Ambiguity in interpretation of seismic data from modern convergent margins: An example from the IPOD Japan Trench transect

In the plate tectonic theory, crust generated at mid-ocean ridges is disposed beneath the continents at convergent margins. The first half of the theory is firmly substantiated by seemingly uncontrovertible evidence for generation of ocean crust. Consequently, crustal disposal is often assumed per se. The absence of early Mesozoic and older crust in the ocean basins is a strong indication that ocean crust is indeed disposed, but the processes of this disposal are poorly known. Structural models of convergent margins are largely deduced from what are presumed to be ancient convergent margins and subaerially exposed collision zones. The structure of modern convergent margins is observed in marine seismic reflection records most of which have poor resolution (50-100 m) even in the zone 1-5 km below the sea floor. Geological structure commonly featured in models or cross-sections of the large volume of rock lying between that detected in seismic reflection records and the seismically delineated Benioff Zone are necessarily based more on inference than observation (e.g. Hamilton 1977). One model has been used extensively to guide interpretation. This model involves an imbricate stack of thrust sheets lying above a master thrust, which is the interface between material being added to the upper plate and the top of the downgoing slab (e.g. Coates 1962; Seely et al. 1974; Karig & Sharman 1975). It has since been discussed and applied by many authors, and the necessary concepts were recently summarized by Dickinson & Seely (1979). We will refer to this model as the