provide the mesoscale observations needed to connect the regional view provided
by seismic reflection work with the fine-scale details of drilling data.
In order to bridge the gap between local observations in drill holes and cores and
the regional seismic reflection data, we now summarize our new observations made
using SHINKAI 6500 submersible. We mapped and sampled a turbiditic sequence of
Plio-Pleistocene (mostly Pleistocene based on the results of Kawamura et al. 2009 )
mud, silt, and fine-grained sand. Higher metamorphic grade rocks associated with
the Tokai thrust (as documented by Kawamura et al. 2009 ) were observed locally
(on Dive 1055, for example), but were not widespread. Rather, samples were gener-
ally relatively weak, unconsolidated sediments, even within the Tokai thrust zone.
The first order structure of the accretionary prism in the vicinity of Tenryu can-
yon comprises east-west trending folds that are associated with slip along an under-
lying array of thrust faults. However, we observed a number of second order
structures that trend at a high angle to the trench. Second order structures include
both tight outcrop-scale folds and broad map-scale folds with north plunging
hinges. These structures appear to be associated with transpressional deformation
that appears to intensify in the vicinity of the Tokai thrust. Where observed, the
Tokai thrust zone appears to comprise folded strata and overturned imbricate thrust
slices, which reflect some degree of along-strike displacement history.
In order to tie our observations together, we present a simple geometric, concep-
tual model for the evolution of this part of the Nankai accretionary prism (Fig. 8 ).
In this model we rely on a seamount or a proto-Zenisu ridge to impose deviations
from plane strain, though there are alternative hypotheses described in the follow-
ing paragraph. The subduction of a seamount or ridge imparts a map-scale,
indenter-controlled curve in the trends of first order structures within the accretion-
ary prism (c.f., Marshak 2004 ) and causes trench-parallel shortening, which mani-
fest as second order folds with north plunging hinges such as observed on Dives
1056 and 1057.
In our model, ultimately this east-west shortening gave way to strike-slip fault-
ing. Evidence for this strike-slip faulting includes small-scale strike-slip faults
observed during Dive 1058, and the slivers within the duplex structure in the Tokai
imbricate zone appear to have undergone some strike-slip offset. Such mesoscale
evidence reinforces inferences from bathymetry (Le Pichon et al. 1987a, b ) and
seismic reflection data (Takahashi et al. 2002 ) for strike-slip deformation. Such
strike-slip faulting also offers a possible mechanism for exhumation of deeper
prism materials; transpressive deformation typically has a strong component of
uplift (Karig 1980 ).
There are alternative hypotheses for the causes of trench-parallel shortening and
strike-slip faulting. It is possible that out-of-plane deformation is necessary to
accommodate plate-convergence and accretion near the triple junction of the Nankai
and Sagama troughs and the Izu-Bonin arc. Indeed, velocity models indicate that the
top of the subducting Philippine Sea plate is dipping toward the northwest in this
part of the prism (Nakanishi et al. 2002a, b ). A subset of this hypothesis is that
transpressional deformation is related to the plate-boundary reorganization in the
early Pliocene that gave rise to the current plate-boundary geometry (Hirono 2003 ).