Download Paper 3.2 Mb pdf - Miles F Osmaston

International Geology Review, Vol. 50, 2008, p. 685–754. DOI: 10.2747/0020-6814.50.8.685
Copyright © 2008 by Bellwether Publishing, Ltd. All rights reserved.
Basal Subduction Tectonic Erosion (STE), Butter Mélanges,
and the Construction and Exhumation of HP-UHP Belts:
The Alps Example and Some Comparisons
MILES F. OSMASTON1
The White Cottage, Sendmarsh, Ripley, Woking, Surrey GU23 6JT, United Kingdom
Abstract
Studies of high-pressure–ultrahigh-pressure (HP-UHP) belts stand to benefit from having a
better idea of how subduction constructs them. Their exhumation, too, merits a closer look at epeirogenic processes. Progress in both these matters now appears possible and their relevance tested,
using the Alps, the world’s most comprehensively studied such belt, as the prime example. Its
comparative youth means that less of the potentially pertinent surface detail has been lost by erosion
or burial.
Much evidence suggests that subduction downbend is actually not a flexure but is a seismically
evident, escalator-like, through-plate step-faulting process. Especially where the subducting plate
is younger than about 90 Ma, this gives it the property (STE) of incorporating upper-plate material
in the step-angles, thus moving forward the position of the downbend surface that it has carved in
the upper plate rapidly, and at a shallow angle for hundreds of kilometers. This yields a nominally
“flat” interface profile, above which, at the shallow end of the system, the upper-plate crust will have
been chilled and perhaps had its lower part removed, thus providing an origin for basement thrust
sheets and (as subducted imbricate slices) those seen in HP-UHP belts.
Seismology shows that “flat” interface profiles often pass through an inflection nearer the trench.
This means that STE is “taking a second cut” and then disgorges the step-angle material along the
succeeding flat part of the interface. When a STE-thinned margin imbricates, this diagnostic material appears as a fluid-overpressured tectonic mélange “buttered” (no shearing at the contact) onto
the underside of each. Included floaters (up to >5 km in size) exhibit many lithologies, even slickensided, in an ex-oceanic, usually pelitic matrix that protected them from deformation. This “butter”
is a focus of attention in HP-UHP belts, especially for its mafic-ultramafic floaters, while the water
in its matrix seems important for reaction kinetics, for eventual epeirogenic action, and even to flux
partial melting.
Subduction of imbricate slices, already cooled by STE beneath them, and their successive lodgement beneath the “roof” and angle of the distant downbend profile carved by STE into the hanging
wall, builds a subcreted triangular section extending far down the back wall, much further than
hitherto thought possible. So this part exhumes the most, breaking through the “roof” above as a
fault-line prone to strike-slip (Insubric, Møre-Trøndelag, Median Tectonic Line).
In the Alps, and following extensive STE from the north, successive imbrication marked by
Cretaceous flysch sequences transposed the paleogeographic order, so the Piemont ophiolites were
from an oceanic-crusted northern forearc, not a separate ocean. Such transpositions cannot happen
during collision; they require too much transport. During Eocene collision, the old block-and-basin
structure in the European autochthon was epeirogenically rejuvenated (external massifs, Tauern
window) by deep-crust heating and petrological density effects caused by the overthrusting,
ultimately obstructing surface closure and causing back-thrusting. At an intermediate stage (early
Oligocene) Adria, with the Austroalpine “roof” still attached, moved >200 km to WNW, creating the
Western Alps as a tectonically distinct entity. Unroofing was caused by epeirogenic action at depth,
not by “extensional collapse.” Brief comparisons with six other HP-UHP belts—Kokchetav, DabieSulu, Franciscan, Sanbagawa, Maksyutov, and the Western Gneiss Region—are encouraging. In the
latter case, Norway and Greenland must have been closely juxtaposed by collision, not separated by
1Email:
[email protected]
0020-6814/08/1012/685-70 $25.00
685
686
MILES F. OSMASTON
the present shelves built shortly thereafter, their present epeirogenic characteristics being consistent with such construction.
Exhumation of the assemblage, mainly attributable to self-reheating of the deepest parts, which
in most cases may not even now be exposed, cannot begin until cooling by ongoing subduction stops
(collision) or markedly slows. The resultant volume increase (and incipient melting?) at such depths
may provide much of the required buoyancy, but also pushes the plates apart at depth and imposes
late-stage extension upon the superincumbent material. To the extent that this volume increase is
laterally constrained at depth, it must enhance exhumation rate and attenuate the steeply subducted
slices. The subducted assemblage has been kept so cool by continuing subduction that not until this
reheating becomes sufficiently established can zircons within it begin to provide a record. The subsolidus overgrowths they then develop record the progress of heating and cooling during exhumation; they do not even record the moment of collision and provide no insight as to the constructional
history—a function reserved for low-closure-temperature techniques that record the chilling by
STE. That history, in the case of the Alps, from the onset of STE to the onset of active exhumation,
occupied a period of at least 70 m.y. and is unlikely to have taken much less in comparable orogens.
Introduction
Studies of low-temperature HP-UHP metamorphic belts (HP-UHP belts) have concentrated on
their metamorphism and exhumation, being handicapped by having unclear ideas as to how they are
assembled and attain the depths. Proposed exhumation mechanisms have included: tectonic squeezing
(Platt, 1986, 1987); underthrust by continental crust
of passive margin (Hsü, 1991); indentation and
extrusion (Hacker et al., 1995); buoyancy control
(Reinecke, 1991; Chemenda et al., 1997); reheating
when subduction ceases (Draper and Bone, 1981);
extension (Platt, 1993); extension or underplating
(Harley and Carswell, 1995); complicated slab
drop-off to “let it exhume” (Froitzheim et al., 2003);
and slab pull to subduct buoyant slices, followed by
break-off to exhume (Ernst et al., 1997; Ernst, 2001,
2006; Brueckner and van Roermund, 2004). Many
of these authors overlook the observation that most
assemblages exhibit coherent sialic slices of limited
thickness (5 ± 3 km, according to Ernst and Liou,
1999, or even thinner, Ernst, 2006) and great extent
(Parkinson and Maruyama, 2004), not entire continental crust, so a means of providing these is
required; this is one of the primary motives of this
paper, although the thinness may partly arise from
mutual compression at depth during exhumation. A
related motive is to draw a distinction between the
behavior of inherited buoyancy and the dimensional
effects of developing it thermally in situ.
On the construction aspect, apart from those
working in the Alps, few, if any, authors seem to
have considered the role, abundantly evident in the
Alps, of a rooted Austroalpine-equivalent “roof,”
beneath which the HP-UHP construction may have
been done, but of which in other cases all vestige
has been lost through subsequent erosion. The construction of multi-slice belts with cross-strike
widths attaining 150 km, such as the Western
Gneiss Region (WGR, Norway) and Dabie Shan
(China), suggests the need to think about this. How
otherwise was burial maintained during construction? This question assumes much more significance in light of how long, as we will show, such
construction actually takes. Another, but related
problem, is that of closely juxtaposing major differences of exhumation, to which attention was originally drawn by Ernst (1971).
In 1978, well before HP-UHP belts had been
made an issue by the coesite discoveries (Chopin,
1984; Smith, 1984), the attention of the author was
attracted to the apparently severe décollement problem, from an engineering perspective, presented by
the presence in the Eastern Alps of crystalline basement thrust sheets of large cross-strike extent and
limited thickness (large aspect ratio). The reconstructions of their former extent then being inferred
by Tollmann (1978) seemed to bring the problem to
a head (Fig. 1). Hubbert and Rubey (1959; Rubey
and Hubbert, 1959) had been concerned with the
possibility of allochthon movement assisted by fluid
overpressure, but Forristall (1972) showed that to
avoid structural failure of the allochthon while being
pushed, severe limits upon allochthon size and surface gradient exist, even in these favorable conditions. Armstrong and Dick (1974), recognizing the
severity of the décollement problem in the case of
crystalline rock, noted that ophiolite emplacement
presents a similar problem but involves high
temperatures at shallow depth, so their solution was
one of softening by high heat flow, also found in
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
687
FIG. 1. The perceived problem of creating basement thrust sheets of large aspect ratio, detached from their lower
crust and mantle. This is a fundamental problem for constructing the Alps.
back-arc basins. But this was obviously of no use for
the Eastern Alps, there being scarcely any sign of
arc-type magmatism anywhere in the Alps during
Alpine times, until the very end (Periadriatic intrusions, Taveyannaz Sandstone) (e.g. Trümpy, 1960,
1982). Subsequently, Bally (1981) suggested that
the presence of seismic low-velocity zones within
the crust might be a useful approach to this matter,
and Laubscher and Bernoulli (1982) favored some
kind of subductive action to remove the lower crust.
Originally, Ampferer (1906) had sought to
explain the disappearance of the lower crust from
below the Austroalpine thrust sheets, some now
superimposed, by a process of “ünterströmung”
(understreaming) to swallow it, but not necessarily
all in one piece. Similarly, although the problem
portrayed in Figure 1 might not be regarded as
wholly resistant to special pleading, the author felt
that a more secure solution might be had by some
kind of tectonic erosion by the subduction process,
but not of the kind then offered by Scholl et al.
(1980), whose concern lay at the edges of plates (the
loss of terranes being then an active topic), not on
their undersides, far from the margin. More recent
and extensive work by this group (von Huene et al.,
2004; Ranero et al., 2006 and references therein),
by Clift et al. (2003), and by Kukowski and Oncken
(2006) has indeed shifted attention to the basal
erosion of the near-trench forearc as an explanation
of observed cases of subsidence, but has continued
to focus upon the availability of graben on the subducting plate, to act as receptacles for hanging-wall
material to fall into, in a process largely devoid of
positive mechanical action to detach that material.
Although, as we have noted, the Alpine/UHP
belt problem is primarily concerned with the
removal of middle and lower crustal material, but
necessarily removing its underlying mantle too, the
seismological evidence elsewhere, e.g., in the
Andean “flat-slab” segments, north-central Peru
and central Chile, already confirmed that the undercutting had proceeded to depths far into the mantle
domain. This excluded gravitational action and
suggested something much more forceful, possibly
associated with important seismicity, perhaps associated with the phenomenon known as “seismic
coupling” (Kanamori, 1977; Ruff and Kanamori,
1980) and even with the major tectonic action that
had been identified in the so-called flat-slab
segments (see below).
Accordingly, to avoid confusion with the term
“subduction erosion” adopted by the abovementioned authors, the distinguishing term
“subduction tectonic erosion (STE),” additionally
qualified on occasion by the term “basal” (as in the
title of the present paper), was adopted from the start
in the present author’s publications on this topic
(Osmaston, 1985, 1988a, 1991, 1992, 1994, 1995a,
1997). We will show that the tectonic aspect of STE
is distinguished by a rather close relationship to the
incidence and fault mechanisms of seismicity where
it is active. A serious aspect of our originally identified STE problem was therefore to achieve the
removal of tough lower crust and mantle material
with an agent consisting primarily of oceanic crust.
In fact, the effective dip angle achieved at
trenches is never more than 8° and usually is much
less, so most of a representative ca. 25–30° downbend slope of the subducting plate must develop
under the forearc as hanging wall. Accordingly, the
place to start was obviously to review the process of
plate downbend, since it was already known
688
MILES F. OSMASTON
(Stauder, 1975; Barazangi and Isacks, 1979) that, in
the north-central Peru and central Chile segments of
the Andes, downbend is currently displaced eastward beneath the continent by >600 km from the
trench, and is clearly associated with forelanddirected thrusting, a recognition (Molnar and
Atwater, 1978; Uyeda and Kanamori, 1979; Jordan
et al., 1983) that led the latter to propose that the
Sevier–Laramide sequence of tectonism in the
southwestern USA (see also Bird, 1984; Jordan and
Allmendinger, 1986) had involved an even greater
such flat-slab distance, a comparison repeated by
Osmaston (1995a) in an STE context.
This paper outlines an escalator-like, throughplate step-faulting view of the downbend mechanism and we develop its propensity for basal subduction tectonic erosion and advancement of the
downbend location carved into the hanging wall by
this process. Although the subducting plate is not,
in this context, actually bending, the term “downbend” is retained in this paper as descriptive of the
shape carved into the hanging wall. From the outset
this involves major departure from the view that the
form of the subduction interface profile is wholly a
function of the physical properties of the subducting
plate. Important among the diagnostics both for STE
and for the “subduction erosion” envisaged by von
Huene at al. (2004) is that it is expected to produce
a mechanically assembled and fluid-overpressured
kind of tectonic mélange. But in the case of the STE
process, we will show that such mélange will be
spread along the undersurface of the tectonically
thinned margin, itself provided by STE, to become
exposed in various tectonic circumstances including, it appears, in HP-UHP belts, where both its
now-boudin-like floaters2 and its matrix have been a
major focus of interest.
Inasmuch as every observable HP-UHP belt has
necessarily undergone a great deal of epeirogenic
action, the effects of that process need to be recognized and allowed for before we can properly assess
how it was constructed. So we begin by taking a new
look at thermally induced epeirogenic action, based
on deep-crustal metamorphism. We will suggest that
2First used in a mélange context by Pettinga (1982), “floater”
emphasizes the matrix-supported character of the occurrence
without imputing how it got there. In this paper, we refer to
“blocs de toute origine” (Marthaler and Stampfli, 1989) and a
wide range of sizes, potentially drawn from the gamut of
coherent crustal and upper mantle lithologies.
not only may such a process be relevant to HP-UHP
exhumation, but that it also offers the possibility of
differential epeirogenic action after an allochthon
has been thrust over the block-and-basin structure
of an autochthon, thus affecting the tectonics of
closure.
Armed with these two new additions (for STE and
for epeirogeny) to the plate tectonics armory, we will
then derive an interpretation of Alpine history, concluding with brief comparative remarks on other
HP-UHP belts.
To complement the understanding of our proposed mechanism for differential thermal epeirogenic action in the context of block-and-basin
autochthons and elsewhere, we also introduce, at a
late stage in the paper, a recognition of the genesis
and restricted epeirogenic behavior of what we call
“intermediate crust” (IC), as distinct from that of
mature continental crust (MCC). We regard IC as
being formed in the course of early plate separation
(not stretching) by a mid-ocean ridge (MOR)–type
process, but heavily modified by the simultaneous
presence of copious sedimentation. Initially this will
be pertinent to understanding the differential
epeirogenic behavior of the External Massifs of the
Alps, but our outline of its nature is postponed until
the section on the Western Gneiss Region (WGR) of
Norway, where it appears that, at the moment of
Scandian collision, Norway and East Greenland
need to have been closely juxtaposed, not separated
by the present extent of the continental shelves;
these have displayed a distinctively low thermal
epeirogenic response during the Neogene.
Was the former lack of these tools the reason
“Why plate tectonics was not invented in the Alps”
(Trümpy, 2001)? The reason for choosing the Alps as
an example is not just that the geology is so well
explored but because of the unparalleled degree of
control provided by the fact that, at the current level
of erosion, it displays a full range from UHP metamorphism at one end (Western Alps) to subsiding
basin (Pannonian) at the other, all of which need,
under the plate tectonics paradigm, to have been
achieved with a common set of relative plate
motions. This very diversity of result has, as recently
discussed by Kurz (2006), long led to strongly
dichotomous inferred histories for the differing
segments. Our primary aim is therefore to seek an
underlying unity; if that proves possible it should
widen the relevance of an Alpine prototype for HPUHP belts elsewhere.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
689
TABLE 1. Thermodynamics of a Representative Mid-crust Dehydration Reaction
Thermal expansivity (from Osmaston, 1973)
Volume change/unit heat in or out … Factor B = 7.7 × 10–3 mm3/joule
Metamorphism reaction
Muscovite1 = K-feldspar + corundum + H2O
(Muscovite may comprise up to 50% of the rock)
(i) At 807°C, 9 kbar
Volume increase of muscovite content … 10.5%
Factor B = 97 × 10–3 mm3/joule
i.e., 12.5 times more efficient dilatation than thermal expansion
(ii) If the H2O from dehydration of muscovite at 9 kbar moves up into non-reacting rock at 5 kbar, 700°C,
the effective B-value more than doubles to B ~ 227 × 10–3 mm3/joule,
i.e., ~30 times more efficient dilatation than thermal expansion
(iii) If the rock volume thus invaded were assumed equal to the original muscovite source volume,
the dilatation of that invaded rock volume would be 24.4%.
Moreover, the upward migration of the H2O would efficiently advect high temperatures to shallower levels,
promoting other dehydration reactions there. This cascade effect could greatly increase the total dilatation.
1In
the presence of quartz, muscovite dehydrates differently (to ky/sil/and), but the thermodynamics are more complicated
to calculate and the results are not expected to differ greatly from those given above. With acknowledgement to G. T. R.
Droop, pers. commun., 2000).
Thermal Epeirogeny of the Crust
Of the four mechanisms of thermally induced
volume change in silicates— (i) thermal expansion,
(ii) fusion/solidification, (iii) metamorphism with no
loss or gain of water from the system, (iv) solid-state
phase change—the two latter are hugely more efficient than thermal expansivity and (iv) is mainly
relevant in the mantle (Osmaston, 1973). On that
occasion the author applied (iii) to the rapid thermal
subsidence of certain sedimentary basins which, if
explained by thermal contraction, would have
involved unacceptably high heat flow and temperature in the lower crust. Application of (iii) in the
reverse direction means abandoning the standard
petrologic precept that the released water escapes
from the system. If, on the contrary, the water merely
migrates up the crustal column, where it may or may
not react to form lower density minerals, it appears
that the mean effective density of the column may be
reduced (despite the higher density of the dry rocks
left behind) and uplift should ensue. The results of a
representative calculation, from among the huge
range of possible metamorphic reactions, are outlined in Table 1. The objective there is merely to
outline the principle, not to offer quantitative evaluation in a real case. Note particularly the thermal
efficiency of the volume changes in the example, as
compared with pure thermal expansivity.
If the migrating water is silica-rich, the lower
temperature at higher levels will make the silica
precipitate, blocking any fractures that had been
produced by the intruding water, thus trapping the
fluid at depth. There is some observational support
for this. Large and unexpected volumes of water
were found at depth in high-grade rocks in the Kola
Superdeep Well (Kozlovsky, 1987). Wenzel and
Sandmeier (1992) noted a seismic low-velocity zone
(LVZ) beneath the Black Forest (eastern) flank of the
Rhine Graben and attributed it to a mid-crustal
layer enriched in fluids derived from thermal retrograde metamorphism caused by the heat from
Tertiary rift processes. Hall (2001) reported magnetotelluric detection of a high electrical conductivity
mid-crustal layer in the Colorado Plateau, well
known for the rapidity of its epeirogenic uplift.
Other examples of rapidly elevated plateaus
beneath which mid-crustal LVZs have been detected
are Tibet (Kind et al., 1996) and the Altiplano-Puna
plateaus of the central Andes (Yuan et al., 2000;
Schilling et al., 2006), but in both of these cases the
authors attributed the LVZ to partial melting, without considering hydrous fluid. Schilling et al. (2006),
however, also noted mid-crustal high electrical
conductivity by magnetotelluric observation, and a
similar feature has been found in Tibet (Chen et al.,
1996). Accordingly, we suggest that in all these
690
MILES F. OSMASTON
cases the crustal LVZ is actually hydrous in nature
and is the product of the sensitive epeirogenic mechanism outlined in Table 1. For a given degree of
heating, a hydrous LVZ would yield a much bigger
reduction of density than melting, contributing to the
fact (Yuan et al., 2000) that the Andean plateaus
have higher elevations than standard isostasy, based
on their crustal thicknesses, would predict.
A pertinent property of interstitial but nonmigrating fluid is that it will lower the thermal
conductivity of the rock (e.g., Osmaston, 1977). A
possible example of this in the Central Alps traverse
is that three features coincide: an LVZ at 15–25 km
depth (Mueller, 1982); a slight minimum in the surface heat flow, relative to that in the foreland basin;
and a maximum in the current exhumation rate (1
mm/yr) (Rybach et al., 1977), suggesting that heating below the crustal LVZ, concealed by its low
conductivity, is causing the high exhumation rate.
In a block-and-basin continental crust such as
characterizes much of western Europe, differences
in crustal thickness and deep constitution will, on a
Table 1 model, make the epeirogenic response to a
given thermal stimulus vary from place to place.
One such mode of thermal stimulus is to cover it
with a major allochthon, providing a thermal blanket
that will also further impede the loss of released
metamorphic fluid.
Although the loading by such an allochthon will
initially depress the crust isostatically, facilitating
the overriding, the greater but thermally delayed
epeirogenic response of the thicker crustal blocks
will then make them rise the most and exhume
through the allochthon, creating tectonic windows.
The rapid subaerial erosion thus set in train will
usually remove the top more quickly than will allow
readjustment of the geotherm at depth in the crust.
In effect, removing the cool part of the crust makes
the mean temperature of the rest higher, so that
“exhumational overshoot” will occur. We will argue
that this process has had a big impact in the Alps; it
is thus no accident that the External Massifs and the
Tauern Window now exhibit many of the highest
peaks in the Alps. Elsewhere, the Grandfather
Mountain Window (Southern Appalachians) and
Jebel Akhdar (through the Oman Ophiolite) may
offer further examples. Note that the same mechanism—isostatic depression by an allochthon, followed by epeirogenic uplift—would perhaps go far
to resolving the difficulty, otherwise presented, of
pushing ophiolites, or indeed any allochthon, far
uphill onto elevated positions on continental crust.
The problem of so doing has not always been appreciated but, as noted above, severe structural limitations are entailed. In fact, in many cases of
thrusting, geological investigation shows that the
autochthon surface was at or below sea level at the
time, so the need for subsequent epeirogenic action
is then explicit.
The mechanism may also have a bearing on that
which must be responsible for metamorphic core
complexes; in this respect, it is relevant that the
mechanism we are considering is one of volume
increase, so that, if tectonic circumstances do not
permit lateral relief, the entire volume change will
be expressed vertically, perhaps as gneiss domes. In
the case of crustal slices subducted to HP-UHP conditions, a similar mechanism will attenuate the
slices, pehaps with boudinage, and enhance exhumation rates, which have exceeded 20 mm/yr in
some cases (Ernst, 2006).
Conversely, in regional structural situations that
do not provide lateral restraint, it is important to
note (Osmaston, 2000c) that the volume increase is
deep within the crust, not at the surface, so the surface crust is thereby put into extension. In my oral
presentation (Osmaston, 2000c), I suggested that
the world’s most famous extensional example, the
Basin and Range region of the western United
States, was the product of the access from below of
Laramide subduction–inherited heat, evidence of
whose presence lies in the previous widespread
post-Laramide silicic volcanism (Coney and
Reynolds, 1977; Dickinson and Snyder, 1978). The
continued underlying presence of the formerly ultrayoung and rapidly subducted Farallon plate
(Atwater, 1989), reponsible both for the Laramide
tectonism and for this heat (see below), has been
demonstrated by the occurrence of 81 Ma eclogite
xenoliths in a 33 Ma Colorado Plateau diatreme
(Helmstaedt and Doig, 1975; Usui et al., 2003).
In respect of the metamorphic core complexes
that also characterize the wider Basin and Range
region, Lister and Davis (1989, p. 65) defined their
origin as due to “when the middle and lower crust is
dragged out from beneath the fracturing and extending crust.” If we replace this “dragged out” definition with the lateral component of our envisaged
thermal dilatation of the material, we also have a
possible explanation of the central doming that
these authors were forced to attribute to isostasy
following tectonic denudation.
This line of reasoning bears directly upon studies
of the exhumation of HP-UHP belts in which,
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
presumably unaware of the phenomenon described
in Table 1, the extension has popularly been
ascribed to “post-orogenic extensional collapse”
(Dewey, 1988), stemming from the suggestion of
Ramberg (1980), and regarded as the cause of exhumation. Because, however, some other cause of
uplift would be necessary for such “collapse” to
begin, it appears to carry the implicit assumption
favored by some authors, as previously noted, that
exhumation is, at least initially, a matter of restoring
originally buoyant material, in which case the material does not increase in volume and none of the consequences noted above (or in the next paragraph)
would be present.
In the case of exhumation that is caused by volume increase at depth, it follows, if any of that
increase is expressed laterally, that extensional
features will be produced at shallow levels, the
orientation of these being a function of the lateral
constraints upon dilatation at depth. In the case that
the growth in volume is progressive, as the material
reaches shallower levels and lower pressures, the
current erosion surface will always be subject to
such extension for as long as denudation is active,
no matter how much has been taken off the top. In
the case of the Western Gneiss Region, to be
discussed near the end of this paper, Milnes et al.
(1997) also concluded that the idea of extensional
orogenic collapse is not supported, the main extension in that case being related to Devonian plate
motions.
Granted the existence of any mechanism for
sharply differentiated vertical uplift, we may recall
that there arises an additional cause for shallow
extensional features. It has long been recognized
(Sanford, 1959; Stearns, 1978) that it will result in
shallow divergent peripheral thrusts, with compensatory normal faulting interior to that.
Overall, these considerations make clear that
extensional features are an inescapable corollary of
the kind of epeirogenic mechanism under consideration, so they are devoid of plate kinematic significance. Notably, the deformation of the rapidly
uplifting Altiplano-Puna plateaus has switched in
the past 2 m.y. from compression to extension,
despite the continuance of subduction (Schoenbohm
and Strecker, 2005). Accordingly, the fact (Champagnac et al., 2006 and references therein) that
Miocene and continuing extensional action has been
widespread in the Alps is taken here to support the
proposed mode of epeirogenic action, but will have
no place in our broader kinematic analysis.
691
We reason later that the primary mechanism of
HP-UHP exhumation is self-heating of the deep
subduction-constructed wedge of crustal material,
made possible by the cessation of the subduction
that had kept it cool, roughly as originally proposed
by Draper and Bone (1981). In these circumstances,
the release of metamorphic fluids at great depth in
the wedge would be an inevitable result, thus greatly
enhancing the epeirogenic action and helping to
explain the very fast exhumation rates now becoming evident in some cases (Hacker et al., 2003;
Ernst, 2006). The metamorphic incorporation of that
fluid in the first place could owe much to the structural incorporation of the STE-specific, although
widely observed, type of mélange that forms a main
focus of this paper.
Subduction Downbend and STE
The support of outer rises
The argument (Hanks, 1971, 1979, 1981; Caldwell et al., 1976) that the presence of outer rises
requires the elastic support of the downbending
plate has been persistently pursued (e.g., Watts,
2001), but in various ways it is not satisfactory. It
was shown (Melosh, 1978; Melosh and Raefsky,
1980) that, provided the rolling hinge on the subducting plate at the downbend produces compression at the inside of the hinge, this would demand
active expulsion of material from the inside of the
bend, creating a supralithostatic pressure gradient
in the plate that would generate an outer rise profile
dynamically (Fig. 2).
There is strong qualitative support for this view
(Osmaston, 2006). In the angle of the Arica Bight on
the Peru-Chile margin the subducting plate has, in
effect, to bend down in two directions, thereby
restricting the egress of material from the hinge and
increasing the pressure gradient needed for its
expulsion. Consistent with this, the Arica Bight displays the world’s highest outer rise (~1000 m), rising
abruptly from nearly zero along the Chile Trench
and no more than 200–300 m along the Peru Trench
(Prince et al., 1980; Coulburn, 1981; Schweller et
al., 1981). Moreover, the rise follows around the
angle of the Bight, whereas an elastic bend would
produce a plunging fold whose axis bisects the
angle. Positive free-air gravity anomalies are
present at most outer rises, amounting to 80 mgal in
this case (Schweller et al., 1981), but Segawa and
Tomoda (1976) showed, in respect of outer rises in
the Japan region, that when all had been considered,
692
MILES F. OSMASTON
FIG. 2. Portrayal of the dynamic genesis and support of outer rises as a product of compression within the angle of
plate downbend, as inferred by Melosh and Raefsky (1980). This opened the door to consideration of downbend mechanisms other than elastic flexure. Elevation of a mantle phase-change boundary, due to the supralithostatic pressure
gradient beneath the rise, is a suggested possibility, see text. To regard the oceanic plate as stationary is purely for
convenience when reading the diagram; in practice, all that matters is the relative motion. After Osmaston (2006).
a gravity excess still exists that, with remarkable
prescience, they attributed to elevation of the
garnet-to-spinel peridotite phase change. The pressure gradient in the Melosh model might do that
(Fig. 2). It may explain discrepancies between
sounding data and outer rise heights inferred from
satellite gravity observations, attributed to random
error by Levitt and Sandwell (1995). A valuable
control on this interpretation of the outer rise in the
Arica Bight is that the Mariana Trench, which
curves the other way, displays no outer rise (Uyeda
and Kanamori, 1979; Uyeda, 1983; Levitt and Sandwell, 1995) despite the greater elastic thickness
suggested by the ~ threefold greater age of the subducting plate in the latter case.
Note also that microseismicity occurs to a depth
of 70 km in the subducting plate below the flat
interface beneath the NE Japan (Tohoku) forearc
(Hasegawa et al., 1991), nearly three times the elastic thickness (28 km) calculated by Caldwell et al.
(1976) from the outer rise nearby, where the plate is
of similar age. We conclude that the size and indeed
the existence of outer rises has little bearing upon
the flexural rigidity of subducting plates, although
the rheological gradient within them is clearly one of
the factors, subduction rate being another.
Mechanism of plate downbend and of STE
Downbend and seismic coupling. We are now in a
position to devise a downbend mechanism that
accords with observations in trenches and meets the
requirement of compression inside the bend. Seismic rupture right through the plate in the outer rise
area was inferred by Kanamori (1971) for the 1933,
M = 8.9 Sanriku earthquake, off northeastern Japan.
Since then similar ruptures, long as well as deep,
have occurred near the Tonga and Sunda trenches
(Lundgren and Okal, 1988) and near the Kuril
Trench (Ammon et al., 2008). Despite this, these
slabs are clearly not in the process of “dropping off”
at this point. Seismic tomography transects of
subduction zones (e.g., Fukao et al., 2001; E. R.
Engdahl, pers. commun., 2001, 2005) all show continuity of the cold-slab signature (i.e., elevated Vp)
to depths of at least 180 km. If one extrapolates the
recorded rate, since 1933, of such along-strike rupture, all the world’s subducting slabs would have
dropped off in about 5,000 years. This suggests that
through-plate rupturing is part of the downbend
mechanism. Based on the evidence then available
from the Peru-Chile Trench, Osmaston (1977, p.
28), following Lliboutry (1969) and Lister (1971),
proposed a through-plate vertical step-faulting
mode of downbend, but this would have produced no
compression. Limited evidence that outer rise earthquakes have mechanisms that are extensional at
shallow depth but compressional deeper in the plate
(Chapple and Forsyth, 1979) suggests a cylindrical
form of faulting (Fig. 3A); this would provide the
required compression for outer rise genesis. It
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
693
FIG. 3. Sketches of the proposed mechanism of subducting plate downbend and of downbend advance by basal
subduction tectonic erosion (STE). At any given moment, where below the upper plate, all the step-angles will be full of
material, much of it acquired from the hanging wall as the result of successive increments in step-fault throw (A) with
intervening forward movements. Details of the action are in (B). Step-offset and shear-off actions do not necessarily
alternate; more than one step-offset increment may occur before enough ridge-push stress has accumulated to result in
shear-off. After Osmaston (2006).
appears that further seismological support for this
form of rupture may be hard to get. Typical faultplane solutions in this area are steeply dipping
towards the trench axis, largely, it appears, because
most of the energy comes from the strong shallow
levels. Enquiries of seismologists as to whether it
would be possible to reach a “fault cylinder” solution have been discouraging.
The steps in the Peru-Chile Trench reach as
much as 1 km in height and in some cases the
‘treads’ on the deeper steps do slope axis-ward more
than those above (Coulburn, 1981; Schweller et al.,
1981), as cylindrical faulting would do. Typically
the treads are 5–8 km wide and hardly ever less.
Along the Peru-Chile Trench the subducting
plate is nowhere more than 50 Ma old (e.g., Asch et
al., 2006). Elsewhere, where the plate is much older
and (presumably) mechanically thicker, graben
between the “treads” are increasingly common.
These go down to zero width and might be due to
yawning of imperfectly cylindrical faults when displacement occurs. This suggests that these graben
are secondary and not a symptom of extensional
stress upon the plate, but it may be that the cylindrical form is not always achieved where the subducting plate is so thick, as suggested by the recent
through-plate rupture near the central Kurils
(Ammon et al., 2008). Cylindrically curved faults
would not be expected if the lithosphere were elastically uniform, so we need to appeal to the rheological gradient in the plate for this, and imperfection of
the curvature then comes as no surprise.
The behavior of this step-faulting mode of downbend where it is overlain by the upper plate is
694
MILES F. OSMASTON
illustrated in Figure 3B. Each individual step-fault
increment offsets and locks the interface until
subduction-directed stress is enough to shear
through the obstruction. This stress is probably generated by the mid-ocean ridge system, not by slab
pull (Osmaston, 2006, and Fig. 6, below), so that the
forward movement thus permitted causes the upper
plate to load one of the steps until failure re-occurs,
thus producing an alternation of these two types of
rupture. Buoyancy of the subducting plate is
required for this loading by the upper plate to be
effective. Lay et al. (1989), from a study of 1130
earthquakes, reported “compelling evidence” of a
link between the two types. In the recent Kuril
example (Ammon et al., 2008), a major rupture (M =
8.3) of a previously locked interface was followed 60
days later by a major through-plate rupture, but this
was on the outer rise, so cannot have resulted in
locking the interface once more. Another step-fault
displacement, sufficiently beneath the upper plate,
will be required for this to happen. Accordingly this
process is here identified as the cause of the phenomenon known as seismic coupling in subduction
zones (Kanamori, 1977; Ruff and Kanamori, 1980;
Christensen and Ruff, 1983, 1988; Lay et al., 1989;
Osmaston, 1999, 2005a). That the interface-slip/
thrust-type earthquakes are characteristically bigger than the other type (Ruff, 1996) seems to be
much better explained by the positive obstruction
due to step-offset during accumulation of the strain
energy for earthquakes that may reach magnitude
9.0 or more, than by mere stick-slip on an already
slip-prone interface.
This duality of mechanism also offers an explanation as to why big subduction earthquakes do not
always produce tsunamis (Osmaston, 2005a). In this
respect the >1000 yr sequence of tsunami deposits
to the west of Puerto Montt in southern Chile
(Cisternas et al., 2005) provides interesting support.
