate will stop as soon as India collides. Similarly, the spreading ridge in the NE Pacific is going to push itself under the Aleutians, when upon the slab-pull will also stop. This leaves the northerly pull forces on the Pacific plate very weak, and very vulnerable to change in plate motion direction. So about 40 my ago the Pacific Plate changed motion from northwards to westward (c.f. kink in Hawaiian-Emperor seamount chain). The sequence of events can be tracked as follows:

(1) The Kyushu-Palau Ridge is thought to mark the position of one of these major transform faults, with younger, warmer and thinner ocean ocean lithosphere to the west, and older, cooler and denser lithosphere to east. Drawn to scale, the position immediately before the change in plate motion probably looked like this:

It can easily be envisaged how the eastern side would easily subduct under the new young warm lithosphere to the west that had recently formed at a spreading ridge. After the change in plate motion direction, the map then looked like:

A new volcanic arc forms at the site of the easternmost transform, and many complications develop in SE Asia (Philippines, etc.) because of transforms turning into arcs, and various subduction-flips as thick (plateau-type) ocean crust refuses to subduct. A new subduction zone develops north of Australia.

(2) Rapid build-up of Kyushu-Palau Arc in late Eocene – Oligocene through voluminous eruption of island arc tholeiites and high-Mg boninites. Activity continued for ca 10 my. So what happened to bring about such a rapid rate of magma production. It is possible that the earliest stages of subduction looked as follows:

Note that the downgoing plate not only has "conveyor-belt" motion, but also a strong vertical component so that it is sinking into the mantle. At this point hot asthenosphere mantle rushes in to replace it. So in a rather unique rapidly extensional tectonic environment, wet altered ocean crust is juxtaposed next to very hot asthenospheric mantle. With an abundance of heat and water, it is not surprising that huge amounts of magma are generated. This tectonic situation is actually even more extensional than at a mid-ocean ridge, so it may be expected that all the features of a "type" mid-ocean ridge are reproduced: pillow lavas, sheeted dykes, gabbros, etc. This is shown below:

(to come)


(3) Splitting of K-P Arc in half about 30 my ago with formation of Parece-Vela & Shikoku Basins by back-arc spreading. Spreading stopped about 16 my ago.


(4) Formation of West Mariana Arc between about 17 and 8 my ago through eruption of calc-alkaline basalts and basaltic andesites.


(5) Splitting of West Mariana Arc abut 6 my ago to form Mariana Trough by back-arc spreading, and leaving West Mariana Ridge as remnant arc.


(6) Formation of new Mariana Arc 5 my ago to present. Now erupting lavas with mixed calc-alkaline - island arc tholeiite characteristics.

Presumably the Mariana Arc will continue migrating eastwards into the Pacific.


Magma Compositions

Arc Magmatism

The magmas erupted at the Mariana Arc show a gradual evolution in composition with time. Note that the whole arc system has evolved entirely within the oceanic regime (no continental crust or sub-continental lithosphere involved).

The earliest lavas erupted (now seen on Kyushu-Palau Ridge and Mariana Fore-arc) are island arc tholeiites (IAT) and boninites. These are characteristic of very primitive oceanic island arcs, and are not usually erupted on continents or in the later stages of arc development. IAT have similarities with mid-ocean ridge basalts (MORB), in having depleted rare-earth element (REE) patterns, but are usually more Fe-rich and with low Cr and Ni contents, very low Nb and Ta, higher K contents and high K/Rb ratios. Boninites are high-Mg lavas, but have high silica contents more typical of andesites; they have high Cr and Ni contents, but have lower Ti contents and higher K, Rb, Ba and Sr contents than would normally be expected of high-Mg rocks.

Boninites are thought to result from wet melting of the rather refractory Mg-rich mantle wedge beneath the developing arc - with the wedge being contaminated with elements such as K, Rb, Ba, Sr transported from the subduction zone during dehydration of the hydrous ocean crust.

IAT could be melts of the more fertile asthenosphere, the magmas then undergoing extensive crystal fractionation en route to the surface. Or they could represent melts of subducted ocean basalt crust (only possible at the very start of subduction when the ocean lithosphere is pushed down into hot mantle).

After opening of the Parece Vela basin by back-arc spreading, arc volcanic activity was transferred 17 my ago to the what is now the West Mariana Ridge, and continued building up that arc for ca. 9 my. The lavas erupted however were mainly calc-alkaline basalts (CAB) and basaltic andesites, with higher Al contents, much higher Sr and Ba contents and light rare-earth enriched rather than depleted REE patterns. These lavas are more similar to calc-alkaline lavas erupted at continental margins (though the latter are usually dominated by andesite rather than basaltic andesites).

These CAB magmas may have been derived from the mantle wedge. But if so there is an implication that the wedge may have been enriched in Ba, Sr, light REE, etc., perhaps as a result of continued fluid transport of these elements into the wedge from the dehydrating subducting slab.

Modern lavas erupted at the active Mariana Arc tend to be mainly andesites and basaltic andesites having characteristics in between those of IAT and CAB. There is some evidence that a small component (ca. 0.5%) of subducted abyssal sediment is involved in their source regions.

Perhaps the most interesting aspect of the Mariana arc is that at least three distinct magma types appear to have been generated from the one subduction zone. Yet the whole arc system evolved entirely within the oceanic environment.

Back-arc Basalts

In many respects marginal basin basalts (MBB) are similar to normal mid-ocean ridge basalts (N-type MORB). However during the early stages of back-arc spreading, when the uprising mantle diapir splits the volcanic arc, the basalt magmas are derived from the sub-arc mantle. These basalts tend to have an arc-like geochemical signature. Thus their REE patterns may be slightly light REE enriched, they have higher Ba, Sr, K and Rb, but low Nb and Ta. Moreover they tend to have higher water contents and be vesicular - a consequence of fluids distilled from the subducting slab. These features are useful discriminants in trying to characterise ophiolites as being derived from either obducted ocean floor or marginal basin crust. See Saunders & Tarney (1984; 1991) for summary.


Addition: Schematic cross-section across the Mariana Arc showing the components involved in magma generation.

Fluids are released from the sub-ducting slab as "wet" amphibolite recrystallises at ca. 100km depth to dry dense eclogite. These fluids migrate upwards into the mantle wedge and induce melting of the sub-arc lithosphere. (The more water, the more melting, and higher the magma production?). However, this mantle varies in it's fertility because of previous metasomatic events affecting the deeper lithosphere.

More active mantle diapirism occurs in the back-arc region, and this results in much more melting and active spreading. Hydrous fluids are still involved in these mamgas, but to a lesser extent than in the arc rocks.



If we bear in mind that plate motions are dominantly controlled by 'slab pull', then anything which reduces the slab-pull force will encourage changes in the direction and speed of plate motion. It is notable that in the southeastern Pacific the Aluk Ridge (spreading centre) began to progressively subduct along the Antarctic Peninsula; at the same time, the northwestern Pacific the Kula Ridge began to subduct beneath the Aleutians - Kamchatka. A result was a marked reduction in the N™S slab-pull, because recently formed hot lithosphere is not very dense and not keen to subduct. In combination with other plate re-configuring events worldwide, this may have been enough to cause switch in Pacific Plate motion from N – S to E – W. But see Richards et al. (1996)


REFERENCES: Arcs and Marginal Basins


The references below lead to most aspects of interest to island arcs, even if you just look at the abstracts & diagrams!


BLOOMER, S.H. 1987. Geochemical characteristics of boninite- and tholeiite-series volcanic rocks from the Mariana forearc and the role of an incompatible element-enriched fluid in arc petrogenesis. Geological Society of America, Special Paper 215, 151-164.

CARLSON, R.L., HILDE, T.W.C. & UYEDA, S. 1983. The driving mechanism of plate tectonics: relation to age of the lithosphere at trenches. Geophysics Research Letters 10, 297-300.

