IODP

Previous   |    Next

doi:10.2204/iodp.sp.350.2013

Schedule for Expedition 350

Project IBM: arc evolution and continental crust formation

Key questions for comprehending arc crust formation include

  1. What is the nature of the crust and mantle in the region prior to the beginning of subduction?

  2. How does subduction initiate and initial arc crust form?

  3. What are the spatial changes of arc magma and crust composition of the entire arc?

  4. How does the middle arc crust evolve?

The best possible strategy for answering these questions is drilling by the International Ocean Discovery Program (IODP) at the Izu-Bonin-Mariana (IBM) arc system (Fig. F1). Four proposals to IODP to drill at the IBM arc system, including three riserless holes (proposed Sites IBM-1, IBM-2, and IBM-3) and one ultradeep riser hole (proposed Site IBM-4), focus on addressing these questions. Drilling at the four sites will result in comprehensive understanding of arc evolution and continental crust formation. Ultradeep drilling (proposed Site IBM-4) will follow riserless drilling with the R/V JOIDES Resolution at three sites (proposed Sites IBM-1–IBM-3) scheduled for 2014.

Site IBM-1: nature of the original crust and mantle (IODP Expedition 351)

Preexisting nonarc oceanic crustal components should contribute to arc magma chemistry through assimilation and partial melting triggered during passage of arc magmas; oceanic crustal remnants could also make up an important part of the lower arc crust. At the IBM arc system, preexisting oceanic crust is present west of the arc, under 1–1.5 km of sediment in the Amami Sankaku Basin adjacent to the Kyushu-Palau Ridge (KPR) remnant arc (Fig. F1) (Taylor and Goodliffe, 2004).

Site IBM-2: initial arc crust and subduction initiation (IODP Expedition 352)

A section through the volcanic stratigraphy of the outer fore arc of the IBM system will be drilled in order to trace the processes of magmatism, tectonics, and crustal accretion associated with subduction initiation. The result of this drilling will be used to test hypotheses for subduction initiation and arc crust formation processes. This in turn has implications for understanding the origin of the many ophiolites that are now believed to form in this type of setting. The drilling will provide an opportunity to test the supra-subduction zone ophiolite model and involve the land-based geological community in IODP.

Site IBM-3: the rear arc: the missing half of the subduction factory (IODP Expedition 350)

The spatial and temporal evolution of arc magmas within a single oceanic arc is fundamental to understanding the initiation and evolution of oceanic arcs and the genesis of continental crust, which is one key objective of the IODP Initial Science Plan (ISP) (Bickle et al., 2011). The Izu-Bonin-Mariana arc has been a target for this task for many years, but previous drilling efforts have focused mainly on the IBM fore arc and thus the magmatic evolution of the volcanic front through 50 Ma. Rear-arc IBM magmatic history has not been similarly well studied in spite of its importance in mass balance and flux calculations for crustal evolution, in establishing whether and why arc-related crust has inherent chemical asymmetry, in testing models of mantle flow and the history of mantle depletions and enrichments during arc evolution, and in testing models of intracrustal differentiation.

Site IBM-4: continental crust formation at intraoceanic arc: ultradeep drilling to the middle crust of the Izu-Bonin-Mariana arc

This proposal is for the ultradeep drilling site of a series of IODP proposals in the IBM arc that aim at comprehensive understanding of arc evolution and continental crust formation. We propose to drill a deep hole that penetrates through a complete sequence of intraoceanic arc upper crust and into the in situ middle crust that may be a nucleus of continental crust.

The IODP Initial Science Plan states that “The creation and growth of continental crust remains one of the fundamental, unsolved problems in Earth science” (Bickle et al., 2011). The formation and evolution of continental crust is a first-order problem of terrestrial geochemistry because for many trace and minor elements, this reservoir is quantitatively important despite its volumetric insignificance on a planetary scale. In the latter part of the 1960s, Ross Taylor (1967) proposed the “andesite model” for the origins of continental crust on the basis of similarities between “calc-alkaline” or orogenic andesite formed in island arcs and the “intermediate” bulk composition (~60 wt% SiO2) of this crustal type. For many arc enthusiasts, this observation has been a prime motivation for studies of island and continental arc systems. Subsequent studies have substantiated Taylor’s estimate of continental crust bulk composition and have noted that distinctive continental trace element fractionations (e.g., high U/Nb and Pb/Ce) are only found in suprasubduction-zone magma types (Hofmann, 1988).

