The IBM (Izu-Bonin-Mariana) Arc System is a 2500 km long convergent margin that extends south from Tokyo, Japan, to Guam, U.S.A. This system is a natural laboratory for understanding convergent plate margin processes and products, for the following reasons:

• It is an endmember arc system, the type example of a decoupled arc, characterized by a deep, steep subduction zone due to subduction of the world1s oldest seafloor.
• The tectonic and magmatic history, from initiation of subduction during the mid-Eocene onward, is better known than that of any other convergent margin.
• It is entirely intra-oceanic, so that anatexis of continental crust is precluded. This allows a much more confident assessment of mantle-to-crust fluxes.
• A high-quality OBS refraction profile provides the best image of crustal structure available for any intra-oceanic convergent margin.
• A cross-section of the arc is preserved on land in Japan, where the northernmost IBM arc has been colliding with Japan for the past 15 million years.
• Because it is far removed from continents, sediment fluxes to the forearc and the subducting plate are low and the arc infrastructure is largely exposed, and there is no accretionary prism. This simplifies studying the forearc lithosphere, sampling of slab-derived fluids venting in the forearc, and estimating sediment input fluxes.
• The sedimentary sequence on the downgoing Pacific plate changes systematically parallel to the trench, principally involving changes in the relative proportions of pelagic sediments and intraplate lavas and volcaniclastic sediments.
• Convergence rates vary from 3 cm/year in the south to 7 cm/year in the north, providing an opportunity to examine relationships between arc products and processes vs. convergence rate.
• The distribution of strain and tectonic style varies simply along strike of the arc system, with back-arc spreading in the southern third, intra-arc rifting in the northern third, and an unrifted arc in the middle.
• Rifting episodes that preserve fossil arc crust in the Palau-Kyushu Ridge and West Mariana, along with preservation of tephra in pelagic sequences provide excellent opportunities to temporal variability of IBM arc volcanic activity and composition.
• In addition to being the simplest and best-known intra-oceanic arc system on the planet and a logical setting in which to focus studies of convergent margin processes, favorable logistical and political considerations exist and are likely to continue. In particular, efforts by the U.S. and Japanese scientific communities to carry out thoughtful, cooperative studies are likely to be favorably received.

A workshop designed to stimulate and co-ordinate studies of the IBM arc system was supported by the U.S. National Science Foundation and the Japan Society for the Promotion of Science and held July 27 to August 2, 1996, at the Shonan Village Center, Hayama, outside of Yokohama, Japan. The purpose of the workshop was to stimulate international efforts to study fundamental convergent margin processes, focusing on the IBM arc system. The workshop was divided into 2 parts. The first part (July 27-30) consisted of scientific presentations and discussion of problems that need to be resolved, along with a visit to the Japan Marine Science and Technology Center (JAMSTEC) in Yokohama. This part of the workshop was attended by 10 U.S. and 75 Japanese scientists along with a scientist from Germany, Britain, and Australia. Recommendations come from discussions of the working groups during this part. The second part of the workshop (July 31-August 1) consisted of a field excursion to study exposures of the IBM arc in the Tanzawa Mountains uplifted by continuing collision between the northernmost IBM arc and Honshu. The last day consisted of a visit to the Hakone extinct volcanic complex. During the first part of the workshop, four working groups (Tectonics & Geophysics, Magmatic Evolution, Subduction Fluxes, and Hydrothermal Activity & Mineralization) discussed the important research priorities and hypotheses that might be addressed by a collaborative effort and made recommendations as to how these objectives could be met. There was general agreement among participants on three points:

• Understanding the products and processes of convergent plate margins is an issue of fundamental scientific and societal significance, and the IBM arc system is a natural focus for such studies.
• Comprehensive studies of the IBM arc system have suffered because of political boundaries such that U.S. investigators have concentrated in the southern half while Japanese investigators have concentrated in the northern half. Significant advances have occurred where the arc system has been subjected to focused and co-ordinated study by U.S., Japanese, and other scientific groups, such as ALVIN diving and ODP drilling programs undertaken during the late 19801s.
• Future studies of the IBM arc system should be increasingly based on a commitment between U.S. and Japanese scientists to a systematic program of joint study that will involve other scientists wherever possible and appropriate.

The scientific priorities and recommendations of the 4 working groups are summarized - but not ranked - below. These recommendations are presented in greater detail along with a short bibliography in "Working Group Reports".

