Why Are Island Arcs Arcs

Question 2.6: Why do subduction zones have arcuate structures?

One of the striking features of subduction zones is their arcuate structure in map view or planform. Subduction zones are made up of a sequence of arc structures with a clear planform curvature; this is the origin of the term "island arc." A good example is the Aleutian arc, shown in Figures 2.1 and 2.5. Just as accretionary margins are characterized by their orthogonal ridge-transform geometry, subduction zones are characterized by their arc configuration.

Frank (1968a) proposed a simple model for the curvature of island arcs based on a ping-pong ball analogy. If an indentation is made on a ping-pong ball, there is a simple analytical relation between the angle of dip and the radius of the indentation. Frank proposed that this relation also could be used to relate the dip angle of the subducted lithosphere to the planform radius of curvature of the island arc. The assumption was that the rigidity of the subducted lithosphere controlled the geometry of subduction in direct analogy to a ping-pong ball. Clearly this problem is related to the problem of the angle of dip considered in the previous section.

Several authors have tested Frank's hypothesis (DeFazio, 1974; Tovish and Schubert, 1978) and have found that it is a fair approximation in some cases and a poor approximation in other cases. It is generally accepted that the arcuate structure of island arcs can be attributed to the rigidity of the descending plate (Laravie, 1975), but the detailed mechanism remains controversial. Yamaoka et al. (1986) and Yamaoka and Fukao (1987) attribute the island arc cusps to lithospheric buckling. It is clear from seismic observations that the cusps represent tears in the descending lithosphere.

Any completely successful numerical model for mantle convection must reproduce the observed arcuate structure of subduction zones.

2.5.5 Subduction Zone Volcanism

Question 2.7: What is the mechanism for subduction zone volcanism?

Volcanism is also associated with subduction (Tatsumi and Eggins, 1995). A line of regularly spaced volcanoes closely parallels the trend of almost all the ocean trenches. These volcanoes may result in an island arc or they may occur within continental crust (Figure 2.21). The volcanoes generally lie above where the descending plate is 125 km deep, as illustrated in Figure 2.17. It is far from obvious why volcanism is associated with subduction. The descending lithosphere is cold compared with the surrounding mantle, and thus it acts as a heat sink rather than as a heat source. The downward flow of the descending slab is expected to entrain flow in the overlying mantle wedge. However, this flow will be primarily downward; thus, magma cannot be produced by pressure-release melting. One possible source of heat is frictional dissipation on the fault plane between the descending lithosphere and the overlying mantle (McKenzie and Sclater, 1968; Oxburgh and Turcotte, 1968; Turcotte and Oxburgh, 1968). However, there are several problems with generating island arc magmas by frictional heating. When rocks are cold, frictional stresses can be high and significant heating can occur. However, when the rocks become hot, the stresses are small, and it may be difficult to produce significant melting simply by frictional heating (Yuen et al., 1978). On the other hand, Kanamori et al. (1998) have used the unusual properties of the 1994 Bolivian earthquake, including a slow rupture velocity, a high stress drop (about 100 MPa), and a low ratio of radiated seismic energy to total strain energy, to infer that melting may have occurred on the fault plane during this earthquake. They suggested a minimum frictional stress of about 55 MPa and calculated a minimum amount of nonradiated seismic

Island Arc Basalts
Figure 2.21. Schematic of (a) oceanic lithosphere subducting beneath oceanic lithosphere and the creation of a volcanic island arc, and (b) oceanic lithosphere subducting beneath continental lithosphere and creation of a volcanic chain on the continent. After Tarbuck and Lutgens (1988).

energy equal to about 1018 J, sufficient to have melted a layer on the fault plane about 300 mm thick.

One proposed explanation for arc volcanism involves interactions between the descending slab and the induced flow in the overlying mantle wedge, leading to heating of the descending oceanic crust and melting (Marsh, 1979). Many thermal models of the subduction zone have been produced (e.g., Oxburgh and Turcotte, 1970; Toksoz et al., 1971; Turcotte and Schubert, 1973; Hsui and Toksoz, 1979; Hsui et al., 1983; Peacock et al., 1994; Ponko and Peacock, 1995; Iwamori, 1997; Kincaid and Sacks, 1997). All these models show that there is great difficulty in producing enough heat to generate the observed volcanism, since the subducted cold lithospheric slab is a very strong heat sink and depresses the isotherms above the slab.

Water released when hydrated minerals in the subducted oceanic crust are heated can contribute to melting by depressing the solidus temperature of the crustal rocks and adjacent mantle wedge rocks (Anderson et al., 1976; Ringwood, 1977a; Bird, 1978a). However, the bulk of the volcanic rocks at island arcs have near-basaltic compositions and erupt at temperatures very similar to eruption temperatures at accretional margins. Studies of the petrology of island arc magmas (Hawkesworth et al., 1994) indicate that they are primarily the result of the partial melting of fertile mantle rocks in the mantle wedge above the descending slab.