There the tsunami deposit from the currently biggest
earthquake on seismogram record, the Valdivia M =
9.5 of 1960, is underlain (with only peat-bog intervention) by another, carbon 14 dated at 1575,
although records show there were big earthquakes in
1739 and 1839. If these were of step-fault type, the
lack of tsunamis would seem understandable, and
they could have provided the secure locking of the
interface that enabled the accumulation of all the
elastic strain in preparation for the Valdivia interface rupture, thought to have released some 250
years’ worth of plate convergence (Cisternas et al.,
2005).
STE and mélange construction. The STE aspect
of this downbend mechanism (Fig. 3B) is its incorporation of hanging-wall material into the stepangles. Notionally this is in the form of sheared-off
slivers, but the evidence (below) is that a large
amount of plucking goes on. Each step-fault earthquake offsets the interface downwards for only a
limited distance down-dip, so the line of subsequent
shear-off is guided by the existing shear zone lithology at all the rest of the undisplaced interface, both
above and beyond the step-displaced part. Consequently the dislodged material is wholly hangingwall material that has entered the step-angles, and
none is transferred in the other direction. For the
same reason, the erosional capability of the mechanism is no longer just a matter of “relative toughness” of hanging-wall and subducting-plate
material, but also (and mainly?) of the relative
dimensions of the offset and not-offset parts of the
interface. It is this asymmetrical behavior of the
STE process that gives it its downbend advance
capability. The downbend advance rates inferrable
from the progression of thrusting show that this is a
systematic property, not just a matter of chance.
The subducting plate will commonly be topped
by pelitic sediment, but other material (turbidite?)
will also be present where there is active sedimentation into the trench, although this stands a strong
chance of being removed from its substrate by innerwall accretion as the plate enters the interface. Thus
one expects the material filling the step-angles to be
a matrix-supported tectonic mélange of upper-plate
lumps in a usually pelitic matrix. The lumps/blocks
in the mélange occasionally demonstrate (see later)
that they were detached by seismic shearing but
they are not olistoliths, so are referred to herein as
“floaters.” The dominant lithology of the floaters
will vary according to the crustal/lithospheric level
in the upper plate at which the STE is active. Since
the step-throw builds over a considerable depth
range, around the downbend, the lithology of floaters
in successive “slivers” will vary accordingly. Based
on a typical 5–8 km width of steps in trenches, the
cross-strike size of floaters could conceivably run up
to a comparable dimension, unless broken during
detachment, and to rather more than that in the
other direction.
Because mélange construction takes place
entirely at depth, the impressing of floaters into the
matrix will ensure that the latter acquires a fluidoverpressured state, giving it a mobility supported
by the author’s observations (e.g., the Esk Head
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
695
FIG. 4. The large-scale action of STE. An initial volcanic arc, although indicated here, may not always be developed
before rapid downbend advance becomes established, inhibiting the surface expression of magmatism so long as
advance remains rapid (see text). If an arc does develop, the resulting heat content and melt-impregnation of the mantle
wedge may enable the arc to remain active for a considerable period, until STE has removed that mantle. The up-front
inflection of the interface probably does not develop until the main downbend has advanced a considerable distance.
The possibility of interface locking at either location can then put all the intervening part of the upper plate into compression, a situation confirmed in the case of the Central Chile flat-interface segment (Beck et al., 2005). Main advance
in individual cases may have attained as much as 1000 km (see text). To indicate that the process is not seen as specific
to continental margins, two possible upper-plate Moho levels are suggested. STE and downbend advance may proceed
to deeper levels and for greater ultimate distances if the initially undercut margin is foreshortened by imbrication.
Foreland-directed thrusting depends on the interplate coupling associated with STE at the downbend, so it migrates as
the downbend advances. As the downbend progresses to deeper levels, the nature of the thrusting changes to Laramidetype upthrusting, the vergence of which may be unclear. The level of the downbend relative to the upper-plate Moho
apparently depends primarily on the amount of downbend advance, not upon the nature of the rock (see Fig 5).
Abbreviations: UM = ultramafic.
mélange [Bradshaw, 1973], north of Christchurch,
New Zealand) of places where a split in the floater
has undergone slight separation in shear and the
foliation shows that the matrix has intruded orthogonally to fill the space. This state results in any deformation of the zone, including the shear transport of
floaters for long distances along the interface, being
wholly accommodated by the matrix until exhumation releases the fluid overpressure.
STE as a Large-Scale Process
Downbend advance and flat subduction
The progress of downbend advance through STE
action, as outlined above, seems to be the mechanism by which so-called “flat-slab” subduction
develops (Fig. 4) (Osmaston, 1992, 1994). Because
the main erosional carving is concentrated at the
downbend, one may reason that this part of the interface should advance faster than the deeper part,
thus steepening the downbend angle, as shown. In
fact, in the two Andean flat segments, north-central
Peru and central Chile, where both the downbends
are now about 650 km from their respective
trenches, the downbend angle is about 60 degrees in
each case (Barazangi and Isacks, 1979; Jordan et
al., 1983; Cahill and Isacks, 1992), much steeper
than in many other places, including the adjacent
Andean segments. We discuss later a mantle serpentinization mechanism that may be the means
whereby this steep and apparently straight part of
the interface manages to advance, albeit less readily
than at the downbend locus of STE.
Every time the interface is locked by a stepincrement, ridge push is temporarily coupled to the
upper plate, so foreland-directed thrusting may be
expected if the push is strong enough, and this will
migrate as the downbend does (Fig. 4). For this
purpose, among others, a new MOR (mid-oceanic
ridge) model has been developed (Osmaston, 1995b,
2000a, 2005a, 2006) with a far greater, although not
yet quantifiable, push-capability than that of the
standard divergent mantle flow model. Although
already recognized (Molnar and Atwater, 1978;
Uyeda and Kanamori, 1979), the existence of such
thrusting and its migration was further emphasized
by Jordan et al. (1983) for the Andean examples,
and they compared it with the similar SevierLaramide migration of thrusting in the southwestern
USA, suggesting that flat subduction (“downbend
advance” in the present context) may have extended
twice as far in that case. More recently, Gutscher et
696
MILES F. OSMASTON
al. (2000), pointing out that flat segments are much
commoner than has been appreciated hitherto,
remarked upon their association with forelanddirected thrusting reaching far into the hinterland.
In principle, the rate of migration of thrusting offers
a general indication of the rate of downbend
advance, because the STE site is the source of the
force, and suggests rates of the order of 30 km/m.y.
for the past 10 m.y. in both the current Andean
cases.
At that rate the locus of the would-be arc is probably migrating too fast for the magmas from the subduction interface to heat the crust enough for them
to reach its surface, so they underplate the crust, or
intrude its base, instead. The faster the downbend
advance, the thinner will be the magmatic underplating and crustal thickening. This explanation for
the well-known gaps in young volcanism in the “flatslab” segments of the Andes seems more satisfactory than that of subducted oceanic plateaus (e.g.,
Nur and Ben-Avraham, 1981; Pilger, 1981; Gutscher et al., 2000) for two reasons. (1) A collision belt
such as the Alps, with an extensive nappe system,
which we later relate to STE, exhibits a nearly total
absence of arc-type volcanism over a long period
until the very end (Taveyannaz Sandstone; Periadriatic intrusions) (e.g., Trümpy, 1960), which is quite
reasonable because any underplating would have
been removed by STE when it got there. (2) The
hypothesized link between the buoyancy of such an
oceanic plateau and the shutting-off of volcanism is
obscure. But it may well be that the presence of this
buoyancy, added to that of the young subducting
plate (see below), is currently, at the present stage of
the thermal evolution of the Earth, the critical factor
which provides the upward mechanical contact with
the upper plate necessary to trigger the action
of STE and downbend advance in a particular
segment.
This interpretation is pertinent, but also not
without an Alpine significance, to the ongoing
debate as to how, in the Andes, the crust of the Altiplano and Puna plateaus became so thick (~70 km
and 50 km; James, 1978; Isacks, 1988; Allmendinger et al., 1997; Kley and Monaldi, 1998; Oncken
et al., 2006), in that crustal shortening evidence
falls well short of a sufficient explanation. The
author’s comparison (Osmaston, 1995a) of the
Andes with Laramide-related events in the western
United States was based upon earlier discussions
with R. L. Armstrong (1987) and W.R. Muehlberger
(1992), from which it emerged that, following a
magmatic gap while the downbend of the very young
Farallon plate was rapidly advancing by STE toward
its sub-Colorado limit, the process became unsustainable at about 45 Ma, being promptly replaced by
a new “short-cut” steep, but much slower, subduction interface in the west, which took some 18 m.y.
to reach a magmatic stage in the Sierra Nevada.
Consistent with the conclusion (Osmaston, 2006)
that even young oceanic plates should properly be
regarded as including a substantial thickness of the
underlying, heat-containing LVZ material, this left a
still thermally buoyant oceanic plate in place
beneath the region, slab drop-off not being an
option, because such buoyancy cannot suddenly be
lost. The heat from its LVZ material then conducted
upward and melted the former oceanic crust at the
interface, causing the widespread “sweep-back”
(Coney and Reynolds, 1977) of rhyolitic-ignimbritic
magmatism referred to above. This sequence of
events offers two stages of magmatic crustal thickening, of which the second, involving melting of the
entire thickness of the subducted crust, would
surely be the most voluminous and not restricted
thermally to underplating.
Data assembled and discussed by Oncken et al.
(2006) for the Central Andes to the north of the
present Chilean flat segment (see also Coira et al.,
1993) yield the inference that an identical sequence
of events, but on a more restricted E-W scale,
occurred there in the period 26–5 Ma. These examples demonstrate that such flat-slab segmentation
can be impermanent. Importantly for our study of
the Alps, moreover, they also demonstrate that such
events have a major magmatic impact within the
upper crust, so we can be quite sure, on the
evidence, that no such interrupted segmentation
happened during the subduction that led to construction of the Alps.
A downbend advance rate of 30 km/m.y. in the
Andean flat segments is nearly 35% of the plate
convergence rates determined by Somoza (1998). It
is easily shown, however, that the bucket-space
available on a sequence of representative steps on
the subducting plate is only a fraction of the crosssection being removed, so plucking must be a major
feature of the detachment process, as indicated in
Figure 3B. A further relief on this point is that this
estimate of advance rate, based on the progress of
tectonism, is likely to be an overestimate because,
as the downbend advances, it also deepens, thus
enabling the thrusting to emerge even further ahead.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
The transect sketched in Figure 4 recognizes the
possibility of STE beneath a margin in an intraoceanic domain, at least the forearc then being formed
on oceanic crust. In such a case, the shallowly
undercut margin could then suffer imbricate shortening, perhaps followed by a further stage of STE
and downbend advance, and so on (Osmaston,
1992). This would have the effect ultimately of moving any intraoceanic subduction to a continental
margin, perhaps generating imbricate terranes in
the process, but without the need to invoke a second
subduction zone, as Dalziel (1981) found necessary
for the southern Andes example. Successive stages
of STE would then also be at increasing entry depth,
explaining why, after an initial descent, we have
rather deep flat interfaces in the Andes, combined
with general geologic indications at the present
coasts that do not accord with being so close to the
present trench. On this basis it becomes possible
that all or most of the subduction zones, currently at
continental margins, may actually have started life
in the intraoceanic domain.
The up-front inflection of the interface (Fig. 4) is
an apparent feature of many of the seismicity
transects examined by the author, although the
faulting associated with passing through the inflection results in a cloud of seismicity foci that tends to
obscure the actual interface path. Remember that
the inferred cylindrical faulting developed at the
trench effectively articulates the plate so that it can
follow around a bend in either direction. The implication is that STE is “taking a second cut” at entry
to the inflection and then—most important to the
topic of this paper—the reversal of step-offsets on
emerging from it discharges the mélange from the
step-angles, to be spread (“buttered”) onto the flat
part of the interface. It is this mélange, derived from
near the shallow end of the system and with floater
lithologies to match, that becomes evident when the
STE-undercut margin undergoes imbrication; the
mélange derived by the STE action at the main
downbend is carried into the mantle. The geometry
suggests that much of the main downbend advance
may occur before an up-front inflection forms; until
that stage, therefore, the undersurface of the upper
plate will lack a coating of butter mélange. We will
return to this point when we consider the Eastern
Alps, because its presence, with its ultramafic floaters, has always been regarded as the defining
distinction between Penninic and Austroalpine, and
thereby to assign them differing origins, with ocean
floor between.
697
Another kind of mélange, having floaters like
those mentioned here, but distinguished by its serpentinite matrix, also occurs in HP-UHP belts and
is discussed later (p. 707).
Two Pacific examples
The N-Central Peru transect (Fig. 5), selected
from a number of transects constructed by the
author for different subduction zones, illustrates
three of the foregoing points: the up-front inflection;
eastward younging and tapering-off of volcanism as
downbend advance accelerated (although the reason
for acceleration is unclear, but the buoyancy of the
Nazca Ridge may have helped); progress of foreland-directed thrusting. The Lima embayment in the
Peruvian coastline corresponds closely to the extent
of the flat segment and is clearly due to the subsidence produced by the STE at the inflection. Clift et
al. (2003) attributed this particular subsidence to
subduction erosion also. The Lima and related
basins in the embayment consist of late Cenozoic–
Pleistocene carbonates on the subsided and muchfaulted northwestward continuation of the Arequipa
Coastal Massif of Precambrian–Paleozoic age.
Deposited at less than 500 m depth, they are now in
deep water (Hussong and Wipperman, 1981; Kulm
et al., 1981; Thornburg and Kulm, 1981). Note that
STE produces major subsidence where crust is
being removed and replaced in the gravity column
by mantle but, where STE operates below the Moho,
oceanic crust is being introduced into the mantle
column and some uplift may result.
An E-W (Tohoku) transect of northern Japan at
40° N (not shown here) illustrates STE at shallow
level also, but in this case it is the product of action
at the main downbend. Here the downbend action
now occurs beneath the coast, generating important
upper-plate seismicity offshore and in a narrow
coastal strip (Hasegawa et al., 1991), ceasing above
the point where the downbend is complete. To reach
this position, STE appears to have advanced the
downbend westward about 200 km at the end of the
Oligocene, removing in the process the lower half of
the former arc crust, now far out in the forearc
(DSDP holes 438, 439), 50 km from the present
trench axis. This arc had been constructed upon a
terrane locally called the Oyashio Cretaceous Landmass (a former source of sediment), but STE set in
train its major subsidence, so there is little doubt as
to the continental magnitude of the former crustal
thickness. Clearly this is directly pertinent to the
Alpine problem; the removal of crust rules out the
698
MILES F. OSMASTON
FIG. 5. A transect of the north-central Peru flat-subduction segment constructed, for dynamical reasons, approximately in line with the plate convergence (see inset) despite the complication that seismologists traditionally present
transects orthogonal to trenches. Information drawn from a band lying between 7° and 11° S at the coastline. Lima Basin
is used as a generality for the basins along this part of the margin, the coastline of which appears to be embayed because
of STE-induced subsidence. Sources include Noble et al. (1974, 1984, 1990), Stauder (1975), James (1978), Barazangi
and Isacks (1979), Couch et al. (1981), Coulburn (1981), Hasegawa and Sacks (1981), Kulm et al. (1981), Thornburg
and Kulm (1981), McKee and Noble (1982), Jordan et al. (1983), Mégard (1984), Atherton and Petford (1996), and E.
R. Engdahl (pers. commun., 2001). The seismic mechanism for the indicated 1970 earthquake is from Abe (1972).
Abbreviations: L = Lima; A = Arica.
explanation often offered, that the within-mantle dip
of the subduction interface had passively “shallowed” in such cases. Early Miocene west-directed
thrusting near the Japan Sea coast, investigated for
oil (Fig. 5.5 of Hashimoto, 1991), marked the arrival
of the downbend near its present position. Forelanddirected thrust motions also typify seismicity farther
south along the Japan Sea coast (Fukao and
Furimoto, 1975).
Whereas the popular geophysical view has been
that development of flat subduction is attributable
solely to properties of the subducting plate, this has
had little regard to what was being geologically
recorded in the upper plate. In fact, as outlined
above, that record can be tied intimately to the
development of the interface profile. In the STE
result, the shape of a flat interface profile is a record
of what STE has done to the hanging wall, the latter
being seen to have preserved a degree of mechanical
integrity. On a world scene and backward through
time, STE is perhaps to be seen as the only adequate
mechanism (so far) by which large thrust sheets of
metamorphic basement, but lacking their lower
crust and mantle, could have been produced.
Driving forces
In view of the possibility (Laramide case) that
STE may have been responsible for flat interface
profiles as extensive as 1000 km, the matter of what
drives subduction is clearly important. To achieve
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
the necessary, and evident, upward mechanical
coupling for STE it is clear that it cannot be driven
by slab pull in such cases. The subducting plate
must have buoyancy to do the job. Nor can roll-back
be invoked because the upper plate is clearly in
compression during STE and downbend advance.
The seismological study by Yuan et al. (2000) of the
“back-arc” region of the Central Chile flat-slab segment has emphasized that the entire region is under
compression. Seismic coupling appears mainly, but
not exclusively, to be limited to young-plate (<70
Ma) subduction (Christensen and Ruff, 1983, 1988),
and the author’s studies indicate that STE appears to
be, and to have been, similarly restricted, supporting that the required plate buoyancy is primarily of
thermal origin.
This means not that the “slab” is somehow less
dense but, as mentioned above, that the oceanic
plate actually incorporates, mechanically integral
with the slab above it, at least part of the seismic
low-velocity zone and the thermal resource inherent
within it. This is made possible by the recognition
(Karato, 1986; Hirth and Kohlstedt, 1996) that
interstitial melt in the LVZ actually makes the mineral structure very much more creep-resistant, not
less, as previously thought, because of its effective
removal of the water-weakening from the crystalline
structure. A way of achieving this at MORs and at
the same time to much increase the ridge push
available, as both the observed foreland-directed
thrusting and the large magnitude of thrust-type
earthquakes implies, has been given by Osmaston
(1995b, 2000a, 2005a, 2006) but is not repeated
here. Note that the lateral support of Andean topography requires a five-fold greater ridge push than
that provided by the standard divergent mantle flow
model of MORs (Silver and Russo, 1995).
Increased ridge push is also needed to drive the
subduction of the resulting thermally buoyant young
plates; the East Pacific Rise clearly has no difficulty
in doing that; nowhere along the Peru-Chile margin
does the age of the subducting plate exceed 50 Ma,
and is less than 5 Ma in places (e.g., Asch et al.,
2006). Recognition of this fact has an important
implication, particularly in the Alps situation. The
maximum distance between two converging continents does not, if there was an active MOR between
them, set a limit upon the ocean floor consumption
available for the execution of extensive STE and
downbend advance at the margins. The efforts
(Dewey et al., 1973; Smith, 2006, and references
therein) to assess the ocean width available between
699
Africa and northern Europe during the period of
Alpine evolution have overlooked this factor as a
relief upon tectonic limitation.
In any case, whatever the means of providing the
subducting plate with thermal buoyancy, it militates
strongly against the idea of slab drop-off; thermal
buoyancy cannot suddenly be lost. We will return to
this in the context of HP-UHP belt exhumation. The
increasingly widely observed fading, in the 200–400
km depth interval, of the tomographic signature of
“slabs” (e.g., Fukao et al., 2001), and many
transects supplied by E. R. Engdahl (pers. commun., 2001, 2005) is therefore better explained as
due to reheating of the slab, rather than as slab
drop-off. A mineralogical explanation is available
for the revival of an apparent slab signature beyond
that (Osmaston, 2000b, 2003), an explanation
directly relevant to the formation and history of the
D" zone at the base of the mantle (Osmaston,
2005c). A mineralogical explanation has the further
benefit of explaining more readily how it is that
these traces penetrate so far into the lower mantle,
without fading if they were thermal. Another possible sign of slab reheating is that the slabs of 28
subduction zones, identified as warmer on the basis
of exhibiting lower intraslab brittle seismicity, are
distinguished by having slow or young-plate subduction or both (Kirby et al., 2002). Since it has long
been known (e.g., Minear and Toksöz, 1970) that
slab pull mainly derives from the bit that gets
further than 400 km depth, interpretation of the
tomographic fading as slab drop-off, the alternative
favored by Fukao et al. (2001), leaves us in a difficulty with explaining the bowing of arcs to form
back-arc basins. A reheated slab can give little or no
slab pull either.
Fortunately, as Figure 6 suggests, the anticlastic
curvature mechanism offers a resolution of the
apparent paradox of back-arc opening but compressive environments produced by subduction. Note
that anticlastic curvature, like the dynamical
support of outer rises, discussed above, depends
upon the properties of the material in the angle of
the downbending plate. This enables observations to
provide a variety of controls upon the possible calculation of those properties, wholly independent of
the elastic response approach used hitherto. In this
context, we note that the formation of moats around
large intra-oceanic volcanoes is also not necessarily
an elastic response, but a reversal of phase change–
related thermal epeirogenic action around the magmatic channel as volcanism wanes (Osmaston,
700
MILES F. OSMASTON
FIG. 6. The anticlastic curvature mechanism, illustrated here, after Osmaston (2006), will make curved arcs; rollback does not. The lines are to show surface form; they are not fractures. Anticlastic curvature is a phenomenon familiar
to anyone bending a thick sheet of metal or plastic by hand and is well known to engineers, e.g., Case et al. (1999). It is
due to along-axis extrusion of compressed material in the inside of the bend. Most back-arc basins that formed while
subduction was active—Mariana, Okinawa, Kurile, Banda, Cretan Sea—demonstrate that the ends of the arc have
remained pinned and did not participate in the basin-forming motion. This selectivity is not to be expected of “rollback.” The thicker the plate, the stronger the effect, so the generally greater effective thicknesses of plates recognized
by Osmaston (2006) increases the relevance of the anticlastic curvature mechanism. The curvature generated seems to
vary appropriately with subducting-plate age. Japan Sea and Lau Basin are exceptions for which another subductionrelated mechanism is available. This is the ability of subduction, following segmented STE of the upper plate, to drive
the arc terrane sideways when the convergence vector changes (Osmaston, 1992; see the initial discussion of Fig. 8 in
the text). The origin of the Aleutian arc is different yet again, because the basin behind it is older than the arc, i.e., it is
“trapped” ocean floor (Vallier et al., 1994). See discussion in Osmaston (2006).
2006), so does not conflict with the perception of
thick plates which favors the occurrence of anticlastic curvature.
In the context of all the above-mentioned
evidence about how subduction works and the
compressive nature of the resulting environment, a
phenomenon originally recognized by Molnar and
Atwater (1978), stressed by Uyeda and Kanamori
(1979), and fully substantiated in the trail-blazing
paper of Jordan et al. (1983), with further emphasis
by Uyeda (1983), it is suggested that the influential
and curiously unequivocal assertion (Hamilton,
1988, p. 3–4), “The common case … is that the
hinge rolls back, away from the overriding plate.…
a consequence [of which] is that the dominant
regime in an overriding plate above a sinking slab is
one of extension, not of shortening” now be assigned
to history. Hamilton (1988) did note the Sevier–
Laramide progression of thrusting, a feature to
which Molnar and Atwater (1978) had referred, but
he chose to sideline it, presumably because of the
conflict, and made no mention of the Andean evidence. Finally, we may note that Karig (1971), often
cited as the originator of the extensional interpretation of back-arc basins, actually recognized the
compressive environment provided by subduction,
so he suggested that they were pushed apart by
diapiric intrusion forces. Once established in a
back-arc position, a new MOR would certainly help
in this regard.
Examples of Unmetamorphosed
Butter Mélange
In that this paper seeks to establish how HPUHP belts may have been constructed, recognition
of starting materials is valuable. Figure 7 illustrates
examples of butter mélange, at two locations: southern Crete and Hokkaido, Japan. This type of
mélange seems to be of quite wide occurrence, but
its significance has not been realized. The Cretan
example is proposed as the type example that justifies the application of the “butter” epithet to this
kind of mélange, on the analogy that when you
butter a slice of bread there is no shear zone at the
surface of the bread or at the knife; all deformation
is accommodated within the butter matrix. This is
one of the demonstrations of the fluid-overpressured
state of the matrix; another, as mentioned earlier, is
the presence of injection phenomena where minor
shear splitting and separation of a floater has occasionally occurred. Such splitting can give information as to the originating subduction direction;
otherwise the scaly envelopment of floaters gives
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
701
FIG. 7. Examples of butter mélange. A and B. These images were taken a few meters apart on opposite sides of the
narrow road along the Asteroussia coast of southern Crete, just east of Platia Peramata. Location map (C) shows position
of Asteroussia nappe; for more detail on its structural evolution and location, see Fassoulas (1999). (A) shows the
unsheared basal contact between the gneiss and amphibolite Asteroussia nappe and the matrix of the underlying Arvi
mélange. Seidel et al. (1976, 1981) recorded a Late Cretaceous K–Ar (73–70 Ma) age for the Asteroussia nappe, seen
here as an STE-chilling age, similar but slightly younger ages being widespread in the Cyclades, suggesting a northward
progress of STE. The remarkable slickensides along the carbonate block (B) demonstrate its seismic detachment, under
conditions of major rock-pressure contact, not just gravity (as in the case of an olistostrome), possibly some 150 km updip along a shallow-dipping STE-cut interface, all of which has been accommodated by distributed shear in the matrix.
Bedding runs across the line of sight. Rucsac for scale, about 35 cm. D and E. Carbonate floaters (D) in Homanosawa
mélange (Paleogene) of the Kamiokoppe Formation being quarried for agriculture; Hidariyama quarry, northern Hidaka
belt, eastern Hokkaido, Japan. Extraction levels for vehicles give scale. Note the angular shape of the carbonate floater.
Slickensided floater (E) in the same quarry.
very little clue. The overpressured state may materially assist the imbricate sliding of one part of an
STE-undercut margin beneath another (see the next
section), or even of the motion over underlying
structures of the entire STE-undercut “roof,” as
seems to have happened in the Alps (p. 717), Dabie
(p. 728), the WGR (p. 734), and perhaps elsewhere.
Exposure of the mélange depicted in D and E of
Figure 7 arises from SW-directed Kuril-Hokkaido
convergence (Kiminami et al., 1992) and appears to
702
MILES F. OSMASTON
demonstrate imbrication of a former continuation of
the STE-undercut Tohoku forearc further south,
mentioned earlier (p. 697), in that magnetic mapping of the Tohoku forearc shows N-S lineaments
converging northwestward towards the Hokkaido
foldbelt. The causative motion of the Kuril arc and
Kamchatka is an example of convergence partitioning, discussed briefly below.
In one respect the Figure 7 examples are not
wholly representative, because, in the author’s
experience, slickensiding is comparatively rare.
Some eclogite “knockers” in the Franciscan, however, were recorded by Coleman and Lanphere
(1971) as being “grooved and striated.” If this was
slickensiding in an eclogitic state, it suggests STE
detachment at HP depths, although not necessarily
at the depth ultimately experienced. Metamorphic
sialic-basement floaters also seem to be relatively
uncommon, consistent with the up-front position
and shallow depth of the interface inflection inferred
to be the source of the floaters, although they occur
in the inter-Penninic mélange zones (Schmid et al.,
1990) and have been reported from the WGR (Skår
and Pedersen, 2003).
In its unmetamorphosed state, the kind of
mélange under consideration may correspond to the
Type II of Cowan (1985), in that he mentions “scaly
clay matrix” and “blocks up to a km or more in
size.” But his failure to mention the presence of
mafic-ultramafic floaters, which will emerge as a
frequent component of the butter mélange
addressed in this paper, may be because, as we will
show, their incorporation is in situations that destine
them for HP metamorphism, whereas Cowan’s attention appears to have been directed at examples
relatively free of metamorphic overprint.
For other descriptions of the type of tectonic
mélange seen to be at issue in this paper, although
mostly more metamorphosed, see, for example, Hsü
(1968, 1992, 1994). For an excellent review of those
in UHP terranes see Compagnoni and Rolfo (1999)
and, for an early description, see Smith (1980). The
importance of hydrous fluid being available to speed
the kinetics of reaction, particularly of retrogression
during exhumation (Schreyer, 1995; Maruyama et
al., 1996; Ernst, 1999; Ernst and Liou, 1999;
Proyer, 2003) means that the significance of the
incorporation of butter mélange into HP-UHP
terranes is not just as an indicator of the mode of
assembly but also as a source of water. One result is
the pegmatitic necks (e.g., Krogh et al., 2003)
apparently marking the attenuation of butter
mélange floaters into the boudin-like shapes that
characterize them in exhumed UHP terranes.
Imbrication and Subduction of
STE-Undercut Flat-Subduction Margins
Styles and environment of imbrication
Imbrication of an extensively undercut margin
does not require collision with another margin but
may occur in open ocean conditions (Osmaston,
1992). The reason is that STE is inevitably nonuniform along strike, even if it does not result in
clear segmentation. Such megaslickensiding of the
hanging wall will tend thereafter to guide, or tramline, the direction of subduction slip. Consequently,
if the plate convergence vector changes, one of two
things will happen. Either the change component of
motion will be accommodated as strike-slip faulting
outside the tramlined zone (e.g., along the magmasoftened line of the arc) or the margin will fail structurally and imbricate.
The first is well known as convergence partitioning, such as that in the Philippines and Sumatra
(Fitch, 1970), the powerful North Sumatra earthquake of December 26, 2004 being an extreme
example. The plate convergence there is towards
N08oE, but the large (272 mm) post-seismic elastic
rebound at Phuket in Thailand was towards S68oW
(Vigny et al., 2005), so compression had been deviated/tramlined through 60o.
The second, however, is now our concern, the
alternatives for which are outlined in Figure 8. In
Option 2 (Fig. 8) the subducting plate has to be
regarded as a conveyor belt; simple compression
will not provide enough transport, so any HP-UHP
belt in which transposition of the paleogeographic
order appears to have taken place must have been
constructed during open-ocean subduction. Bearing
in mind the foregoing arguments that, in the light of
the STE process, subduction zones may commonly
have started life with ocean-crusted forearcs, this
would mean that ophiolitic slices would be the first
to be imbricated and lodged at the downbend. We
will be referring to the Alps, the WGR, and Dabie
Shan HP-UHP belts, among others, in this particular regard.
If, however, as seems likely, the thicker part of
the undercut margin has not imbricated when the
opposing continental rise is encountered, any imbricate slices from this remaining part may now stack
with retention of their paleogeographic order. The
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
703
FIG. 8. Alternative imbrication styles after STE—open-ocean conditions. In Option (1) the foreland-directed vergence may be encouraged by the presence of latent foreland-directed thrusts that had been produced during the STE
process (Fig. 4). Option (1) is not the topic of this paper, but in either case imbrication is likely to happen first at the
thinner, frontal part of the margin, least able to withstand the disruptive stress applied by the subducting plate below it.
This means, in (2), that the stacking of slices reverses the paleogeographic order, but not in (1). In sedimentary paleogeography terms, this reversal means that a source on one slice will now be on the wrong side of the recipient slice.
FIG. 9. Some of the closure and crunch tectonic options when subduction and downbend advance have occurred
successively under both margins. The ultimate limit of closure is when the former downbends of both subduction zones
are juxtaposed. This will normally need a late-stage reversal of vergence and brief reactivation of the interface below B.
If that reactivation occurs sooner, at the stage shown in the lower diagram, the HP/LT assemblage will be driven under
the imbricated slices of margin B; this may be what happened to the WGR (see text and Fig. 18).
thick Austroalpine thrust sheets of the Eastern Alps
may offer possible examples of this.
A further, and very important, variant arises if
the opposing margin has subsequently become the
one with active subduction beneath it. In that case,
the vergence of that subduction will control the vergence of the imbrication imposed upon the original
STE-undercut margin when they meet (Fig. 9), thus
inhibiting the subduction of slices to HP-UHP on
that side of the system. We return to this later (p.
734), in our discussion of the Western Gneiss
Region (WGR) of Norway.
We note here, however, following Osmaston
(1997), that the Apennines may be a further, but
simpler, example of the lower part of Figure 9. The
western margin of “greater Italy”/Adria (side B),
including an oceanic-crusted/ophiolitic edge
(Ligurian ophiolites), already undercut roughly
704
MILES F. OSMASTON
northeastward (present coordinates) during an earlier (Late(?) Cretaceous) STE sequence, was met by
the E-ENE–vergent Oligo-Miocene convergence of
Corsica–Sardinia (side A in the figure), imposing
that vergence upon the Apennine imbrication. In
that case, however, subduction under side A
(Corsica–Sardinia) had been too short-lived for
appreciable STE and the subduction of imbricate
slices to HP-UHP, but the backthrusting is seen in
NE Corsica, where its HP/LT imprint shows that this
did occur to a limited extent. We show later that, in
the early to mid-Oligocene, Adria moved at least
200 km westward relative to “Europe,” so this
appears to have been the main Apennine-building
movement, rather than an eastward motion by
Corsica–Sardinia.
The southward-increasing incompleteness of that
“collision” means that a large part of the undercut
Italian/Adria margin was left resting on the ocean
floor involved in the earlier STE, with the resulting
subsidence being responsible for much of the
Tyrrhenian Sea Basin and the scattered allochthonous slices of continent on its floor. In this scenario
the original western edge of Adria, bordered by
ocean crust of ~170 Ma age (Lombardo et al., 2002;
Smith, 2006) may have lined up all the way from a
pre-imbricate western Alps to western Sicily, a unity
long suspected. Thus the present coastline of Italy,
apart from volcanic modifications, probably approximates the position of the early STE-advanced
downbend, i.e., the eastern limit of STE-thinned
crust.
In view of our earlier debilitation of the slab-pull
concept as a generality (Fig. 6), it is appropriate to
recognize that the generation of young crust in parts
of the floor of the Tyrrhenian Sea (Marsili and
Vavilov basins) may be an unique or very rare example of its occurrence. The ocean floor left in place
beneath the STE-undercut western margin of
greater-Italy/Adria experienced a substantial cooling interval before reactivation of subduction by
Corsica–Sardinia closure; this interval, marked by
the observed subsidence of the now-superincumbent STE-thinned margin, would have enabled it to
lose much of its former thermal buoyancy, so that
slab pull would arise when closure pushed it beyond
the downbend.