CHASE, C.G. 1978. Extension behind island arcs and motions relative to hot spots. Journal of Geophysical Research 83, 5385-5387.

CHASE. C.G. 1979. Asthenospheric counterflow: a kinematic model. Geophysical Journal of the Royal Astronomical Society 56, 1-18.

CRAWFORD, A.J., BECCALUVA, L. & SERRI, G. 1981. Tectono-magmatic evolution of the West Philippine-Mariana region and the origin of boninites. Earth and Planetary Science Letters 54, 346-356.

DAVIES, J.H. & STEVENSON, D.J. 1992. Physical model of source region of subduction zone magmatism. Journal of Geophysical Research 97, 2037-2070.

GARFUNKEL, Z., ANDERSON, C.A. & SCHUBERT, G. 1986. Mantle circulation and the lateral migration of subducted slabs. Journal of Geophysical Research 91, 7205-7223.

HAMILTON, W.B. 1988. Plate tectonics and island arcs. Geological Society of America Bulletin 100, 1503-1527.

HASTON, R. & FULLER, M. 1991. Palaeomagnetic data from the Philippine Sea plate and their significance. Journal of Geophysical Research 96, 6073-6098.

HAWKINS, J.W., BLOOMER, S.H., EVANS, C.A. & MELCHIOR, J.T. 1984. Evolution of intra-oceanic arc-trench systems. Tectonophysics 102, 174-205.

HICKEY, R.L. & FREY, F.A. 1982. Geochemical characteristics of boninite series volcanics: implications for their source. Geochimica et Cosmochimica Acta 46, 2099-2115.

HILDE, T.W., UYEDA, S. & KROENKE, L. 1977. Evolution of the western Pacific and its margin. Tectonophysics 38, 145-167.

HOLE, M. J., SAUNDERS, A. D., MARRINER, G. F. & TARNEY, J. 1984. Subduction of pelagic sediment: implications for the origin of Ce-anomalous basalts from the Mariana Islands. Journal of the Geological Society, London 141, 453-472.

HSUI, A.T., MARSH, B.D. & TOKSOZ, M.N. 1983. On melting of the subducted ocean crust: effects of subduction induced mantle flow. Tectonophysics 99, 207-220.

IDA, Y. 1983. Convection in the mantle wedge above the slab and tectonic processes in subduction zones. Journal of Geophysical Research 88, 7449-7456.

JURDY, D.M. 1979. Relative plate motions and the formation of marginal basins. Journal of Geophysical Research 84, 6796-6802.

JURDY, D.M. & STEFANICK, M. 1983. Flow models for back-arc spreading. Tectonophysics 99, 191-200.

KARIG, D.E. 1974. Evolution of arc systems in the Western Pacific. Annual Reviews of Earth and Planetary Sciences 2, 51-78.

KARIG, D.E. 1971. Structural history of the Mariana island arc system. Geological Society of America Bulletin 82, 323-344.

KARIG, D.E. 1971. Origin and development of marginal basins in the Western Pacific. Journal of Geophysical Research 76, 2542-2561.

KARIG, D.E. 1982. Initiation of subduction zones - Implications for arc evolution and ophiolite development. Geological Society of London, Special Publication 10, 563-576.

KINCAID, C. & OLSON, P. 1987. An experimental study of subduction and slab migration. Journal of Geophysical Research 92, 13832-13840.

KUSHIRO, I. 1990. Partial melting of mantle wedge and evolution of island arc crust. Journal of Geophysical Research 95, 15929-15939.

LAWVER, L.A. & HAWKINS, J.W. 1978. Diffuse magnetic anomalies in marginal basins: their possible tectonic and petrologic significance. Tectonophysics 45, 323-339.

MARSH, B.D. 1979. Island arc development: some observations, experiments and speculations. Journal of Geology 87, 687-713.

MOLNAR, P. & ATWATER, T. 1978. Interarc spreading and cordilleran tectonics as alternates related to the age of subducted ocean lithosphere. Earth and Planetary Science Letters 41, 330-340.

MUELLER, S. & PHILLIPS, R.J. 1991. On the initiation of subduction. Journal of Geophysical Research 96, 651-665.

NATLAND, J.H. & TARNEY, J. 1982. Petrological evolution of the Mariana Arc and Back-arc Basin System: a synthesis of drilling results in the South Philippine Sea. Initial Reports of the Deep Sea Drilling Project 60, 877-908 (Washington: U.S. Government Printing Office).

PEACOCK, S. M. 1990. Fluid processes in subduction zones. Science 248, 329-337.

RICHARDS, M.A. & LITHGOW-BERTELLONI, C. 1996. Plate motion changes, the Hawaiian™Emperor bend, and the apparent success and failure of geodynamic models. Earth and Planetary Science Letters 137, 19-27.

RINGWOOD, A.E. 1974. The petrological evolution of island arc systems. Journal of the Geological Society, London 130, 183-204.

SAUNDERS, A.D. & TARNEY, J. 1984. Geochemical characteristics of basaltic volcanism within back-arc basins. In KOKELAAR, B.P. & HOWELLS, M.F. (eds) Marginal Basin Geology. Geological Society of London, Special Publication 16, 59-76.

SAUNDERS, A.D. & TARNEY, J. 1991. Back-arc basalts. In FLOYD, P.A. (ed) Oceanic Basalts. Blackie, Glasgow, pp. 219-263.

SHEMENDA, A.I. 1993. Subduction of the lithosphere and back arc dynamics: insights from physical modeling. Journal of Geophysical Research 98, 16167-16185.

SPENCE, W. 1987. Slab pull and the seismotectonics of subducting lithosphere. Reviews of Geophysics 25, 55-69.

STERN, R.J. & BLOOMER, S.H. 1992. Subduction-zone infancy - Examples from the Eocene Izu-Bonin-Mariana and Jurassic California arcs. Geological Society of America Bulletin 104, 1621-1636.

STERN, R.J., BLOOMER, S.H., LIN, P.-N. & SMOOT, N.C. 1989. Submarine arc volcanism in the southern Mariana arc as an ophiolite analogue. Tectonophysics 168, 151-170.

TARNEY, J., SAUNDERS, A.D. & WEAVER, S.D. 1977. Geochemistry of volcanic rocks from the island arcs and marginal basins of the Scotia Arc region. In: TALWANI, M. & PITMAN, W.C. (eds) Island Arcs, Deep Sea Trenches and Back-arc Basins. American Geophysical Union, Maurice Ewing Series 1, 367-378.

TARNEY, J., SAUNDERS, A. D., MATTEY, D. P., WOOD, D. A. & MARSH, N. G. 1981. Geochemical aspects of back-arc spreading in the Scotia Sea and Western Pacific. Philosophical Transactions of the Royal Society of London A300, 263-285.

TARNEY, J., PICKERING, K.T., KNIPE, R.J. & DEWEY, J.F. 1991. Fluids and subduction zone processes. In TARNEY, J., PICKERING, K.T., KNIPE, R.J. & DEWEY, J.F. (eds) Behaviour and Influence of Fluids in Subduction Zones. The Royal Society, London. (i-vi)

TATSUMI, Y., MURASAKI, M. & NOHDA, S. 1992. Across-arc variation of lava chemistry in the Izu-Bonin Arc: identification of subduction components. Journal of Volcanology and Geothermal Research 49, 179-190.

TAYLOR, B. & KARNER, G.D. 1983. On the evolution of marginal basins. Reviews of Geophysics 21, 1721-1741.

TOKSOZ, N. & BIRD, P. 1977. Formation and evolution of marginal basins and continental plateaus.In TALWANI, M. & PITMAN, W.C. (eds) Island Arcs, Deep Sea Trenches and Back-arc Basins. American Geophysical Union, Maurice Ewing Series 1, 379-393.

