Australia: The Land Where Time Began

A biography of the Australian continent 

Accretionary Prisms

Accretionary prisms form on the inner wall of ocean trenches, though not all trenches have accretionary prisms. Seismic reflection profiles, and drilling at active subduction zones, have been used to deduce the internal structure and form of these features. Evidence has also been obtained by the study of ancient subduction complexes that are now found on the continents.

Places where the leading edge of the overriding plate scrapes flysch (trench fill turbidites), as well as a percentage of the pelagic sediment, from the top of subducting oceanic plates, the scraped off material being accreted to the leading edge of the overriding plate. Many of the features, structural, lithologic and hydrological, of a large, active accretionary prisms, with a thick sedimentary section, are present in the Nankai Trough, south of Japan (Moore et al., 2001, 2005). The plate boundary is defined by a gently dipping fault or shear zone, that is 20-30 m thick, beneath the prism, separates a deformed sedimentary wedge from a deeper section of subducted trench sediment that is little deformed, volcaniclastic rock and basaltic crust are beneath it (Fig. 9.20b, source 1). A 'dcollement', this boundary, develops in a weak layer of sediment of hemipelagic mud that is typically of low permeability, that is overlain by trench turbidites that are stronger and more permeable. Listric thrust ramps rising through the stratigraphic section to form imbricated arrays, comprise a fold and thrust belt above the dcollement. Internally folded and cleaved wedge-shaped lenses are defined by these faults. The dcollement, at the base of the imbricate series, slopes downwards towards the volcanic arc, becoming better developed as it approaches the arc. It extends a short distance seaward from the deformation front marked by the first small protothrusts and folds, situated inward of the trench, away from the arc. Farther towards the sea, the incipient or proto-dcollement zone, where the incoming sedimentary section is only weakly deformed, is the stratigraphic horizon hosting the dcollement (Kearey et al., Source 1).

The youngest faults in an accretionary prism are suggested by seismic reflection data and deformed sediment ages to occur at the deformation front, generally becoming older away from the trench (Moore et al., 2001, 2005) (Fig. 9.20b, Source 1). Old thrust wedges move upwards gradually, and rotate toward the arc as new wedges are added to the toe of the prism, as shortening occurs. As a result of this process of 'frontal accretion' older thrusts become more steeply dipping over time, and it causes the prism to grow laterally. The most intense deformation is required at the oceanward base of the sedimentary pile for lateral growth to occur. During this rotation some older thrusts may remain active, and some new thrusts may form that cut across older imbricate thrusts. These cross-cutting faults are called out-of-sequence thrusts (Fig. 9.20b, Source 1), as they do not conform to the common arcward progression of faulting. As well as frontal accretion, the downward movement of some incoming material past the deformation front may result in it being underplated, or transferred to, the prism base by thrust faulting above the dcollement. This material that is underplated may become buried deeply and undergo high pressure metamorphism, which does not occur to sediments scraped off at the toe. The wedge is thickened by tectonic underplating and internal shortening, increasing the slope of the upper surface (Konstantinovs-kaia & Malavielle, 2005).

A relatively abrupt slope decrease, the trench slope break, defines the top of an accretionary prism. A forearc basin may develop between the island arc and this break, which is filled by sediment eroded from the volcanic arc and its substrate. Tranquil sedimentation occurs in this basin where older thrust slices in the wedge are covered by flat-lying units. Small pockets of sediment accumulates on the tops of old thrust slices on the trench basin slope, to the seaward of the forearc basin (Fig. 9.20b, source 1). Lateral growth rates are estimated from the ages of the old slices, as well as their distance from the prism toe. An example is drilling in the Nankai Prism, at sites 1175 and 1176, has shown that thrust slices that may be 1-2 Ma are overlain unconformably by trench slope sands (Moore et al., 2001; Underwood et al., 2003). Lateral growth rates of up to 40 km over the last 1-2 Ma are indicated by the distance of these thrust slices from the deformation front, provided the seaward growth is steady state. Compared with this, off the Mexican coast, the width of the Middle American accretionary prism has grown ~23 km during the past 10 million years (Moore et al., 1982) and the eastern Aleutian accretionary prism has grown by 20 km in 3 million years (von Huene et al., 1998).

Slump deposits and debris flows, in which material can be carried as far as the trench, often result from trench slope erosion and other landward material. Once in the trench, it is offscraped and recycled back into the wedge. The accumulation of large amounts of sediment originating in exposed rock on Shikoku Island at the time is indicated by the presence of thrust slices composed of turbidites from the Miocene, as found in the Nankai prism, at site 1178 (Moore et al., 2005). During transport, large blocks of slumped material, 100-1000 m long, 'olistostromes', remain semi-coherent, providing much of the material that allow accretionary prisms to increase in width (Silver, 2000). Long-term circulation of the material of the wedge results from erosion, deformation and sedimentary recycling (Platt, 1986). Material that has been scraped off moves down toward the prism base then back towards the surface. A general increase in the metamorphic grade of the rock from the trench to the arc results from this pattern, the oldest, high grade rocks are structurally higher and uplifted compared to the younger deposits. A chaotic mix of igneous, sedimentary and metamorphic rock types, a mlange, may be created by this process. Blueschist or eclogite facies may be recorded by some of the oldest rock fragments in the the mlange metamorphism, which indicates it was buried to a depth of at least 30 km (section 9.9, Source 1).