This group identified the following tectonic research priorities:

1) Understand processes and products of subduction initiation, including identification of pre-existing crust and formation of fore-arc crust.

2) Reconstruct plate kinematics, especially how much rotation of the Philippine Sea plate has there been since subduction began.

3) Understand processes of terrane accretion associated with the end-on collision of the northern IBM arc with Honshu, including: when subduction began along the Nankai Trough, how deformation in the IBM-Japan collision zone occurs, and what the seismic risks associated with this collision are for the greater Tokyo metropolitan area.

4) Determine the relationships between tectonics and igneous activity, particularly regarding the cause of shoshonitic volcanism in the central IBM arc and that associated with the rifting/spreading progression of the Mariana Trough back-arc basin.

5) Understand the transition from unrifted arc to intra-arc rifting to back-arc basin spreading.

The following geophysical research priorities were identified:

1) Determine, in as much detail as possible, the crustal and upper mantle velocity and density structure for unrifted and rifted sectors of the IBM arc.

2) Constrain the zone of serpentinization beneath the forearc, and the boundary between lithosphere and asthenosphere beneath the arc.

3) Generate models for convective flow and thermal structure of the mantle wedge, and determine how diapirs supplying the arc ascend against induced mantle downwelling.

Related to these research objectives is the need of the scientific community to obtain existing bathymetric and magnetic data, now in the archives of the US Navy and the Japan Hydrographic Department.

This group dealt with both short-term and long-term aspects of magmatic evolution. Research priorities include:

1) Understand the nature and composition of Philippine Sea crust and asthenospheric sources before IBM arc initiation.

2) Constrain the relationship between subduction initiation and generation of boninites.

3) Understanding the thermal and material controls responsible for stabilizing the magmatic front, about 10 million years after subduction initiation.

4) Understand the changes in magmatic compositions over the life of the arc, and what controls these changes.

5) Determine the composition of primitive IBM arc magmas, and how they are generated and fractionate.

6) Determine how IBM felsic melts are generated.

7) Evaluate controls on along-arc and across-arc compositional variations.

This group emphasized the importance of quantifying fluxes to and from the IBM convergent margin 'factory'. Input fluxes include subducted sediments, altered oceanic crust, and convecting asthenosphere. Output fluxes include fluid flow from the trench and forearc, serpentinization of the forearc lithospheric mantle, alteration of the convecting asthenosphere, and magmatic outputs from the arc to the back-arc basin. Tests of the input-control model are favored by the strong variations observed along and across the IBM arc system. Both Magmatic Evolution and Subduction Fluxes working groups called for development and implementation of a petrologic-geochemical-isotopic database for IBM lavas. Testing of flux models will require critical new data obtained by drilling the incoming plate, submersible studies of forearc fluids, microbeam studies of magmatic volatiles, and experimental simulations of subduction reactions. The ultimate aim is the development and testing of a quantitative flux model for the IBM convergent margin system.

This group emphasized the need to continue discovering hydrothermal sites in the IBM arc. It is also important to investigate submarine arc hydrothermal activity there with a level of commitment that approaches studies of hydrothermal activity at mid-ocean ridges. Variations in tectonic style lead to different styles of hydrothermal activity which may be evaluated by detailed studies of representative of arc, intra-arc rift, and back-arc basin spreading hydrothermal systems. Hydrothermal systems associated with mafic, felsic, and shoshonitic submarine volcanic centers should be investigated. Complete characterization of a representative submarine arc hydrothermal system (including estimating the dimensions of the convection cell, periodicity of fluid flow, its relationship to seismic and volcanic activity, the role of phase separation, and effective water-rock ratios), is proposed. This will lead to the determination of hydrothermal fluid end members and of hydrothermal fluxes for the IBM arc system. This research program will benefit from ODP drilling of hydrothermal fields.

Establishing firmer links between interested U.S. and Japanese scientists is the next step in studying the IBM Arc System. Proposals to establish NSF-JSPS funding for exchange visits of U.S. and Japanese scientists are now being prepared by the co-convenors, and interested scientists are encouraged to contact them:

Robert J. Stern (rjstern@utdallas.edu), Center for Lithospheric Studies, University of Texas at Dallas, Box 830688, Richardson TX 75083-0688 USA, Telephone:972-883-2442, FAX:972-883-2829 or 2537

Makoto Arima (arima@ed.ynu.ac.jp), Geological Institute, Yokohama National University, 156 Tokiwadai, Hodogaya-ku Yokohama 240, Japan FAX:81-45-333-1536