Some new terminology
The foregoing comments show that extensive
STE beneath one or both margins means recognizing, as never before, apparently, that much subduc-
tion-related tectonics can occur before, and perhaps
a long time before, ultimate collision. Accordingly it
would seem useful in this paper to distinguish four
nominal stages; namely “closure,” “encounter,”
“crunch,” and “collision.” Closure would refer to all
tectonic effects introduced by open-ocean subduction. Within “closure” we can distinguish the possibility of two substages: STE and its associated
tectonics, and imbrication—broadly Figures 4 and
8, respectively, although some of “closure” may not
involve either. The moment that divides “closure”
from “crunch” might reasonably be called the
moment of “encounter.” This will, of course, vary
slightly in time along an iregular margin, but if precision is required the term is still appropriate for
any individual transect. “Crunch” would refer to all
tectonic effects arising from encounter with an opposite margin that do not actually bring convergent
motion, of either vergence, to a halt. The more
extensive the prior STE, particularly if STE has
operated under both margins (Fig. 9), the more
protracted and potentially complicated will crunch
be, potentially only reaching finality (collision)
when the post-downbend tectospheres meet. And
“collision” would here be the moment that divides
the completion of “crunch” from the onset of all the
effects thereby set in train, mostly of indefinite
duration, and notably the reheating-induced exhumation tectonic sequence of any HP-UHP assemblage upon effective cessation of subduction.
A closer look at imbrication and HP-UHP belt
construction
Although Figure 10, to which we now refer,
draws attention to various Alps-related matters, and
indeed has been constructed with the Alps in mind,
these are to be discussed specifically later. Here we
discuss a few general aspects of the figure. Justification for assuming the initial presence of an oceancrusted forearc was discussed in a previous section
(p. 697). Downbend advance by STE produces in
general a gently dipping pre-downbend interface. At
some point the downbend will reach a level where
STE is removing superincumbent continental lower
crust that was at or above the closure temperature of
low-closure-temperature (LCT) dating systems (K–
Ar, 40Ar/39Ar, Rb–Sr, and some forms of Sm–Nd).
This juxtaposing of oceanic crust will chill a date
into it, providing a valuable record of the progress of
downbend advance. This chilling, cutting off the
heat supply from its lower crust and mantle and continuing thereafter so long as subduction beneath it
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
705
FIG. 10. Stages, during open-ocean subduction, in the potential construction of a Penninic-style HP/LT assemblage
following major (e.g. >600 km) undercutting and downbend advance by STE. An oceanic-crusted forearc is assumed.
An early arc, prior to the establishment of rapid downbend advance, is uncertain. The presence of an up-front inflection
of the interface (see Figs. 4 and 5), developed late in the STE sequence, is assumed but has been omitted for reasons of
clarity. The presence of such an inflection appears to be important for the imbrication process because the STE action
there intermittently couples ridge push to the undercut region of the upper plate. All sections in the figure are greatly
foreshortened by the excision of several hundred kilometers. So as to illustrate what is going on near the front of the
system, even after much imbricate shortening, the notional amount excised decreases from A to D. The details of the
behavior in C are speculative. In the Alpine connotation, north is on the left.
continues, is seen here as an essential prerequisite
to enable imbricate slices of the upper plate readily
to assume HP mineralogy. The very low heat flow on
the clearly STE-undercut Tohoku forearc of NE
Japan (Horai and Uyeda, 1963), demonstrates this
effect.
Thus the STE process achieves two of our main
requirements: the genesis of thin sheets of metamorphic basement and the preparatory chilling of them
for HP-UHP metamorphism. This probably means
that, so long as they are kept cool, there is little
problem in keeping the slices down in the positions
to which subduction has carried them, although the
possible preservation of a minor degree of residual
buoyancy, as favored by Ernst (2006), for example,
needs to be assessed in specific instances. The continued coolness of the assemblage is central to
understanding the zircon dating problems that have
arisen and we discuss this below. It also means that
exhumation hypotheses must concentrate on reheating to restore buoyancy, ideally with lots of help
from retrogression, rather than on the “release” of a
still-buoyant assemblage. Compagnoni (2006)
stresses that the UHP index minerals “are totally
706
MILES F. OSMASTON
destroyed during the post-climactic exhumation”
and replaced by lower-grade assemblages, only
being protected from retrogression when occurring
as micro-inclusions inside stronger phases.
Failure to appreciate the significance of these
LCT chilling dates, which typically turn out to be
very precisely recorded in the material, has resulted
in a huge amount of confusion. Their precision
suggests that the chilling rapidly carries the temperature far below the relevant closure temperature and
that continuing subduction has kept it that way. On
one hand, LCT dates have been seen as dating early
structural metamorphism, and on the other they
have been discounted in favor of supposedly
“better,” but much younger, high-temperature U-Pb
dates on zircons. From what has been said so far it
will already be clear that the construction of a HPUHP assemblage is inevitably a protracted affair,
involving a large amount of subduction. We will
show that in the case of the Alps this duration can be
quantified stratigraphically with some assurance,
rendering LCT dates as desirable corroboration of
preparation and construction timings.
The evident failure of zircons and garnets to
record any of that earlier history appears to emphasize the low temperature of the STE-generated
HP-UHP assemblage. Evidently (Parkinson, 2000;
Hermann et al., 2001; Parkinson et al., 2002;
Compagnoni, 2006; Ernst, 2006) both these minerals possess the ability to armor UHP inclusions
within them against pressure relief and retrogression during exhumation. The example (Liou et al.,
2007) of a zircon within whose old inherited core a
low-pressure inclusion had been preserved, but
which had then successively acquired a UHP mantle and HP rim during exhumation, suggests (as one
might well expect even more surely on mechanical
grounds) that this armoring capability also extends
to the protection of LP relics against the environmental pressure when subducted to UHP conditions. If the temperature there is low enough, and is
maintained so by the continuation of subduction,
any zircon growth will be impossible. As a result,
the only record of their stay in the assemblage will
be during the temperature rise associated with thermally driven exhumation. This confusion, reinforced
by the narrow U–Pb date ranges ubiquitously
obtained from UHP belts, has clearly been responsible for a belief that HP-UHP belt construction, as
well as exhumation, all happen within a very short
time frame (e.g., Platt, 1986, 1987, and many subsequent authors).
Another potential source of dating at the STE-cut
under-surface of the margin is the possible presence
of pseudotachylite. During the early downbend
advance stage, the frontal inflection of the interface
may not yet have developed so, as noted earlier, the
surface will not be lubricated by butter mélange.
But the highly intermittent subduction slip associated with STE and seismic coupling, wherever that
action is occurring, means that the instantaneous
slip rate may be around seven orders higher than the
mean subduction rate, with correspondingly high
local heating. Clearly this jerky motion (Osmaston,
2005a) could also be relevant to arc magmagenesis,
but at much deeper parts of the interface.
Sedimentation may, via rapid deepening, offer
evidence of the progress of STE at the underlying
inflection of the interface, as in the Lima Basin case
(Fig. 5). Sedimentation also offers evidence of constructional progress and, via paleontology, of its
exact timing. Up-front accretion may expose the
ophiolitic forearc, shedding chromitiferous sediment into the forearc basin (Fig. 10, section A). The
character and direction of sedimentation would then
be changed by a thrust exposure with resulting
flysch associated with imbrication (section B). The
imbrication process may thrust the flysch forward,
together with some of its substrate, whose presence
thus proving that it was not generated in a subduction trench. If the frontal (ophiolitic) slice underwent complete subduction, this would shove the
assembled flyschs virtually into a trench position, at
the very front of the margin.
We have drawn attention to the progressive
steepening of the downbend angle as downbend
advance proceeds. There may have been cases
(Laramide?) in which the angle became so steep
(vertical or even reverse-dipping?) that continued
subduction passage around the bend became
mechanically non-viable. Although the angle
depicted in Figure 10 is less steep than that, this
possibility should be borne in mind because it
makes the lodgment of slices more likely, controls
the shape of the downbend-lodged assemblage, and
affects its exhumation pattern. The lodgment of
slices across the angle will, of course, reduce the
active downbend angle (see Figs. 13 and 14), facilitating the continuation of subduction.
There seems no obvious limit to the depth down
the back wall of the subduction interface to which
the triangular assemblage might be built, except
that set by the depth to which the back wall may still
behave as a physical discontinuity. This will depend
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
on the effective tectosphere thickness on the continental side, which may be very much greater than
many realize (Agee, 1998; Gu et al., 1998; Osmaston, 2005b, 2006, 2007), especially if, as noted earlier, even the suboceanic LVZ may not have the
“asthenospheric” behavior usually attributed to it.
In the scenario presented in Figure 10, based on
reasoning given earlier, the first slice(s) to be
stacked against the back wall of the interface will be
ophiolitic, but this will not signify the former presence of an ocean in that paleogeographic position.
On the contrary, the entire HP-UHP assemblage,
together with its “roof,” comes from the same side of
the ocean—apart, that is, from the ocean-floor sediment associated with forming the butter mélange.
Because of the drag of ongoing subduction past
the base of a downbend-stacked slice, it is likely
that the basement of each slice will become displaced down-dip relative to its cover, now stuck to
the current ceiling. This means that the next slice to
arrive will catch the front edge of those cover rocks
and shear them out below their former basement.
This is crudely indicated in section D, being a feature actually observed in the Alps (e.g., Fig. 13).
Butter mélange will, of course, be likely to intervene
at many places, partly because each slice will tend
to import some on its base, and partly because there
may be an ongoing smearing of it by the subduction,
making it possible for butter mélange of slightly
different origins to be present at the same place.
Another kind of mélange—serpentinite
mélange—also occurs in many HP-UHP metamorphic belts, but typically exhumed close to the former
“back wall” of the subduction zone. As described by
Dobretsov and Buslov (2004) and by Moore and
Lockner (2007), the matrix in these is serpentinite
but the floaters are like those in butter mélanges,
typically including mafic, now eclogite, ones. These
authors suggest, very reasonably, that the serpentinite arises from hydrous mobilization of the hanging wall of the subduction profile. Wherever, in STE
flat-slab situations, the interface is within the
mantle, butter mélange would provide both the
water for the serpentinization and be the source of
the floaters it encapsulates. Such interaction would
almost inevitably introduce shearing of floater
lithologies and interleaving of matrices. In the
northern part of the Western Gneiss Region the
descriptions of Griffin et al. (1985) suggest that both
serpentinite and butter mélanges, with eclogite
floaters, are present but that, westward along the
Norwegian coast, the serpentinite matrix has under-
707
gone prograde metamorphism to garnet peridotite.
As these authors point out (see also BucherNurminen, 1991), the latter feature has long given
rise to debate as to how mafic materials of rather
shallow origin could have got pushed, or otherwise,
into the mantle at such depths.
The fact that serpentinite mélanges occur at all
confirms the occurrence of such mobilization. So
this may be an important additional STE agent for
downbend advance within the mantle domain, especially the advance and steepening of the post-downbend part of the interface, remarkable for the rarity
of seismicity upon it until the deep end is reached at
>500 km depth (Cahill and Isacks, 1992 and references therein).
Section A in Figure 10 assumes that STE has
progressed successively through sub-oceanic mantle and then through subcontinental mantle, so any
ultramafic floaters in the resulting butter mélanges
may exhibit isotopic features (e.g., Sm–Nd) distinctive of these two domains. The same may apply to
the matrix of serpentinite mélanges.
In considering the imbrication events sketched
in Figure 10, there is no reason to adopt a cylindrist
expectation that imbrication events would be coeval
and continuous along strike; that would require a far
more uniform force-transfer system and mechanical
properties than is at all likely. For example, even if
STE had been in progress all along the margin, it
could easily have left some segment(s) thicker than
others, thus affecting the susceptibility to imbrication. This may be what distinguished Eastern Alps
construction from further west.
Exhumation of the Assemblage
from HP-UHP
Effects of the configuration
Three things emerge from the preceding discussion of Figure 10.
1. The HP-UHP assemblage is constructed
beneath an upper plate/roof consisting of the thickest part of an STE-undercut margin, perhaps 30–50
km thick at this point but still extending a further
100 km or so before reaching the truncation wrought
by imbrication. Until this has been eroded off or tectonically removed, the near-level character of much
of that ceiling must inhibit exposure of the HP-UHP
assemblage by buoyant sliding (or “regurgitation”;
Ernst, 2001) from beneath it.
2. The assemblage is likely to be broadly triangular in cross-section, with a very big jump in effective
708
MILES F. OSMASTON
Moho depth between its deep tip and that present in
the upper plate/back wall of the subduction zone.
On the assumption of buoyancy generation by selfreheating, this means that maximum exhumation,
and maybe the fastest, will occur roughly above the
bottom tip of the triangle, tilting the assemblage as it
does so and exposing, in that location, rocks of the
highest pressure origin by differential erosion. This
means that, when seeking to interpret an actual
structural transect, this tilt would need to be undone
to achieve a view of structural relationships during
construction. Note, however, that the first slice to be
lodged as the triangular-shaped assemblage is built
cannot be the one that reaches the bottom tip of the
triangle. This has to be built downward and away
from the wall by successive slices. At the time of
exhumation, interslice mobility provided by
mélange between them may be enough to enable
differential exhumation, leaving behind any (earlylodged) slices in contact with the upper part of the
back wall. This is seen in the Dabie-Sulu belt,
where the slice(s) next to what we infer later to have
been the back wall are only at greenschist grade
(Zheng et al., 2005). A further constraint on thermal
exhumation rate of those parts of the assemblage
close to the back wall is its possible cooling influence when they reach shallower levels. A corollary
of this differential exhumation behavior will be that
the radiometric cooling dates recorded by any one
method will differ across strike.
3. When crunch is nearing completion, and
exhumation has not yet begun, the passive margin of
the other plate will begin to underthrust what
remains (non-imbricated) of the opposing STEundercut upper plate. At this stage the steep back
wall (former downbend line), probably rather linear
in plan, may become susceptible to accommodating
transcurrent motion between the two plates. In this
sense the HP-UHP assemblage, slightly underthrust
by the opposing, passive plate, now becomes part of
that plate, while the “lid,” thick and still integral
with the rest of the upper plate, will be slid alongstrike by it over the HP-UHP assemblage, partly
lubricated, perhaps, by the intervening layer of
fluid-overpressured butter mélange. Widespread
“top-to-the- …” fabrics, quasi-orthogonal to the
former convergence, will thus be impressed upon
the assemblage. If something of this sort is seen to
have occurred, it immediately calls into question a
commonly-made assumption—that of broadly
orthogonal compression throughout the late stages.
Self-reheating and the restoration of buoyancy
By thinning and chilling the slices before subducting them, STE opens the possibility of building
UHP assemblages that extend much further into the
mantle than workers, disturbed by the problem of
inherent buoyancy, have dared to envisage without
invoking slab pull. Where micro-diamonds now
occur at the surface (see later) and the crust, formed
of HP-UHP slices, is still, say, 50 km thick, as it is
in the Swiss Alps (Mueller, 1982), it must mean that
the original depth to which some assemblages
extended was approaching 200 km. At the bottom of
this a small part (6–10? km) is likely to be that of
oceanic crust of the European passive margin, but it
seems unlikely that collision could have pushed
uncooled and undiminished European continental
crust that far. If it did, it would have assisted reheating of the HP-UHP assemblage.
On the suggested terminology (p. 704), “collision” occurs when subduction during crunch has
halted or slowed enough for subductive cooling of
the assemblage to become insignificant in relation to
its internal (radioactive) heat generation. Based on
the reasoning set out in the second section of this
paper, the thermal release of metamorphic water—
water originally taken to depth in the matrix of butter mélange—is likely to be a major contributor to
the development of buoyancy, both volumetrically
and in assisting retrogression to lower density
parageneses, not to mention the possible promotion
of melting (another volume increase mechanism).
For this reason the site/depth at which buoyancy is
most effectively developed probably depends more
upon the conditions required for these changes than
upon physical thermal expansion of the rock. For
this reason, although calculations need to be done,
we suggest that the deepest and never subsequently
exposed parts of the assemblage may become the
biggest contributors to buoyancy, driving the exhumation of the column above. With such a very tall
column it is possible also that the cascade effect
mentioned in the lower part of Table 1 could become
a significant factor. Probably, therefore, the rocks at
the present level of exposure may not be those that
were mainly responsible for the buoyancy that drove
the exhumation.
It might be argued that the upper part of such an
immensely thick crustal section would receive a
considerable heat flow from below and thereby
acquire thermal buoyancy. But the establishment of
such conditions by thermal conduction would take
very much longer than the high exhumation rates
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
currently being inferred demand, so we would need,
instead, to invoke the advective transfer of heat by
the upward migration of fluids for this to be the case.
It remains for us to consider, and perhaps eliminate, one further potential source of heat for the
exhumation process—the LVZ material now
regarded (Osmaston, 2006) as included in the subducted plate. We suggested earlier (p. 696) that this
was the heat responsible for melting the crust of the
subducted plate that sourced the voluminous postsubduction silicic magmatism (PSM) that immediately followed the flat-slab developments both in the
western United States and in north-central Chile.
Where STE has been extensive and the subducted
plate was young, reheating (from below) of the thin
lithosphere of that plate has already been in
progress for some time when it reaches the faradvanced downbend. So any crustal melting will
begin at the far end, with little or no delay, a feature
supported by the westward migration of onset in the
post-Laramide US example (Coney and Reynolds,
1977). Such PSM-diagnostic migration is also
present in the late Caledonian Siluro-Devonian
(428–392 Ma) southward migration of granitoid and
arc-like magmatism across Scotland (Watson, 1984),
which began very shortly (on stratigraphic evidence)
before the Southern Uplands arrived in its present
final proximity to the English Lake District. The
snag in invoking such heating for the exhumation of
HP-UHP assemblages is that, except in very
restricted circumstances, the melting and magmatism would be expected to spread right across the
transect, and of this there has been no sign in any
case known to the author.
If the favored source of buoyancy, as outlined
above, is in the comparatively small volumes of rock
near the bottom tip of the triangular section, there is
a possibility of differential epeirogenic action developing in adjacent slices of differing constitution, so
that some, lubricated by intervening butter, rise relative to their neighbors in the assemblage, bringing
disparate pressure parageneses into contact. It may
not always be appropriate to regard the assemblage
as a coherent lump during exhumation.
If this differential movement of slices were to
develop before the roof had been eroded, the upward
motion could be guided by the curve of the roof
(former downbend), perhaps giving the impression
of thrusting in which a slice of higher pressure
origin now overlies one of lower pressure character.
Conversely, the occurrence of such a relationship
709
would offer support for the former presence of such
a roof.
The Alps as an Example of
HP-UHP Construction
Preliminary overview
As introduction to wide-ranging treatments of the
geological and structural history of the Alps and the
related Pannonian-Carpathian system, and to some
of the problems, the reader is referred to Argand
(1922), Trümpy (1960, 1973, 1975, 1980, 1982,
1988, 1997), Oxburgh (1974), Tollman (1977, 1980,
1986), Janoschek and Matura (1980), Laubscher
and Bernoulli (1982), Flügel and Faupl (1987),
Frank (1987), Polino et al. (1990), Horváth (1993),
Hsü (1994), Wagreich (1995) and Kurz (2006). A
simplified tectonic map appears in Figure 11.
In light of the coverage in preceding sections of
this paper, the scenario we will pursue is a development from that briefly outlined by Osmaston (1997),
itself a development from Osmaston (1985).
Broadly, our proposal runs, the differences along the
Alpine chain, from UHP in the west to a subsiding
basin in the east, owe much less to the degree of
prior STE beneath the Adria-Apulia margin than to
the nature of the opposing European margin that it
encountered. In the east, the European passive margin of Tethys appears to have been embayed so that
the undercut Adria margin did not imbricate until it
met the coastline/shelf edge, to form the curve of the
Carpathians. Westward along the Alps, the European continental margin was encountered sooner,
and its structure already incorporated an increasing
number of positive crustal blocks whose subsequent
epeirogenic action through the Adria-Austroalpine
allochthon would have been responsible for the tectonic windows (Tauern, Engadine) and their exposure as the external massifs (Aar-Gotthard, Mont
Blanc, Aiguilles Rouges, Belledonne, Pelvoux, etc).
Because the Penninic nappes include major
slices of metamorphic basement, indeed somewhat
thinner than the ca. 10–12 km of the Ötztal-Silvretta
Austroalpine (AA) thrust sheets that evoked the
author’s interest (Fig. 1), the scenario we require
must include the progress of STE beneath the
Penninic realm also. The structural intervention of
ophiolites, the Piemont3 Ophiolites, between the
two, hitherto always thought to imply the presence of
a “South Penninic Ocean,” is now seen as due to the
paleogeographic reversal of the latter brought about
by sequential imbrication. This reversal also bears
710
MILES F. OSMASTON
FIG. 11. Tectonic map of the Alps. Legend: 1 = Cretaceous metamorphism, mostly dated by low-closure-temperature
methods: a = blueschist, B = eclogitic; 2 = Cretaceous metamorphism, similarly dated, in the eastern Austroalpine cover
and basement units: a = very low grade, b = greenschist to amphibolite facies; 3 = Cretaceous to Eocene flysch and
Gosau beds; 4 = ophiolitic-type units with HP/LT imprint in Western Alps and Tauern-Rechnitz windows; 5 = main
Periadriatic plutons (early Oligocene but Eocene, southern Adamello); 6 = peri-alpine Oligo-Miocene basins, but
molasse of the Po Valley is unmarked. Abbreviations: A = Adula nappe; Ad = Adamello; AG = Argentera massif; AR =
Aar massif; AU = eastern Austroalpine sheets (Si = Silvretta, Ot = Oetztal); B = Bergell; Bi = Biella; CL = Canavese
Line; CS = Centovalli–Simplon Line; DI = Dinarides; Ef = Engadine fault; EW/TW/RW = Engadine, Tauern, and Rechnitz windows; GL = Giudicaria Line; GO = Gotthard massif; HE = Ultrahelvetic, Helvetic, and Dauphinois thrust units;
IL = Insubric Line; IVR = Ivrea zone; LIG = Ligurian (internal) ophiolites; LPN = Lower Penninic nappes; MR/GP/DM
= upper Penninic units (Monte Rosa, Gran Paradiso, Dora Maira); NCA = Northern Calcareous Alps; PF = Penninic
front; PGL = Pustertal–Gailtal Line; R = Rieserferner; S = Suretta nappe (upper Penninic); SA = Southern Alps; SB =
Grand St. Bernard nappe; SC = subalpine chains; SL/DB = Sesia Lanzo and Dent Blanche nappes (western Austroalpine); V = Voltri block; VB = Vienna Basin. Modified from Polino et al. (1990).
importantly upon the paleogeographic position of
the Briançonnais “ribbon continent” (Trümpy, 1982)
which, on account of its structural position to the
north of the Piemont ophiolites, has widely been
considered to have been a detached, or semidetached, strip adjacent to the European plate, but
3Two other spellings (Piedmont, Piemonte) are in use in the
literature. We adopt here the one used by Trümpy (1980), one
advantage of which is there can be no confusion with the
Piedmont of the Southern Appalachians.
which now is to be recognized as having been a principal part of the STE-undercut Adria continental
margin. Such a position also explains its narrowness. In its previously supposed capacity as representative of northward subduction under the
European plate (see discussion in Polino et al.,
1990) close proximity of the Briançonnais to the
European margin proper is denied by the sedimentary record of that margin provided in the Ultrahelvetic and Helvetic nappes, continental rise and
shallow shelf, respectively.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
A further differentiation along the Alps is that
the Eastern Alps display a very high proportion of
Austroalpine allochthon, whereas virtually all of it,
with the exception of the Dent Blanche klippen, has
been eroded off the structures further west. Proof
that there was such an overall allochthonous “roof”
in the west is to be had (inter alia) in the thrusting of
the Prealps flysch nappes, many with Penninic
affinity (see p. 713), to their present foreland position (Fig. 11). A temptation to regard this difference
in uplift and exposure merely as a fortunate circumstance that can tell us, on traditional cylindrist principles, what is present under the Eastern Alps
overlooks that epeirogenic action needs a cause, its
lesser incidence in the east being an indication that
the constructed overall crustal thickness is less.
So the preference here is that in the Eastern Alps
sector there was (for an unclear reason) less subcretion of Penninic slices and the depth at which STE
acted under the AA was greater than further west;
this meant that the AA allochthon was thicker and
more resistant to imbrication, although not in the
Northern Calcareous Alps (NCA). As discussed
later, only two Penninic slices can certainly be identified in the Tauern Window (e.g., Tollmann, 1977,
1980) whereas more, although laterally overlapping,
can be identified further west (e.g., Trümpy, 1960,
1980), where the bottom of the assemblage laps onto
the Gotthard massif. Further west still, the gneissic
domes that constitute the Dora Maira, Gran Paradiso, and Monte Rosa internal “massifs” are suspected by the author of being relatively coherent
multi-slice assemblages, although details have yet
to emerge.
In contrast to the Eastern and Central Alps, it is
in the Western Alps, and Dora Maira in particular,
that the real UHP metamorphism has been found,
beginning with the first finding of coesite (Chopin,
1984), but leading to the evidence (Hermann, 2003;
Groppo et al., 2007) that diamond-stability conditions had formerly existed, now retrogressed to
graphite, in whiteschist estimated at deriving from
43–45 kbar conditions, nearly 140 km depth. It can
be estimated that, for the sequence of STE, imbrication and downbend-lodgement to 140 km depth
portrayed in Figures 4, 8, and 10 to have operated,
there must have been more, probably much more,
than 500 km of broadly E-directed subduction
closure, especially as this is a multislice assemblage
(Lardeaux et al., 2006). Such a closure distance
almost certainly precludes (geometrically) that the
closure was during crunch, starting from the well
711
established approximately mid-Eocene moment of
encounter in the north (e.g., Hsü, 1994). So the
Eocene Rb-Sr ages in the nearby Zermatt-Saas area
(Dal Piaz et al., 2001) cannot possibly relate to time
of high-pressure burial. This, in turn, resurrects the
significance of the Cretaceous LCT ages, widespread not only here and near the western Penninic
Front (Fig. 11) (Hunziker et al., 1989, 1992; Polino
et al., 1990) but in Dora Maira also (Biino and
Compagnoni, 1992). That significance (Fig. 10 and
our discussion of it) is that they record the chilling
progress of downbend advance during open-ocean
subduction. It was this dichotomy of ages that made
Hunziker (1986) and Neubauer et al. (2000) view
them as signifying “a two-stage collision.” The argument that such Cretaceous ages reflect “excess
argon” (Scaillet, 1998; Dal Piaz et al., 2001) may
need reconsideration in this case. In any case, this
cannot apply to the Rb-Sr ages, and its influence is
largely eliminated by the stepwise heating in the
Ar-Ar method. The dichotomy arises because the
dates relate to two different cooling processes, STE
and exhumation.
In this regard, a close association between
tectonism and high-pressure metamorphism in the
Zermatt-Saas area led Fry and Barnicoat (1987, p.
657) to infer continuous refrigeration by underthrusting of thrust slices “in a zone of continuing
subduction.” This reasoning does nothing, however,
to explain the very existence of the change of strike
in the westward curve of the Western Alps which,
even within the region west of the Penninic front, is
associated with at least 140 km of post-Eocene
WNW-directed closure (Butler et al., 1986). The
Voltri block, at the southern end of the Western Alps
bulge, also exhibits ubiquitous W-directed deformation (R. L. M. Vissers, pers. commun., 2005). Faced
with this amount of thrusting, attempts to explain
the bulge as the product of N-S compression (Ricou
and Siddans, 1986; Vialon et al., 1989) seem unsatisfactory and overlook the multiple evidence for
major W-directed motion of Adria along the Insubric
Line (Fig. 15) at about this time—an important matter that we develop in some detail later (p. 717).
Constructing the Alps by STE and its sequels
We turn now to the specifically Alpine aspects of
Figure 10, in continuation of the remarks on it made
in the preceding subsection (p. 704) entitled “A
Closer Look at Imbrication and HP-UHP Belt Construction.” Ages refer to the timescale of Gradstein
712
MILES F. OSMASTON
et al. (2004). For descriptive purposes, present
geographic orientations are used.
Oceanic layout. A primary structural feature of
the Central and Western Alps is a concentration of
ophiolitic slices and material at the AA-Penninic
junction which turns sharply downward close to the
north side of the Insubric Line. With the advent of
plate tectonics, this was immediately seen as relating to a suture. Dewey and Bird (1970) set the suture
at the Insubric Line, whereas Ernst (1973) more
precisely set it at the AA-Penninic junction. This
was taken up by Frisch (1976) and since that time
the standard concept has regarded these South
Penninic Piemont ophiolites as marking the closure
of a main oceanic strip, with a seemingly secondary
strip, not marked by ophiolite but delineated by
large amounts of apparently accretionary sediment
(Bündnerschiefer), in a North Penninic/Valais position, now seen structurally south of the Aar-Gothard
massifs and evident further north in the Eastern
Alps, overthrust by the AA.
But Osmaston (1985) showed that, by recognizing STE, an alternative layout becomes possible,
essentially that shown in Figure 10, in which the
only ocean had lain in the North Penninic position
with the entire Penninic domain being derived from
a northward continuation of the Austroalpine
margin. Subsequent imbrication would not only
generate the up-front flyschs seen in the Prealps
but could remove from view the ophiolites whose
original presence up-front was evident in some of
the early flyschs, but whose principal remnant now
appears to be the Ybbsitz ophiolite, in front of the
northeastern edge of the NCA (Decker, 1990 and
references therein), although a small western representative, with a similar relationship, appears in the
Iberg klippen, north of Drusberg (Trümpy, 1980,
p. 76).
Quite independently, Polino et al. (1990) saw a
postulated position of the Briançonnais/Gd St
Bernard unit adjacent to the supposedly passive
European margin (Lemoine and Trümpy, 1987) as
unacceptable in view of its Cretaceous metamorphism (LCT dates), so they opted for a layout
broadly similar to the one given here, using what
they called an “orogenic wedge” mechanism to subduct the units. That argument, subsequently thought
to have been invalidated by the discounting of LCT
dates as due to “argon excess” (Dal Piaz et al.,
2001), is now sustainable again, so its implication
will be briefly discussed here.
In the Western Alps the sedimentary cover of the
Briançonnais uniquely reaches up to the midEocene (~44 Ma) but it is Cretaceous elsewhere
(Michard et al., 1996). This means that the thick
Briançonnais unit must have remained part of the
AA roof until encounter with the European margin
caused further imbrication and pushed it under the
rest/root of the roof. Subsequent encounter with the
external massifs (Pelvoux-Belledonne) apparently
detached a further imbrication from the roof and
drove it around the downbend curve of the roof so
that it now forms the Sesia-Lanzo zone and Dent
Blanche klippen, separated by subsequent erosion.
The true root of the AA roof at this transect thus lies
in the Ivrea zone, much upheaved by exhumational
drag nearby. A speculative interpretive section to
illustrate these events is given later (see Fig. 14).
Sedimentary features. In section A of Figure 10,
the mention of Lombardian flysch refers to the Lombardian basin south of the Insubric Line. This was
N-derived and of clearly AA-sourced origin, the
main flush being of late Cenomanian age, with
repeats in early Turonian and again (Bergamo
Flysch) in late Santonian–Campanian (Bernoulli
and Winkler, 1990; Bernoulli, in Trümpy, 1980),
roughly 94, 91, and 83 Ma. This sequence could
record either the southward migration of downbend
thrusting in the AA or the rejuvenation of the same
one, but it suggests that, at this transect, STE at the
primary downbend under the AA may have arrived
near its eventual southernmost limit by one of these
dates. In that case LCT dates chilled into the undercut margin would all be older than that.
The switch of source direction and lithology
mentioned in section B of Figure 10 has long been
recognized in the Upper Gosau Group basins on the
NCA (Faupl et al., 1987; Wagreich, 1995, and references therein; Faupl and Wagreich, 2000). The
switch, in early Campanian (83 Ma) was from a
chromitiferous northerly source (presumed to be
ophiolitic forearc crust uplifted by an accretionary
structure) to a southerly source of garnet and staurolite and was accompanied by a big increase in subsidence rate (Wagreich, 1995). The switch of source
must surely record the uplift of AA basement by the
onset of an imbrication, in which case the deepening
could be due to depression in front of the thrust
as subduction of the slice began, rather than to
subduction tectonic erosion below the margin as
suggested by Wagreich (1995) and by Faupl and
Wagreich (2000).
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
For imbrication to occur, ridge push needs to be
coupled to the frontal part of the undercut margin by
locking, as indicated in section A. This needs stepfaulting of the subducting plate at an up-front inflection of the interface (not shown, see caption). This
83 Ma imbrication must have been a sequel to
others, further north, that marked the subduction of
the Penninic slices seen in the Tauern Window and
which juxtaposed the ocean-crusted northern
margin and accretionary prism to the NCA as source
for the earlier Gosau sediments.
Section D of Figure 10 refers to the gathering
together of many flysch sequences in the Prealps,
but a similar collection is distributed all along the
northern margin to the eastern NCA. The bulge-like
shape of the Prealps appears to have no significance
other than easier thrust transport through the gap
between rising external massifs. That many are of
Penninic affinity has been recognized at least since
Staub (1937) suggested this for the Niesen flysch, so
it is inappropriate to ignore them when discussing
the origin of the Penninic nappes, as stressed by
Stampfli and Marthaler (1990). Although in this
paper we see flysch generation as particularly diagnostic of the imbrication process illustrated in
Figure 10, the incorporation of flysch at many places
in the Penninic structure shows that it was not all
retained on the top of the system, but large volumes
of it were subducted with the imbrication slices. A
good example is the Cretaceous Prättigau Flysch,
which emerges westwards from beneath the Silvretta
thrust sheet (AA) and the intervening ophiolitic
Arosa Zone (e.g., Trümpy, 1980), the latter being a
superb example of the kind of butter mélange with
which this paper is concerned.
Of the flysch sequences seen in the Prealps, several rest on pre-flysch substrates, some going back
to Early Triassic (Caron et al., 1989; Hsü, 1994).
This makes it impossible for those that predate the
Eocene encounter with the European continental
rise to be seen as slices accreted in a subduction
trench. The imbrication site provided by our imbrication model seems much better. It also provides for
progressive leap-frogging, due to imbricate foreshortening, that might be able to explain their
observed structural order, with the most ophiolitic
one (Gets-Simme) now on top. Most of the flysch
nappes exhibit quite a long succession of flysch
groupings, showing that the site remained within the
depositional reach of later imbrication events and
perhaps setting a size for their spacing. As to age, 6
of the 10 described by Matter et al. (1980) are of
713
Cretaceous age, leaving no doubt that major aspects
of Alpine construction were already in progress by
then. In view of the early arc suggested in section A
of Figure 10, we note that one of these, the Schlieren
Flysch, is reported by Homewood and Caron (1982)
to contain calc-alkaline volcanic clasts, but it seems
uncertain if their age has yet been determined (C.