UYEDA, S. & KANAMORI, H. 1979. Back-arc opening and mode of subduction. Journal of Geophysical Research 84, 1049-1061.

WYLLIE, P.J. 1988. Magma genesis, plate tectonics and chemical differentiation of the earth. Reviews of Geophysics 26, 370-404.

ZHAO, D., HASEGAWA, A. & HORIUCHI, S. 1992. Tomographic imaging of P and S wave velocity structure beneath northeastern Japan. Journal of Geophysical Research 97, 19909-19928.



Contrasts between West & East Pacific

Uyeda & Kanamori (1979) emphasised that there were two contrasting types of subduction zone: Mariana Type and Chilean Type - with of course many intermediate types. The Mariana Type is characterised by a very steeply dipping slab; the Chilean Type by a shallow-dipping slab. These differences were further amplified by Dewey (1981).

Mariana Type has:

1. Deep open trench (up to 11 km deep) that subducts old cold Jurassic crust.

2. A very steep Benioff Zone

3. Extensive faulting, subsidence and tectonic erosion of the outer trench wall.

4. Widespread intra-arc extension and back-arc spreading.

5. More earthquakes in the under-riding than in the over-riding plate.

6. A rather thin mafic-intermediate composition volcanic-plutonic crust.

7. Extensive volcanism; mainly basaltic with only minor andesites.

8. Little or no sedimentary accretion at the trench.

9. Subdued morphological expression.

10 Lavas have quiet eruptive style.

11 Volcanoes are mainly submerged cones with fringing reefs.

12 Poorly developed volcaniclastic dispersal fans.


Chilean Type has:

1. Shallower trench (up to 6 km) that subducts younger, warmer, Eocene age oceanic crust.

2. Thrust faulting common on outer trench wall.

3. Major thrust faulting in the under-riding Nazca Plate up to 200 km west of the trench.

4. A Benioff Zone with a very shallow dip down to about 200 km, and then a steeper deeper portion

below a seismic gap.

5. Widespread intra-arc compression and back-arc thrusting over a foreland trough.

6. More, and higher energy, earthquakes in the over-riding than in the under-riding plate.

7. Plutonism is dominant over volcanism.

8. Volcanism is dominantly of andesite-dacite-rhyolite type; basalts being much rarer.

9. Thick (ca 70km) continental crust gradually tapering trenchward to less than 10 km.

10 Because of dominant compression, continental arc has high uplift rates.

11 Violent eruptive style. High viscosity lavas. Extensive volcaniclastic dispersal fans.

12 Spectacular geomorphological expression.


Difference in seismic characteristics: The steep dip of the Benioff Zone in the Mariana type means that the contact interface between the subducting slab and the mantle wedge lithosphere is less than 100 km, hence not much frictional drag. In any case tectonic conditions are extensional. In Chilean type however, the shallow slab dip and greater thickness of continental lithosphere means that the contact interface can be as much as 400 km. Hence considerable resistance and friction and much greater seismic activity.


Tectonic Erosion and Accretion: In the Mariana Arc there is no accretion of abyssal sediments at the trench. Yet considerable volumes of sediment are entering the trench: sediments are 0.5km thick on Pacific Plate entering the trench, subduction rate 10 cm/yr for ca. 40 m.y. (work out how many cubic km per unit length of arc!). Instead forearc is undergoing tectonic erosion ("subcretion"). Most of the sediment is being subducted - only a small proportion of it is re-cycled into arc volcanics. Along Chilean margin the sediment supply varies: very little in north where desert conditions, but much more in south where rainfall is high. It has been suggested that the continental basement may be eroding by subcretion in Northern Chile, but growing by sediment accretion in Southern Chile. Where sediment supply is high, sediments may fill the trench and flood over on to the oceanic plate; thus depressing it so that it approaches subduction zone at a shallow angle.


Explanation for differences between East and West Pacific Margins

Contrast cannot be explained simply by differences in convergence rate, since Chilean, Mariana, Japanese and Tonga arcs all have head-on convergence rate of about 10 cm/yr. Contrast must be related to balance between "roll-back" of hinge and convergence rate. If roll-back is faster than convergence rate then back-arc extension results; if slower, then back-arc compression.

Roll-back may be determined by age of subducting lithosphere (Molnar & Atwater 1978). Old cold lithosphere is denser and subducts at steeper angle . . presumably takes less time to reach 650 km discontinuity. If it cannot penetrate discontinuity then splays back (see experiments of Kinkaid & Olsen (1987)) and induces roll-back of hinge at subduction zone, giving extensional tectonics. However, with shallower angle subduction of younger warmer lithosphere the slab will take longer to reach 650km discontinuity, and will warm up more and become less coherent and less able to induce roll-back effect. So no extension. An additional factor is that in the Eastern Pacific the American Plate is over-riding the Pacific (Nazca) Plate due to the opening of the Atlantic . . although the rate is quite small.

Wider implications: If the balance between compression and extension at convergent plate margins is related to dip of slab (and hence age of lithosphere subducting), then it may explain why intraoceanic island arcs are essentially a Phanerozoic phenomenon, and become rare or absent in the middle to early Precambrian. Higher thermal gradients in Precambrian would mean greater ridge length and smaller plates (see Hargraves 1986), so subducting plates would be younger and warmer, and less likely to subduct at steep angle. Hence much less likely to induce extensional conditions at convergent plate boundaries. Is it only when there is extension that island arcs are produced?



DEWEY, J.F. 1981. Episodicity, sequence and style at convergent plate boundaries. In: The Continental Crust and its Mineral Deposits. Geological Association of Canada, Special Paper 20, 553-572.




For the last 2 decades, geologists, geophysicists and geochemists have argued about the physical and chemical conditions which allow melting to occur in subduction zones. Whereas it is easy to explain magmatism at ocean ridges where hot mantle is rising, it is not at first easy to explain why magmas appear in abundance when a cold slab is pushed into the mantle at subduction zones. It used to be thought that friction between the overriding and under-riding plates was responsible, but calculations have showed that friction is most unlikely: there is probably too much hydrous fluid and soft oozy subducted sediment that act as a lubricant. It is important to try to understand the thermal structure of subduction zones.

In a classic review paper, Ringwood (1974) suggested that the most primitive island arc lavas (IAT), which are basaltic, could be related to dehydration of the hydrated ocean crust (amphibolite) as it transforms to dense eclogite at depths of ca. 100km. The hydrous fluids rise up into the peridotite mantle wedge, promoting melting (magmas form at much lower temperatures in the presence of water). These magmas then rise slowly up to the arc volcanoes above, and crystallise Mg-rich olivines and pyroxenes as they ascend, so the magmas become more iron-rich. The eruption of basalt (tholeiite) is non-violent. This is shown in cartoon form:

For the calc-alkaline, more silicic andesitic and dacitic magmas or more mature arcs, Ringwood suggested a slightly different mechanism based on his experimental work on eclogite. Hydrous melting of eclogite (if Si-poor garnet stays in the residue produces silica-rich dacitic magmas. These then react with the mantle wedge and rise up as diapirs and erupt as much more violent hydrous magmas, of which Mt. St. Helens is a good example.

However, there are a number of problems with these simple models, and it is now accepted that they only acount for a minor number of features of subduction zone magmas. For instance, the primitive Mariana arc tholeiites are really a result of fore-arc diapirism connected with the initiation of a new subduction zone, following a change in plate motion, as outlined in the last lecture.

How can subduction zones give rise to the following range of magmas? Surely this must imply a range of P-T conditions that involve both slab and wedge melting?


Boninites (High-Mg andesites): usually formed at early stage of island arcs

Island Arc Tholeiites (IAT): normally restricted to primitive island arcs

Calc-alkaline basalts & andesites: found in mature island arcs and continental margins

Bajaites (Adakites)High-Mg andesites (but different from boninites) where ridge subduction occurs or mafic rocks have been underplated.