The andesites of most young oceanic arcs, however, have been found to be more depleted elementally and isotopically than average continental crust, at least at the volcanic front. An old idea, still generally valid, that may explain part of this problem is the Kuno-Dickinson-Hatherton K-h relationship (Kuno, 1959; Dickinson and Hatherton, 1967): at a given SiO2 content, the K2O of related volcanic series is positively correlated with depth (h) to the Wadati-Benioff Zone. Thus, geochemical asymmetry in arcs was known prior to the advent of plate tectonics and may be what makes juvenile arc crust “continental” in key elements like Th and light rare-earth elements (LREEs) as well as intermediate in silica content. Potential processes to produce this asymmetry might be that

  1. Magma is first extracted from the mantle in the rear arc and is then extracted a second time as the depleted mantle moves toward the volcanic front where it is fluxed by fluid from the slab (Hochstaedter et al., 2000, 2001),
  2. The rear-arc magma source contains more slab-derived sediment melt (Ishizuka et al., 2003a, 2006; Tamura et al., 2007),
  3. The degree of melting is smaller beneath the rear arc (e.g., Tatsumi et al., 1983; Sakuyama and Nesbitt, 1986; Kushiro, 1994), or
  4. There is more recycling of old crust at the magmatic front (Kimura and Yoshida, 2006).

In addition to these petrological considerations, paleo-arcs are an essential part of most mountain belts from the Archean to Tertiary. Their geological architecture represents a fundamental aspect of the formation of continental crust. Most rocks in the upper crust of arcs are volcaniclastic, highly vesicular, and vitric. Models for their facies architecture have relatively good constraints for the subaerial arc front and for some aspects of submerged arc-front volcanoes, but knowledge of the rear-arc facies architecture is limited by the paucity of mapping, drilling, and sampling from modern rear-arc settings. Facies models depend largely on ancient successions now uplifted and dissected (e.g., McPhie and Allen, 1992), but incomplete preservation and exposure limit the value of such models. Nevertheless, facies models underpin a great deal of fold belt research worldwide, particularly tectonic interpretations, structural analyses, paleogeography reconstructions, and resource assessments. For example, in ancient successions, aligned volcanic centers and proximal associations are commonly the basis for reconstructing the arc trend and polarity. The presence of almost orthogonal rear-arc seamount chains in arcs like Izu has serious implications for such reconstructions. If the rear-arc affinities of volcaniclastic sediments can be better defined through this expedition, similar features may be recognized in ancient foldbelt successions. Likewise, knowledge of the eruption and depositional processes at submarine arc volcanoes depends largely on these same ancient successions (Fisher and Schmincke, 1984) and are weighted toward shallow-water volcanic front edifices. The effects of voluminous vesicular glass on water chemistry and microbiology are potentially enormous but largely unknown. Drilling is the only way to obtain information about volcanic eruption, sedimentation, and stratigraphy in oceanic arcs without looking through the effects of collision and accretion (e.g., Haeckel, et al., 2001; Wiesner et al., 2004).

Full crustal velocity profiles for the IBM arc obtained in the last decade show a velocity structure with continuous layers extending ~200 km across the arc (Fig. F2). We do not yet know how these layers developed, but testable hypotheses are being developed (Kodaira et al., 2007a, 2007b; Tatsumi et al., 2008). Given the relatively large volume of crust now in a rear-arc position (Fig. F2) and its present-day geochemical contrast with the magmas erupted at the volcanic front, it is vital that we understand the structure and compositional variations of the rear arc through time in the same way that we have unraveled the temporal evolution of the fore arc and volcanic front magmas.

A basic premise of this project is that time series of geochemical information about igneous rocks can be obtained from volcaniclastic sediment, mostly turbidites. This potential was realized during earlier Ocean Drilling Program (ODP) drilling in the Izu fore arc (e.g., Gill et al., 1994), and subsequent advances in microanalytical methods (e.g., analysis of clinopyroxene and zircons to obtain igneous geochemical information) make this even more likely now. The fore-arc turbidites preserve a faithful record of arc evolution, which parallels that seen in tephra (e.g., Straub, 2003; Bryant et al., 2003). The mass wasting of submarine edifices guarantees lots of volcaniclastic sediment. Even though mixing makes the signal more of a running average than in tephra, first-order changes (say on a few hundreds of thousands-year scale) are well resolved. Therefore, although many of our ends are igneous in nature, our means are sediments.

Top of page   |    Previous   |    Next