Heinrich and W. H. Winkler, pers. commun., 2008).
All we are left with is that the volcanism must predate the (late Maastrichtian; Matter et al., 1980) age
of the flysch. What is more helpful, in relation to the
order of imbrication predicted in Figure 10, ophiolitic/chromitiferous content is characteristic of a
majority of the older flysch sequences, going back at
least to the Cenomanian or late Albian (~99 Ma)
(Simme, Gets) (Matter et al., 1980; Homewood and
Caron, 1982). And it is fully consistent with the conclusion (Trümpy, 1980, p. 89) “that as early as by
Cenomanian time the Upper Austroalpine sheets of
the Calcareous Alps had overridden the Lower
Austroalpine nappes and were lying along Upper
Penninic elements,” i.e., the highly ophiolitic Arosa
Zone, interpreted here as a major example of butter
mélange. Conversely, some of the flysch are totally
devoid of ophiolitic content but are rich in basement
lithologies—nearly impossible to achieve if they
had had a trench-accretion origin.
If STE and advance of the main downbend is, as
seems logical, halted by the downbend-emplacement of a first imbricate slice, it follows that most or
all of the flysch-generating events must post-date
the main STE sequence, which must therefore have
begun well before 100 Ma. For example, if (and this
has yet to be reassessed from the scenario presented
in this paper) one guesses that the amount of downbend advance and STE in the Alps amounted to 600
km, this would take us back a further 15–20 m.y. (to
mid-Aptian?) on the basis of Andean downbend
advance rates given earlier. This further underlines
the error of ignoring the Cretaceous LCT dates.
In this connection we note that, in a wide-ranging
sampling of the AA nappes in the Eastern Alps,
Dallmeyer et al. (1998) obtained tightly defined
40Ar/39Ar and Rb–Sr ages in the range 100–70 Ma,
but a surprising feature was that the youngest ages
were from the structurally lowest levels. This puzzle
will not be analyzed here, but the general picture
confirms much previous work to the effect that
Cretaceous LCT ages are orogen-wide, seen here as
demonstration that STE was orogen-wide too. Frank
et al. (1987) attributed the 95–70 Ma dates in the
Eastern Alps AA to chilling when underthrust by
714
MILES F. OSMASTON
Penninic material—not quite identical to that proposed here, in which the chilling was by contact
with the subducting ocean floor and the actual
Penninic slices did not travel south until later.
The structural result at the deep end
As sketched in section D of Figure 10, the successive emplacement of slices shifts the active interface downward, filling in the downbend angle. This
may mean that later slices become lodged at greater
depth, rather than building wholly horizontally
trenchward beneath the roof. This is important if the
observed high metamorphic pressures are to be
achieved. The presence of ophiolitic mélange–like
materials between the Penninic nappes of the
Central Alps has long been recognized (e.g., Hsü,
1994, and references therein), to the extent that
Trümpy (1980) recognized at least three such zones
(Mesocco/Misox, Avers, and Fex, in addition to the
highly ophiolitic Arosa Zone that underlies the AA
nappes), which he interpreted as sutures between
former intra-oceanic microcontinents, successively
subducted.
A notable example of butter mélange, containing
“blocs de toute origine” that remained beneath the
AA roof, without being smeared to the deep end, is
the Tsaté Nappe of “Schistes lustrés with Ophiolites” which immediately underlies the Dent
Blanche klippe and is little metamorphosed (Marthaler and Stampfli, 1989; Stampfli and Marthaler,
1990), although it is clear that some of the nappe is
sedimentary, having escaped the STE-related
“mélangification” process, as presaged in our earlier discussion of the STE process. Further south,
half way between Dora Maira and the Pelvoux
massifs, in approximately the same structural position, is the large (7 km+) but partly dismembered
Montgenèvre Ophiolite body behaving as a klippen
on a mélange that exhibits a pumpellyite-riebeckitelawsonite metamorphism that was not developed by
the ophiolite (Bertrand et al., 1980, 1982; J.
Bertrand, pers. commun., 2004), perhaps because of
its relative dryness. The size of this ophiolite body
sets it in the uncertain range between an STE product and an imbricate slice.
Moving eastward, to the Central Alps, the east
slope of the Lepontine (semi)dome has exposed a
structural cross section through the Penninic
nappes (Fig. 12). We consider later the late top-S
deformation that affected the Schams nappes and
may have pushed the Schams-Suretta-Tambo
assemblage southward relative to Adula; here our
interest is in the structure before that. The Schams
nappes are the former Mezozoic cover of the Suretta
and Tambo nappes (Schmid et al., 1990) and must
formerly have lain somewhat further to the north, as
predicted in the Figure 10 emplacement sequence
discussed earlier, with the basement having been
dragged southward as successive slices came in
below. Note how this action has (and for the Adula
Nappe also) smeared out the cover around the noses
and beneath the nappes. In the southern part of the
picture, the exhumational gradient toward the
Insubric Line has removed or reduced the original
southward dip of the nappes at the time of emplacement near the subduction downbend, so the back
wall of the interface would have been much steeper,
as discussed earlier. Initial emplacement of the
Suretta nappe (Avers phase) is known stratigraphically from the cover to be of mid-Cretaceous age
(Milnes and Schmutz, 1978; Milnes and Pfiffner,
1980), so this is positive proof that UHP construction had begun by then.
Moving eastward again, this strong attenuation of
the “lower limb” was also remarked upon by
Tollmann (1977, 1980) in regard to the two huge Nvergent Penninic “folds” exposed in the Tauern
Window, draped across the Zentralgneis uplift
(“Sonnblick dome”). In this case the Matrei Zone,
directly beneath the AA and exposed around much
of the window margin (Frisch et al., 1987; Höck and
Miller, 1987), offers a good example of butter
mélange as envisaged in this paper, equivalent to
the Arosa Zone further west. In the Matrei Zone, the
mélange part merges with the underlying Bündnerschiefer (Frisch et al., 1987; Kurz et al., 1998), as
predicted for the STE process where the sediment
imported on the subducting plate is thick and/or less
argillaceous. Notably, however, the mélange floaters
here are in a pelitic matrix, inconsistent with
creation of the mélange in an olistostrome or flysch
situation. Further, and bigger, ophiolite slices occur
in the underlying Glockner nappe (Kurz et al.,
1998), suggesting that we have here a double layer
of butter mélange.
The Zentralgneis, a Variscan granite gneiss,
forms the core of the Sonnblick/Granatspitz doming
associated with the exhumation of the Tauern
Window and is seen here, not as having Penninic
origin, but as being the top of an autochthonous
massif within the former European plate and similar
in this respect to the external massifs further west.
The epeirogenic mechanism of this exhumation, as
the result of being overthrust by the AA roof, was
FIG. 12. Interpretation of structural phases in the Suretta and Schams nappes (modified from Milnes and Pfiffner, 1980, plate). Phase names and dates according to these
authors were: Av = Avers (mid-Cretaceous); Fe = Ferrera (mid-Cretaceous, 110 Ma); Sch = Schams (mid-Eocene); Nm = Niemet (late Eocene–early Oligocene). According to the
present paper an early Oligocene top-WNW deformation intervenes between Schams and Niemet, and Niemet is later (late Oligocene/early Miocene). Black star (near center)
marks present location of Andeer.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
715
716
MILES F. OSMASTON
FIG. 13. Proposed mechanism of late backthrust phase during continued Adria-Europe closure at depth: differential
epeirogenic rise of blocks in a pre-existing block-and-basin structure, due to self-heating under an allochthon. Generally applicable, the section is drawn with Aar massif in mind; 1, 2, 3 represent present observed attitudes of successive
overthrusts, the last being the Glarus thrust (3). The Aar massif—a positive crustal block—would probably have had a
deeper Moho (not shown) than adjacent European crust. Although the section is drawn NW-SE, this may not have been
the actual late-closure direction. In the case of the uplift that produced the Tauern Window, the Austroalpine lid
escaped laterally eastward (Fig. 15) instead of being backthrust.
discussed in the second section of this paper. There,
it was pointed out that the resulting rapid erosion
steepens the thermal gradient in the remainder.
Supporting this, young coals on the Tauern Window
area and the Rechnitz Window exhibit rapid maturation, attributable to high heat flow (Sachsenhofer,
2000). Because the proposed thermal epeirogenic
process takes only limited time to get going, this is
consistent with not overriding the “Zentralgneis
Massif” until the middle or late Tertiary, as the midEocene initial encounter with the European margin
requires, but which rules out a Penninic origin for
the “massif.” The 36–32 Ma 40Ar/39Ar phengite
ages from the Schieferhulle sheets in the window
(Zimmermann et al., 1994) support this.
The Crunch Sequence—Major Movements
The sequence of Alpine events now to be set out
is far from the simplistic form, such as “continuing
N–S compression,” as traditionally assumed for
collision belts. We reason that the HP-UHP exhumation process has thereby been tectonically much
affected in the Alps, so this could give insight as to
what happened to other HP-UHP belts and be crucial to understanding them and their exhumation.
The governing principle is this: once the frontal
edge of the system has become fully engaged with
the opposing margin (Europe in this case) the locus
of interplate freedom shifts to the exhumation site
(in this case the Insubric Line and its equivalents
along strike). This is not surprising, because the
process of STE development of a very steep downbend and “back wall” interface to the subduction
zone provides a structural discontinuity right
through the “upper” plate, apart from the thickness
of the STE-undercut roof, potentially of the order of
30–40 km at the interface downbend.
Backthrusting—Part 1. The mechanism
The structures in Figure 12 show clearly that
backthrusting was primarily a top-S affair, affecting
the AA roof and relatively shallow levels of the
structure below it, not a squeezing of the entire
structure such as that envisaged by Platt (1986,
1987). A way to achieve this (Fig. 13) is to recognize
the effect of the progressive epeirogenic rise of the
external massifs, the Aar-Gotthard massifs in this
transect. This rise process will, on the basis of Table
1, have been initiated, although with some thermal
delay, by the passage of the allochthon over it. In the
case of the Aar massif (R. Trümpy, pers. commun.,
1980) the rise was about 8 km, commencing soon
after the earliest overthrusting, and is explicit (personal observations) in the dips of the succeeding
thrust planes on its north side. Figure 13 shows what
happened when commencement of exhumation at
the Insubric Line detached and removed the thrusting drive by Adria and the AA roof could no longer
surmount the massif.
Other support for this positive kind of structural
behavior, past and present, by the Aar massif is the
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
following. Its Mesozoic cover is thin and of shallowwater character (Hsü, 1994) and rests on a Variscan
basement exhumed from 10–12 km depth (Frey et
al., 1979). Currently the basement-cover contact
rises from 6–7 km below s.l. in the foreland to up to
4–5 km above it on the Aar and Aiguilles Rouges
massifs (Pfiffner et al., 1990).
The Figure 13 transect at the Aar massif is a relatively simple one. A much more complex situation
occurs further west, on a Mont Blanc–Dent
Blanche–Sesia transect. Although almost certainly
in need of further development, Figure 14 is an
attempt, using the same motions and principles, to
illustrate how the observed dispositions could have
been brought about. This may be corrected, as necessary, by comparison with the recent account of
observed end-result relationships given by Pleuger
et al. (2007) and by taking into account the independent(?) motions apparently responsible for the
construction of the Western Alps, which we discuss
in the next section. Before discussing the time of the
backthrusting and its effects south of the Insubric
Line, we now deal with a very important intervening
motion—Oligocene dextral motion along the Insubric Line, prior to its exhumation.
Oligocene WNW-directed motion of Adria and the
AA roof
We referred earlier to the need for well over 140
km of Oligocene WNW-directed thrusting in the
Western Alps. Ceriani et al. (2001) found that
thrusting at the Penninic front (Fig. 11) changed
abruptly to WNW in mid-Oligocene after being Nor NNW-directed previously, but Ceriani and
Schmid (2004) preferred a date of change at the
Eocene–Oligocene boundary. To achieve this thrusting, some sort of decoupling from the rest of the Alps
is required. The thick S-dipping shear zone at the
Simplon Line (Mancktelow, 1985) then deeply
buried seems, with its eastward continuation, the
Centovalli Line, to offer a possible link from the
Insubric Line to the western Penninic front.
But it appears that this motion was not confined
to the Insubric Line (IL) but affected superficial
structures much further north, i.e., the AA roof was
moved too, implying that exhumation and its detachment at the Insubric Line had not yet begun. Steck
(1980) reported top–W deformation in the central
Alps, particularly near the SW end of the Aar
massif, and Laubscher and Bernoulli (1982) saw
such deformation as widespread. A final top–W
deformation of the western Helvetic nappes is
717
present to some distance north of the Simplon Line
latitude (Dietrich and Durney, 1986; Ramsay,
1989). The Bergell Granite, dated at 32–30 Ma
(base Eocene = 33.9 ± 0.1 Ma; Gradstein et al.,
2004), and centered 13 km N of the IL, suffered top–
W deformation during intrusion (Trommsdorff and
Nievergelt, 1983) and a “tail” of it was dragged
some 48 km dextrally parallel to the IL, with additional shearing between it and the IL that Lacassin
(1989) thought might bring the total dextral motion
to well over 200 km. The 25 Ma Novate leucogranite, which intrudes the seemingly intrusive Gruf
migmatite, adjacent to the Bergell, suffered no
top-W deformation (Trommsdorff and Nievergelt,
1983). Thus the top-W and dextral motion was confined to a less than 5 m.y. interval in the early to
mid-Oligocene, which suits most of the Western
Alps observations but emphasizes the speed of the
motion.
To the north of Bergell and 35 km from the IL,
the detailed structure of the Schams nappes
(Schreurs, 1993) yields, on close inspection, evidence of top-W deformation that preceded the top-S
shown in Figure 12. Fifty kilometers farther west of
Bergell, a gneiss-limbed west-vergent schuppen
zone at Cima di Gagnone has infolded butter
mélange with mafic-ultramafic floaters (Trommsdorff, 1990; Pfiffner and Trommsdorff, 1998). The
axes of these folds curve downward as the IL is
approached southward (field demonstration by V.
Trommsdorff, 2002), linking this deformation to the
dextral motion along the IL.
Figure 15 offers a major, but speculative, extrapolation of these results, encouraged by the way in
which the AA roof further east was evidently split
apart in the Miocene to form the Vienna Basin
(Royden, 1985) presumably facilitated by the butter
mélange cushion beneath it. The ultimate goal is to
explain the otherwise unexplained post-Gosau/postDanian ca. 30 degrees. CW rotation (relative to
Adria/Apulia/Africa) (or about 45° relative to
Europe) of much, but not all, of the NCA, the exception being the southerly part of the NCA west of the
Inn valley (Mauritsch and Frisch, 1978; Tollmann,
1986, p. 155; Mauritsch and Becke, 1987; Flügel et
al., 1987; Heller et al., 1989; Channell et al., 1992;
Channell, 1996; Thöny et al., 2006). The idea is that
the NCA would be caught in a dextral shear zone
between a westward-moving AA complex, to the
south, and a stationary European margin, resulting
in the rotation of NCA strips from roughly N–S to
NE–SW, a conclusion reached by Thöny et al.
FIG. 14. Closure tectonic effects of the epeirogenic rise of the External massifs: Mont Blanc–Dent Blanche–Ivrea transect. Speculative section, farther west than Figure 13,
drawn in an attempt to illustrate an intermediate stage in the development of more complex relationships such as are present there. As in Figure 13, the assumed fundamental
mechanism is that the external massifs, as former blocks (with thicker crust) within a block-and-basin European autochthon, are considered to have risen epeirogenically in
selective metamorphic response to the deep-crustal heating that ensued after being overthrust by the Austroalpine roof. A generalized “NW-SE” direction of relative motion is
shown because it has not been possible here, at the intersection of the main-range Alps with the Western Alps, to distinguish the Europe-to-SSE-directed closure of the former
and the later (see Fig. 15) top-WNW closure that completed the Western Alps. Superimposed upon this, but not portrayed here, will have been the major differential epeirogenic
rise, close to the Insubric–Canavese Line, of formerly deepest-extending components of the Penninic assemblage, e.g., Monte Rosa, followed by further top-ESE back-thrusting
as Adria moved westward still farther, indenting the Western Alps structures at depth (see section on “Construction of the Western Alps,” p. 721). For locations of the principal
units involved, see Figure 11.
718
MILES F. OSMASTON
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
(2006) whose date for the motion, but on stratigraphic evidence, is precisely the same as ours,
namely intra-early Oligocene (Rupelian—33.9–
28.4 Ma). This would involve sinistral motion at the
contacts of adjacent strips, which is supported by
the detailed fault pattern of the NCA (Ratschbacher
et al., 1991; Linzer et al., 1995, 2002; Thöny et al.,
2006). Some of the abundant NE-striking sinistral
faulting in the NCA may have been mistakenly
attributed to the later (late Oligocene–early
Miocene) eastward lateral wedging, or “extrusion,”
of the Austroalpine roof of the Eastern Alps under
broadly N–S compression (Ratschbacher et al.,
1991). This would be consistent with the suggestion
(Ratschbacher et al., 1991; Kurz, 2006) that as the
Tauern Window lies at the geometrical apex of the
extrusion its incipient exhumation may have
enabled the extrusion by rupturing the AA roof. In
that case the NCA lay north of the field of lateral
extrusion activity (Fig. 15). But there are two major
problems.
The first is that since Ratschbacher (1986) and
Neubauer (1987) reported top–WNW deformation
in the other parts of the AA nappe structure, especially east of the Tauern Window, much further corroborative work has appeared, extending also to the
western edge of the main AA nappe system in the
Margna nappe–Oetztal nappe area (e.g., Schmid and
Haas, 1989) (east of the Schams area discussed
above, see Fig. 11), all of which has been interpreted by these authors as being of early-to-midCretaceous age. Much W-directed slicing in the
latter area (the Margna nappe lies just to the east of
Bergell) has long been recognized (e.g., Trümpy,
1975, 1980; Hsü, 1994, and references therein) but
the matter at issue is the age. From what has been
said earlier in this paper, neither the Cretaceous
LCT ages nor the ages of pseudotachylites at nappe
bases, both of which can relate to the progress of
STE, has any certain bearing on the age of the deformation at issue. Laubscher (1983), indeed, thought
the presence of pseudotachylites and mid-Cretaceous mylonites at the base of the Silvretta-Oetztal
sheet could mark the detachment of their original
substrate and we concur. It is clear, however, (M.
Wagreich, pers. commun., 2005) that a Cretaceous
age for at least some of the top-W thrusting further
east is secured by overlying sedimentation. But to
regard it as a regional top-W deformation in the
early or mid-Cretaceous would raise the problem of
how it could have been done without the involvement of a regional lid, attached to another plate. To
719
be sure, such a situation could occur more locally
during imbrication, if the direction of so doing were
appropriate. Is this a bit of evidence for the early
east-directed subduction, STE, and west-directed
imbrication that must have been the forerunner to
the Oligocene events (see p. 721) that built the
Western Alps?
The second problem (M. Wagreich, pers. commun., 2005) relates to evidence of NCA paleosurface development, thought to be of Eocene–early
Oligocene age, and apparently inconsistent with an
early-to-mid-Oligocene disruption of the Eastern
Alps of the kind proposed here.
As Figure 15 indicates, it is envisaged and is
mechanically essential that the top-WNW motion,
regardless of how much of the AA roof was involved,
took place when the IL and the similarly dextral
Pustertal-Gailtal (PGL) fault lines were in alignment. This means that the Giudicaria offset was due
to NNE-directed thrusting and compression of the
Eastern Alps, post-dating the top-WNW motion and
causing the widespread lateral, eastward extrusion
of the roof, as proposed by Ratschbacher et al.
(1991; see also Kurz, 2006), which they thought to
have begun in the late Oligocene. The IL must
therefore have had a WNW strike also, so we infer
that its current E–W strike resulted from more
northward (s. l.) underthrusting (“backthrusting” in
the sense of Figs. 12, 13, and 14) of it by the deep
Adria crust in the east than in the west, as seen in
the deep seismic results (Frei et al., 1990; Schmid et
al., 1997).
This linking of the IL to the PGL does not necessarily imply that the latter marks the exhumed back
wall of the subduction zone in the manner we have
inferred for the IL. In fact the geology suggests that
this particular (faulted) boundary lies at some
distance to the N of, and not exactly parallel to, the
PGL. This suggests that the PGL fault line was a
cut-through rupture to enable the WNW-directed
forces to reach the weakness offered by the IL, and
thus to build the Western Alps. In this way, perhaps,
we can dispose of one of the disparities between the
Central and Eastern Alps underlying the dichotomous interpretations referred to by Kurz (2006).
Origin of the WNW-directed motion. To provide a
major, broadly westward, movement of Adria such
as this may seem hard to reconcile with the picture
of broadly N-S relative motion of Africa and
Europe, varying from compression in the east to
former extension in the west. Osmaston (2007),
however, showed that this movement of Adria might
720
MILES F. OSMASTON
FIG. 15. The crunch sequence of inferred plate motions following the initial NNW(?)-directed closure at encounter
with Europe in the Eocene. Double-shafted arrows indicate early Oligocene WNW-directed motion of Apulia, driving
compression of the Western Alps, and sliding the still-attached Austroalpine lid above the Western Alps, (most of?) the
Central Alps, and an uncertain amount of the Eastern Alps (see discussion in the text). Strips of the Northern Calcareous
Alps appear to have undergone the indicated type of rotations, causing its many NE-striking sinistral faults (not shown,
but see Thöny et al., 2006). At this time the Insubric and Pustertal–Gailtal lines were co-linear as shown, but unexposed
beneath the AA-Adria roof, with dextral slip motion at depth. Succeeding (late Oligocene or after) motions shown by
light arrows (in generalized form for “lateral extrusion” of Eastern Alps). Abbreviations: En = Enntal fault and continuations. The sinistral movement on the Engadine fault (E) may relate to the additional northward deflection of the
Insubric Line near its eastern end. Other abbreviations: A = Aar massif; CL = Canavese Line; CSL = Centovalli–
Simplon Line; DM = Dora Maira; IL = Insubric Line; GL = Giudicaria Line; NCA = Northern Calcareous Alps; P =
Pelvoux-Belledonne massifs; PGL = Pustertal–Gailtal Line; TW = Tauern Window.
reasonably be the indirect result of that convergence of two cratons—Arabia and Russia—in the
Caucasus area. He investigated the double proposition that the Earth currently has a both a two-layer
mantle, bounded at the base of the upper mantle,
and that cratons have tectospheric keels that extend
nearly that deep. In such circumstances, the separation or mutual approach of cratons must set up
major horizontal movements of upper mantle material. Previously, Osmaston (2006) found that in
respect of the opening of the largely craton-surrounded Eurasian part of the Arctic Ocean, the
intra-Eocene N-directed compressive movements
of Greenland (another cratonic keel) were consistent with being driven by the mantle flow needed to
put beneath the Arctic Ocean.
In the converse situation due to Caucasus
convergence (Osmaston, 2007) the upper mantle
material between the deep keels, even while they
were far apart, appears to have been expelled west-
ward, impinging upon the tectospheric keel of
the Moesian microcraton that underlies much of
Romania, driving it westward. Such indenter motion
of Moesia has long been inferred both from studies
of the South Carpathians and from the existence, on
its western side, of a major calc-alkaline arc, running N-S all the way from the Apuseni Window
(within the Carpathian arc) to Bulgaria. This has
Late Cretaceous granites recently dated at 92–78.5
Ma (von Quadt et al., 2005). Burchfiel (1980)
reported the presence of a dextral shear, linking the
South Carpathians with the PGL to the Giudicaria
offset, and which he inferred to have been active in
the Miocene. Accordingly we propose that this was
the source of the force that drove Adria westward to
construct the Western Alps in the Oligocene, before
the Giudicaria offset had taken place. We surmise
that this motion was also what started Apennine
construction at about this time, rather than eastward
convergence by the Corsica-Sardinia microplate.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
That this flow-force is still active, but is obstructed
by the Giudicaria offset, is evidenced by the current
compressive seismicity across the GL (Viganò et al.,
2007), by a broadly similar degree of seismicity in
the Western Alps (Lardeaux et al., 2006), and perhaps by the concentrated deep seismicity (80–180
km) of the southeast Carpathian Vrancea area (e.g.,
Sperner et al., 2001).
Backthrusting—Part 2. Timing and products
On the basis of Figure 13, the failure of thrusting
to pass over the rising Aar massif should mark the
start of backthrusting. The last and major movement
(>35 km) on the Glarus thrust was apparently
early(?) Miocene (Schmid, 1975). After that any further N-S closure would have produced backthrusting in the Southern Alps. South of the IL, in the
South Alpine foredeep, deposition of massively conglomeratic material from the Central Alps began
about 30 Ma (Di Giulio et al., 2001), culminating in
the Gonfolite Lombarda, of latest Oligocene age,
containing Bergell boulders in a position 60 km
dextrally offset from that body (Schmid et al., 1989).
This is unlikely to limit the total dextral motion on
the IL to 60 km, but probably records the amount of
motion after the Bergell Granite became exposed by
erosional removal of its AA roof.
These data strongly suggest that, although late
back-thrusting was not until later, massive exhumation at the IL, leading eventually to mechanical
decoupling of the AA roof from its Adria root, was
already under way by the middle Oligocene. That
will seem entirely reasonable when we come to consider the origin of the many Periadriatic intrusions,
of which Bergell is one.
Construction of the Western Alps
In continuation of our initial discussion of UHP
metamorphism in the Western Alps (p. 711) and in
light of our preceding analysis of the Oligocene
westward motion of Adria and of the subsequent
back-thrusting at the Insubric Line, we now discuss
briefly the resulting structural history of the Western
Alps. Strictly, the Western Alps, with their recognized UHP features (Hermann, 2003; Groppo et al.,
2007), dating from the first identification of metamorphic coesite (Chopin, 1984) and regarded by
Compagnoni and Rolfo (2003) as making it the most
representative UHP belt in the world, ought to have
been the main objective of this paper. But, for the
purpose of setting up a comprehensive Alpine
model, they lack the wide variety of constraints
721
provided by the main Alpine range. They also suffer
from being, in some respects, a constructional afterthought relative to the main Alps events and must
structurally interfere with the latter at their junction.
For an instructive map of the Western Alps showing the main structure and metamorphic zonation,
see Compagnoni (2003). The admirable deep structural transects across the belt at the latitude of the
UHP Dora Maira (DM in Fig. 11) massif (Lardeaux
et al., 2006) show the following, encouraging the
view that the constructional sequence, but in a
broadly E-W direction, was of the same character as
that which we have inferred above for the main
Alpine range.
The lithospheric mantle of Adria has intruded
westward at a deep level below DM, producing
back-thrusting like that seen along the Insubric
Line, but clearly when exhumation of the former was
already far advanced. Westward from the top of the
DM dome—probably, it seems, an incompletely
deciphered Penninic-type assemblage—to the
Briançonnais Penninic front virtually all structural
contacts dip westward and have top–W sense of
shear. This is consistent with an E-directed Figure
10 mode of assembly, upon which has then been
imposed a strong E–W exhumation gradient, consistent with the deepest corner of the section being in
the east.
How much of the top-W shearing originated
during E-directed subductive construction, and how
much arose during the Oligocene movement of the
AA roof, is an interesting question. In this regard,
the Monviso Ophiolite, one of the traditional
Piemont ophiolites, now rests on the western summit
of the DM dome and appears to have been thrust
westward over the top of it from a position consistent
with its having been the first unit subducted and
therefore from an ocean-crusted forearc, as in Figure 10. This relationship closely replicates that of
the Zermatt-Saas Fee ophiolite to the Monte Rosa
dome (e.g., Trümpy, 1980). Lodgement would have
been beneath an AA-equivalent roof (now all lost by
erosion except for the Dent Blanche klippen) but
which was a western part of Adria. These features
imply that we need to have had a margin extensively
STE-undercut by east-directed subduction, possibly
simultaneously (although we have few flysch controls here) with the broadly S-directed system operative along the main Alpine front. A resolution of this
problem is offered in the conclusions (B9, p. 726).
Compagnoni (2003) was at pains to emphasize
that there are two U-Pb age clusters of high-pressure
722
MILES F. OSMASTON
metamorphism (65 Ma and 35 Ma) to be explained,
which he accepted as representing distinct events
but of the same tectonic kind, the former being at
only half the pressure of the latter. The clockwise
paths confirm that temperatures were higher during
exhumation. The 35 Ma date seems a trifle too early
to correlate with the 28–32 Ma age we have inferred
above for the main westwards motion of Adria, but
the motivation to do so is strong. The 65 Ma date is
from the Sesia-Lanzo Zone, apparently a near-root
imbricate portion of the AA roof (Fig. 14) that got
pushed around the interface downbend into a back
wall position, between the AA roof and the Penninic
assemblage. This would explain the lower pressure.
If, as suggested earlier, this structure was a variant
on the back-thrusting theme associated with the AA
being obstructed by the rising Pelvoux-Belledonne
external massifs, this pre-encounter date is a
mystery. Is this a rare (unique?) U–Pb record of STE
chilling of the roof?
The Pannonian Basin
Although not strictly a crunch matter, it is convenient to treat the Pannonian Basin here as an example of the versatility of the STE process, as to end
result. Argand (1922) saw the floor of the Pannonian
Basin as an AA “mega-nappe,” a conclusion supported by Horváth (1993). The question is: What
lies beneath it? Instead of basement-stretching, with
its severe geometrical and causation difficulties, to
produce the thin overall crust, high heat flow, and
subsidence (Horváth and Royden, 1981; Horváth,
1993) we favor, as mentioned earlier, that the STEundercut AA margin did not encounter European
continental crust until the edges of the present
basin, so it rests on former ocean floor, much as we
suggested earlier for the continental units resting on
the floor of the Tyrrhenian Sea.
A natural association of high heat flow with subsidence occurs when the heat being lost was already
there. Horváth and Royden (1981) rejected this
because the subsidence rate seemed too high to
relate to a cooling plate of relevant age, on the standard cooling-ocean-plate model. Wood and Yuen
(1983), however, showed that the Clapeyron slope of
the spinel peridote to garnet peridotite phase change
is reversed when it occurs at lower temperature,
which would, if the phase change occurs within the
thickness of the plate, as it certainly does with the
thicker ocean plate model (Osmaston, 2006) discussed earlier, explain the apparent completion of
cooling subsidence of oceanic plates at around 60
Ma. In turn this means that oceanic plate cooling is
actually not substantially complete by 60 Ma.
Beyond that point the subsidence caused by continuing thermal contraction would be negated by the
volume increase produced by the phase change. This
could open a door to interpreting the Pannonian
Basin as an AA sheet resting on former Tethyan
floor, because of the high subsidence rate, in relation to heat loss, engendered by phase-change
action within the underlying mantle.
Proceeding on this basis, the oceanic crust and
sediment, trapped under the allochthon and heated
by the restriction of its heat loss, might have provided the source for the andesitic and then rhyolitic
volcanism which ensued in the Miocene (Boccaletti
et al., 1976; Ebner and Sachsenhofer, 1995), without having to invoke subduction. Adoption of our
thick-plate picture also probably provides, without
appealing to delamination, for the even younger
mafic volcanism of the area, because of the new
intraplate magmagenetic mechanism which then
becomes available (Osmaston, 2005d).
Self-Reheating and Exhumation of the
Alps HP-UHP Assemblage
The Periadriatic intrusions (Fig. 11), mainly
Biella (syenite), Bergell, Adamello, Rieserferner,
are mostly granitic, and have ages which cluster
around 32–30 Ma (Schmid et al., 1989; von Blanckenburg et al., 1998). These are seen here as direct
evidence of sufficient self-reheating at the deepest
part of the HP-UHP assemblage for the occurrence
of melting, fluxed by the availability of hydrous
fluid, mostly from the matrix of the butter mélanges
liberally incorporated into the assemblage. The fact
that these ages post-date those recorded in the HPUHP metamorphic rocks now exposed at the surface
indicates that heating at even deeper levels was still
in progress, and it may show that melting had not yet
developed there at the time the recovered rocks
were at their final metamorphic state. Adamello is
interesting in being a succession of intrusions that
young from 42–30 Ma NNE-wards over 35 km
towards the IL (Del Moro et al., 1983). Our expectation has been that the back wall of the STE-cut subduction interface, now represented by the IL fault,
would have been very steep but southward-dipping.
This progression would be consistent with the onset
of fusion at a very deep level near that sloping
interface, progressing with time to shallower levels
in the assemblage. The resultant positions of these
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
intrusions, wholly to the south of the HP-UHP metamorphic rocks, suggests that most of any melt generated at the source levels had avoided migrating
upward in those rocks and being evident among
them. The fact that the Insubric fault now dips
steeply to the north is due to the northward “intrusion” by Adria at depth, already discussed. Von
Blanckenburg et al. (1998) were unable to discuss
the 25 Ma Novate leucogranite (Trommsdorff and
Nievergelt, 1983), for their hypothesis of exhumation by slab drop-off, but its position, together with
the antecedent Bergell Granite, north of the Insubric
Line seems consistent with coming from a yet shallower level as heating progressed.
Nor does slab drop-off have much to offer for the
high-temperature Barrovian-type metamorphism, of
which, in the Alps, the Lepontine uplift lying west of
the Adula nappe is a well-known example (e.g., Frey
et al., 1980) and whose wide occurrence in exhumed
UHP belts is emphasized by Parkinson and
Maruyama (2004), However, it fits well into the
wholly thermal exhumation favored here, perhaps
with a degree of advective transport by rising melt.
The importance of the aqueous and potentially
carbonatic fluid imported in the matrix of butter
mélange as an aid to retrogression during exhumation has been mentioned. That the presence of such
fluid is a primary control on the formation of
microdiamond under UHP conditions (Korsakov et
al., 2004; Korsakov and Hermann, 2006; Sokol and
Pal’yanov, 2008) confirms its presence in many
UHP belts. This makes it essential to recognize this
retrogression in all but the driest objects when
trying to assess the deepest point in the burialexhumation cycle and the date at which it was there.
Very importantly, Parkinson (2000) was able to show
from the chemistry of inclusion residues in garnet
cores in whiteschist from the Kokchetav UHP massif that retrogression had been so complete that this
was the only remaining proof that the material, as
butter mélange matrix, had indeed been to the same
depth as the “boudins” it encloses. If that were not
the case generally, the entire thesis upon which this
paper is based, that of butter mélange assembly and
the significance of its location, must fall apart. The
recognition of partial melting of the assemblage,
seen here as stemming perhaps from near the greatest depth within it, may mean that the only depth
and date record remaining when we see such material is the history after leaving those conditions. This
underlines, even more than the arguments previously given, that U-Pb zircon ages, for example,
723
however carefully determined, may well not tell us
the moment at which exhumation even began, let
alone the time of original burial (Hermann et al.,
2001).