The overall profile shape of an accretionary wedge is a tapered wedge-shape, the upper surface sloping in the opposite direction to that of the underlying dcollement. It has been shown that the tapered wedge shape is required if the entire wedge moves together and the Mohr-Coulomb fracture criterion is followed by the behaviour of the system (Davies et al., 1983; Dahlen, 1990). The interaction between sliding resistance on the dcollement and the rock strength in the thrust wedge determines the surface slope (α). Pore fluid pressure (λ), the dcollement dip (β) and the weight of overlying rock all influence both these latter factors. The wedge is thickened by tectonic shortening and underplating, leading to steepening of the surface slope. Various mechanical adjustments take place if the slope becomes oversteep to return the slope angle to a steady state. Lengthening of the dcollement and/or normal faulting may be involved in these mechanisms, resulting from the same forces driving the gravitational collapse of large topographic uplifts. The mechanical behaviour of the wedge is especially sensitive to the redistribution of mass by surface erosion and deposition (Konstantinovs-kaia & Malavielle, 2005; Stolar et al., 2006), leading to changes of topographic gradients, and affect the thermal evolution of the crust at large scales (section 8.6.3, source 1).

The importance of fluid flow and fluid pore pressure changes in accretionary prisms have been shown by unequivocal evidence from drilling into active prisms. Evidence from measurements of porosity, density, resistivity, as well as other physical characteristics, suggest that the accreted sediments do not not dewater before burial because their descent is too rapid for dewatering to occur (Silver, 2000; Saffer, 2003; Moore et al., 2005). 

Effective stress is reduced by elevated pore pressures and lower rock shear strength, as well as the allowing sliding on the dcollement, are the result of the typical low permeabilities of marine sediments, as well as the process in which sediments are not dewatered.  The dcollement may also be allowed to propagate laterally beneath the wedge, by the former flow paths collapsing and episodic fluid flow (Ujiie et al., 2003). The generally small taper angles found in most accretionary wedges, that can occur only if the material in it are very weak and there are very low shear stresses on the dcollement, are explained by this process (Davies et al., 1983; Saffer & Bekins, 2002). Many other phenomena associated with prisms, such as mud volcanoes and diapirs, are also explained by high pore pressure (Westbrook et al., 1984), as well as unique chemical and biological environments that develop at the leading edge of the prism (Schoonmaker, 1986; Ritger et al., 1987).

There are also mechanisms that tend to decrease fluid pore pressure in the wedge, such as the flowing of fluids along narrow channels of high permeability exiting through conduits, that are vertical or lateral, to the seafloor and the dcollement (Silver, 2000; Morris & Villinger, 2006). Some of the conduits coincide with thrust faults above the zone of the dcollement. Fluid is allowed to escape by the high fracture permeability of these thrust faults (Gulick et al., 2004; Tsuji et al., 2006). The dcollement zone is implied to have a lower fluid pressure than the surrounding material by the way in which fluid escapes, which appears to contradict evidence of high fluid pore pressures in the dcollement zone. In some models there is variation in the wedge, spatially and temporally, of the dcollement zone fluid pressure, that reconciles the apparent conflict. Factors, such as the rate of convergence and the stratigraphy, lithology, mineralogy, as well as the hydrologic properties, of the incoming sediments, influences to a large degree the nature of these variations and the deforming wedge evolution (Saffer, 2003).

Mechanical and numerical models have been used to determine how sensitive accretionary prisms are to fluid flow and pore fluid pressure fluctuations. It has been concluded from the combination of critical taper theory with a groundwater flow model that poorly drained systems are sustained by low permeability, low pore pressure and rapid convergence, resulting in shallow tapers, well-drained systems and steeper taper geometry resulting from conditions of high permeability, low pore fluid pressure and slow convergence (Saffer & Bekins, 2002). It has also been shown by the same authors that some of the most important factors governing wedge pore fluid pressure are the sediment composition incorporated into the wedge and stratigraphic thickness (Saffer & Bekins, 2006). Large prisms capable of sustaining higher pore fluid pressure and low taper angles that are stable result from thick sedimentary sections. It is also suggested by the results that prisms, such as the northern Antilles (Barbados) and eastern Nankai (Ashizuri), where the composition is mostly fine-grained sediment with low permeability, have thin taper angles. Prisms such as Cascadia, Chile and Mexico, with turbidites of high permeability, have steep taper angles. Any long-strike sediment lithology or thickness variation is implied by the sensitivity of the physical properties of accreted and subducted sediment to strongly influence the geometry and mechanical behaviour of accretionary prisms. The accretionary complex is forced to adjust to a new dynamic balance by any sediment thickness or composition variation (Kearey et al., Source 1).

See Source 1 for more detailed information on all aspects of plate tectonics

Sources & Further reading

  1. Kearey, Philip, Klepeis, Keith A. & Vine, Frederick J., 2009, Global Tectonics, 3rd Edition, Wiley-Blackwell.
Author: M. H. Monroe
Last Updated  17/03/2011



Journey Back Through Time
Experience Australia
Aboriginal Australia
National Parks
Photo Galleries
Site Map
                                                                                           Author: M.H.Monroe  Email:     Sources & Further reading