Nevertheless, there is geochemical evidence that the subducted oceanic crust does play an important role in island arc volcanism. Beryllium isotopic studies of volcanic rocks in subduction settings have revealed 10Be enrichments relative to mid-ocean ridge and ocean island basalts that are attributed to sediment subduction (Tera et al., 1986; Sigmarsson et al., 1990). One way to incorporate 10Be from subducted sediments into island arc magmas is through dehydration of the sediments and transport of beryllium with the liberated water (Tatsumi and Isoyama, 1988). Thus, direct melting of the subducted oceanic crust and lithosphere is not required to explain the 10Be excess in island arc volcanic rocks. Other evidence that the subducted oceanic crust is important in island arc magmatism is the location of the surface volcanic lines, which has a direct relationship to the geometry of subduction. In some cases two flaps of slab subduct at different angles, as in the Aleutians. For the shallower dipping slab, the volcanic line is further from the trench, keeping the depth to the slab beneath the volcanic line nearly constant (Kay et al., 1982).

The basic physical processes associated with subduction zone volcanism remain enigmatic, though it is apparent that the subducted oceanic crust triggers this volcanism. However, substantial melting of the subducted crust only occurs when young and relatively hot lithosphere is being subducted (Drummond and Defant, 1990; Kay et al., 1993). The bulk of the volcanism is directly associated with the melting of the mantle wedge similar to the melting beneath an accretional plate margin. A possible explanation for island arc volcanism has been given by Davies and Stevenson (1992). They suggest that "fluids" from the descending oceanic crust induce melting and create sufficient buoyancy in the partially melted mantle wedge rock to generate an ascending flow and further melting through pressure release. This process may be three dimensional with along-strike ascending diapirs associated with individual volcanic centers. Sisson and Bronto (1998) have analyzed the volatile content of primitive magmas from Galunggung volcano in the Indonesian arc and concluded that the magmas were derived from the pressure-release melting of hot mantle peridotite. There is no evidence that volatiles from the subducted oceanic crust were directly involved in the formation of these magmas. We conclude that many aspects of island arc volcanism remain unexplained.

2.5.6 Back-arc Basins

Question 2.8: Why do back-arc basins form?

In some subduction zones, a secondary accretionary plate margin lies behind the volcanic line (Karig, 1971). This back-arc spreading is similar to the seafloor spreading that is occurring at ocean ridges. The composition and the structure of the ocean crust that is being created are the same. Behind-arc spreading has created marginal basins such as the Sea of Japan.

Volcanic Behind-Arc Line Spreading

Volcanic Behind-Arc Line Spreading

Volcanic Behind-Arc Line Spreading

Volcanic Behind-Arc Line Spreading

Figure 2.22. Models for the formation of marginal basins. The descending slab, volcanic line, and behind-arc spreading axis are shown. The mantle wedge is the region above the descending slab. (a) Secondary mantle convection induced by the descending lithosphere. (b) Ascending convection generated by the foundering of the sinking lithosphere and the seaward migration of the trench.

A number of explanations have been given for behind-arc spreading (Hynes and Mott, 1985). One hypothesis is that the descending lithosphere induces a secondary convection cell, as illustrated in Figure 2.22a (Toksoz and Hsui, 1978a; Hsui and Toksoz, 1981). An alternative hypothesis is trench rollback, where the ocean trench migrates away from an adjacent continent because of the transverse motion of the descending lithosphere. Behind-arc spreading occurs in response to the rollback, as illustrated in Figure 2.22b (Chase, 1978; Garfunkel et al., 1986).

A number of authors have proposed that there are basically two types of subduction zones (Wilson and Burke, 1972; Molnar and Atwater, 1978; Uyeda and Kanamori, 1979). If the adjacent continent is being driven up against the trench, as in Chile, marginal basins do not develop. If the adjacent continent is stationary relative to the trench, as in the Marianas, the foundering of the lithosphere leads to a series of marginal basins as the trench migrates seaward. Jarrard (1986) has provided a more extensive classification of subduction zones. There is evidence that behind-arc spreading centers are initiated at volcanic lines (Karig, 1971). The lithosphere at the volcanic line may be sufficiently weakened by heating that it fails under tensional stress.

Was this article helpful?

+1 -1


  • duenna
    Why are volcanic island arcs arcuate?
    9 years ago
    Why do island arcs Yamaoka?
    8 years ago
  • monica
    Do island arcs subduct?
    8 years ago
  • berylla
    What leads to the creation of island arcs?
    11 months ago
  • Nestore
    Why do island arcs from in an arc shape?
    15 days ago

Post a comment