Shoshonites: Often late-subduction or post-subduction: high-Ba, Sr magmas

Archaean TTG suite: Distinctive, and thought to be derived from subducted ocean crust (they resemble adakites).


The critical points of issue are:

(a) under what conditions does the slab melt?

(b) what is the difference between subducting old ocean crust and young ocean crust?

(c) do magmas originate instead in the mantle wedge?

(d) what is the mineralogy of the wedge. Are minerals like hornblende, phlogopite & K-richterite stable in subduction zones?

(e) how can the difference between primitive island arcs (e.g. Marianas) that tend to erupt basalts, and mature arcs (e.g. Andean margin) that tend to erupt andesite, be explained?


Anderson et al. (1978; 1980) were the first to consider the thermal structure of subduction zones seriously. Wyllie & co-workers, in a series of papers (e.g. Wyllie, 1988), used experimental petrology to try to constrain what will melt under hydrous conditions, and what the magma compositions would be. He produced some useful cartoon models, one of which is shown below:

The important points to note are that the ocean crust reaching a subduction zone will be relatively "cold" and "wet". Just how cold it will be will depend on just how many hundreds or thousands of km it has travelled from the spreading ridge. It will be wet as a result of hydrothermal alteration near the ridge axis. As the plate subducts the basaltic crust will undergo a progressive increase in metamorphic grade – Greenshist > Amphibolite > Eclogite facies – which is also a series of dehydration reactions to about 100km depth.


More recently, Peacock (1991) and Bickle & Davies (1991) have produced much better thermal models. For instance, Peacock (1991) has produced useful thermal numerical models. He explores the thermal effect of:


(a) Age of the oceanic crust being subducted. Clearly young warm ocean crust will be more likely to melt if subducted than old cold lithosphere. The diagram below shows the increase in temperature at 1 my intervals (dots) as ocean crust ranging in age from 5 my [A] to 200 my [D] is subducted to 200 km. The surprising result is that only when quite young crust is being subducted is there a possibility of melting (i.e. temperatures reach >900°C beneath the arc). So as subduction continues and older and older crust begins to be subducted, it is less easily melted:

The blocks labelled eclogite and blueschist show the P-T conditions found in exhumed subduction complexes (e.g. the Franciscan of California) which are consistent with an average age of 50 Ma old for subucted ocean crust.


(b) Amount of previously subducted lithosphere. Clearly the more you stuff cool oceanic lithosphere into the upper mantle, the more it will cool it (the iced drink analogy!). With subduction rates of 10 cm/yr it is possible to subduct 100 km of ocean crust per m.y. [the diagrams are drawn for a much slower subduction rate of 3 cm/yr]


The implication from the diagram below is that the cooling effect of continued subduction is quite severe, So after less than 600 km (= 6 m.y.) of ocean crust subduction, temperatures are below those at which the slab melts. But what about the mantle wedge?

The diagram also shows how the top and base of oceanic crust heats up. The base is initially hotter, but the top eventually gets hotter because of heat conducted from the mantle wedge.

(c) Magmas from the mantle wedge? Curves E and F show the temperatures of the mantle wedge (straddling the depths at which magmas are generated below arc volcanoes) at 10 m.y. and 20 m.y. after the start of subduction, but without allowing any convection in the mantle wedge. The cooling effect of the slab is very important, quickly taking the wedge below temperatures at which magmas would be generated.


However, Curve G shows the effect of allowing induced convection in the mantle wedge (a similar curve linked with E would be even higher temperature . . . ). In this case the temperatures stay above 950 °C as the wedge material is dragged down, and so hydrous melting would be possible. An important implication from this diagram is that it is much more likely that arc magmas are derived from the mantle wedge: conditions for slab melting are very restricted.


(d) Effect of induced convection on slab. Briefly, the modelling shows that induced convection can enhance the meltability of older slab, but the effect on young ocean crust is not important:


(e) Temperature or pressure control on magma generation?

The diagram above shows effect of water on the melting behaviour of basaltic oceanic crust. Under dry conditions melting increases with pressure (red dashed line). However, under water-saturated conditions (red full line) melting temperatures plummet by almost 400°C at depths of 50 km.

Importantly, the blue curve shows how hornblende becomes an important mineral under hydrous conditions; however, note that the curve turns over at ~70-80 km to become pressure-sensitive – hornblende in mantle breaks down at ~100+ km. This means that a lot of fluid will be released from hornblende at these depths, which could promote melting. Is this why most arc volcanoes lie ~100 km above the Benioff Zone?


(f) Upward and Downward flow in mantle wedge

There is increasng interest in subduction-induced flow in the mantle wedge. At shallow levels (25-50km) the massive amounts of water entering the subduction zone may hydrate the mantle wedge to give serpentinite: this rock contains >12% water and is significantly less dense than normal mantle, and so can rise diapirically and "intrude" (solid state flow) the fore-arc regions, whether formed of arc volcanics or accreted sediment.

Further down, cooling of the wedge by the subduction zone itself may make it negatively buoyant (i.e. denser) and help drag the wedge down, promoting hornblende breakdown and fluid release. This will enhance induced convection effects. The amount of coupling between slab and mantle wedge would however be reduced by soft sediment at the interfact between the two.

Upward flow further back in the mantle wedge would compensate these effects, particularly if enhanced by low-density fluid and magmas. (to be continued . . . )



ANDERSON, R.N., DELONG, S.E. & SCHWARTZ, W.M. 1978. Thermal model for subduction with dehydration in the downgoing slab. Journal of Geology 86, 731-739.

ANDERSON, R.N., DELONG, S.E. & SCHWARTZ, W.M. 1980. Dehydration, asthenospheric convection and seismicity in subduction zones. Journal of Geology 88, 445-451.

CARLSON, R.L., HILDE, T.W.C. & UYEDA, S. 1983. The driving mechanism of plate tectonics: relation to age of the lithosphere at trenches. Geophysics Research Letters 10, 297-300.

DAVIES, J.H. & BICKLE, M.J. 1991. A physical model for the volume and composition of melt produced by hydrous fluxing above subduction zones. Philosophical Transactions of the Royal Society, London A335, 355-364.

DAVIES, J.H. & STEVENSON, D.J. 1992. Physical model of source region of subduction zone magmatism. Journal of Geophysical Research 97, 2037-2070.

DEFANT, M.J. & DRUMMOND, M.S. 1990. Derivation of some modern arc magmas by melting of young subducted lithosphere. Nature 347, 662-665.

DEWEY, J.F. 1981. Episodicity, sequence and style at convergent plate boundaries. In The Continental Crust and its Mineral Deposits. Geological Association of Canada, Special Paper 20, 553-572.

GARFUNKEL, Z., ANDERSON, C.A. & SCHUBERT, G. 1986. Mantle circulation and the lateral migration of subducted slabs. Journal of Geophysical Research 91, 7205-7223.

HARGRAVES, R.B. 1986. Faster spreading or greater ridge length in the Archean? Geology 14, 750-752.

KINCAID, C. & OLSON, P. 1987. An experimental study of subduction and slab migration. Journal of Geophysical Research 92, 13832-13840.

MOLNAR, P. & ATWATER, T. 1978. Interarc spreading and cordilleran tectonics as alternates related to the age of subducted ocean lithosphere. Earth and Planetary Science Letters 41, 330-340.

PEACOCK, S.M. 1987. Thermal effects of metamorphic fluids in subduction zones. Geology 15, 1057-1060.

PEACOCK, S.M. 1991. Numerical simulations of subduction zone pressure-temperature-time paths: constraints on fluid production and arc magmatism. Philosophical Transactions of the Royal Society, London A335, 341-353.