Some Initial Conclusions
Conclusions are given at this point to provide a
summary background for the comments on other
HP-UHP belts that follow. Terminology used here is
that proposed under this heading in a previous section (p. 704).
A. The general case
A1. Objective review of how subduction actually
works and of its effects upon the upper plate leads to
the recognition of STE as a major plate tectonic process. Such review includes and necessitates recognizing (i) the hanging wall/upper plate as having
mechanical integrity, (ii) that subducting plate
downbend is by escalator-like step-faulting, especially where the subducting plate is young and STE
is active, (iii) that subduction places the upper plate
into compression generated by ridge push, and (iv)
the opening of back-arc basins as being due to the
anticlastic curvature effect (Fig. 7), not to slab pull.
A2. Recognition of STE raises the possibility
that the resulting undercut margin may undergo
imbrication, the slices being either subducted or
stacked together at the surface. This could, by repetition if necessary, progressively bring an initially
intra-oceanic subduction system to a continental
margin (Osmaston, 1992). So although a majority of
subduction zones are now at continental margins, we
should be prepared to envisage that an oceaniccrusted forearc may have been an early feature of
their evolution and of those now extinct.
A3. STE during open-ocean subduction is the
essential preliminary to constructing a HP-UHP
metamorphic assemblage: it creates margins from
which slices, both thin enough and pre-chilled,
become available for subduction to a downbend site,
itself prepared within the upper plate by the STE
process. Without the compressive upper-plate environment engendered by the step-faulting mode of
subduction downbend and the presence of ridge
push, such imbrication would not occur.
A4. STE generates, wholly subterraneously, a
diagnostic fluid-overpressured butter mélange that
then becomes incorporated into the assemblage. Its
floaters, often referred to as “boudins” on account of
their shapes when seen after HP-UHP exhumation,
724
MILES F. OSMASTON
although boudinage is not a likely feature of their
inclusion, are a focus of much attention in HP-UHP
belt studies. Its commonly pelitic matrix is an
important source of water at the time of exhumation
(see A17 below) and perhaps in assisting the kinetics of UHP metamorphism at the time of subduction.
A5. Although STE is responsible, overall, for the
undercutting of margins for many hundreds of kilometers, the butter mélange of interest here is that
which is generated by STE, not at the main downbend, but at an up-front inflection of the interface
which develops late in the sequence. So its floaters
derive from the shallower lithologies encountered
there and the mélange is smeared/buttered along the
ensuing flat part of the interface, where it enters our
picture when this is imbricated.
A6. In passing to sub-Moho levels, STE may
incorporate ultramafic floaters into the mélange.
These may differ in composition, according to
whether they were culled from suboceanic or from
subcontinental mantle, and in mineralogy according
to the depth from which they came.
A7. Until an inflection develops, the “flat” part
of the interface is not lubricated by butter mélange
and the jerky nature of the subduction process may
produce pseudotachylite there.
A8. Serpentinite-matrix mélange is formed when
the smearing of butter mélange along the mantle
hanging wall of the interface serpentinizes it and
provides floaters that may be impressed into it.
A9. The most significant parts of HP-UHP belts
are constructed by open-ocean subduction before
encounter, crunch, or collision, their rock slices,
including ophiolitic ones, being derived wholly from
the upper plate.
A10. Any ophiolitic slices were part of the
forearc when subduction began. They therefore differ from standard MOR-generated crust for two
reasons; (a) they were formed at a margin, not in
mid-ocean, probably in the presence of active sedimentation; or (b) STE may have cut away their lower
part.
A11. Each successive imbrication event of an
STE-undercut margin generates a flysch, and these
form an up-front assemblage on, but not in front of,
the margin as its foreshortening proceeds. The
paleontological age of any such flysch, where still
preserved, provides important evidence of constructional timing, and thereby of the early history of the
belt concerned, which has proved remarkably
elusive to discover by radiometric dating methods.
A12. Because imbrication and subductive transport begins at the thinner part of the STE-undercut
margin, and progresses towards its thicker part, subductive assembly transposes the paleogeographic
order, any ophiolitic slices being the first to enter
the assemblage.
A13. This transposition, together with the prior
STE-undercutting of the margin, requires the
subduction of a lot of ocean floor. At recognized subduction rates, therefore, construction of a typical
UHP assemblage must take many tens of m.y. to
accomplish.
A14. Subductive assembly at depth generates a
broadly triangular cross-section of material, the
deepest point being against the back wall of the subduction zone and reaching greater depth as subcretion assembly proceeds. It is this part that exhumes
the most, rupturing the overlying “roof” or “lid,”
formed by the remaining part of the STE-undercut
upper plate, and exposing the highest pressure
material. It is not the location of a suture, even
though it may display a concentration of ophiolitic
slices, serpentinite mélange, and butter mélange.
A15. The back wall of the former subduction
zone constitutes a deeply extending structural break
in the tectosphere of the upper plate that becomes
prone to transcurrent use during crunch, even
before exhumation has ruptured the “lid” projecting
from it.
A16. Exhumation is primarily thermal, confined
to when subduction has stopped or greatly slowed,
and is due to self-reheating of the assemblage, especially its deepest part. Consequently, the surface
above the deepest part of the triangular assemblage
tends to exhume the most, and perhaps the most rapidly, recording radiometric cooling dates that differ
from those recorded in lower pressure exposures.
A17. As a chemical substance, water from the
butter mélange matrix may speed exhumation in
four ways: it assists retrogression to lower density
minerals; it fluxes melting, melts being of lower density than the parent solid; these melts or the water
itself may advect heat to shallower level and speed
up reheating there; and the water itself is of lower
density when in a free interstitial state, or it may
recombine to form lower density minerals than those
from which it came.
A18. The volume increases at depth, associated
with the thermal mode of exhumation, may achieve
dimensional expression in any of the three directions, according to the operative constraints. Horizontal expression of that dilatation, at the level
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
concerned, inevitably results in extension of all
superincumbent material and is not to be seen as
extensional collapse.
A19. To the extent that they are constrained
laterally, those same volume increases will be
expressed predominantly in the vertical direction,
creating gneiss domes and enhancing rates of exhumation. This vertical extension at depth is likely the
cause both of the thinness of subducted slices when
they re-emerge and of the boudin-like shapes of
mélange floaters they display (see A4 above).
A20. Dating by low-closure-temperature methods of parts in the structure that have escaped the
high exhumation temperature can yield precious
information about the time of original chilling by the
STE process, before the imbrication and HP-UHP
construction began. Dates of this type can, of
course, also be set in the later stages of exhumation
cooling from higher temperature, so the choice of
significance needs to be carefully made.
A21. Apart from those referred to in A20, above,
most radiometric dates, by whatever method, record,
not the first moment of HP or UHP metamorphism of
that particular bit of rock, but various stages in the
course of its thermally driven exhumation and subsequent cooling. Although the reheating that causes
the latter may be said to begin at the moment subduction ceases (collision), and therefore at the same
moment all along a belt, it may proceed rapidly or
slowly at different places, according to the constitution of the subducted assemblage, giving disparate
dates of radiometric setting.
A22. The great strength of zircon and garnet as
casings for included relics means both that UHP relics are armored against reacting to lower pressures
during thermal exhumation, and that low-pressure
relics in inherited zircons are protected from the rise
in environmental pressure during HP-UHP assembly under cool conditions.
A23. The low temperature of the assemblage
resulting from the STE process and continued
subduction, suggested by preserved occurrences,
globally, of lawsonite eclogites (Tsujimori et al.,
2006), ensures that the inherited zircons themselves
remain unreactive until their temperature rises
during exhumation. Consequently, the pressure
recorded by inclusions in any mantle overgrowth
they might acquire at this time will not record the
maximum pressure experienced, nor the mantle
itself the date at which collision initiated the reheating. That slow, early part of the exhumation
sequence remains unrecorded by these techniques,
725
as does the preceding duration spent at UHP. The
maximum pressure/depth which that involved is not
wholly inaccessible, however; observations from
rare inclusions (Ye et al., 2000) indicate that it may
have exceeded 200 km, even at the levels now
exposed.
B. The Alps example
B1. The essential HP-UHP part of the construction of the Penninic belt north of the Insubric–
Guidicaria–Pusterta–Gailtal line was accomplished
during broadly south-directed Cretaceous openocean subduction (present coordinates).
B2. The preparatory undercutting of the Adria
margin by STE is recorded in the widespread Cretaceous ages determined by low-closure-temperature
methods. This STE advanced the subduction downbend surface carved into the upper plate to a position now occupied by the southern margin of the
Penninic belt. The imbrication events subsequently
affecting that undercut margin are recorded in the
mainly Cretaceous flyschs of the foreland flysch belt.
B3. The Penninic realm was an integral part of
Adria, not separated from it by a Piemont or South
Penninic ocean, but forming a northern continuation
of the Austroalpine realm. The only ocean lay in a
North Penninic position and, even then, the oceanic
rocks observed were not a part of that ocean, but of
an oceanic-crusted forearc. The almost total lack of
ophiolite now in this position shows it was nearly all
subducted, much of it being lodged at the position of
the STE-advanced interface downbend.
B4. Those (upper) levels in the Austroalpine
stack that lack a mélange-buttered underside
appear to have been too far south for buttering to
reach from the up-front inflection site at which it
was generated. At this thicker part of the undercut
margin, particularly in the east, imbrication and the
lodgement of slices to the HP-UHP site at the interface downbend appears to have been mechanically
inhibited, resulting in the preservation of the Upper
Austroalpine roof over a smaller subducted assemblage, causing less exhumation.
B5. Encounter with the European continental
rise in about mid-Eocene (~40 Ma) was followed by
a series of major intracrunch tectonic motions
because the residual Austroalpine roof was still
extensive. Provisionally, these motions were: (i)
north- or NNW-directed compression, not necessarily continuous, until (ii) major (probably >150 km)
and rapid WNW-directed motion of Adria at close to
30 Ma (Early Oligocene) with dextral motion on the
726
MILES F. OSMASTON
not-yet-exhumed Insubric–Pustertal–Gailtal Line
(then in alignment) and transporting the stillattached Austroalpine roof in top-WNW motion over
much of the Central and Western Alps;4 and (iii)
NNE-directed compression, probably starting in the
early Miocene, that produced the Giudicaria offset,
triggered major eastward lateral escape in the Eastern Alps, and intruded the Adria plate northward
below the Insubric Line (Fig. 15), together with the
top-S back-thrusting of the Austroalpine “lid”
across it (Figs. 12 and 13).
B6. The occurrence of the Periadriatic granitic
intrusions in the 42–25 Ma interval and their
proximity to the Periadriatic Line, together with the
Barrovian metamorphism of the Lepontine Dome,
are secure indications of major reheating of the
deepest tip of the UHP assemblage, stacked against
the back wall of the former subduction zone.
B7. A thermal epeirogenic uplift mechanism
(Table 1), associated with the thermal release, but
non-escape, of metamorphic water in deep crustal
rock, is consistent with the behavior of the external
massifs and the core of the Tauern Window after
being covered by the allochthon. This suggests that
these were all former blocks of mature continental
crust within a block-and-basin region of the European autochthon, and not primarily the product of
selective thrusting, although their uplift exposed
them subsequently to that.
B8. Construction of the Alps in the way outlined
raises a paleogeographic problem—we need enough
ocean to subduct, to execute the required amount of
STE, and then transport the imbricate slices one by
one. Detailed discussion is out of place, but the following suggestion is made. A Triassic or older part
of Tethys is not denied by the lack of ophiolites of
that age. By analogy, the floor of the Western Mediterranean is nowhere exposed, yet the basin, with a
floor widely accepted as being of oceanic character,
undoubtedly exists, and closure would not necessarily expose it either. The records of Early Jurassic
normal faulting widely present in the Alps are not
necessarily progenitors of ocean floor production,
but may well be its after-effects caused by the rapid
thermal subsidence of the adjacent recently generated ocean plate, laterally attached/welded to the
4The
extent of this transport in the Eastern Alps is uncertain.
This motion was apparently driven by the westward expulsion
of deep upper mantle from between the converging cratonic
tectospheres in the Caucasus region, impinging upon the
tectosphere of the Moesian microcraton (Osmaston, 2007).
margin. Pursuing this reasoning, the scarce or shallow-water Upper Triassic deposition that preceded
this at these locations could record the marginal
thermal uplift that had resulted from juxtaposition of
hot new oceanic plate.
Once MOR action is under way, the amount of
plate available for subduction is not confined to the
original continental separation, but the ridge push
developed may cause it to be exceeded several-fold.
For mid-Jurassic ocean floor to have been incorporated into the passive North Penninic margin before
subduction began suggests that a fresh addition to
the ocean occurred at that time, close to the former
margin. The many Balkan ophiolites of just this age
appear to record that same episode of floor production. Their sedimentary cover, where preserved,
abundantly records that they too were generated in
close proximity to a continental margin. That they
were not subducted likewise but were launched into
supracrustal positions marks another variant on the
plate tectonics theme. The deep-water substrate
upon which most of them rested first does suggest
the presence of basins of Triassic or earlier oceanic
character.
B9. A final paleogeographic question concerns
the provision of broadly east-dipping subduction
and STE in preparation of the western side of Adria
for the construction of the Western Alps and, probably, the Apennines too. In that sense, this paper has
not fully provided for the UHP metamorphism specific to the Western Alps. We speculate that this
subduction, of Cretaceous age, was driven, as in the
Oligocene (Conclusion B5(ii)), by cratonic convergence in the Caucasus, evident in the Cretaceous
westward motion of the Moesian microcraton. This
mechanism does not work without N-S convergence,
but it does not necessitate that the main cratons
were close together, so it could have been simultaneous or closely interleaved with the southward
subduction of Conclusion B1.
B10. The hypotheses of slab pull, slab drop-off,
post-orogenic extensional collapse, and delamination do not seem to have useful parts to play in
respect of the wide range of Alpine observations
discussed in this paper, though still only a fraction
of the available observational base.
Other HP-UHP Metamorphic Belts
The intention here is, relatively briefly, to compare the foregoing conclusions to prime features in
other belts. Note first that microdiamonds have been
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
reported from Kokchetav, Dabie Shan, North
Qaidam, Western Gneiss Region, Erzegebirge,
Sulawesi, and perhaps Rhodope (Ogasawara, 2005)
and that there were conditions for their former presence in the Western Alps (Hermann, 2003; Groppo
et al., 2007). Partial melting has been reported from
WGR (Labrousse et al., 2002) and Kokchetav
(Hermann et al., 2001; Korsakov et al., 2004); and
from the Alps, if the Periadriatic intrusions, discussed above, are accepted as such. The continuing
discovery of more HP-UHP belts must surely add to
this list.
Kokchetav, northwest Kazakhstan
The comprehensive set of accounts (Parkinson et
al., 2002) suggest butter mélange everywhere, showing little sign of sialic metamorphic slices. None of
the recorded lumps lie outside the probable size
limits for mélange floaters, so one gets the impression that the butter may have been squeezed upward
out of the deeper terrane by ongoing tectonic events
in the region. For this reason, any systematic
arrangement of the structure seems to be lacking. A
related consideration is that its Eocambrian age has
given time for much more complete exhumation than
has been available for the other belts discussed in
this paper, thus exposing deeper and less familiar
parts of our envisaged constructed triangular section
of subducted materials (section D in Fig. 10). These
may also have achieved more mobility than is seen
at shallower levels, thus losing the assembly relationships, The reports of interstitial melting seem to
support this. Hermann et al. (2001) stressed that the
presence of in situ partial melting implied loss of the
preceding age record.
Dabie-Sulu, eastern China
Since the reporting of coesite from Dabie Shan
(Wang et al., 1989), this SE-striking UHP belt, of
which the eastern part (Sulu) was offset to the northeast along the Tan-Lu fault by some 500 km at least
50 m.y. after the ~240–230 Ma date of UHP metamorphism (Wang, 2006; Webb et al., 2006), this belt
has attracted a great deal of attention. Eastern
Dabie, near the Tan-Lu offset, attains a cross-strike
width of ~150 km, similar to that of the WGR. As
noted above, microdiamonds have been reported.
Shutian et al. (2005, p. 1207) recognized “decompressional partial melting and intense retrogressive
metamorphism.” The belt is complicated by multiple suturing (see below) and by widespread intrusion of granitoids at 130–120 Ma, which overprinted
727
metamorphism and interrupted structural boundaries. Our remarks are based mainly on Ratschbacher et al. (2006) and Wang et al. (2008) and
references therein.
Superficially the belt occupies a suture between
the Sino-Korean craton, with basement ages in the
2.6–2.0 Ga bracket, and the Yangtze craton to the
south, characterized by 700–800 Ma ages. Sections
by Suo et al. (1999, 2002) show a concentration of
mafic-ultramafic bodies and steep structure in the
north, with a presumed (by them) suture at the steep
Balifan–Mozitan–Xaoitian (BMX) fault in an adjacent position closely analogous to the Insubric Line
of the Alps. The structure is broadly domed with
steep northern limbs and flattening southern limbs
with SW-directed vergence onto the Yangtze plate.
The UHP rocks, with two lower-pressure slices on
top, are draped off the dome, which is migmatitic
and widely exposes Cretaceous granite, but was
epeirogenically active very much earlier, suggesting
that this was the epeirogenic action of the MCC of
the subducted Yangtze craton. The up to 150 km
cross-strike width and pre-doming horizontal disposition of those three slices, although now only ~10
km thick, suggests widely spaced imbrication of a
thick STE-undercut margin, so that the thick slices
would not flex enough on reaching the interface
downbend. The pressure differences between slices
are too great for their present thicknesses, so substantial late-stage lateral extension is implied.
All rocks of the structure display Yangtze-characteristic ages, so Hacker et al. (2004) inferred that
the UHP belt had been constructed from the leading
edge of that plate. Our Alps model, however, would
prescribe that the entire SW-vergent structure was
derived from an STE-undercut former extension of
the Sino-Korean craton (Adria-equivalent) that lies
on the north side of the BMX fault.
Previous suturings. This apparent difficulty is
resolved as follows. Ratschbacher et al. (2006), following earlier work (Liou et al., 1996; Hacker et al.,
1996), identified the presence of a long and narrow
intervening Qinling Microcontinent (QM) along the
north side of the belt. This exhibits Yangtze craton
ages and shows arc-magmatism evidence of being
sutured with the Sino-Korean craton at about 400
Ma. It follows that the QM separated northward
somewhat earlier from the main body of the Yangtze
craton and suggests that it could have been the
ocean then created which was closed to form the
Dabie-Sulu UHP structure. STE and imbrication of
the southern edge of the QM would explain its
728
MILES F. OSMASTON
narrowness, in the manner discussed earlier in
respect of the “ribbon” shape of the Alpine
Briançonnais terrane.
But the south side of the QM is marked by the Ndipping, 5–10 km wide “Franciscan-like” Huwan
HP mélange, with eclogites and other floaters typical of butter mélange from a continental margin,
going back to Ordovician in age, and a probably
end-Carboniferous (300 Ma) age of metamorphism.
So this was the closure that undercut and narrowed
the QM, but it cannot be the one responsible for the
Dabie-Sulu UHP. The steep Mozitan–Xiaotian fault
that bounds the south side of the Huwan mélange
records the differential exhumation of the Dabie
HP-UHP structures (Insubric analogue). For those
structures we need a second continental strip (call it
the Dabie-Sulu Microcontinent—DSM) to have
been separated from the main Yangtze craton. If it
sutured with the QM at 300 Ma, we may reason that
closure then transferred to the south side of the
DSM; so the Dabie UHP belt would have been the
product of STE and imbrication in the 60 m.y. period
300 Ma to ~240 Ma, comparable with what we have
inferred for the Alps. A basis for this reasoning is
that, in the author’s ongoing re-analysis of the
British Caledonides, it has emerged that a temporary obstruction at the moment of post-Iapetus
encounter caused closure to switch briefly to the
Moine Thrust Zone in the northwest at ~430 Ma.
This would only have happened if convergence had
been driven by forces external to the region, not by
slab pull. The Yangtze–Qinling closure may have
been similarly driven.
Evidence for a roof. Wang et al. (2008) record that
the Dabie belt, particularly the eastern segment,
displays a late-stage ubiquitous top-NW (i.e., alongstrike) deformation. This suggests the movement of a
former roof, as in the Alps and (see below) in the
WGR. But there is no similarly striking fault that
could have guided this motion, as did the Insubric
Line. Wang et al. (2008) attribute the deformation to
southeastward extrusion of the HP-UHP assemblage, but fail to note that this is indication that a
non-extruding roof was present. Imposition of this
deformation on the top of the dome shows that
doming beneath the roof was already well under way,
so it was probably this general thermal dilatation,
combined with the vertical squashing action of the
rising dome, that caused the lateral extrusion and
the above-mentioned thinning of the slices. If this
halved the slice thicknesses from 60 km total, the
UHP requirement yields a roof thickness of at least
40 km.
Dating the UHP metamorphism. Leech et al.
(2006) reported zircon ages for the Sulu component
that are ~10 m.y. younger than those for Dabie,
which raises plate kinematic difficulty if both sets of
ages are regarded as recording a supposedly brief
period of UHP burial. We concluded (A21–A23, p.
725) that such age determinations must actually
record points in the course of thermally induced
exhumation, and do not record the moment of transport to UHP conditions, which, for constructional
reasons, would have been very much earlier. This
reheating will proceed at different rates, according
to the radiogenic content and bulk of the most
deeply subducted material, which could well vary
along strike in an orogen. Further evidence that
appears to tell a similar story is that Xu et al. (2006)
reported from the Sulu belt 40Ar/39Ar plateau ages in
biotite that are as much as 50 m.y. younger for UHP
rocks than for HP rocks.
Franciscan Complex and Coast Range Ophiolite,
western California
A special case? Here both the main concentration
of peridotite bodies and the highest metamorphic
pressures occur adjacent to or near the Coast Range
Thrust (CRT) which, in the north, borders the east
side of the terrane and was interpreted as a suture
(Ernst, 1970, 1971). So let us see how well our Alps
scenario, with its new view of the function of the
Insubric Line, can be applied. The Coast Range
Ophiolite (CRO), up-edged into view as the east wall
of the CRT, is here in the role of the Alpine Adria–
Austroalpine roof, beneath which the Franciscan
Complex (FC) appears to have been assembled by
what (with current orientations) can be described as
“eastward” subduction. There is no report of UHP
metamorphism being present. Figure 16 is a sketch
of the proposed scenario underlying the remarks
that follow.
Important differences from the Alps situation
are: (1) that the roof, having been made of oceanic
crust, was much thinner, even at its near-CRT root,
than the continental-crusted roofs in the other belts
studied in this paper; and (2) that the slices that
built the FC lacked the thickness given by continental content, so many more slices would have been
needed for it to have been built downward into the
UHP domain. A further consequence of (1) is that
slices tucked up beneath the thin roof, itself with
low radiogenic heating, would have undergone far
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
729
FIG. 16. Proposed scenario in which the CRO and Franciscan assemblage were built, based in part on the present
layout of the Lau (back-arc) basin, in which (see text) crustal construction has changed, with time, from back-arc chemistry to MORB chemistry. The northern (Elder Creek) expression of the CRO was constructed with back-arc chemistry
but underwent final intrusion by MORB dikes (Shervais et al., 2004), so is inferred to have been constructed close to
the transition in the basin. The CRO–Franciscan “plate” was docked against the Sierra Nevada by the action of subduction zone b, with a suture now beneath the Great Valley. Continued activity of subduction zone a, this being the rapid
young-plate subduction of the Farallon plate (Atwater, 1989), then established a fresh STE and downbend advance
sequence from an interface level below the Franciscan, progressing beneath the Sierra Nevada and then away far to the
east, successively executing the Sevier thrusting (~95 Ma) and Laramide events (75–45 Ma).
less metamorphism than is habitually expected to
distinguish slices in this situation. These reasons
make it doubtful that the FC should continue to be
regarded as “the world’s type subduction complex”
(Wakabayashi and Dumitru, 2007), venerable
though that distinction has been.
The combined CRO-FC has been much sliced by
younger dextral movements associated with those on
the San Andreas fault, but original relationships are
best preserved in the north, where we will consider
the CRO evidence of Shervais et al. (2004) at Elder
Creek. Some justification for concentrating there is
that in light of the foregoing remarks the great
majority, if not all, of the widely distributed exposures of CRO elsewhere (see map in Shervais et al.,
2004) may in fact not be too big to have been floaters
incorporated into proto-Franciscan butter mélange
and therefore derive, not from the back-arc position
inferred by these authors for Elder Creek (and incorporated into Fig. 16), but from a totally different,
forearc position.
Remote construction. The layout crudely indicated in Figure 16 recognizes that the CRO, of
Jurassic age, was not formed in a forearc position
relative to the Sierra Nevada, but (on paleomagnetic
evidence) far away, in the open ocean (Hopson et al.,
1996). It also recognizes the findings of the Ocean
Drilling Program in the Lau Basin. This basin documents a 17o CW angular separation by the TongaKermadec arc from the Fiji Islands block (Sager et
al., 1994), so the spreading is segmented (Hawkins,
1995; Parson and Wright, 1996). In essence, the
northern segment has therefore opened the most and
has MORB-chemistry floor in the middle, but backarc chemistry at the margins, whereas the more
southerly segment still exhibits mixed chemistry at
its spreading center. The reason for this transition is
unclear. One possibility is that it might record the
ingress of Pacific mantle from the north (Hawkins,
1995).
Application of this layout can explain why
Shervais et al. (2004) found the Elder Creek CRO
was mainly of back-arc chemical construction, but
was finished off by MORB dike intrusions. They
liken this to the sequence widely found in hotemplaced ophiolites, but in those cases, where
present, the transition is in the reverse direction,
from MORB to supra-subduction chemistry (e.g. in
730
MILES F. OSMASTON
Oman; Coleman, 1981), and there is no sign in the
CRO of either a hot mantle tectonite—commonly
the biggest component of such ophiolites—or a
metamorphic sole, so such a correlation seems
insecure.
Docking, consequences, and timing. Figure 16
also implements the suggestion (Hopson et al.,
1996) that the resulting CRO-FC assemblage was
then drawn toward—and, in late Jurassic, docked
beneath—the Great Valley, by the pre-existing, separate, subduction under the Sierra Nevada (SN). The
trench for the subduction whereby it approached the
SN arc must lie beneath the axis of the Great Valley
and is well marked by a positive gravity anomaly
(Hopson et al., 1996). The subduction (zone a in Fig.
16) beneath the FC-CRO was then reactivated and
may have linked up with the former sub-SN interface (b in Fig. 16), causing SN magmatism to jump
eastward at this time (Bateman and Dodge, 1970)
and enabling it to proceed by rapid STE nearly to
Colorado, to carry out the Sevier-Laramide
sequence of events, as discussed earlier in this
paper. A hiatus in Zone a subduction during the
approach to this docking is suggested by the exhumation responsible for massive erosion of the CRO
at Elder Creek and the deposition of up to >1000 m
of the Crowfoot Point Breccia before being overlain
by Great Valley Group sediment of Kimmeridgian
age (Shervais et al., 2004).
Ernst (1988) singled out the Franciscan as being
unusual in having little thermal overprint. This
might be because of the low radiogenic content that
is to be expected.
As to timing, high-grade “exotic” blocks
immersed in a low-grade matrix were considered by
Mattinson (1988) to have been transported into the
FC matrix, but Wakabayashi and Dumitru (2007)
reported overlapping 40Ar/39Ar ages from blocks
and matrix and concluded that a common age and
origin is likely, retrogression of the matrix having
been undetected. These authors, moreover, concluded that the overall duration of high-pressure
metamorphism was at least 80 m.y. This is consistent with our finding for the Alps.
Sanbagawa terrane, central Japan
Here the Median Tectonic Line (MTL) looks like
a clear counterpart of the Insubric Line, even having
undergone strike-slip motion, and having highest
metamorphic pressure near the MTL and a generally
S-vergent structure (Ernst, 1971). The belt itself is
of Cretaceous age. Banno (2004) reported metamor-
phism pressures approaching the UHP field and the
presence of ophiolitic bodies in the terrane, ranging
up to a size of 5 km × 1 km, but not evidently concentrated near the MTL. Terabayashi et al. (2005)
mapped eclogitic mafic-ultramafic bodies surrounded by epidite-amphibolite-facies schists—
butter mélange?—and evidence for differential
exhumation of slices. But the structural arrangement
near to the MTL appears to have undergone some
modification, perhaps by backthrusting of terranes
now lying well to the north, and this is confirmed in
central Shikoku by Masago et al. (2005). The basal
zone of the Tamba Nappe group, northwest of Kyoto,
(author’s observations, 1992) betrays top-S deformation (fish in pelitic schist) suggestive of having been
such a roof to N-directed subduction. The Mesozoic
ophiolite forming part of the Tamba group would
then represent the remains of an oceanic-crusted
forearc, which might have been the source of the
limited ophiolitic material in the Sanbagawa terrane
and explain where the bulk of it finished up. Such
structural behavior repeats the Alpine deformation
shown in Figures 12 and 13.
Maksyutov Complex (MC), southern Urals
This is a broadly W-vergent HP-metamorphic
complex of mid-Paleozoic (Late Devonian) age
backed on its east side by the Siluro-Devonian Magnitogorsk volcanic arc (MVA), but the main Uralian
collision was not until much later, with the suture
lying even further west, apparently confirming our
conclusion that high-pressure metamorphic belts
are constructed in the course of open-ocean subduction. The origin of that ocean may be marked by the
Middle or Lower Ordovician evidence of rifting
present along most of the Urals (Savelieva and
Nesbitt, 1996). The analysis of Hetzel (1999, and
references therein) emphasizes various points that
seem pertinent to an Alps-Maksyutov comparison,
so we base the following remarks upon it. The eastdipping “Main Uralian Normal Fault” (Hetzel,
1999) (MUNF), west of the MVA, marks the eastern
boundary of the much-exhumed HP MC, strongly
reminiscent of the Insubric Line. It is, moreover, a
“several kilometer-wide mélange zone with serpentinites, basalts” (one with an Early Ordovician K-Ar
age) (Hetzel, 1999, p. 578) and a wide range of other
lithologies. In more detail, this mélange zone comprises serpentinite mélange and butter mélange in
the expected relationship, with the former contituting the Upper Unit, next to the fault representing the
back wall/hanging wall, and the latter constituting
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
the Lower Unit, farther from it (Dobretsov and
Buslov, 2004).
Westward beyond some antiformal structure
close to and parallel with the MUNF (analogous to
that near the IL in the Alps) the MC exhibits W-dipping structural boundaries, consistent with the differential exhumation expected in our model, and it
then dips westward beneath the Suvanjak Complex,
here equated with the uneroded remainder of the
former (AA-equivalent) roof beneath which the MC
was constructed. At the western edge of the Suvanjak Complex lies the Zilair Flysch nappe, a position
analogous to the Prealps and other Alpine flysch, so
it probably marks the imbrication events from which
the slices of the MC derive. Those slices predominantly contain Proterozoic-aged protolith material,
which has been thought to imply that the MC was
constructed by driving the Russian craton below the
MVA, thereby regarding the MUNF as the suture
(e.g., Matte, 1998)—again reminiscent of the Insubric Line interpretations. The significance of the
Ordovician(?) rifting needed to produce the Uralian
Paleo-ocean is that it could easily have detached a
major piece from the Russian craton, and this was
the platform on which the MVA was built with its
well-known mineralizations, and from which the MC
was constructed from a former more-western part of
it, after eastward Siluro-Devonian STE and major
downbend advance beneath it.
This situation bears close similarity to that of
Dabie-Sulu, above. In both of these cases, the ocean
subducted would have been quite young, confirming
yet again that youth is no bar to subduction (making
ridge push essential) and that STE is then very
efficient.
Western Gneiss Region (WGR), southwestern
Norway
The tectonic frame of this UHP metamorphic belt
has proved especially hard to decipher, due, in large
measure, to the apparent lack of exposure evidence
from the other side of the system. We will show, however, that this can be overcome and, in the event, our
Alps-based model appears to have much relevance.
Just as in the Alps, the differing structural outcomes
along the once-continuous Caledonian–Northern
Appalachian belt offer valuable guidance upon the
understanding of events. Of necessity, the documentation is brief. In conformity with widespread usage,
we refer specifically to this complex of HP-UHP
rocks as the “WGR,” although “region” is normally
a less precise term.
731
An excellent description of the features of butter
mélange in a UHP terrane, as discussed in this
paper, is that of Smith (1980) from the northern part
of the WGR. The presence of garnet peridotites and
pyroxenites in the northern parts of the WGR
attracted the early attention of O’Hara and Mercy
(1963). Bryhni (1966) described the presence of
peridotitic bodies in linear belts or swarms, mostly
in the structurally steep northwestern part (Bryhni,
1989; Krogh and Carswell, 1995; Carswell et al.,
2006), parallel to the coast, which is also the high-P
side of the exhumation gradient (Cuthbert and Carswell, 1990). Griffin et al. (1985) recorded that
metamorphic T (and by implication, P also)
increases to the northwest and that the foliation of
gneisses wraps around ultramafic bodies of a “few to
several kilometers” size. A demonstration of butterlike behavior is the observation of mylonites surrounding “lenses or layers of undeformed to weakly
deformed Precambrian protoliths” (Skår and Pedersen, 2003, p. 937). Dobrzinetskaya et al. (1995)
recorded that the microdiamonds occur in pelitic
and migmatitic gneiss (former butter mélange
matrix?).
Pre- and post-Scandian events—Greenland
margin. This structural arrangement suggests that
the WGR was constructed beneath the overriding
Laurentian plate (Cuthbert et al., 1983, 2000; Carswell et al., 2003; Roberts, 2003), a view consistent
with the widespread evidence, some of it faunal, that
the Uppermost Allochthon of Norway is of Laurentian origin (Roberts, 2003), and which we favor also.