RINGWOOD, A.E. 1974. The petrological evolution of island arc systems. Journal of the Geological Society, London 130, 183-204.

STERN, R.J. & BLOOMER, S.H. 1992. Subduction-zone infancy - Examples from the Eocene Izu-Bonin-Mariana and Jurassic California arcs. Geological Society of America Bulletin 104, 1621-1636.

SUDO, A. & TATSUMI, Y. 1990. Phlogopite and K-amphibole in the upper mantle: implications for magma genesis in subduction zones. Geophysics Research Letters 17, 29-32.

UYEDA, S. & KANAMORI, H. 1979. Back-arc opening and mode of subduction. Journal of Geophysical Research 84, 1049-1061.

WYLLIE, P.J. 1988. Magma genesis, plate tectonics and chemical differentiation of the earth. Reviews of Geophysics 26, 370-404.




The floors of the world's oceans are covered by sediment up to 1 km thick (age dependent) as a result of slow accumulation of calcareous and siliceous biogenic oozes capped by fine clays that have been carried in suspension to the middle of oceans. Additionally, nearer continents there may be much thicker accumulations of clastic sediments brought in by deltas and turbidity currents, and further re-distributed by strong bottom water currents. Sooner or later this sedimenty must finish up at a subduction zone. What happens to it? Does it get scraped-off, or does it get dragged down the subduction zone? If the latter, does it just disappear into the deep mantle, or does it get recycled into island arc magmas? The balance is shown as follows:


Effectively, subduction at active margins can be likened to a conveyor belt carrying a lot of loose rubbish moving against a buttress: some material is going to get scraped-off:

There are many variables in the whole process. So it is important to look at a number at different tectonic situations.

(1) Primitive Island Arcs: no sediment accretion


At intraoceanic island arcs, such as the Marianas, there is no sediment supply from the continent (this is trapped by the back-arc basin), and the arc itself produces only a minor amount of volcanic ash (the eruptions are basaltic and not violent). Most of the sediment arriving at the subduction zone is abyssal ooze and clay carried on the subducting plate (on old ocean crust, at least 0.5km thick). It used to be thought that this abyssal sediment was scraped off to form an accretionary wedge in the fore-arc. However, dredging and drilling in the Mariana forearc and trench has shown that there is little on no sediment in the Mariana trench. Yet during the 40 my since the arc system has been in existence, up to 40 km3 of sedimen / km length of arc should have been scraped off the subducting plate (which is subducting at 10 cm/year).

The sediment must be subducted - but how? The answer seems to be that, as the subducting plate bends over to become vertical, the flexure causes horsts and graben to develop. Sediments are scraped off from the horsts into the graben and thus encased as the ocean lithosphere deforms (for this reason it was thought this would be a good place to dispose of nuclear waste!) In fact the ocean crust acts as a gigantic rasp on the arc too - the forearc is gradually, but slowly, eroded:

However geochemical studies have shown that very little of the sediment is actually incorporated into the arc volcanics, so most of it must be cycled into the deeper mantle. Presumably, as the slab at the Marianas is avalanching into the lower mantle, the sediments may be taken down also.


(2) Northern Chile: no sediment to subduct


Here the sediment supply is also very limited because of the arid climate. Many of the rivers from the high Andes never quite make it to the ocean, and in any case there few floods (which produce the turbidity currents that carry the sediment out into the ocean proper). Also, major faults parallel to the coast tend to obstruct the rivers, forming saline lakes (were common in N. Chile).

So the situation is similar to that in the Marianas, although the dip of the subducting slab is not so great. Some geologists have suggested that the rasping action of the subducting slab has actually eroded back the continental margin of N. Chile and Peru. Is this why the locus of volcanic activity continually moves eastwards with time in the N. Andes? And why Palaeozoic batholiths are exposed right at the coast, close to the trench? (Although difficult to prove it was there when it has gone!).

Where sediment supply is a little higher, trench gets partly filled with sediment. Some of this sediment may get scraped off. But drilling in the Middle America Trench suggests that the abyssal ocean floor sediments are still subducted (soft oozes act as a lubricant)

(3) S. Chile and Alaska: high sediment input

Here the climate is temperate and wet. Abundant rivers, some deriving from glaciers. Floods common. High rate of sediment supply to the ocean. Sediment supply was even higher during the Pleistocene (and there has not been time yet to subduct them).

Result is that large amount of sediment is carried into the trench. Trench quickly gets filled, and sediment then carried out onto subducting plate. As this continues the weight of sediment actually depresses the plate as it approaches the trench so that angle of dip is smaller (dip increases under the continental margin proper).

With a shallower dip, no horsts & graben form, and sediment is scraped off. This can readily be seen from reflection profiles. Layering of sediments disappears as continent is approached. Low angle thrusts appear. Younger sediments are progressively underplated. If sedimentation rates are high (as they are in high northern/southern latitudes) this can give rise to lateral growth of continents. The process is called subduction-accretion and the structures are called Accretionary Prisms. The general features are shown below:

(4) Characteristics of Accretionary Wedges/Prisms

Lateral continental growth by subduction-accretion is dependent on (a) the supply of material from the ocean, and (b) the sediment supply from the continent. These two might vary over a large range.


(a) Material accreted from oceans

The ocean floor is not smooth. Study of the Pacific map shows that the pre-Tertiary ocean floor is considerably rougher than that generated in the Tertiary. There are more oceanic plateaus, aseismic ridges, ocean island chains and arcs – in large part this results from the spate of mantle plumes which punched through the Pacific ocean plate in the late Cretaceous (120 - 80 Ma).. Many of these upstanding structures are capped by carbonate banks, because they stayed above the carbonate compensation depth (CCD) much longer than normal ocean floor.

Ocean floor that is rough and upstanding is more likely to be scraped off when it reaches subduction zones at active margins. So this sceaped-off material will be a mixture of mafic rocks (metamorphosed to amphibolite) associated with thick limestone (marble) sequences, as well as sileceous and carbonate oozes (= "cypoline schists") and lithified cherts. Large oceanic structures such as plateaus and arcs may "choke" the subduction zone, causing back-stepping of the subduction zone, the arcs being left as an ophiolite (e.g. the classic Troodos complex on Cyprus). However, normal ocean floor, which is smooth and cold, may not be scraped off at all (it is th is that converts to eclogite to provide the slab-pull force), so the soft carbonate-siliceous oozes and cloay may not be scraped off quite so readily.


(b) Material supplied from the continents

This is largely material supplied by river systems feeding active continental margins. Of course at the present day there are not many rivers feeding active continental margins -- they are mostly still feeding the passive margins of the Atlantic, the Indian ocean and around Antarctica/Australia.

It is important to note that in the Upper Palaeozoic and early Mesozoic, the southern continents formed part of Gondwanaland - a very large continental landmass. Moreover much of Gondwanaland was rimmed by active margins. The margin had low relief (the present high Andes is not typical, and results from Miocene deformation and uplift). So it is probable that very large rivers were dumping sediment onto the subducting plate, and the sediment was then accreted back on to the continental margin . . . now exhumed and exposed, particularly in southern Chile, where they are of late Palaeozoic age (before the Andean magmatic cycle), and South Island, New Zealand. But they can also be seen in Alaska, and of course occur in older mountain belts (commonly termed Flysch).

Compared with the partly-lithified material scraped off from the oceanic plate, the material coming from the continent is unlithified clastic sediment. The two get tectonically intermixed and intensely deformed (the subduction interface allows thousands of km of relative movement in just a few tens of Ma ™ far more than with continental collision), so most rocks from this environment have strong penetrative foliations and linear fabrics (see New Harbour Group on Anglesey) and finish up as teconic melanges -- lenses of oceanic rocks in deformed soft sediment.