The Uppermost Allochthon, displaying metaperidotites particularly abundantly near the coast, extends
eastward to up to 100 km from the Norwegian coast
for much of the way from the WGR to the Lofoten
Islands (e.g., Bucher-Nurminen, 1988). To arrive in
this overriding position suggests that Greenland had
a W- or NW-dipping subduction margin at the time
of Iapetus ocean closure, just as was the case
beneath the Southern Uplands in the British Isles. A
close tectonic link with Greenland is made the more
apposite and indeed necessary in the light of the
extensive (200 km or more) WNW-directed late
Caledonian multiple thrusting at many points along
the entire length of the East Greenland orogen,
recently recognized in Higgins and Kalsbeek
(2004), among which, in Dronning Louise Land,
Strachan et al. (1992) found it to have been NW- to
NNW-directed. On the other hand, the WGR has
long been regarded (Gorbatschev, 1985, and references therein) as an uplift of Baltica basement,
732
MILES F. OSMASTON
FIG. 17. Index geological map of western Norway for features discussed in the text, redrawn from Austrheim and
Mørk (1988), with suggested new subdivision nomenclature for the “Uppermost Allochthon,” due to identifying the
WGR as its lowest member in the region. Abbreviations: MTFZ = Møre-Trøndelag fault zone; Val = Valdres nappe.
Approximate position of MTFZ is from Figure 4.3 of Evans et al. (2003).
seemingly supported by subsequent Proterozoic age
determinations, but these are not definitive, even if
no comparable-aged terrane is now to be seen on the
Laurentian side today. Our Alps model, together
with the other examples discussed in this paper,
teaches that the construction of an UHP assemblage, together with the subsequent erosion of the
roof beneath which that was done, has the effect of
removing from the map a major swath of the continent concerned.
With the Alps perspective in mind, noting especially the SW-striking sinistral Møre-Trøndelag fault
zone (MTFZ) system (Chauvet and Séranne, 1989;
Séranne, 1992) that runs offshore (Fig. 17) parallel
to the WGR structure (a possible Insubric equivalent), the former close presence of an Adria-equivalent continent seems entirely reasonable; but how?
Post-Scandian splitting of Greenland-Norway.
The distance between these two structural features,
if enforced by the present width of the continental
shelves, with the NE Atlantic closed (e.g., the reconstruction map of Stephens, 1986), is, at ~750 km, far
too great for tectonic reality. So it is necessary to
envisage that at the moment that Scandian collision
was complete, the plates, represented by the present
land exposures (or some approximation thereto),
were formerly closely juxtaposed, having been
followed shortly thereafter, probably in the Early
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Devonian, by a separation, or sequence of separations, forming crust that now comprises the present
shelves and forming an otherwise-unknown depocenter for all the erosion products. The fact that
Svalbard, displaying a transect of Caledonian structures (Harland, 1985), remained adjacent to NE
Greenland until the NE Atlantic began to open
means that the route of our supposedly Devonian
separation traversed the Barents Sea shelf also. That
would explain this gap in the Caledonian orogen, a
gap traversed by several Late Paleozoic rift structures (Faleide et al., 2008). Reconstruction with this
in mind suggests that Greenland’s southwestward
motion along the MTFZ may be marked on Greenland by the NE-trending line of Cretaceous basalts
that passes through Wollaston Foreland, central
East Greenland (Escher and Pulvertaft, 1995); this
movement, perhaps limited to <100 km, was followed by a much bigger northward (present Scandinavian coordinates) one, which would provide the
Svalbard-Tromsø(?) separation.
Intermediate Crust (IC) and thermal epeirogenic
action. The possibility of making such crust has long
been studied by the present author, based on recognizing that the MOR process behaves quite differently in the presence of heavy synseparation
sedimentation, of which in this case there would
have been plenty in the aftermath of the Scandian
collision. In support of this idea, we mention here
the following four points. (1) It has been noted
(Larson et al., 1972; Klitgord et al., 1974; Larter and
Barker, 1991) that active axis sedimentation, even
amounting to only a few tens of meters, inhibits the
generation of spreading-type magnetic anomalies,
so Osmaston (1977) suggested that this is due to
blocking the hydrothermal cooling systems, responsible on MORs for freezing in the anomaly before
the field reverses. (2) The Salton Sea Scientific
Drilling Project recorded intersill metamorphosed
syn-rift deposits with a corrected P-wave seismic
velocity of 5.95 km/s (Tarif et al., 1988), from which
it is clear that an assemblage of such sills and metamorphics would be seismically indistinguishable
from the canonical “stretched continental basement.” Interestingly, “One of the most spectacular
appearances of seismic reflections is the densely
laminated reflectivity in the lower crust that is
widely observed in the shallow crust of Phanerozoic
extensional areas” (Meissner et al., 2006, p. 81), but
here they opt for the only explanation they saw as
available to them—that of ductile stretching. (3)
The generally much greater plate thicknesses now
733
recognized (Osmaston, 2006, 2007) greatly prolong
the subsidence-prone timescale for constructed
areas, so sedimentary thickening may long continue.
(4) Because of its differing deep constitution and its
possibly shallower Moho the resulting “intermediate
crust” (IC) is likely to have a much lesser thermal
epeirogenic sensitivity, of the kind outlined in Table
1, than mature continental crust.
This latter point is of immediate significance
here. Superimposed upon the extreme exhumation
of the WGR, and explicit both in the present drainage system and in offshore sedimentation, is a Cenozoic southeastward tilting of the south Norwegian
landmass, which both Rohrman et al. (2002) and
Lidmar-Bergström and Näslund (2002) stress did
not begin until the Neogene, pointing to the 30 m.y.
delay since NE Atlantic opening began in the early
Eocene. Such a delay, in the thick-plate context of
Osmaston (2006), is to be expected of the lateral
spread of heat from the new plate at the continental
margin. The seismic LVZ at 16 km depth below the
WGR (Mykkeltveit et al., 1980), the drop in density
at 16 km depth below the Jotun nappe (Smithson et
al., 1974) and the concentration of uplift and a
strong (-80 mgal) Bouguer anomaly in the area
(Rohrman et al., 2002) all suggest the epeirogenic
action of that heat. The failure of the shelves, much
nearer to this heat source, to uplift in the same
manner, is a strong indication that they are of IC
character, as prescribed above.
Using the same reasoning, the eastward transfer
of heat at the time of the post-collision splittingaway of Greenland would have greatly accelerated
exhumation of the WGR and may account for the
westward-increasing metamorphism temperature
now exposed (i.e., more exhumation) and the eastward plunge of the Surnadal synform which crosses
it (Krill, 1985).
Appalachian-Caledonian correlation. To grasp a
picture of how Greenland and Norway came to be
juxtaposed, we jump briefly to farther south along
the orogen. In New England, Robinson et al. (1998)
distinguished on structural and faunal grounds that
the Taconian (broadly 450 Ma) and Acadian
(broadly 400 Ma) closures were at different sites,
with an intervening strip terrane which they named
“Medial New England,” the Taconian closure being
along its western side. A similar interpretation had
previously been offered by Osmaston (1988b) for the
Newfoundland segment of the orogen. There is obvious correspondence here to the functions of the
Qinling and Dabie-Sulu terranes in the Dabie-Sulu
734
MILES F. OSMASTON
example discussed above. In the author’s ongoing
re-analysis of the British Caledonides, the counterpart is identifiable as the Scottish Midland Valley–
Southern Uplands terrane. In this case, the Taconian
closure is replicated by placing more emphasis upon
closure along the Highland Border. Far to the North,
this closure is again seen in the Tromsø Nappe, comprising the uppermost slice of the Uppermost
Allochthon, and displaying UHP metamorphism but
with datings that cluster around 450 Ma (Corfu et
al., 2003; Ravna and Roux, 2006)—just right for the
Taconian closure.
In accepting the former existence of such a
medial terrane in this part of the orogen, we secure
two important constraints on the sources of nappes.
(1) It provides a source for allochthons displaying
intermediate faunal affinity, among which we refer
below to the Otta Conglomerate and the Støren ophiolite nappe (Bruton and Harper, 1985; Spjeldnaes,
1985). (2) Almost all the ophiolite fragments embedded in the Scandes are of an age—early Arenig
(Stephens et al., 1985)—closely correlative with the
Ballantrae Ophiolite of southwestern Scotland,
which exhibits a full range of chemistry (Smellie and
Stone, 2001). Ballantrae Ophiolite was on the northwest side of Iapetus (as generally recognized in the
British Isles), so it seems probable that the Scandian
ophiolites were too. The Støren ophiolite nappe satisfies both these constraints.
Scandian juxtaposition of Greenland and Norway.
To speculate about how this juxtaposition of a
Greenland already incorporating these items could
work structurally in the Trondheim-WGR-Jotun
region of Norway let us work backward (Fig. 18),
with an overall scenario corresponding to that
sketched in Figure 9. The lack of any magmatic
indication that subduction and STE had previously
been active beneath the margin of Baltica corresponds to its absence in the flat-slab Andes, the
Alps, and elsewhere when this was taking place, as
already discussed. But in the British Caledonides
the magmatism evidence for subduction at the
southeast margin of Iapetus in the Ordovician is
abundant.
The Støren nappe is the uppermost of several
that have been infolded into the strongly NNWvergent, WSW-striking Surnadal-Moldefjord synform that runs parallel to the WGR structure at some
distance from the coast (Gee et al., 1985; Krill,
1985; Rickard, 1985; Bryhni, 1989) (Fig. 17). This
means that, as suggested in connection with Figure
9, the WGR must have been inserted below the
existing nappe system and was exhumed through
them. An attractive alternative to this structurally
tricky proposition, but which we do not explore here
(or consider in the caption to Figure 18), is that all
of the other Trondheim nappes that underlie the
Støren in the Surnadal synform actually derive from
on top of the Laurentian side of the system. Suffice
to mention that hot-emplaced ophiolites often rest
upon quasi-synchronously emplaced thrust sheets.
The vergence of the synform and the presence of
Støren nappe correlative on the island of Smøla
(Roberts, 1980), just north of the MTFZ, means that
there was late back-thrusting to the NW across the
top of it and the MTFZ—another Alps analogy. But
the contact between these nappes and the underlying WGR is a metamorphic pressure jump, and
the widespread top-SW shear deformation in it was
imposed at ~395 Ma at a depth of 37 km (Terry et
al., 2000). This suggests that the top-SW deformation was imposed by SW-directed motion of the 37
km thick roof (Austroalpine analogue) beneath
which the WGR had been constructed, but which
was still attached to Greenland in a manner identical with that inferred for the Alps (Fig. 15). This roof
must have been eroded or removed southwestward
by the departure of the Greenland plate along the
MTFZ before the WGR was driven further beneath
the Norwegian margin and this back-thrusting
occurred. There is an intimate association between
the top-SW deformation of the WGR and the sinistral shearing at the MTFZ (Chauvet and Séranne,
1989; Séranne, 1992), just like that present in the
Alps.
That the Jotun nappe and the slices forming the
WGR could have been parts of the same STE-undercut margin is supported by their mutual content of
anorthositic massif materials in the 970–900 Ma
range (Lundmark and Corfu, 2008). The same
applies to the similar basement of the Valdres
nappe, which underlies it and extends southeastward from it, where a substantial cover is preserved.
We propose in Figure 18 that the thick Jotun basement nappe, together with the Valdres nappe (not
distinguished in Fig. 18), was originally a middle
part of this roof, somewhat like one of the Austroalpine sheets, but that backthrusting (as in Fig. 13),
when Greenland approached still further, subsequently cut out the distance between it and the
downbend-assembled WGR. The slices of the latter
had been imbricated from the more distal part of
that undercut margin, just as inferred herein for the
Penninic slices in the Alps, so the Jotun/Valdres
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
735
FIG. 18. Tectonics of Scandian closure in the Jotun–Western Gneiss region. This composite diagram illustrates, first
and separately, the envisaged STE-dominated structural evolution of the two margins: Greenland (left) and Norway
(lower right). The upper part then attempts to show what happened when the two came together in the Late Silurian. For
this purpose, envisage the Norwegian side to be stationary. The basis for the 437 Ma date for encounter is explained in
the text. The other dates arise from comparison with the author’s continuing analysis of the British Caledonides. EE’ is
the notional present exposure level of WGR; E’ is now at Faltungsgraben. The following late actions (crunch) are
inferred. 1. SE-vergent closure by continuing NW-directed subduction is halted when the southeast edge of JotunValdres is obstructed (by pile-up of Scandian nappes?). Continued plate closure—limited only when mantle “back
walls” of interface downbends meet—requires reversal of vergence and reactivation of older interface under Norway.
Much foreshortening of the remainder of Greenland’s STE-undercut margin occurs. The new (~NW) vergence, with
major transport distances, passing across the top of the then-WGR roof, not yet exhuming, is seen all the way up the
Caledonides of East Greenland (see text). 2. Southwestward initial separation of Greenland; MTFZ sinistral motion (395
Ma). WGR, with its roof still in place, is now decoupled from the Jotun proper but remains part of the Greenland plate,
so the roof is moved strongly to the southwest, giving the Solund detachment, not due to “extensional collapse.” Coupling
of this roof motion to the Trondheim nappes, which it had partly under-ridden, drew them or stretched them across the
top of the unroofed WGR, where the renewal of northwest vergence folded them into the NW-vergent Surnadal syncline.
The lid presumably broke from its Greenland root as separation increased and WGR exhumation (3) got under way. 3.
Differential exhumation of WGR. Maximum exhumation (microdiamonds) is on the northwest side, where the tip of the
triangular-section assemblage went deepest, so WGR, as a unit, is now strongly tilted to the southeast (though with some
differential behavior) reversing the original northwestward dips of the structure, as constructed beneath the roof. This
roof has been eroded completely except southeast of the SW-striking Faltungsgraben. On the northeast side of the WGR,
WGR exhumation has verticalized the southwest edges of the Trondheim nappes, conveying the former impression that
the WGR is an uplift of Baltica basement.
crust and that in the WGR were originally adjacent.
Southeast of the Valdres lies a series of thrust sheets
(Synnfjell, Aurdal, etc.) formed from a complex SEvergent imbricate zone (Nickelsen et al., 1985),
whose imbrication might be a record of that excision, in that its sedimentation extends no further
than mid-Ordovician and includes the proximal turbiditic (flysch-analogue?) Strondafjord Formation,
thought to have come from a thrust front. On this
view, the SW-striking structure defining the northwestern edge of the Jotun-Valdres assemblage,
known as the Faltungsgraben (Fig. 17), marks where
the Jotun nappe remained behind when the root
zone of the roof over the WGR departed southwestwards with Greenland. Speculatively, this motion
might have been responsible for the Bergen Arcs.
On this scenario the Jotun nappe should display
opposite vergences on its SE and NW sides, and this
is indeed the case (Milnes and Koestler, 1985).
Below its SE margin there is, as noted, clear
evidence of top-to-SE displacement, as would be
expected, both in order to get it there and, if preserved, during bottom-to-NW subduction and STE.
Meeting with resistance to any further such motion,
736
MILES F. OSMASTON
the continuing closure motion of the Greenland
plate, together with the deeply “engaged” WGR,
would have carried the WGR, with its roof, to the
southeast beneath the now-stationary northward
continuation of the Jotun sheet. The resulting NWvergent deformation is well recorded in the Synnfjell
area, at the exposed northern junction of the Jotun
Nappe and the least-exhumed southern edge of the
WGR (Gibbs, 1979; Banham et al., 1979). Indeed
the intervening material, at the Jotun base, resembles an ophiolitic mélange with a “schisty pelitic
matrix and very variable strain partitioning” (A.
D.Gibbs, pers. commun., 2003), corresponding well
to the butter mélange we would expect to have been
pasted, during STE (Fig. 18), onto the underside of
the (Jotun-related) roof below which the WGR was
constructed. The floaters/clasts in this mélange,
otherwise known as the Otta Conglomerate, contain
an Early Ordovician fauna of quasi-Laurentian
affinity (Bruton and Harper, 1985; Spjeldnaes,
1985), an affinity also displayed by that in the
Støren nappe on Smøla, mentioned above.
Dating WGR construction. Looking back still
further, in reporting U-Pb dates of 410–400 Ma for
the WGR, Carswell et al. (2003) discarded the older
(~425 Ma and older) LCT dates (Griffin and Brueckner, 1980; Kullerud et al., 1986) as irrelevant, just
as had been done in the Alps. We, in turn, here suggest that these dates may in fact record the Silurian
time of STE undercutting beneath the Greenland
margin, corresponding to the long-recognized period
of NW-directed subduction beneath the Southern
Uplands in the British Isles (Clarkson et al., 2001).
The dates suggested on Figure 18 are based upon
this. The 437 Ma (early turriculatus graptolite biozone) date for “encounter” given on Figure 18 is
based on the author’s ongoing analysis of the British
Isles sector of the orogen. But its correctness in the
Scandinavia–Greenland sector seems to be secured
by almost identical dates, although in differing
sedimentary facies, both for the onset of voluminous
easterly-derived sedimentation into the North
Greenland Trough (Hurst et al., 1983; F. Surlyk,
pers. commun., 1986) and for the first westerlyderived sedimentation onto elements of the Scandian nappe system (Bassett, 1985). This seems to
remove the much-invoked (misfit) tectonic significance of the Tornquist Line at this time, in agreement with the analysis of Berthelsen (1998) for the
part near its southeastern end. In this respect, the
author’s independent, detailed, and conservative
assessment of post-encounter closure in the British
Caledonides yields the surprising figure of >700
km; no longer so disparate from what evidently
occurred during the Scandian collision.
Achievements and problems. This new insight on
the WGR problem brings to light a close resemblance to our model for building the Alps and
appears to open up new opportunities for intraorogen correlation of Caledonian-Appalachian plate
movements. In the other direction, it offers a link
with the UHP metamorphism farther north on the
East Greenland margin (McClelland et al., 2006,
and references therein), although the problem(s)
raised by its apparently delayed exhumation and the
unusually high metamorphism temperature (>950o
C) now exposed (Gilotti and McClelland, 2007) will
need to be addressed. The former might represent
slow self-heating if the assemblage has low radioactive content, as in the Franciscan. On the other hand
(see Conclusions A18, A19, p. 724–725), a lack of
lateral constraint upon thermal dilatation at depth
could remove a major contribution to exhumation
rate, demanding a higher temperature to achieve the
required buoyancy.
Concluding Discussion
The preceding remarks on six other HP-UHP
metamorphic belts enable some more embracing
conclusions to be adduced than was appropriate in
our General (A1–A23) and Alps-specific (B1–B10)
sets of conclusions given previously (p. 723–726).
Conclusion A2 is confirmed, that the action of STE,
followed by imbrication of the undercut margin, has
the propensity, originally proposed by Osmaston
(1992), progressively to move any subduction zone
that formed with an oceanic-crusted forearc eventually to the continental margin position that is common today. This makes it a realistic option that a
subduction zone that has led to formation of an HPUHP belt may have started life with an oceaniccrusted forearc area. The concentration of maficultramafic material at the high-pressure side of the
structural layout of six of the eight belts studied
herein provides, by paleogeographical transposition, a record that this was indeed the case. In none
of these cases, therefore, was the ophiolitic material
“obducted” from the subducting plate—it was
within the upper plate from the start.
On the global tectonics scene, this appears to
shift the still-debated problem of how subduction is
started to considering the intra-oceanic domain
rather than the continent-ocean boundary. In view of
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
this uncertainty, it is clearly desirable to avoid having to invoke it too liberally, an aim partly achieved
here. As envisaged by Osmaston (1992), repeated
STE and imbrication reduces the need to invoke
multiple subduction systems to move arcs toward a
continental margin. In each of the HP-UHP belts
studied here, we have removed the identification of
the zone marked by HP-metamorphosed ophiolite as
the site of a former subduction zone.
We have adduced good evidence in the Franciscan, Dabie, and WGR belts for the former presence of a thick STE-generated roof, or lid,
abundantly evident in the Alps, beneath which the
HP-UHP belt was assembled, and through which it
is exhumed. It would probably be correct to assume
the same in other cases. Uniquely, the Ivrea Zone of
the Alps appears to be the only place where there is
exposure of continental crust and mantle to which
the roof was attached. In the Franciscan case, the
corresponding exposure is the easternmost CRO
emerging from beneath the Great Valley sequence.
The constitution-dependent crustal mechanism
for thermally induced differential epeirogenic uplift
identified in Table 1 bears valuably both upon the
exhumation of HP-UHP metamorphic assemblages
and on the behavior of autochthonous continental
block-and-basin layouts when overthrust by an
allochthon. The result in the latter case—major
epeirogenic uplift, though after a thermal delay—is
widely confirmed in the Alps and elsewhere, but is
at complete variance with previously embraced geophysical expectations based solely upon isostatic
loading. A fascinating and potentially far-reaching
implication is that, far from concealing the tectonic
history of the autochthon, overthrusting may actually bring about the display of its major features.
Butter mélange, named and identified in this
paper as an expected product of STE action, is a
frequent occurrence in the belts studied here, and is
commonly in the expected structural positions, so
appears to constitute a reliable diagnostic of the
former involvement of STE. A potentially distinguishing feature is that, in its unmetamorphosed
state, the clearly fluid-overpressured matrix displays having had a butter-like ability to accommodate major shear deformation without any of it being
concentrated at contacts with its related allochthon
or structural slice, seldom transferring any to the
floaters it encloses, unless and until its fluid overpressure is released during exhumation. The presence, though apparently rare, of slickensiding on the
floaters additionally secures that their detachment
737
was in a seismicity-inducing situation, not in a
relatively low pressure gravity slide/olistostrome
environment.
Butter mélange is constructed by the STE process—a previouly unrecognized mélange-making
process—in cool, low-pressure conditions. In that
state, when exposed in low-grade imbricate orogens,
it does not have outstandingly distinctive features,
so it may be remarkably common. But this study of
its occurrence in HP-UHP situations draws attention to its previously unrecognized tectonic significance. Serpentinite mélange, which also occurs in
these circumstances, appears to derive its floaters
from adjacent butter mélange and its matrix from
serpentinization of the hanging wall.
HP-UHP belts, if constructed in the same way as
the Alps, are constructed by subduction, not by collision, but it is the cessation of subduction, usually
but not essentially as the result of collision, that
brings about their exhumation. Most, but not all,
dating methods record stages in the latter, not the
moment of burial, so even by the same method, they
can vary considerably within the same belt without
reflecting upon the originating plate kinematics. In
the main-range Alps, the duration of the entire
process, from the commencement of STE to the
beginning of exhumation, was around 70 m.y. and is
unlikely to have taken much less in comparable
belts. For the Franciscan Complex, Wakabayashi
and Dumitru (2007) inferred that it was “at least 80
m.y.” For the Dabie orogen we have suggested 60
m.y. In the other belts studied here, adequate controls are, as yet, lacking, but there is hope of this
from the Zilair Flysch belt in front of the Maksyutov
Complex of the Urals.
A remarkable degree of affinity appears to exist
between the scenario assembled here from the
Alpine example and the features of six other welldescribed HP-UHP belts, although most exhibit
individual complications. While this adds further
strength to the Alps model of HP-UHP belt construction and exhumation, with its many observational constraints outlined here, a rather different
emphasis on the mechanisms we have identified
may have sufficed for the many HP/LT belts
recorded, for example, by Maruyama et al. (1996) in
which pressures did not much exceed 12 kbar.
It appears that one of the reasons “Why plate tectonics was not invented in the Alps” (Trümpy, 2001)
was indeed that the form of that paradigm was inappropriately framed for the job of Alpine construction. It is now capable of rewarding modification,
738
MILES F. OSMASTON
but inevitably that process still has further to go.
Principal among the new perspectives recognized or
endorsed here are: (1) the step-faulting mechanism
of subducting plate downbend; (2) that slab pull is
an inappropriate force for driving STE, resulting in
“flat-slab” subduction profiles, and may not be
needed to drive subduction anywhere (Fig. 6): (3)
the importance of thermally promoted petrological
changes of volume in the crust and mantle as principal agents in both vertical epeirogenic and horizontal extensional movements; and (4) the dynamical
consequences of tectonic plates being demonstrably
much thicker than previously imagined. Any one of
these might prove hard to fit in as a stand-alone
change to the present form of the paradigm. It is only
when they are put together and applied that the real
benefits begin to emerge.
Acknowledgments
This paper, the culmination of an investigation
started in 1978, as made clear in the Introduction,
could not have been constructed without the helpful
and sometimes serendipitously illuminating input
from countless individuals. At the risk of seeming
invidious by singling out a few, I would like to
record and thank the following.
In seismological matters: A. Dziewonski, T. Lay,
G. Masters, and E. R. Engdahl (particularly) for his
liberal provision of nearly 100 of his circum-Pacific
seismicity-cum-tomography transects of subduction
zones. For early discussions of Sevier-Laramide tectonics and magmatism: R. L. Armstrong and W. R.
Muehlberger. For petrological advice and rigorous
thermodynamic calculations: G. T. R. Droop. On
mélange matters, outside the Alps area: H. Williams, for field demonstration (Paleozoic Dunnage
mélange, Newfoundland); S. Miyashita, for field
demonstration and subsequent advice
(Homanosawa mélange, Hokkaido); C. Fassoulas,
for information on outcrop location (Arvi mélange,
Crete); and many individuals from the Universities
of Auckland and Christchurch, and from the New
Zealand Geological Survey, for information on outcrop locations (various New Zealand mélanges). For
non-Alps field demonstrations: north of the Median
Tectonic Line (Tamba nappes), A. Ishiwatari; tsunami deposits (southern Chile), M. Cisternas and B.
Atwater; Northern Appalachians–Newfoundland, P.
Robinson and H. Williams; Norwegian Caledonides,
D. G. Gee, T. Grenne, and A.G. Krill. On the Alpine
area: for written advice, J. Bertrand, J. Hermann,
and W. H. Winkler; for field demonstration and
advice, G. Gosso, C. Heinrich, Ö. Müntener, R.
Vissers, V. Trommsdorff, and R. Trümpy, particularly the two last.
For early correspondence: T. B. Andersen, Ø.
Engen, M. Krabbendam and C. D. Parkinson. For
reading and helpful advice on the manuscript at a
late stage: D. Cowan, A. G. Milnes, and C.-H. Tsai.
M. Wagreich read an early draft of the Alps part of
the paper and is thanked for multiple correspondence and advice. G. Gosso, W. G. Ernst, and B. L.
Isacks critically reviewed earlier versions of the
manuscript and are sincerely thanked for the
improvements to which they led, although responsibility for any error is wholly mine. W. G. Ernst is
also thanked for his editorial patience. Finally, I
thank the Library Staff of the Geological Society for
unfailingly willing support over many years, without
which this paper might never have been assembled.
REFERENCES
Abe, K., 1972, Mechanism and tectonic implications of
the 1966 and 1970 Peru earthquakes: Physics of the
Earth and Planetary Interiors, v. 5, p. 367–379.
Agee, C. B., 1998, Phase transformations and seismic
structure in the upper mantle and transition zone, in
Hemley, R. J., ed., Ultrahigh pressure mineralogy:
Physics and chemistry of the Earth’s deep interior:
Mineralogical Society of America, Reviews in Mineralogy, p. 165–203.
Allmendinger, R. W., Jordan, T. E., Kay, S. M., and Isacks,
B. L., 1997, The evolution of the Altiplano-Puna
Plateau of the Central Andes: Annual Review of Earth
and Planetary Sciences, v. 25, p. 139–174.
Ammon, C. J., Kanamori, H., and Lay, T., 2008, A great
earthquake doublet and seismic stress transfer cycle in
the central Kuril islands: Nature, v. 451, p. 561-568
[doi:10.1038/nature06521].
Ampferer, O., 1906, Über das Bewegungsbild von Faltengebirgen: Jahrbuch d.k.k. geologische Reichanstalt,
v. 56, nos. 3-4, p. 539–622.
Argand, E., 1922, La tectonique de l’Asie, in Proceedings
of 13th International Geological Congress, p. 171–
372.
Armstrong, R. L., and Dick, H. J. B., 1974, A model for
the development of thin overthrust sheets of crystalline
rock: Geology, v. 2, p. 35–40.
Asch, G., Schurr, B., Bohm, M., Yuan, X., Haberland, C.,
Heit, B., Kind, R., Woelburn, I., Bataille, K., Comte,
D., Pardo, M., Viramonte, J., Rietbrock, A., and Giese,
P., 2006, Seismological studies of the Central and
Southern Andes, in Oncken, O., Chong, G., Franz, G.,
Gieze, P., Götze, H.-J., Ramos, V. A., Strecker, M. P.,
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
and Wigger, P., eds., The Andes: Active subduction
orogeny: Berlin/Heidelberg, Germany, Springer, p.
443–457.
Atherton, M. P., and Petford, N., 1996, Plutonism and the
growth of the Andean crust at 9° S from 100 to 3 Ma:
Journal of South American Earth Sciences, v. 9, p. 1–
9.
Atwater, T., 1989, Plate tectonic history of the northeast
Pacific and western North America, in Winterer, E. L.,
Hussong, D. M., and Decker, R. W., eds., The Eastern
Pacific Ocean and Hawaii: Boulder, CO, Geological
Society of America, The Geology of North America
(DNAG), v. N, p. 21–72.
Austrheim, H., and Mørk, M. B. E., 1988, The lower continental crust of the Caledonian mountain chain:
Evidence from former deep crustal sections in western
Norway: Norges Geologiske Undersøkelse, Special
Publication 3, p. 102–113.
Bally, A. W., 1981, Thoughts on the tectonics of folded
belts, in McClay, K. R., and Price, N. J., eds., Thrust
and nappe tectonics: Geological Society of London
Special Publications, v. 9, p. 13–32.
Banham, P. H., Gibbs, A. D., and Hopper, F. W. M., 1979,
Geological evidence in favour of a Jotunheimen Caledonian suture: Nature, v. 277, p. 289–291.
Banno, S., 2004, Ultramafic rocks and associated eclogites of the Sanbagawa schists, Japan: International
Geology Review, v. 46, p. 693–704.
Barazangi, M., and Isacks, B. L., 1979, Subduction of the
Nazca plate beneath Peru: Evidence from spatial distribution of earthquakes: Geophysical Journal of the
Royal Astronomical Society, v. 57, p. 537–555.
Bassett, M. G., 1985, Silurian stratigraphy and facies
development in Scandinavia, in Gee, D. G., and Sturt,
B. A., eds., The Caledonian Orogen—Scandinavia and
related areas: Chichester, UK, Wiley, p. 283–292.
Bateman, P. C., and Dodge, F. W., 1970, Variations of
chemical constituents across the central Sierra Nevada
batholith: Geological Society of America Bulletin, v.
81, p. 409–420.
Beck, S. L., Gilbert, H., Wagner, L., Alvarado, P., Anderson, M., Zandt, G., Araujo, M., and Triep, E., 2005,
The lithospheric structure of the Sierras Pampeanas
region of Argentina [abs.]: Geological Society of America, Abstracts with Programs, v. 37, no. 7, p. 553.
Bernoulli, D., and Winkler, W., 1990, Heavy mineral
assemblages from Upper Cretaceous South- and
Austroalpine flysch sequences (northern Italy and
southern Switzerland): Source terranes and palaeotectonic implications: Eclogae Geologicae Helvetiae, v.
82, p. 287–310.
Berthelsen, A., 1998, The Tornquist Zone northwest of the
Carpathians: an intraplate pseudo-suture, in Sundquist, B., and Stephens, M. B., eds., Tectonics and
geological history of some Phanerozoic orogens (David
G. Gee honorary volume). GFF v. 120, p. 223–230.
739
Bertrand, J., Courtin, B., and Vuagnat, M., 1982, Elaboration d’un secteur de lithosphere océanique LiguroPiemontais d’après les données de l’ophiolite du Montgenèvre: Ofioliti, v. 7, nos. 2/3, p. 155–196.
Bertrand, J., Steen, D., Tinkler, C., and Vuagnat, M., 1980,
The mélange zone of the Col du Chenaillet (Montgenévre ophiolite) Hautes Alpes, France: Archives de
Science, Genève, v. 33, nos. 2-3, p. 117–138.
Biino, G. G., and Compagnoni, R., 1992, Very-high pressure metamorphism of the Brossasco coronite metagranite, southern Dora Maira Massif, Western Alps:
Schweizerische Mineralogische und Petrographische
Mitteilungen, v. 72, p. 347–363.
Bird, P., 1984, Laramide crustal thickening event in the
Rocky Mountain foreland and Great Plains: Tectonics,
v. 3, p. 741–758.
Boccaletti, M., Horváth, F., Loddo, M., Mongelli, F., and
Stegena, L., 1976, The Tyrrhenian and Pannonian
Basins: A comparison of the Mediterranean interarc
basins: Tectonophysics, v. 35, p. 45–69.
Bradshaw, J. D., 1973, Allochthonous Mesozoic fossil
localities in mélange within the Torlesse rocks of north
Canterbury: Journal of the Royal Society of New
Zealand, v. 3, no. 2, p. 161–167.
Brueckner, H. K., and van Roermund, H. L. M., 2004,
Dunk tectonics: A multiple subduction/eduction
model for the evolution of the Scandinavian Caledonides: Tectonics, v. 23, TC2004 [doi:10.1029/
2003TC001502].
Bruton, D. L., and Harper, D. A. T., 1985, Early Ordovician (Arenig-Llanvirn) faunas from oceanic islands in
the Appalachian-Caledonian orogen, in Gee, D. G.,
and Sturt, B. A., eds., The Caledonide Orogen–Scandinavia and related areas: Chichester, UK, Wiley, p.
359–368.
Bryhni, I., 1966, Reconnaissance studies of gneisses,
ultrabasites, eclogites, and anorthosites in outer Nordfjord, Western Norway: Norges Geologiske Undersøkelse Bulletin, v. 241, p. 1–53.
Bryhni, I., 1989, Status of the supracrustal rocks in the
Western Gneiss Region, S Norway, in Gayer, R. A., ed.,
The Caledonide geology of Scandinavia: London, UK,
Graham and Trotman, p. 221–228.
Bucher-Nurminen, K., 1988, Metamorphism of ultramafic
rocks in the Central Scandinavian Caledonides:
Norges Geologiske Undersøkelse, Spec. Publ. 3, p.
86–95.
Bucher-Nurminen, K., 1991, Mantle fragments in the
Scandinavian Caledonides: Tectonophysics, v. 190, p.
173–192.
Burchfiel, B. C., 1980, Eastern European Alpine system
and the Carpathian orocline as an example of collision
tectonics, in Banks, M. R., and Green, D. H., eds.,
Orthodoxy and creativity at the frontiers of Earth
Sciences (Carey Symposium): Tectonophysics v. 63, p.
31–61.
740
MILES F. OSMASTON
Butler, R. W. H., Matthews, S. J., and Parish, M., 1986,
The NW external Alpine Thrust Belt and its implications for the geometry of the Western Alpine Orogen, in
Coward, M. P., and Ries, A. C., eds., Collision tectonics: Geological Society of London Special Publications, v. 19, p. 245–260.
Cahill, T. A., and Isacks, B. L., 1992, Seismicity and
shape of the subducted Nazca Plate: Journal of Geophysical Research, v. 97, p. 17,503–17,529.
Caldwell, J. G., Haxby, W. F., Karig, D. E., and Turcotte,
D. L., 1976, On the applicability of a universal elastic
trench profile: Earth and Planetary Science Letters, v.
31, p. 239–246.
Caron, C., Homewood, P., and Wildi, W., 1989, The original Swiss flysch: A reappraisal of the type deposits in
the Swiss Alps: Earth Science Reviews, v. 26, p. 1–45.
Carswell, D. A., Brueckner, H. K., Cuthbert, S. J., Mehta,
K., and O’Brien, P. J., 2003, The timing of stabilisation
and the exhumation rate for ultra-high pressure rocks
in the Western Gneiss Region of Norway: Journal of
Metamorphic Geology, v. 21, p. 601–612.
Carswell, D. A., van Roermund, H. L. M., and Wiggers de
Vries, D. F., 2006, Scandian ultrahigh-pressure metamorphism of Proterozoic basement rocks on Fjørtoft
and Otrøy, Western Gneiss Region, Norway: International Geology Review, v. 48, p. 957–977.
Case, J., Chilver, A. H., and Ross, C. T. F., 1999, Strength
of materials and structures; 4th ed.: London, UK,
Arnold, 706 p.
Ceriani, S., Fügenschuh, B., and Schmid, S., 2001, Multistage thrusting at the “Penninic Front” in the Western
Alps between Mont Blanc and Pelvoux massifs: International Journal of Earth Science, v. 90, p. 685–702.
Ceriani, S., and Schmid, S., 2004, From N-S collision to
WNW-directed post-collisional thrusting and folding:
Structural study of the frontal Penninic units in Savoie
(Western Alps, France): Eclogae geologicae Helvetiae,
v. 97, p. 347–369.
Champagne, J. D., Sue, C., Delacou, B., Tricart, P.,
Allanic, C., and Burkhard, M., 2006, Miocene lateral
extension in the inner western Alps revealed: Tectonics, v. 25, p. TC3014 [doi:10.1029/2004 TC001779].
Channell, J. E. T., 1996, Palaeomagnetism and the palaeogeography of Adria, in Morris, A., and Tarling, D. H.,
eds., Palaeomagnetism and tectonics of the Mediterranean region: Geological Society Special Publication
105, p. 119–132.
Channell, J. E. T., Brandner, R., Spieler, A., and Stoner, J.
S., 1992, Paleomagnetism and paleogeography of the
Northern Calcareous Alps (Austria): Tectonics, v. 11,
p. 792–810.
Chapple, W. M., and Forsyth, D. W., 1979, Earthquakes
and bending of plates at trenches: Journal of Geophysical Research, v. 84, p. 6729–6749.
Chauvet, A., and Séranne, M., 1989, Microtectonic evidence of Devonian extensional westward shearing in
southwest Norway, in Gayer, R. A., ed., The Caledonide geology of Scandinavia: London, UK, Graham
and Trotman, p. 245–254.
Chemenda, A., Matte, P., and Sokolov, V., 1997, A model
of Palaeozoic obduction and exhumation of highpressure/low-temperatue rocks in the southern Urals:
Tectonophysics, v. 276, p. 217–227.
Chen, L., Booker, J. R., Jopnes, A. R., Wu, N., Unsworth,
M. J., Wei, W., and Tan, H., 1996, Electrically conductive crust in southern Tibet from INDEPTH magnetotelluric surveying: Science, v. 274, p. 1694–1696.
Chopin, C., 1984, Coesite and pure pyrope in high-grade
blueschists of the western Alps: A first record and
some consequences: Contributions to Mineralogy and
Petrology, v. 80, p. 107–118.
Christensen, D. H., and Ruff, L. J., 1983, Outer rise earthquakes and seismic coupling: Geophysical Research
Letters, v. 10, p. 697–700.
Christensen, D. H., and Ruff, L. J., 1988, Seismic coupling and outer rise earthquakes: Journal of Geophysical Research, v. 93, p. 13,421–13,444.
Cisternas, M., Atwater, B., Torrejón, F., Sawai, Y.,
Machuca, G., Lagos, M., and 9 others, 2005, Predecessors of the giant 1960 Chile earthquake: Nature, v.
437, p. 404–407.
Clarkson, E. N. K., Floyd, J. D., and Stone, P., eds., 2001,
The Southern Uplands Terrane; tectonics and biostratigraphy within the Caledonian Orogen: Proceedings of a 1999 conference in Edinburgh: Transactions
of the Royal Society of Edinburgh; Earth Science, v.
91, parts 3-4, p. 323–555.
Clift, P. D., Pecher, I., Kukowski, N., and Hampel, A.,
2003, Tectonic erosion of the Peruvian forearc, Lima
Basin, by subduction and Nazca Ridge collision:
Tectonics, v. 22, no. 3 [doi:10.1029/2002TC001386].
Coira, B., Kay, S. M., and Viramonte, J., 1993, Upper
Cenozoic magmatic evolution of the Argentine Puna—
a model for changing subduction geometry: International Geology Review, v. 35, p. 677–720.
Coleman, R. G., 1981, Tectonic setting for ophiolite
obduction in Oman: Journal of Geophysical Research,
v. 86, p. 2497–2508.
Coleman, R. G., and Lanphere, M. A., 1971, Distribution
and age of high-grade blueschist and associated eclogites from Oregon and California: Geological Society of
America Bulletin, v. 82, p. 2397–2412.
Compagnoni, R., 2003, HP metamorphic belt of the western Alps: Episodes, v. 26, no. 3, p. 200–204.
Compagnoni, R., 2006, The study of ultrahigh pressure
metamorphism (UHPM) in collisional orogens and its
contribution to the Earth sciences [abs.], in The evolution of Alps, Apennine, and Tyrrhenian basin: Contribution from petrology, geochemistry, and structural
geology. Workshop in memory of Piera Spadea,
Giuseppe Cello, and Lauro Morten; Cosenza, June 7–
9, 2007: Ofioliti v. 32, no. 1, p. 71.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Compagnoni, R., and Rolfo, F., 1999, Characteristics of
UHP pelites, gneisses, and other unusual rocks: International Geology Review, v. 41, p. 552-570.
Compagnoni, R., and Rolfo, F., 2003, The most representative UHP metamorphic units in the world: The Western Alps, in Carswell, D. A., and Compagnoni, R.,
eds., Ultrahigh pressure metamorphism: EMU Notes in
Mineralogy, v. 5, Chapter 2.1.
Coney, P. J., and Reynolds, S. J., 1977, Cordilleran Benioff
zones: Nature, v. 270, p. 403–406.
Corfu, F., Ravna, E. J. K., and Kullerud, K., 2003, A Late
Ordovician U-Pb age for the Tromsø nappe eclogites,
uppermost allochthon of the Scandinavian Caledonides: Contributions to Mineralogy and Petrology, v.
145, p. 502–513.
Couch, R., Whitsett, R., Huehn, B., and Briceno-Guarupe,
L., 1981, Structures of the continental margin of Peru
and Chile: Geological Society of America Memoir, v.
154, p. 703–726.
Coulburn, W. T., 1981, Tectonics of the Nazca plate and
the continental margin of western South America, 18°S
to 23°S: Geological Society of America Memoir, v. 154,
p. 587–618.
Cowan, D. S., 1985, Structural styles in Mesozoic and
Cenozoic mélanges in the western Cordillera of North
America: Geological Society of America Bulletin, v.
96, p. 451–462.
Cuthbert, S. J., and Carswell, D. A., 1990, Formation and
exhumation of medium-temperature eclogites in the
Scandinavian Caledonides, in Carswell, D. A., ed.,
Eclogite facies rocks: Glasgow, UK, Blackie, p. 180–
203.
Cuthbert, S. J., Carswell, D. A., Krogh-Ravna, E. J., and
Wain, A., 2000, Eclogites and eclogites in the Western
Gneiss Region, Norwegian Caledonides, in Carswell,
D. A., ed., Ultra-high pressure metamorphic rocks,
Lithos, v. 52, p. 165–195.
Cuthbert, S. J., Harvey, M. A., and Carswell, D. A., 1983,
A tectonic model for the the metamorphic evolution of
the Basal Gneiss Complex, Western South Norway:
Journal of Metamorphic Geology, v. 1, p. 63–90.
Dal Piaz, G. V., Cortiana, G., Del Moro, A., Martin, S.,
Penacchioni, G., and Tartarotti, P., 2001, Tertiary age
and paleostructural inferences of the eclogitic imprint
in the Austroalpine outliers and Zermatt-Saas ophiolite, western Alps: International Journal of Earth
Science, v. 90, p. 668–684.
Dallmeyer, R. D., Handler, R., Neubauer, F., and Fritz, H.,
1998, Sequence of thrusting within a thick-skinned
tectonic wedge: Evidence from 40Ar/39Ar and Rb-Sr
ages from the Austroalpine nappe complex of the Eastern Alps: Journal of Geology, v. 106, p. 71–86.
Dalziel, I. W. D., 1981, Backarc extension in the southern
Andes: A review and critical appraisal: Philosophical
Transactions of the Royal Society of London, ser. A, v.
300, p. 319–335.
741
Decker, K., 1990, Plate tectonics and pelagic facies: Late
Jurassic to Early Cretaceous deep-sea sediments of the
Ybbsitz ophiolite unit (Eastern Alps, Austria): Sedimentary Geology, v. 67, p. 85–99.
Del Moro, A., Pardini, G., Quercioli, C., Villa, I. M., and
Callegari, E., 1983, Rb-Sr and K-Ar chronology of
Adamello granitoids, southern Alps: Memorie della
Societa Geologica Italiana, v. 26, p. 285–299.
Dewey, J. F., 1988, Extensional collapse of orogens. Tectonics, v. 7, 1123–1139.
Dewey, J. F., and Bird, J. M., 1970, Mountain belts and the
new global tectonics: Journal of Geophysical
Research, v. 75, p. 2625–2647.
Dewey, J. F., Pitman, W. C., III., Ryan, W. B. F., and
Bonnin, J., 1973, Plate tectonics and the evolution of
the Alpine system: Bulletin of the Geological Society of
America, v. 84, p. 3137–3180.
Di Giulio, A., Carrapa, B., Fantoni, R., Gorla, L., and
Valdisturbo, A., 2001, Middle Eocene to Early
Miocene sedimentary evolution of the western Lombardian segment of the South Alpine foredeep (Italy):
International Journal of Earth Science, v. 90, p. 534–
548.
Dickinson, W. R., and Snyder, W. S., 1978, Plate tectonics
of the Laramide orogeny, in Matthews, V., III, ed.,
Laramide folding associated with basement block
faulting in the western United States: Geological Society of America Memoir 151, p. 355–366.
Dietrich, D., and Durney, D. W., 1986, Change of direction
of overthrust shear in the Helvetic nappes of western
Switzerland, in Platt, J. P., Coward, M. P., Deramond,
J., and Hossack, J., eds., Thrusting and deformation:
Journal of Structural Geology. v. 8, nos. 3–4, p. 389–
398.
Dobretsov, N. L., and Buslov, M. M., 2004, Serpentinite
mélanges associated with HP and UHP rocks in
Central Asia: International Geology Review, v. 46, p.
957–980.
Dobrzhinetskaya, L. F., Eide, E. A., Larsen, R. B., Sturt,
B. A., Trønnes, R. G., Taylor, W. R., and Posukhova,
T. V., 1995, Microdiamonds in high-grade metamorphic rocks of the Western Gneiss Region, Norway:
Geology, v. 23, p. 597–600.
Draper, G., and Bone, R., 1981, Denudation rates, thermal
evolution, and preservation of blueschist terrains:
Journal of Geology, v. 89, p. 601–613.
Ebner, F., and Sachsenhofer, R. F., 1995, Paleogeography,
subsidence, and thermal history of the Neogene Styrian Basin (Pannonian basin system, Austria): Tectonophysics, v. 243, p. 133–150.
Ernst, W. G., 1970, Tectonic contact between the Franciscan mélange and the Great Valley sequence, crustal
expression of a Late Mesozoic subduction zone: Journal of Geophysical Research, v. 75, p. 886–901.
Ernst, W. G., 1971, Metamorphic zonations on presumably
subducted lithospheric plates from Japan, California
742
MILES F. OSMASTON
and the Alps: Contributions to Mineralogy and Petrology, v. 34, p. 43–59.
Ernst, W. G., 1973, Interpretative synthesis of metamorphism in the Alps: Geological Society of America Bulletin, v. 84, p. 2053–2058.
Ernst, W. G., 1988, Tectonic history of subduction zones
inferred from retrograde blueschist P-T paths: Geology, v. 16, p. 1081–1084.
Ernst, W. G., 1999, H2O and ultrahigh-pressure subsolidus phase relations for mafic and ultramafic systems:
International Geology Review, v. 41, p. 886–894.
Ernst, W. G., 2001, Subduction, ultrahigh-pressure metamorphism, and regurgitation of buoyant crustal
slices—implications for arcs and continental growth:
Physics of the Earth and Planetary Interiors, v. 127, p.
253–275.
Ernst, W. G., 2006, Preservation/exhumation of ultrahighpressure subduction complexes, in Sajeev, K., and
Santosh, M., eds., Extreme crustal metamorphism and
related crust-mantle processes: Lithos, v. 92, p. 321–
335.
Ernst, W. G., and Liou, J. G., 1999, Overview of UHP
metamorphism and tectonics in well-studied orogens:
International Geology Review, v. 41, p. 477–493.
Ernst, W. G., Maruyama, S., and Wallis, S., 1997, Buoyancy-driven rapid exhumation of ultrahigh-pressure
metamorphosed continental crust: U.S. Proceedings of
the National Academy of Sciences, v. A 94, p. 9532–
9537.
Escher, J. C., and Pulvertaft, T. C. R., 1995, Geological
map of Greenland: 1:2,500,000: Geological Survey of
Greenland.
Evans, D., Graham, C., Armour, A., and Bathurst, P., eds.
2003, The Millennium Atlas: Petroleum geology of the
central and northern North Sea: London, UK, Geological Society of London, 398 p.
Faleide, J. I., Tsikalas, F., Breivik, A. J., Mjelde, R., Ritzmann, O., Engen, Ø., Wilson, J., and Eldholm, O.,
2008, Structure and evolution of the continental margin off Norway and the Barents Sea: Episodes, v. 31,
no. 1, p. 82-91.
Fassoulas, C., 1999, The structural evolution of central
Crete: Insight into the tectonic evolution of the South
Aegean: Journal of Geodynamics, v. 27, p. 23–43.
Faupl, P., Pober, E., and Wagreich, M., 1987, Facies
development of the Gosau group of the eastern parts of
the Northern Calcareous Alps during the Cretaceous
and Paleogene, in Flügel, H. W., and Faupl, P., eds.,
Geodynamics of the Eastern Alps: Vienna, Austria,
Deuticke, p. 142–155.
Faupl, P., and Wagreich, M., 2000, Late Jurassic to
Eocene palaeogography and geodynamic evolution of
the Eastern Alps, in Neubauer, F., and V., H., eds.,
Aspects of geology in Austria: Mitteilungen der Österreichischen Geologischen Gesellschaft, v. 92, p. 79–
94.
Fitch, T., 1970, Earthquake mechanisms and island arc
tectonics in the Indonesian-Philippine region: Bulletin
of the Geological Society of America, v. 60, p. 565–
591.
Flügel, H. W., and Faupl, P., 1987, Geodynamics of the
Eastern Alps: Vienna, Austria, Deuticke.
Flügel, H. W., Faupl, P., and Mauritsch, H. J., 1987,
Implications on the Alpidic evolution of the eastern
parts of the Eastern Alps, in Flügel, H. W., and Faupl,
P., eds., Geodynamics of the Eastern Alps: Vienna,
Austria, Deuticke, p. 407–414.
Forristall, G. Z., 1972, Stress distributions and overthrust
faulting: Geological Society of America Bulletin, v. 83,
p. 3073–3082.
Frank, W., 1987, Evolution of the Austroalpine elements
in the Cretaceous, in Flügel, H. W., and Faupl, P., eds.,
Geodynamics of the Eastern Alps: Vienna, Austria,
Deuticke.
Frank, W., Kralik, M., Scharbert, S., and Thöni, M., 1987,
Geochronological data from the Eastern Alps, in
Flügel, H. W., and Faupl, P., eds., Geodynamics of the
Eastern Alps: Vienna, Austria, Deuticke, p. 272–281.
Frei, W., Heitzmann, P., and Lehner, P., 1990, Swiss NFP20 research program of the deep structure of the Alps,
in Roure, F., Heitzmann, P., and Polino, R., eds., Deep
structure of the Alps: Société Géologique de France,
Mémoire 156, p. 29–46.
Frey, M., Jaeger, E., and Niggli, E., 1979, Gesteinmetamorphose im Bereich der Geotraverse Basel-Chiasso:
Schweizerische Mineralogische und Petrographische
Mitteilungen, v. 56, p. 649–659.
Frey, M., Trommsdorff, V., and Wenk, E., 1980, Alpine
metamorphism in the Central Alps. Excursion No VI
in Geology of Switzerland: A guide book (G10): International Geological Congress, 26th, Paris: Basel, Switzerland, Wepf & Co., p. 295–316.
Frisch, W., 1976, Ein modell zur alpidischen Evolution
und Orogenese des Tauernfensters: Geologische
Rundschau, v. 65, p. 373–393.
Frisch, W., Gommeringer, K., Kelm, U., and Popp, F.,
1987, The Upper Bündnershiefer of the Tauern
Window—a key to understanding Eoalpine orogenic
processes in the Eastern Alps, in Flügel, H. W., and
Faupl, P., eds., Geodynamics of the Eastern Alps:
Vienna, Austria, Deuticke, p. 55–69.
Froitzheim, N., Pleuger, J., Roller, S., and Nagel, T., 2003,
Exhumation of high- and ultrahigh-pressure metamorphic rocks by slab extraction: Geology, v. 31, no. 10, p.
925–928.
Fry, N., and Barnicoat, A. C., 1987, The tectonic implications of high-pressure metamorphism in the western
Alps: Journal of the Geological Society, v. 144, p. 653–
659.
Fukao, Y., and Furimoto, M., 1975, Mechanism of large
earthquakes along the eastern margin of the Japan
Sea.: Tectonophysics, v. 26, p. 247–266.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Fukao, Y., Widyantoro, S., and Obayashi, M., 2001, Stagnant slabs in the upper and lower mantle transition
zone: Reviews of Geophysics, v. 39, no. 3, p. 291–323.
Gee, D. G., Guezou, J.-C., Roberts, D., and Wolff, F. C.,
1985, The central-southern part of the Scandinavian
Caledonides, in Gee, D. G., and Sturt, B. A., eds., The
Caledonian Orogen—Scandinavia and related areas:
Chichester, UK, Wiley, p. 109–133.
Gibbs, A. D., 1979, Structural evolution of part of the
Jotun nappe in Sygnefjell and some speculation on the
regional implications, in Cooper, M. A., and Garton,
M. R., eds., Tectonic evolution of the Scandinavian
Caledonides—proceedings of a conference at Department of Geology, City of London Polytechnic, 27 April
1978, p. 17–20.
Gilotti, J. A., and McClelland, W. C., 2007, Characteristics of, and a tectonic model for, ultrahigh-pressure
metamorphism in the overriding plate of the Caledonian Orogen: International Geology Review, v. 49, p.
777–797.
Gorbatschev, R., 1985, Precambrian basement of the
Scandinavian Caledonides, in Gee, D. G., and Sturt,
B. A., eds., The Caledonide Orogen—Scandinavia and
related areas: Chichester, UK, Wiley, p. 197–212.
Gradstein, F. M., Ogg, J. G., Smith, A. G., Bleeker, W., and
Lourens, L. J., 2004, A new geologic timescale with
special reference to Precambrian and Neogene: Episodes, v. 27, no. 2, p. 83–100.
Griffin, W. L., Austrheim, H., Brastad, K., Bryhni, I., Krill,
A. G., Krogh, E. J., Mørk, M. B. E., Qvale, H., and
Tørudbakken, B., 1985, High pressure metamorphism
in the Scandinavian Caledonides, in Gee, D. G., and
Sturt, B. A., eds., The Caledonide Orogen—Scandinavia and related areas: Chichester, UK, Wiley, p. 783–
801.
Griffin, W. L., and Brueckner, H. K., 1980, Caledonian
Sm-Nd ages and a crustal origin for Norwegian eclogites: Nature, v. 285, p. 319–321.
Groppo, C., Lombardo, B., Castelli, D., and Compagnoni,
R., 2007, Exhumation history of the UHPM BrossascoIsasca Unit, Dora Maira Massif, as inferred from a
phengite-amphibole eclogite: International Geology
Review, v. 49, p. 142–168.
Gu, Y., Dziewonski, A. M., and Agee, C. B., 1998, Global
de-correlation of the topography of transition zone discontinuities: Earth and Planetary Science Letters, v.
157, p. 57–67.
Gutscher, M.-A., Spakman, W., Bijwaard, H., and
Engdahl, E. R., 2000, Geodynamics of flat subduction:
Seismicity and tomographic constraints from the
Andean margin: Tectonics, v. 19, no. 5, p. 814–843.
Hacker, B. R., Calvert, A., Zhang, R. Y., Ernst, W. G., and
Liou, J. G., 2003, Ultrarapid exhumation of ultrahighpressure diamond-bearing sedimentary of the
Kokchetav Massif, Kazakhstan: Lithos, v. 70, p. 61–
75.
743
Hacker, B. R., Wang, X., Eide, E. A., and Ratschbacher,
L., 1996, The Qinling-Dabie ultra-high-pressure orogen, in Yin, A., and Harrison, M., eds., Tectonic evolution of Asia: Cambridge, UK, Cambridge University
Press, p. 345–370.
Hacker, B. R., Ratschbacher, L., and Liou, J. G., 2004,
Subduction, collision and exhumation in the ultrahighpressure Qinling-Dabie orogen, in Malpas, J.,
Fletcher, C. J. N., Ali, J. R., and Aitchison, J. C., eds.,
Aspects of the tectonic evolution of China: Geological
Society of London, Special Publications, v. 226, p.
157–175.
Hacker, B. R., Ratschbacher, L., Webb, L., and Dong, S.,
1995, What brought them up? Exhumation of the
Dabie Shan ultrahigh-pressure rocks: Geology, v. 23,
p. 743–746.
Hall, D. B., 2001, Deep physical state of the western Colorado Plateau and its transition to the eastern Great
Basin inferred from deep electrical resistivity (magnetotelluric) data: Unpubl. masters thesis, University of
Utah, 86 p.
Hamilton, W. B., 1988, Tectonic setting and variations
with depth of some Cretaceous and Cenozoic structural
and magmatic systems of the western United States, in
Ernst, W. G., ed., Metamorphism and crustal evolution
of the western United States. Rubey Volume VII:
Englewood Cliffs, NJ, Prentice Hall, p. 1–40.
Hanks, T. C., 1971, The Kuril Trench-Hokkaido rise system: Large shallow earthquakes and simple models of
deformation: Geophysical Journal of the Royal Astronomical Society, v. 23, p. 173–189.
Hanks, T. C., 1979, Deviatoric stresses and earthquake
occurrence at the outer rise: Journal of Geophysical
Research, v. 84, p. 2343–2347.
Hanks, T. C., 1981, Reply (to Molnar): Journal of Geophysical Research, v. 86B, p. 3947–3948.
Harland, W. B., 1985, Caledonide Svalbard, in Gee, D. G.,
and Sturt, B. A., eds., The Caledonide Orogen—Scandinavia and related areas: Chichester, UK, Wiley, p.
999–1016.
Harley, S. L., and Carswell, D. A., 1995, Ultradeep crustal
metamorphism: a prospective view: Journal of Geophysical Research, v. 100B, p. 8367–8380.
Hasegawa, A., and Sacks, I. S., 1981, Subduction of the
Nazca Plate beneath Peru as determined from seismic
observations: Journal of Geophysical Research, v. 86B,
p. 4971–4980.
Hasegawa, A., Zhao, D., Yamamoto, A., and Horiuchi, S.,
1991, Deep structure of the northeastern Japan arc and
its relationship to seismic and volcanic activity:
Nature, v. 352, p. 683–689.
Hashimoto, M., 1991, Geology of Japan (English edition):
Dordrecht, The Netherlands, Terra Scientific Publishing/Kluwer Academic.
Hawkins, J. W., 1995, Evolution of the Lau Basin—
insights from ODP Leg 135, in Taylor, B., and Natland,
J., eds., Active margins and marginal basins of the
744
MILES F. OSMASTON
Western Pacific: American Geophysical Union, Geophysical Monograph 88, p. 125–173.
Heller, F., Lowrie, W., and Hirt, A. M., 1989, A review of
palaeomagnetic and magmatic anisotropy results from
the Alps, in Coward, M. P., Dietrich, D., and Park,
R. G., eds., Alpine tectonics, Geological Society of
London, Special Publications, v. 45, p. 399–420.
Helmstaedt, H., and Doig, R., 1975, Eclogite nodules from
kimberlite pipes of the Colorado Plateau—samples of
subducted Franciscan-type oceanic lithosphere: Physics and Chemistry of the Earth, v. 9, p. 95–112.
Hermann, J., 2003, Experimental evidence for diamondfacies metamorphism in the Dora Maira massif: Lithos,
v. 70, p. 163–182.
Hermann, J., Rubatto, D., Korsakov, A., and Shatsky, V. S.,
2001, Multiple zircon growth during fast exhumation
of diamondiferous, deeply subducted continental crust
(Kokchetav Massif, Kazakhstan): Contributions to
Mineralogy and Petrology, v. 141, p. 66–82.
Hetzel, R., 1999, Geology and geodynamic evolution of
the high-P/low-T Maksyutov Complex, southern
Russia: Geologische Rundschau, v. 87, p. 577–588.
Higgins, A. K., and Kalsbeek, F., eds., 2004, East Greenland Caledonides: Stratigraphy, structure, and geochronology: Geological Survey of Denmark and
Greenland Bulletin, v. 6, 93 p.
Hirth, G., and Kohlstedt, D. L., 1996, Water in the oceanic
upper mantle: implication for rheology, melt extraction, and the evolution of the lithosphere: Earth and
Planetary Science Letters, v. 144, p. 93–108.
Höck, V., and Miller, C., 1987, Mesozoic ophiolitic
sequences and non-ophiolitic metabasites in the Hohe
Tauern, in Flügel, H. W., and Faupl, P., eds., Geodynamics of the Eastern Alps: Vienna, Austria, Deuticke,
p. 16–32.
Homewood, P., and Caron, C., 1982, Flysch of the Western
Alps, in Hsu, K. J., ed., Mountain building processes:
London, UK, Academic Press, p. 157–168.
Hopson, C. A., Pesagno, E. A., Mattinson, J. M.,
Luyendyk, B. P., Beebe, W., Hull, D. M., Muñoz, I. M.,
and Blome, C. D., 1996, 2. Coast Range Ophiolite as
paleoequatorial mid-ocean lithosphere, in Dickinson,
W. R., Hopson, C. A., and Saleeby, J. B., eds., Alternate origins of the Coast Range Ophiolite (California):
Introduction and implications: GSA Today, v. 6, no. 2,
p. 4–7.
Horai, K., and Uyeda, S., 1963, Terrestrial heat flow in
Japan: Nature v. 199, p. 364–365.
Horváth, F., 1993, Towards a mechanical model for the
formation of the Pannonian basin: Tectonophysics, v.
226, p. 333–357.
Horváth, F., and Royden, L., 1981, Mechanism for the formation of the intra-Carpathian basins: A review: Earth
Evolution Sciences, v. 1, p. 307–316.
Hsü, K. J., 1968, The principles of mélanges and their
bearing on the Franciscan-Knoxville paradox: Geological Society of America Bulletin, v. 79, p. 1063–1074.
Hsü, 1991, Exhumation of high-pressure metamorphic
rocks: Geology, v. 19, p. 107–110.
Hsü, 1992, Reply (to Platt): Geology, v. 20, p. 186–187.
Hsü, 1994, The Geology of Switzerland: An introduction to
tectonic facies: Princeton, NJ, Princeton University
Press, 250 p.
Hubbert, M. K., and Rubey, W. W., 1959, Role of fluid
pressure in mechanics of overthrust faulting. I.
Mechanics of fluid-filled porous solids and its application to overthrust faulting: Bulletin of the Geological
Society of America, v. 70, no. 2, p. 115–166.
Hunziker, J. C., 1986, The Alps: A case of multiple collision, in Coward, M. P., and Ries, A. C., eds., Collision
tectonics: Geological Society of London, Special Publications, v. 19, p. 221–227.
Hunziker, J. C., Desmons, J., and Hurford, A. J., 1992,
Thirty-two years of geochronological work in the
central and western Alps: a review on seven maps:
Mémoires Géologiques (Lausanne), v. 13, p. 1–59.
Hunziker, J. C., Desmons, J., and Martinotti, G., 1989,
Alpine thermal evolution in the central and western
Alps, in Coward, M. P., Dietrich, D., and Park, R. G.,
eds., Alpine tectonics: Geological Society of London,
Special Publications, v. 45, p. 353–367.
Hurst, J. M., McKerrow, W. S., Soper, N. J., and Surlyk, F.,
1983, The relationship between Caledonian nappe tectonics and Silurian turbidite deposition in North
Greenland: Journal of the Geological Society, London,
v. 140, p. 123–131.
Hussong, D. M., and Wipperman, L. R., 1981, Vertical
movement and tectonic erosion of the continental wall
of the Peru-Chile Trench near11°30'S latitude: Geological Society of America Memoir, v. 154, p. 509–524.
Isacks, B. L., 1988, Uplift of the central Andean plateau
and bending of the Bolivian orocline: Journal of Geophysical Research, v. 93, p. 3211–3231.
James, D., 1978, Subduction of the Nazca Plate beneath
central Peru: Geology, v. 6, p. 174–178.
Janoshek, W. R., and Matura, A., 1980, Outline of the
geology of Austria: International Geological Congress,
26th, Paris. Guidebook G-05: Abhandlungen der geologischen Bundesanstalt, v. 34, p. 7–98.
Jordan, T., Isacks, B. L., Allmendinger, R. W., Brewer,
J. A., Ramos, V. A., and Ando, C. J., 1983, Andean
tectonics related to geometry of subducted Nazca
plate: Geological Society of America Bulletin, v. 94, p.
341–361.
Jordan, T. E., and Allmendinger, R. W., 1986, The Sierra
Pampeanas of Argentina: A modern analogy of Rocky
Mountain foreland deformation: American Journal of
Science, v. 286, p. 737–764.
Kanamori, H., 1971, Great earthquakes at island arcs and
the lithosphere: Tectonophysics, v. 12, p. 187–198.
Kanamori, H., 1977, Seismic and aseismic slip along
subduction zones and their tectonic implications, in
Talwani, M., and Pitman, W. C., III., eds., Island arcs,
deep sea trenches, and back-arc basins: Washington,
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
DC, American Geophysical Union, Maurice Ewing
Series, v. 1, p. 163–174.
Karato, S., 1986, Does partial melting reduce the creep
strength of the upper mantle?: Nature, v. 319, p. 309–
310.
Karig, D. E., 1971, Origin and development of marginal
basins in the western Pacific: Journal of Geophysical
Research, v. 76, p. 2542–2561.
Kiminami, K., Niida, K., Ando, H., Kito, N., Iwata, K.,
Miyashita, S., Tajika, J., and Sokakibara, M., 1992,
Cretaceous–Paleogene arc-trench systems in Hokkaido, in International Geological Congress, 29th,
Kyoto, Japan. Guidebook for Field Trip A-01, 43 p.
Kind, R., Ni, J., Zhao, W., Wu, J., Yuan, X., Zhao, L.,
Sandvol, E., Reese, C., Nabelek, J., and Hearn, T.,
1996, Evidence from earthquake data for a partially
molten crustal layer in southern Tibet: Science, v. 274,
p. 1692–1694.
Kirby, S. H., Engdahl, E. R., and Villaseñor, A., 2002,
Warm-slab subduction as a global process, in Kirby, S.,
Wang, K., and Dunlop, S., eds., The Cascadia subduction zone and related subduction systems: Seismic
structure, intraslab earthquakes, and processes and
earthquake hazards: USGS Open-File Report OF 020328, p. 79–80.
Kley, J., and Monaldi, C. R., 1998, Tectonic shortening
and crustal thickening in the central Andes: How good
is the correlation?: Geology, v. 26, no. 8, p. 723–726.
Klitgord, K. M., Mudie, J. M., Bischoff, J. L., and Henyey,
T. L., 1974, Magnetic anomalies in the northern and
central Gulf of California: Geological Society of America Bulletin, v. 85, p. 815–820.
Korsakov, A. V., and Hermann, J., 2006, Silicate and carbonate melt inclusions associated with diamonds in
deeply subducted carbonate rocks: Earth and Planetary Science Letters, v. 241, nos. 1–2, p. 104–118.
Korsakov, A. V., Theunissen, K., and Smirnova, L. V.,
2004, Intergranular diamonds derived from partial
melting of crustal rocks at ultrahigh-pressure metamorphic conditions: Terra Nova, v. 16, no. 3, p. 146–
151.
Kozlovsky, Y. A., 1987, The Superdeep Well of the Kola
Peninsula: Berlin, Germany, Springer.
Krill, A. G., 1985, Relationships between the Western
Gneiss Region and the Trondheim Region: STOCKWERK tectonics reconsidered, in Gee, D. G., and
Sturt, B. A., eds., The Caledonide Orogen— Scandinavia and related areas: Chichester, UK, Wiley, p. 473–
483.
Krogh, E. J., and Carswell, D. A., 1995, HP and UHP
eclogites and garnet peridotites in the Scandinavian
Caledonides, in Coleman, R. B., and Wang, X., eds.,
Ultrahigh pressure metamorphism: Cambridge, UK,
Cambridge University Press, p. 244–298.