As soft wet sediment (greywacke-shale) is continually underplated beneath the accretionary wedge, it heats up slowly. Water is progressively driven off. Hot water dissolves silica from sandy beds, and deposits it at higher levels as abundant cross-cutting quartz veins. However, because underplating is continuous process, sediments and quartz veins become progressively and very strongly deformed. Can be almost mylonite-like fabric. No bedding remains. Cross-cutting quartz veins are stretched out to become sub-parallel to foliation. Very characteristic rock type. Many tens or even hundreds of km of 'new' crust can accrete laterally onto continental margins in this way.

Erosion of upper part of accretionary wedge may occur, and younger sediments deposited on top in fore-arc basins. These may also become deformed, but less so (could the South Stack Series on Anglesey may represent such fore-arc basin rocks?).


DAVIES, J.H. & von BLANCKENBURG, F. 1995. Slab breakoff: A model of lithosphere detachment and its test in the magmatism and deformation of collisional orogens. Earth and Planetary Science Letters 129, 85-102.

von HUENE, R. & SCHOLL, D.W. 1993. The return of sialic material to the mantle indicated by terrigenous material subducted at convergent margins. Tectonophysics 219, 163-175.



"Terrane" concepts are now quite widely used in interpreting geological relationships in many parts of the world, and in rocks of many ages. Basically plate tectonics can move segments of continental crust or oceanic crust (e.g. ocean plateaus) many thousands of km in just a few tens of m.y., and as plates can change their direction of motion (c.f. kinik in Hawaiian chain), this can lead to the juxtaposition of segments of crust that have a completely different geological histories. So it is not just collision of major continents (e.g. India and Asia to form Himalayas) but also on a much smaller scale. In particular, major transform faults can transport differnt crustal segments laterally for many 1000s of km (e.g. San Andreas Fault). Of course terranes are usually fault- or thrust-bounded.


Terrane Terminology (Jargon)


"A fault-bounded package of strata that has a geological history distinct from the adjoining geologic units"


Howell (1989) divided terranes as follows:

Stratigraphic (1) representing fragments of continents

(2) fragments ofcontinental margin

(3) fragments of volcanic arc

(4) fragments of ocean basins




However, a genetic terminology is also prevalent:

Exotic, Suspect, Displaced or Accreted terranes: this implies that the terrane has

been transported some distance to its current position.

Pericratonic: Contains cratonal detritus and formed on attenuated continental crust.


Terranes are sometimes described in terms of tectonic assemblages, which are rock-stratigraphic units formed in actualistic tectonic settings, such as island arcs or ocean floors. A terrane may consist of one or more tectonic assemblages


Domain: A volume of rock, bounded by compositionalor structural discontinuities, within which there is structural homogeneity; these may contain minor stratigraphic distinctions as well andcan be viewed as subterranes.


Superterranes: A composite terrane, consisting of two or more compound terranes, that were amalgamated prior to subsequent orogenesis.




CONEY, P., JONES, D.L. & MONGER, J.W. 1980. Cordilleran suspect terranes. Nature 288, 329-333.

BEN-AVRAHAM, Z., NUR, A., JONES, D. & COX, A. 1981. Continental accretion: from oceanic plateaus to allochthonous terranes. Science 213, 47-54.

HOWELL, D.G. 1989. Tectonics of Suspect Terranes. Chapman & Hall, NewYork, 232pp.

BEBOUT, G.E. & BARTON, M.D. 1989. Fluid flow and metasomatism in a subduction zone hydrothermal system: Catalina Schist terrane, California. Geology 17, 976-980.


References On Alaskan Terranes

VROLIJK, P., MYERS, G. & MOORE, J.C. 1987. Warm fluid migration along tectonic melanges in the Kodiak accretionary complex, Alaska. Journal of Geophysical Research 93, 10313-10324.

BARKER, F., JONES, D.L., BUDAHN, J.R. & CONEY, P.J. 1988. Oceanic plateau-seamount origin of basaltic rocks, Angayuchan Terrane, Central Alaska. Journal of Geology 96, 368-374.


References On Caledonian Terranes

DEWEY, J.F. & SHACKLETON, R.M. 1984. A model for the evolution of the Grampian tract in the Caledonides and Appalachians. Nature 312, 115-121

MURPHY, F.C. & HUTTON, D.H.W. 1986. Is the Southern Uplands of Scotland really an accretionary prism? Geology 14, 54-57.

HUTTON, D.H.W. 1987, Strike-slip terranes and a model for the evolution of the British and Irish Caledonides. Geological Magazine 124, 405-425.

BENTLEY, M.R., MALTMAN, A.J. & FITCHES, W.R. 1988. Colonsay and Islay: a suspect terrane within the Scottish Caledonides. Geology 16, 26-28.

HAUGHTON, P.D.W. 1988. A cryptic Caledonian flysch terrane in Scotland. Journal of the Geological Society, London 145, 685-703.

MARCANTONIO, F., DICKIN, A.P., McNUTT, R.H., & HEAMAN, L.M. 1988. A 1800 million year old Proterozoic gneiss terrane in Islay with implications for the crustal structure evolution of Britain. Nature 335, 62-64.

SOPER, N.J., GIBBONS, W. & McKERROW, W.S. 1989. Displaced terranes in Britain and Ireland. Journal of the Geological Society, London 146, 365-367.

THIRLWALL, M.F. 1989. Movement on proposed terrane boundaries in northern Britain: constraints from Ordovician-Devonian igneous rocks. Journal of the Geological Society, London 146, 373-376.

BLUCK, B.J. & DEMPSTER, T.J. 1991. Exotic metamorphic terranes in the Caledonides: Tectonic history of the Dalradian block, Scotland. Geology 19, 1133-1136.

RYAN, P.D. & DEWEY, J.F. 1991. A geological and tectonic cross-section of the Caledonides of western Ireland. Journal of the Geological Society, London 148, 173-180.

MURPHY, F.C., ANDERSON, T.B., DALY, J.S. & 16 others, 1991 An appraisal of Caledonian suspect terrains in Ireland. Irish Journal of Earth Sciences 11, 11-41.

SOPER, N.J., ENGLAND, R.W., SNYDER, D.B. & RYAN, P.D. 1992. The Iapetus suture zone in England, Scotland and eastern Ireland: a reconciliation of geological and deep seismic data. Journal of the Geological Society, London 149, 697-700.

BROWN, C. & WHELAN, J.P. 1995. Terrane boundaries in Ireland inferred from the Irish Magnetotelluric Profile and other geophysical data. Journal of the Geological Society, London 152, 523-534.


References On Appalachian Terranes

WILLIAMS, H. & HATCHER, R.D. 1982. Suspect terranes and accretionary history of the Appalachian region. Geology 10, 530-536.


References On Andean Terranes

ASPDEN, J.A. & McCOURT, W.J. 1986. Mesozoic oceanic terrane in the central Andes of Colombia. Geology 14, 415-418.


Baltic Shield Proterozoic Terranes

PARK, A.F. 1991. Continental growth by accretion: a tectonostratigraphic terrane analysis of the evolution of the western and central Baltic Shield, 2.50 to 1.75 Ga. Bulletin of the Geological Society of America 103, 522-537.


References On Archaean Terranes

(to be continued)




The important question arises: how far back in time did plate tectonic processes operate? Is it just a modern phenomena? Or can we interpret global geology in terms of plate tectonic concepts right back to the early Archaean?

The common markers of plate tectonics are:

(1) Ophiolite complexes, that textbooks usually argue represent obducted fragments of ocean floor (though most are probably back-arc basin or young forearc). But these appear to be absent before the latest Proterozoic (ca. 700 Ma)

(2) Blueschists (with the Na-amphibole glaucophane) are indicative of a high-P, low-T hydrous environment that is only achieved in a subduction zone. Again, these appear to be absent before the latest Proterozoic (ca. 700 Ma).

(3) Island Arcs (like the Marianas), formed where oceanic plates subduct beneath one another. They are common at the present day, but are much less easy to recognise back in time, particularly in the Precambrian.


So for a majority of geological time (4.0 b.y. to 0.6 b.y.) some of the common markers of plate teconics are absent. Was the mantle regime too hot to allow preservation of blueschists? Because arcs and back-arcs require a subduction-flip or major change of plate direction to initiate them, perhaps these flips or plate direction changes did not occur, and there was a more regular, small scale pattern of mantle convection in the Precambrian? It has been suggested quite a long time ago (e.g. Fyfe, 1978) that "hot-spot teconics" may have been more important in the past than "plate tectonics". However this does not mean than one excludes the other, because hotspots (e.g. Iceland, Hawaii) occur at the present day, and it may just mean than one was dominant over the other.Plume


That low-density solid bodies can rise up through other solid material as large diapirs has been long known from observing salt domes. Ramberg did laboratory experiments to simulate this and showed that these would have one of the following forms (of which the mushroom shape is generally regarded as the most probable):


If the low density is due to excess heat, as in the mantle, they are usually referred to as plumes. Note that as plumes rise, a "spout" or "balloon" will eventually turn into a "mushroom" as it entrains the surrounding mantle material:

An important feature is that plumes must be initiated at some thermo-mechanical boundary layer within the Earth, for instance at the core – lower-mantle boundary [2900km depth] or at the shallower 700km discontinuity:


It is at these boundaries where either compositional differences (core-mantle) or phase differences (700km discontinuity) permit denser but hotter material to exist beneath lighter cooler rock, so that any instabilities in the convective patterns across the boundary layer may spawn a hot plume. At initiation, such plumes can be several hundred degrees hotter than the surrounding mantle, and will rise, the excess heat then causing a lowering of the viscosity of the surrounding mantle (or even melting it slightly) so allowing it to be entrained into the mushroom head of the plume. in this way the plume head gradually enlarges itself and becomes cooler (relatively), whereas the tail of the plume is narrower, but hotter, the hot material continually rising up into the head.

These features have been described by Campbell, Griffiths, Hill & Co., and in simplified form can be summarised as follows:

These plume heads can become quite large as the ascend to the surface, and diameters of 500 - 1000 km have been suggested.

The question arises: how often are these plumes released? And which discontinuity do they come from? Larson (1991a) showed that there was correlation between the rate of ocean crust production and the magnetic reversal time scale:

This shows that the mafic crust production rate was at a maximum during the Cretaceous magnetic quiet period, between 125 – 80Ma. This leads to 2 important conclusions. First, correlation can only occur if in fact the plume originated at the core-mantle boundary (D" layer), because it is convection in the Earth's core that is responsible for the magnetic field, and hence release of a major plume may have upset core convection (see Larson & Olsen, 1991 for details). Second, this excess crust production could be accounted for almost entirely by that represented by ocean plateaus. These ocean plateaus are regions where the ocean crust is anomalously thick. This reflects the fact that if the energetic mantle plumes rise up beneath the mid-ocean ridges, then all the excess heat is converted into basic magma, and so the ocean crust may be over 15 km thick, compared with ca. 6 km of normal ocean crust. This has two further effects. One is that the plateau may become land, like Iceland, instead of being submerged 2 to 6 km beneath the oceans. A consequence is that the ocean water displaced (sea level rise) then floods the low-lying continental margins, so that chalk is deposited in abundance in the shallow warm seas (see Larson, 1991b).

The main body of ocean plateaus are in the Western Pacific (Ontong Java; Manihiki Rise; Hess Rise, etc.) are all about 120 my old, which corresponds with the start of the Cretaceous magnetic quiet period. Note that these are only one half of the plateaus – the other half (that formed over the other side of the mid-ocean ridge) may have been subducted beneath South America. A second phase of plume activity occurred at ca. 87 Ma, and corresponds with the end of the magnetic quiet period. Ths biggest plume here came through at the Galapagos hotspot in the eastern Pacific, and bits of it are found scattered around the Caribbean (see below). Note that the Iceland plume started at 60 Ma. Now this may indicate that large plumes are perhaps more frequent than had been thought. Could they occur far back in time? Is their release periodic or cyclic, or infrequent? Do they get more common as we go back in time? How could we recognise them? They do represent a big pulse of energy transferred from the outer core to the Earth's surface, so they should have associated features. For instance, the large phase of diamond-bearing kimberlite pipes occurs around 120Ma, and the phase of global warming occurred then (see Larson, 1991b).


Large Igneous Provinces

What happens when plumes – after ascending almost 2900 kms vertically – approach the surface? Whether they can break through will depend on the thickness of the mechanical boundary layer (MBL), or lithosphere. The MBL is thin at ocean ridges, so plumes can easily break through, and the energy converted to extensive melting and formation of an ocean plateau. Note that the ridge and plume-hostspot cannot stay together for long because ridges are always moving and hotspots are fixed. So eventually plateaus must end up as ocean island chains (e.g. Hawaiian chain), as the hotspot keeps burning through the plate.


However, if the plume rises beneath thick lithosphere, then it cannot easily break through, and must spread out beneath the lithosphere in the manner shown (after ADS):


It may "incubate" here for some time, perhaps causing extensive melting of the base of the lithosphere, and interactions between plume and lithosphere. If there is enough extension during this time, then the magmas may be poured out as continental flood basalts (examples: Deccan - India, Karoo - S. Africa; Ferrar - Antarctica; and many more. Most of these large igneous provinces erupt vast amounts of basalt, apparently in quite a short time (ca. 1 or 2 m.y.), so it may only be the more vigorous plumes that get through in this way. The diagram below shows how lithsophere is domed as the plume spreads out to a diameter of over 1000 km, and then as the plume melts away at the base of the lithosphere, the magmas get through:

In other cases, the plume may search out weak zones or "thin spots". A good example of this may be the British Tertiary Volcanic Province, which was initiated 60Ma ago when the Iceland plume started, but was over within a few m.y. In this case the "thin spot" extended all the way from Iceland, through NW Scotland, to Lundy in the Bristol Channel. The diagram below shows how uprising plumes may be channeled along thinspots and eventually break through:


Note that the British Tertiary Province is characterised by some quite high-Mg magma compositions (perhaps as high as 22%MgO), particularly on Rhum and on Skye. Such high-Mg composition usually mean high degrees of mantle melting, which in turn requires a lot of thermal energy that can really only be attained by rapid uprise of vigorous plumes from depth. These high temperature magmas have the excess thermal energy that can initiate melting of upper crustal rocks, or lower crustal rocks, or lithosphere. So maybe that is why quite a lot of granite magma (e.g. Red Hills, Skye; Northern Granite, Arran) appears within a m.y. or two of the emplacement of the high-temperature mafic magmas.

Now not all hot-spots have high temperature Mg-rich magmas. Some have very different trace element compositions from Iceland or Hawaiin lavas. One possibility for this is that deep mantle plumes don't always rise straight up to the surface at a ridge system. If instead they rise up beneath a down-going convection cell in the upper mantle, then they may be directed sideways and interact with material stored at the 700km discontinuity (the Ringwood "megalith" layer in Lecture 1), and much of their energy may be consumed in mobilising this material, which would then rise up to form a less energetic plume under an ocean island rather than a plateau:


Komatiites and Precambrian Greenstone Belts

Komatiites are very high temperature lavas with MgO contents up to 33%. Uprise from great depths is the easiest way to achieve such high liquid temperatures. Now it used to be thought that such high temperatures (>1600°C) could only be achieved in the mantle of the early Earth (Archaean). However, the discovery of komatiitic lavas on the small island of Gorgona, off Colombia, at Romeral in the Coca belt of Colombia, and equivalent picrites on Curaçao in the Caribbean (see below), all in a Cretaceous lava pile, shows that it is not Precambrian mantle that was especially hot, but that the explanation lies in vigorous ascending deep mantle plumes. So instead it may be that Precambrian greenstone belts actually represent the uprise of Precambrian plumes.

Now the Colombian/Caribbean occurrences actually illustrate another aspect of plumes very well. First there is a lot of mafic material scattered around the Caribbean (shown in black):

All of this mafic material seems to have an age of 87Ma and formed at the Galapagos hotspot at this time. Now whereas most ocean crust of this age has subducted beneath the Andes, it is clear that this plateau was too hot to subduct, and instead was obducted onto the continental margin of Colombia, etc. Could this mean that Precambrian greenstone belts are just obducted ocean plateaus?

For those interested, the most recent (1992) mantle phase diagram illustrating the P-T conditions for uprising deep mantle plumes is shown below:


The blue dashed lines show the paths of uprising diapirs. Once they cross the solidus curve, melting will occur. As melting mantle minerals (with a high latent heat of fusion) absorbs energy, the melting curves for 30% and 20% melting follows the red line paths, to lower temperatures. To some extent the chemistry of the komatiitic liquids is governed by the nature of the residual mantle minerals (olivine or majorite) - hence the two paths shown. In any case it can be seen that even higher degrees of melting than 30% would generate even higher temperature liquids by the time they reached the surface.


Precambrian Plate Tectonics?

Because komatiites are common in Archaean (>2.5 Ga) or early Proterozoic (ca. 2.1 Ga) greenstone belts, it could be argued either that the mantle was hotter in the Precambrian (Burke & Kidd, 1978), or that deep mantle plumes (forming plateaus that are now greenstone belts) were more important in Precambrian tectonics than was plate tectonics (Storey et al. 1991). Arndt (1983) wanted there to be a hot thin komatiitic ocean crust in the Archaean (which would not necessarily have produced much slab-pull). Hargreaves (1986) argued that more heat would mean more ridges in the Archaean, and therefore smaller plates; this also would not mean so much slab-pull. See also Nisbet & Fowler (1983) and Fyfe (1978). So there is a general consensus that without the slab-pull that is so important in modern plate tectonics, there may not have been plate tectonics as such (although there may have been subduction).


Another important argument is that if plateaus are hot and difficult to subduct to depths of 700km, they may just underplate the continents. In which case there is more opportunity for them to heat up later and melt, possibly giving rise to voluminous tonalitic granitoids that are common in the early Precambrian. It is possible that this is the explanation for the 70 km thick crustal keel under the Andes: was this just Pacific ocean plateau?


References (these cover many different aspects of plumes)


AITKEN, B.G. & ECHEVERRIA, L.M. 1984. Petrology and geochemistry of komatiites and tholeiites from Gorgana Island, Colombia. Contributions to Mineralogy and Petrology 86, 94-105.

ANDERSON, D.L., TANIMOTO, T. & ZHANG, Y-S. 1992. Plate tectonics and hotspots: the third dimension. Science 256, 1645-1651.

ARNDT, N.T. 1977. Ultrabasic magma and high-degree melting of the mantle. Contributions to Mineralogy and Petrology 64, 205-221.

ARNDT, N.T. 1983. Role of a thin, komatiite-rich oceanic crust in the Archean plate-tectonic process. Geology 11, 372-375.

BERCOVICI, D. & MAHONEY, J. 1994. Double flood basalts and plume head separation at the 660-kilometer discontinuity. Science 266, 1367-1369.

BURKE, K. & KIDD, W.S.F. 1978. Were Archean continental geothermal gradients much steeper than those of today? Nature 272, 240-241.

CAMPBELL, I.H. & GRIFFITHS, R.W. 1990. Implications of mantle plume structure for the evolution of flood basalts. Earth and Planetary Science Letters 99, 79-93.

CAMPBELL, I.H., GRIFFITHS, R.W. & HILL, R.I. 1989. Melting in an Archaean mantle plume: heads it's basalts, tails it's komatiites. Nature 339, 697-699.

COFFIN, M.F. & ELDHOLM, O. 1993. Scratching the surface: Estimating dimensions of large igneous provinces. Geology 21, 515-518.

DAVIES, G.F. 1992. On the emergence of plate tectonics. Geology 20, 963-966.

DAVIES, G.F. & RICHARDS, M.A. 1992. Mantle convection. Journal of Geology 100, 151-206.

DUNCAN, R.A. & RICHARDS, M.A. 1991. Hotspots, mantle plumes, flood basalts and true polar wander. Reviews of Geophysics 29, 31-50.

FYFE, W.S. 1978. Evolution of the Earth's crust: modern plate tectonics to ancient hot-spot tectonics? Chemical Geology 23, 89-114.

GRIFFITHS, R.W. & CAMPBELL, I.H. 1990. Stirring and structure in mantle starting plumes. Earth and Planetary Science Letters 99, 66-78.

GRIFFITHS, R.W. & CAMPBELL, I.H. 1991. Interaction of mantle plume heads with the Earth's surface, and onset of small-scale convection. Journal of Geophysical Research 96, 18295-18310.

HARGRAVES, R.B. 1986. Faster spreading or greater ridge length in the Archean? Geology 14, 750-752.

HILL, R.I. 1991. Starting plumes and continental break-up. Earth and Planetary Science Letters 104, 398-416

HILL, R.I., CAMPBELL, I.H., DAVIES, G.F. & GRIFFITHS, R.W. 1992. Mantle plumes and continental tectonics. Science 256, 186-193.

KENT, R.W. 1991. Lithospheric uplift in eastern Gondwana: Evidence for a long-lived mantle plume system? Geology 19, 19-23.

KERR, A.C., SAUNDERS, A.D., TARNEY, J., BERRY, N.H & HARDS, V.L. 1995. Depleted mantle-plume geochemical signatures: no paradox for plume theories. Geology 23, 843-846.

KUSKY, T.M. & KIDD, W.S.F. 1990. Remnants of an Archean oceanic plateau, Belingwe greenstone belt, Zimbabwe. Geology 43, 43-46.

LARSON, R.L. 1991. Latest pulse of Earth: Evidence for a mid-Cretaceous superplume. Geology 19, 547-550.

LARSON, R.L. 1991. Geological consequences of superplumes. Geology 19, 963-966.

LARSON, R.L. & OLSON, P. 1991. Mantle plumes control magnetic reversal frequency. Earth and Planetary Science Letters 107, 437-447.

LARSON, R.L. & KINCAID, C. 1996. Onset of mid-Cretaceous volcanism by elevation of the 670 km thermal boundary layer. Geology 24, 551-554.

NISBET, E.G. & FOWLER, C.M.R. 1983. Model for Archean plate tectonics. Geology 11, 376-379.

SAUNDERS, A.D., TARNEY, J., KERR, A.C. & KENT, R.W. 1996. The formation and fate of large oceanic igneous provinces. Lithos 37, 81-95.

STOREY, M., MAHONEY, J.J., KROENKE, L.W. & SAUNDERS, A.D. 1991. Are oceanic plateaus sites of komatiite formation? Geology 19, 376-379.

THOMPSON, R.N. & GIBSON, S.A. 1991. Sub-continental mantle plumes, hotspots, and pre-existing thinspots. Journal of the Geological Society, London 148, 973-977.

WHITE, R. & McKENZIE, D. 1989. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research 94, 7685-7729.

YAMAJI, A. 1992. Periodic hotspot distribution and small-scale convection in the upper mantle. Earth and Planetary Science Letters 109, 107-116.




下一篇:[转载]Grüneisen parameter


该博文允许注册用户评论 请点击登录 评论 (0 个评论)


Archiver|手机版|科学网 ( 京ICP备14006957 )

GMT+8, 2019-12-16 21:07

Powered by

Copyright © 2007- 中国科学报社