Krogh, T., Robinson, P., and Terry, M. P., 2003, Precise UPb zircon ages define 18 and 19 m.y. subduction intervals in the Averøya-Nordøyane area, Western Gneiss
745
Region [abs.], in Eide, E. A., ed., The Alice Wain
Memorial West Norway Eclogite Field Symposium
abstract volume: Norges Geologiske Undersøkelses
Rapport 2003.055, p. 71–72.
Kukowski, N., and Oncken, O., 2006, Subduction erosion—the “normal” mode of fore-arc material transfer
along the Chilean margin?, in Oncken, O., Chong, G.,
Franz, G., Gieze, P., Götze, H.-J., Ramos, V. A.,
Strecker, M. P., and Wigger, P., eds., The Andes:
Active subduction orogeny: Heidelberg, Germany,
Springer, p. 217–236.
Kullerud, L., Tørudbakken, B., and Illebek, S., 1986, A
compilation of radiometric age determinations from
the Western Gneiss Region, south Norway: Norges
Geologiske Undersøkelse Bulletin, v. 406, p. 17–42.
Kulm, L. D., Thornburg, T. M., Schrader, H., Masias, A.,
Resig, J. M., and Johnson, L., 1981, Late Cenozoic
carbonates on the Peru continental margin: Lithostratigraphy, biostratigraphy, and tectonic history, in
Kulm, L. D., Dymond, J., Dach, E. J., and Hussong, D.
M., eds., Nazca Plate: Crustal formation and Andean
convergence. Geological Society of America, Memoir
154, p. 469–508.
Kurz, W., 2006, Penninic paleogeography from the Western toward the Eastern Alps—still open questions?:
International Geology Review, v. 40, p. 996–1022.
Kurz, W., Neubauer, F., Genser, F., and Dachs, E., 1998,
Alpine geodynamic evolution of passive and active
continental margin sequences in the Tauern Window
(Eastern Alps, Austria): Geologische Rundschau, v.
87, p. 225–242.
Labrousse, L., Jolivet, L., Agard, P., Hébert, R., and
Andersen, T. B., 2002, Crustal-scale boudinage and
migmatization of gneiss during their exhumation in the
UHP Province of Western Norway: Terra Nova, v. 14, p.
263–270.
Lacassin, R., 1989, Plate-scale kinematics and compatibility of crustal shear zones, in Coward, M. P., Dietrich,
D., and Park, R. G., eds., Alpine tectonics: Geological
Society of London, Special Publications, v. 45, p. 339–
352.
Lardeaux, J. M., Schwartz, S., Tricart, P., Paul, A., Guillot,
S., Béthoux, N., and Masson, F., 2006, A crustal-scale
cross-section of the southwestern Alps combining geophysical and geological imagery: Terra Nova, v. 18, no.
6, p. 412–422 [doi: 10.1111/j.1365-3121.2006.00706
.x].
Larson, P. A., Mudie, P. J., and Larson, R. L., 1972, Magnetic anomalies and fracture zone trends on the Gulf of
California: Geological Society of America Bulletin, v.
83, p. 3361–368.
Larter, R. D., and Barker, P. F., 1991, Effects of ridge
crest–trench interaction on Antarctic-Phoenix spreading; forces on a young subducting plate: Journal of
Geophysical Research, v. 96, no. B12, p. 19,583–
19,607.
746
MILES F. OSMASTON
Laubscher, H., 1983, Detachment, shear and compression
in the central Alps, in Hatcher, R. D., Jr., Williams,
H., and Zietz, I., eds., Contributions to the tectonics
and geophysics of mountain chains: Geological Society
of America Memoir 158, p. 191–211.
Laubscher, H., and Bernoulli, D., 1982, History and deformation of the Alps, in Hsü, K. J., ed., Mountain building processes: New York, NY, Academic Press, p.
169–180.
Lay, T., Astiz, L., Kanamori, H., and Christensen, D. H.,
1989, Temporal variation of large intraplate earthquakes in coupled subduction zones: Physics of the
Earth and Planetary Interiors, v. 54, p. 258–312.
Leech, M. L., Webb, L. E., and Yang, T. N., 2006, Diachronous histories for the Dabie-Sulu orogen from
high-temperature geochronology, in Hacker, B. R.,
McClelland, W. C., and Liou, J. G., eds., Ultrahighpressure metamorphism: Deep continental subduction. Geological Society of America Special Paper 403,
p. 1–22 [doi: 10.1130/2006.2403(01)].
Lemoine, M., and Trümpy, R., 1987, Pre-oceanic rifting in
the Alps: Tectonophysics, v. 133, p. 305–320.
Levitt, D. A., and Sandwell, D. T., 1995, Lithospheric
bending at subduction zones based on depth soundings
and satellite gravity: Journal of Geophysical Research,
v. 100, no. 1, p. 379–400.
Lidmar-Bergström, K., and Näslund, J. O., 2002, Landforms and uplift in Scandinavia, in Doré, A. H., Cartwright, J. A., Stoker, M. S., Turner, J. P., and White, N.,
eds., Exhumation of the North Atlantic margin: Timing, mechanisms and implications for petroleum
exploration: Geological Society, London, Special Publications, v. 196, p. 103–116.
Linzer, H.-G., Decker, K., Peresson, H., Dell’Mour, R.,
and Frisch, W., 2002, Balancing lateral orogenic float
of the Eastern Alps: Tectonophysics, v. 354, p. 211–
237.
Linzer, H.-G., Ratschbacher, L., and Frisch, W., 1995,
Transpressional collision structures in the upper crust:
The fold-thrust belt of the Northern Calcareous Alps:
Tectonophysics, v. 242, p. 41–61.
Liou, J. G., Zhang, R. Y., Wang, X., Eide, E. A., Ernst,
W. G., and Maruyama, S., 1996, Metamorphism and
tectonics of high-pressure and ultra-high-pressure
belts in the Dabie-Sulu region, China, in Yin, A., and
Harrison, M., eds., Tectonic evolution of Asia: Cambridge, UK, Cambridge University Press, p. 300–344.
Liou, J. G., Zhang, R. Y., and Ernst, W. G., 2007, Very
high-pressure orogenic garnet peridotites, in Hemley,
R. J., ed., High-pressure geoscience feature: Proceedings of the National Academy of Sciences of the USA,
v. 104, no. 22, p. 9116-9121 [doi: 10.1073/pnas
.0607300104].
Lister, C. R. B., 1971, Tectonic movement in the Chile
Trench: Science, v. 173, p. 719–722.
Lister, G. S., and Davis, G. A., 1989, The origin of metamorphic core complexes and detachment faults formed
during Tertiary continental extension of the northern
Colorado region, USA: Journal of Structural Geology, v.
11, p. 65–94.
Lliboutry, L., 1969, Seafloor spreading, continental drift
and lithosphere sinking with an asthenosphere at melting point: Journal of Geophysical Research, v. 74, p.
6525–6540.
Lombardo, B., Rubatto, D., and Castelli, D., 2002, Ion
microprobe U-Pb dating of zircon from a Monviso
plagiogranite: Implications for the evolution of the
Piedmont-Liguria Tethys in the Western Alps: Ofioliti,
v. 27, p. 109–117.
Lundgren, P. R., and Okal, E. A., 1988, Slab decoupling in
the Tonga Arc: The June 22, 1977, earthquake: Journal of Geophysical Research, v. 93, p. 13,355–13,366.
Lundmark, A. M., and Corfu, F., 2008, Late-orogenic
Sveconorwegian massif anorthosite in the Jotun Nappe
Complex, SW Norway, and causes of repeated AMCG
magmatism along the Baltoscandian margin: Contributions to Mineralogy and Petrology, v. 155, no. 2, p.
147–163 [doi: 10.1007?s00420-007-0233-5].
Mancktelow, N., 1985, The Simplon line: A major displacement zone in the western Lepontine Alps:
Eclogae geologicae Helvetiae, v. 78, p. 73–96.
Marthaler, M., and Stampfli, G. M., 1989, Les Schistes
Lustrés à ophiolites de la nappe du Tsaté: Un ancien
prisme d’accrétion issu de la marge active apulienne?:
Schweizerische Mineralogische and Petrographische
Mitteilungen, v. 69, no. 2, p. 211–216.
Maruyama, S., Liou, J. G., and Terabayashi, M., 1996,
Blueschists and eclogites of the world and their exhumation: International Geology Review, v. 38, p. 485–
594.
Masago, H., Okamoto, K., and Terabayashi, M., 2005,
Exhumation tectonics of the Sanbagawa high-pressure
metamorphic belt, Southwest Japan—constraints from
the upper and lower boundary faults: International
Geology Review, v. 47, p. 1194–1206.
Matte, P., 1998, Continental subduction and exhumation
of HP rocks in Paleozoic orogenic belts: Uralides and
Variscides, in Sundquist, B., and Stephens, M. B.,
eds., Tectonics and geological history of some Phanerozoic orogens (David G. Gee honorary volume): GFF, v.
120, p. 209–222.
Matter, A., Homewood, P., Caron, C., Rigassi, D., van Stuijvenberg, J., Weidmann, M., and Winkler, W., 1980,
Flysch and molasse of Western and Central Switzerland; Excursion No. V, in Geology of Switzerland: A
guidebook (G10), International Geological Congress,
26th, Paris: Basel, Switzerland, Wepf & Co,, p. 261–
293.
Mattinson, J. M., 1988, Constraints on the timing of Franciscan metamorphism: Geochronological approaches
and their limitations, in Ernst, W. G., ed., Metamorphism and crustal evolution of the western United
States. Rubey Volume VII: Englewood Cliffs, Prentice
Hall, p. 1023–1034.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Mauritsch, H. J., and Becke, M., 1987, Paleomagnetic
investigations in the Eastern Alps and the southern
border zone, in Flügel, H. W., and Faupl, P., eds.,
Geodynamics of the Eastern Alps: Vienna, Austria,
Deuticke, p. 282–308.
Mauritsch, H. J., and Frisch, W., 1978, Palaeomagnetic
data from the central part of the Northern Calcareous
Alps, Austria: Journal of Geophysics, v. 44, p. 623–
637.
McClelland, W. C., Power, S. E., Gilotti, J. A., Mazdab,
F. K., and Wopenka, B., 2006, U-Pb SHRIMP geochronology and trace-element geochemistry of coesitebearing zircons, North-East Greenland Caledonides,
in Hacker, B. R., McClelland, W. C., and Liou, J. G.,
eds., Ultrahigh pressure metamorphism: Deep continental subduction: Geological Society of America Special Paper 403, p. 23–43 [doi: 10.1130/2006.2403(02)].
McKee, E. H., and Noble, D. C., 1982, Miocene volcanism
and deformation in the western Cordillera and high
plateaus of south-central Peru: Geological Society of
America Bulletin, v. 93, p. 657–662.
Mégard, F., 1984, The Andean orogenic period and its
major structures in central and northern Peru: Journal
of the Geological Society, v. 141, p. 893–900.
Meissner, R., Rabbel, W., and Kern, H., 2006, Seismic
lamination and anisotropy of the Lower Continental
Crust: Tectonophysics, v. 416, p. 81–99.
Melosh, H. J., 1978, Dynamic support of the outer rise:
Geophysical Research Letters, v. 5, p. 321–324.
Melosh, H. J., and Raefsky, A., 1980, The dynamical
origin of subduction zone topography: Geophysical
Journal of the Royal Astronomical Society, v. 60, p.
333–354.
Michard, A., Goffé, B., Chopin, C., and Henry, C., 1996,
Did the Western Alps develop through an Oman-type
stage? The geotectonic setting of high-pressure metamorphism in two contrasting Tethyan transects:
Eclogae Geologicae Helvetiae, v. 89, p. 43–80.
Milnes, A. G., and Koestler, A. G., 1985, Geological structure of Jotunheimen, southern Norway (Sognfjell–
Valdres cross-section), in Gee, D. G., and Sturt, B. A.,
eds., The Caledonide Orogen—Scandinavia and
related areas: Chichester, UK, Wiley, p. 457–474.
Milnes, A. G., and Pfiffner, O. A., 1980, Tectonic evolution of the Central Alps in the cross-section St. GallenComo, in Rybach, L., and Lambert, A., eds., Symposium on the Alpine geotraverses with special emphasis
on the Basel-Chiasso profile: Eclogae Geologicae
Helvetiae, v. 73, p. 617–633.
Milnes, A. G., and Schmutz, H.-U., 1978, Structure and
history of the Suretta nappe (Pennine zone, Central
Alps)—a field study: Eclogae Geologicae Helvetiae, v.
71, no. 1, p. 19–33.
Milnes, A. G., Wennberg, O. P., Skår, Ø., and Koestler,
A. G., 1997, Contraction, extension and timing in the
South Norwegian Caledonides: The Sognefjord
transect, in Burg, J.-P., and Ford, M., eds., Orogeny
747
through time: Geological Society Special Publication
No. 121, p. 123–148.
Minear, J. W., and Toksöz, M. N., 1970, Thermal regime of
a downgoing slab and new global tectonics: Journal of
Geophysical Research, v. 75, p. 1397–1419.
Molnar, P., and Atwater, T., 1978, Interarc spreading and
Cordilleran tectonics as alternates related to the age of
subducted oceanic lithosphere: Earth and Planetary
Science Letters, v. 41, p. 330–340.
Moore, D., and Lockner, D. A., 2007, Comparative deformation behavior of minerals in serpentinized ultramafic rock: Application to the slab-mantle interface in
subduction zones: International Geology Review, v. 49,
p. 401–415.
Mueller, S., 1982, Deep structure and recent dynamics in
the Alps, in Hsü, K. J., ed., Mountain building processes: London, UK, Academic Press, p. 181–199.
Mykkeltveit, S., Husebye, E. S., and Oftedahl, C., 1980,
Subduction of the Iapetus Ocean crust beneath
the Møre Gneiss Region, southern Norway: Nature,
London, v. 288, p. 473–475.
Neubauer, F., 1987, The Gurktal thrust system within the
Austroalpine region—some structural and geometrical
aspects, in Flügel, H. W., and Faupl, P., eds., Geodynamics of the Eastern Alps: Vienna, Austria, Deuticke,
p. 226–236.
Neubauer, F., Genser, F., and Handler, R., 2000, The Eastern Alps: Result of a two-stage collision process, in
Neubauer, F., and Höck, V., eds., Aspects of geology in
Austria: Mitteilungen der Österreichischen Geologischen Gesellschaft, v. 92, p. 117–134.
Nickelsen, R. P., Hossack, J. R., Garton, M., and
Repetsky, J., 1985, Late Precambrian to Ordovician
stratigraphy and correlation in the Valdres and Synnfjell thrust sheets of the Valdres area, southern Norwegian Caledonides; with some comments on
sedimentation, in Gee, D. G., and Sturt, B. A., eds.,
The Caledonide Orogen—Scandinavia and related
areas: Chichester, UK, Wiley, p. 369–378.
Noble, D. C., McKee, E. H., Farrar, E., and Peterman, U.,
1974, Episodic Cenozoic volcanism and tectonism
in the Andes of Peru: Earth and Planetary Science
Letters, v. 21, p. 213–220.
Noble, D. C., McKee, E. H., Mourier, T., and Mégard, F.,
1990, Cenozoic stratigraphy, magmatic activity, compressive deformation and uplift in northern Peru:
Geological Society of America Bulletin, v. 102, p.
1105–1113.
Noble, D. C., Vogel, T. A., Peterson, P. S., Landis, G. P.,
Grant, N. K., Jezek, P. A., and McKee, E. H., 1984,
Rare-element-enriched, S-type ash-flow tuffs containing phenocrysts of muscovite, andalusite, and sillimanite, southeastern Peru: Geology, v. 12, p. 35–39.
Nur, A., and Ben-Avraham, Z., 1981, Volcanic gaps and
the consumption of aseismic ridges in South America:
Geological Society of America Memoir, v. 154, p. 729–
740.
748
MILES F. OSMASTON
Ogasawara, Y., 2005, Microdiamonds in ultrahigh-pressure metamorphic rocks: Elements, v. 1, p. 91–96.
O’Hara, M. J., and Mercy, E. L. P., 1963, Petrology and
and petrogenesis of garnetiferous peridotites: Transactions of the Royal Society of Edinburgh, v. 65, p. 251–
314.
Oncken, O., Hindle, D., Kley, J., Elger, K., Victor, P., and
Schemmann, K., 2006, Deformation of the Central
Andean upper plate system—facts, fiction, and constraints for plateau models, in Oncken, O., Chong,
G., Franz, G., Gieze, P., Götze, H.-J., Ramos, V. A.,
Strecker, M. P., and Wigger, P., eds., The Andes:
Active subduction orogeny: Berlin/Heidelberg,
Germany, Springer, p. 3–27.
Osmaston, M. F., 1973, Limited lithosphere separation as
a main cause of continental basins, continental growth
and epeirogeny, in Tarling, D. H., and Runcorn, S. K.,
eds., Implications of continental drift to the Earth
Sciences: London, UK, Academic Press, p. 649–674.
Osmaston, M. F., 1977, Some fundamental aspects of plate
tectonics bearing on hydrocarbon location, in Hobson,
G. D., ed., Developments in Petroleum Geology I:
Barking, UK, Applied Science Publishers, p. 1–52.
Osmaston, M. F., 1985, Subduction tectonic erosion (STE)
of the South Alpine plate: Essential preliminary to
nappe genesis and tectonics in the Alps, and key control upon earlier Alpine foredeeps [abs.], in International Association of Sedimentologists Foreland
Basins Symposium, Fribourg, Germany, September
1985, Programme and abstracts, p. 101.
Osmaston, M. F., 1988a, Sequence of tectonic environments during plate convergence and collision: The
controlling influence of prior STE beneath the margins, with some Caledonide examples: Geological
Society of London, Tectonic Studies Group, 19th
Annual Meeting, December 1988, Cambridge, UK.
Osmaston, M. F., 1988b, Subduction tectonic erosion
(STE) and post-subduction magmatism (PSM): New
light on Iapetus closure tectonics and synthesis [abs.],
in GAC-MAC Joint Annual Meeting 23–25 May 1988,
Memorial University, St. John’s, Canada (“St. John’s
88”), p. A93.
Osmaston, M. F., 1991, Subduction tectonic erosion leading to forearc (imbricate) collapse: A major process
with implications for terrane juxtaposition, in PerezEstaun, A., and Coward, M. P., eds., Deformation and
plate tectonics: Tectonophysics, v. 191, p. 434.
Osmaston, M. F., 1992, A new model of subducting plate
downbend, subduction tectonic erosion, and the largescale tectonic evolution of convergent margins [abs.],
in International Geological Congress, 29th, Kyoto,
Japan, Abstracts, v. 1, p. 63.
Osmaston, M. F., 1994, A seismicity-compatible new
model of subducting plate downbend, basal subduction tectonic erosion (STE) and the large-scale tectonic
and seismogenic evolution of convergent margins
[abs.], in IASPEI’94, International Association for the
Seismology and Physics of the Earth’s Interior, 27th
General Assembly, January 1994, Wellington, New
Zealand, S8.24.
Osmaston, M. F., 1995a, Comparability of the flat-slab
sectors of the Andes and Sevier-Laramide USA:
Consequences of tectonic erosion and coupling at the
subducting plate downbend [abs.], in IUGG XXI
General Assembly, Boulder, CO, Abstracts, p. A441.
Osmaston, M. F., 1995b, A straightness mechanism for
MORs: A new view of ocean plate genesis and evolution [abs.], in IUGG XXI General Assembly, 2–14 July
1995, Boulder, CO, Abstracts, p. A472.
Osmaston, M. F., 1997, Pacific-style basal subduction tectonic erosion (STE): Essential precursor to construction of the Alps/Carpathians and Apennines [abs.], in
EUG9, March 1997, Strasbourg, France: Terra Nova, v.
9, p. 341.
Osmaston, M. F., 1999, Seismic coupling at subduction
zones; its mechanical cause and far-reaching geological consequences [abs.], in IUGG99, 19–30 July 1999,
Birmingham, UK, International Union of Geodesy and
Geophysics, p. A.157 [A.159 in abstracts update CD].
Osmaston, M. F., 2000a, What goes on beneath MORs? A
reassessment [abs.], in International Geological Congress, 31st, Rio de Janeiro, General Symposium 4-1,
abstracts CD.
Osmaston, M. F., 2000b, An upper mantle source for
plumes and Dupal; result of processes and history that
have shaped the Earth’s interior from core to crust
[abs.], in Goldschmidt 2000, Oxford, UK: Journal of
Conference Abstracts (CD-ROM), v. 5, no. 2, p. 763.
Osmaston, M. F., 2000c, “Lithosphere extension” that
isn’t [abs.], in 31st International Geological Congress,
6–17 August 2000, Rio de Janeiro, General Symposium 4-2.
Osmaston, M. F., 2003, What drives plate tectonics? Slab
pull, ridge push or geomagnetic torque from the CMB?
A new look at the old players vis-a-vis an exciting new
one [abs.], in XXIII IUGG 2003, 30 June–11 July
2003, Sapporo, Japan, p. B129 (abs. no. 016795-2).
Osmaston, M., 2005a, The ridge push mechanism of
MORs as the agent of seismic coupling, tsunami,
convergence partitioning and landward thrusting at
subduction zones; insights on an interactive family of
mostly-jerky mechanisms [abs.], in IASPEI 2005 General Assembly, 2–7 October 2005, Santiago, Chile;
JS04 Subduction zones with emphasis on those in less
developed countries, Abstract no. 303.
Osmaston, M., 2005b, Interrelationships between largescale plate motions as indicators of mantle structure:
New constraints on mantle modelling and compositional layout [abs.], in 3rd Workshop on Earth’s Mantle
Composition, Structure and Phase Transitions, 30th
Aug.–3 Sept. 2005: St Malo, France, Programme and
abstracts, p. 25 [see also http://deep.earth.free.fr/
participants.php].
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Osmaston, M. F., 2005c, Building the Earth’s core and D"
layer by “subduction”: Processes, timings and compositional implications [abs.], in William Smith Meeting,
Geological Society of London, 23-24 November 2005,
London, Abstracts p. 32–33.
Osmaston, M. F., 2005d, A new mechanism for intraplate
magmagenesis and petrogenetic variation: The importance of process [abs.], in Goldschmidt 2005, Moscow,
Idaho: Geochimica et Cosmochimica Acta, v. 69, no.
10S, p. A439.
Osmaston, M. F., 2006, Global tectonic actions emanating
from Arctic opening in the circumstances of a twolayer mantle and a thick-plate paradigm involving
deep cratonic tectospheres: The Eurekan (Eocene)
compressive motion of Greenland and other examples,
in Scott, R., and Thurston, D., eds., ICAM IV, Proceedings of the Fourth International Conference on Arctic
Margins, Halifax, Canada, 2003: OCS Study MMS
2006-003 (also available at http://www.mms.gov/
alaska/icam).
Osmaston, M. F., 2007, Cratonic keels and a two-layer
mantle tested: Mantle expulsion during Arabia-Russia
closure linked to westward enlargement of the Black
Sea, formation of the Western Alps and subduction of
the Tyrrhenian (not the Ionian) Sea [abs.], in XXIV
IUGG 2-13 July 2007, Perugia, Italy (http://
www.iugg2007perugia.it/webbook/), IASPEI Session
JSS 011, Earth Structure and Geodynamics, Abs. No.
2105.
Oxburgh, E. R., 1974, Eastern Alps, in Spencer, A. M.,
ed., Mesozoic–Cenozoic orogenic belts: Data for orogenic studies: Geological Society of London, Special
Publications, v. 4, p. 109–126.
Parkinson, C. D., 2000, Coesite inclusions and prograde
compositional zonation of garnet in whiteschist of the
HP-UHPM Kokchetav massif, Kazakhstan: A record of
progressive UHP metamorphism, in Carswell, D. A.,
ed., Ultra-high pressure metamorphic rocks: Lithos v.
52, p. 215–233.
Parkinson, C. D., Katayama, I., Liou, J. G., and
Maruyama, S., eds., 2002, The diamond-bearing
Kokchetav Massif, Kazakhstan: Petrochemistry and
tectonic evolution of an unique ultrahigh-pressure
metamorphic terrane: Tokyo, Japan, Universal Academy Press.
Parkinson, C. D., and Maruyama, S., 2004, Probable prevalence of coesite-stable metamorphism in collisional
orogens and a reinterpretation of Barrovian metamorphism [abs.], in 32nd International Geological Conference, August 2004, Florence, Italy, Abstracts v. 1, p.
104.
Parson, L. M., and Wright, I. C., 1996, The Lau-HavreTaupo back-arc basin: A southward-propagating, multistage evolution from rifting to spreading: Tectonophysics, v. 263, p. 1–22.
Pettinga, J. R., 1982, Upper Cenozoic structural history,
coastal Southern Hawke’s Bay, New Zealand: New
749
Zealand Journal of Geology and Geophysics, v. 25, p.
149–191.
Pfiffner, M., and Trommsdorff, V., 1998, The high-pressure ultramafic-mafic-carbonate suite of Cima LungaAdula, Central Alps: Excursions to Cima di Gagnone
and Alpe Arami: Schweizerische Mineralogische und
Petrographische Mitteilungen, v. 78, p. 337–354.
Pfiffner, O. A., Klaper, E. M., Mayerat, A.-M., and Heitzmann, P., 1990, Structure of the basement-cover contact in the Swiss Alps, in Roure, F., Heitzmann, P., and
Polino, R., eds., Deep structure of the Alps: Société
Géologique de France, Mémoire 156, p. 247–262.
Pilger, R. H., 1981, Plate reconstructions, aseismic
ridges, and low-angle subduction beneath the Andes:
Geological Society of America Bulletin, v. 92, p. 448–
456.
Platt, J. P., 1986, Dynamics of orogenic wedges and the
uplift of high-pressure metamorphic rocks: Geological
Society of America Bulletin, v. 97, p. 1037–1053.
Platt, J. P., 1987, The uplift of high-pressure–low-temperature metamorphic rocks: Philosophical Transactions
of the Royal Society of London, ser. A, v. 321, p. 87–
103.
Platt, J. P., 1993, Exhumation of high-pressure rocks: a
review of concepts and processes: Terra Nova, v. 5, p.
119–133.
Pleuger, J., Roller, S., Walter, J. M., Jansen, E., and
Froitzheim, N., 2007, Structural evolution of the contact between two Penninic nappes (Zermatt-Saas zone
and Combin zone, Western Alps) and implications for
the exhumation mechanism and palaeogeography:
International Journal of Earth Science, v. 96, p. 229–
252 [doi: 10.1007/s00531-006-0106-6].
Polino, R., Dal Piaz, G. V., and Gosso, G., 1990, Tectonic
erosion at the Adria margin and accretionary processes
for the Cretaceous orogeny of the Alps, in Roure, F.,
Heitzmann, P., and Polino, R., eds., Deep structure of
the Alps: Société Géologique de France, Mémoire 156,
p. 345–367.
Prince, R. A., Schweller, W. J., Coulburn, W. J., Shepherd,
G. L., Ness, G. E., and Masias, A., 1980, Bathymetry
of the Peru-Chile continental margin and trench:
Geological Society of America. Map & Chart Series,
MC-34 Part 1, scale 1:1 000 000, Sheet 5.
Proyer, A., 2003, The preservation of high-pressure rocks
during exhumation: Metagranites and metapelites:
Lithos, v. 70, p. 183–194.
Ramberg, H., 1980, Diapirism and gravitational collapse
in the Scandinavian Caledonides: Journal of the
Geological Society, v. 137, p. 261–270.
Ramsay, J. G., 1989, Fold and fault geometry in the western Helvetic nappes of Switzerland and France and its
implication for the evolution of the arc of the western
Alps, in Coward, M. P., Dietrich, D., and Park, R. G.,
eds., Alpine tectonics: Geological Society of London,
Special Publications, v. 45, p. 33–45.
750
MILES F. OSMASTON
Ranero, C. R., von Huene, R., Weinrebe, W., and
Reichert, C., 2006, Tectonic processes along the Chile
convergent margin, in Oncken, O., Chong, G., Franz,
G., Gieze, P., Götze, H.-J., Ramos, V. A., Strecker, M.
P., and Wigger, P., eds., The Andes: Active subduction
orogeny: Berlin/Heidelberg, Germany, Springer, p. 91–
121.
Ratschbacher, L., 1986, Kinematics of Austro-Alpine
cover nappes: changing path due to transpression: Tectonophysics, v. 125, p. 335–356.
Ratschbacher, L., Franz, L., Enkelmann, E., Jonckheere,
R., Pörschke, A., Hacker, B. R., Dong, S., and Zhang,
Y., 2006, The Sino-Korean-Yangtze suture, the Huwan
detachment, and the Paleozoic-Tertiary exhumation of
(ultra) high-pressure rocks along the Tongbai-XinxianDabie Mountains, in Hacker, B. R., McClelland, W. C.,
and Liou, J. G., eds., Ultrahigh pressure metamorphism: deep continental subduction. Geological Society of America Special Paper 403, p. 45–75 [doi:
10.1130/2006.2403(03)].
Ratschbacher, L., Frisch, W., Linzer, H.-G., and Merle, O.,
1991, Lateral extrusion in the Eastern Alps, Part 2:
Structural analysis: Tectonics, v. 10, p. 257–271.
Ravna, E. J. K., and Roux, M. R. M., 2006, Metamorphic
evolution of the Tønsvika Eclogite, Tromsø Nappe—
evidence for a new UHPM province in the Scandinavian Caledonides: International Geology Review, v. 48,
p. 861–881.
Reinecke, T., 1991, Very-high-pressure metamorphism
and uplift of coesite-bearing metasediments from the
Zermatt-Saas zone, Western Alps: European Journal of
Mineralogy, v. 3, p. 7–17.
Rickard, M. J., 1985, The Surnadal synform and basement
gneisses in the Surnadal-Sunndal district of Norway, in
Gee, D. G., and Sturt, B. A., eds., The Caledonide
Orogen—Scandinavia and related areas: Chichester,
UK, John Wiley, p. 487–497.
Ricou, L. E., and Siddans, A. W. B., 1986, Collision tectonics in the Western Alps, in Coward, M. P., and Ries,
A. C., eds., Collision tectonics: Geological Society of
London, Special Publications, v. 19, p. 229–244.
Roberts, D., 1980, Petrochemistry and palaeogeographic
setting of the Ordovician volcanic rocks of Smøla,
Central Norway: Norges Geologiske Undersøkelse, v.
359, p. 43–60.
Roberts, D., 2003, The Scandinavian Caledonides: event
chronology, palaeogeographic settings and likely modern analogues: Tectonophysics, v. 365, p. 283–299.
Robinson, P., Tucker, R. D., Bradley, D., Berry, H. N. I.,
and Osberg, P. H., 1998, Palaeozoic orogens in New
England, USA, in Sundquist, B., and Stephens, M. B.,
eds., Tectonics and geological history of some Phanerozoic orogens (David G. Gee honorary volume): GFF, v.
120, p. 119–148.
Rohrman, M., van der Beek, P. A., van der Hilst, R. D.,
and Reemst, P., 2002, Timing and mechanisms of
North Atlantic Cenozoic uplift: Evidence for mantle
upwelling, in Doré, A. H., Cartwright, J. A., Stoker,
M. S., Turner, J. P., and White, N., eds., Exhumation of
the North Atlantic margin: Timing, mechanisms and
implications for petroleum exploration: Geological
Society, London, Special Publications, v. 196, p. 27–
43.
Royden, L. H., 1985, The Vienna Basin: A thin-skinned
pull-apart basin: Society of Economic Paleontologists
and Mineralogists, Special Publication, v. 37, p. 319–
338.
Rubey, W. W., and Hubbert, M. K., 1959, Role of fluid
pressure in mechanics of overthrust faulting. II. Overthrust belt in geosynclinal area of western Wyoming in
light of fluid-pressure hypothesis: Bulletin of the
Geological Society of America, v. 70, p. 167–206.
Ruff, L., and Kanamori, H., 1980, Seismicity and the subduction process: Physics of the Earth and Planetary
Interiors, v. 23, p. 240–252.
Ruff, L., and Kanamori, H., 1983, Seismic coupling and
uncoupling at subduction zones: Tectonophysics, v. 99,
p. 99–117.
Ruff, L. J., 1996, Large earthquakes in subduction zones:
Segment interaction and recurrence times, in Bebout,
G. E., Scholl, D. W., Kirby, S. H., and Platt, J. P., eds.,
Subduction: Top to bottom: American Geophysical
Union, Geophysical Monograph 96, p. 91–104.
Rybach, L., Werner, D., Mueller, S., and Berset, G., 1977,
Heat flow, heat production and crustal dynamics in the
Central Alps, in Jessop, A. M., ed., Heat flow and geodynamics: Tectonophysics, v. 41, p. 113–126.
Sachsenhofer, R. F., 2000, Geodynamic controls on deposition and maturation of coal in the Eastern Alps, in
Neubauer, F., and Höck, V., eds., Aspects of geology in
Austria: Mitteilungen der Österreichischen Geologischen Gesellschaft, v. 92, p. 185–194.
Sager, W. W., MacLeod, C. J., and Abrahamsen, N., 1994,
Paleomagnetic constraints on Tonga arc tectonic rotation from sediments drilled at Sites 840 and 841, in
Hawkins, J., Parson, L., Allan, J., et al, eds., Proceedings of the Ocean Drilling Program, Scientific Results,
v. 135: College Station, TX, p. 763–783.
Sanford, A. R., 1959, Analytical and experimental study
of simple geologic structures: Geological Society of
America Bulletin, v. 70, p. 19–51.
Savelieva, G. N., and Nesbitt, R. W., 1996, A synthesis of
the stratigraphy and tectonic setting of the Uralian
ophiolites: Journal of the Geological Society, London,
v. 153, no. 4, p. 525–537.
Scaillet, S., 1998, K-Ar 40Ar/39Ar geochronology of ultrahigh-pressure rocks, in Hacker, B. R., and Liou, J. G.,
eds., When continents collide: Geodynamics of ultrahigh-pressure rocks: Dordrecht, the Netherlands,
Kluwer Academic Publishers, p. 161–202.
Schilling, F. R., Trumbull, R. B., Brasse, H., Haberland,
C., Asch, G., Bruhn, D., Mai, K., Haak, V., Giese, P.,
Muñoz, M., Ramelow, J., Rietbrock, A., Ricaldi, E.,
and Vietor, T., 2006, Partial melting in the Central
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Andean crust: A review of geophysical, petrophysical,
and petrologic evidence, in Oncken, O., Chong, G.,
Franz, G., Gieze, P., Götze, H.-J., Ramos, V. A.,
Strecker, M. P., and Wigger, P., eds., The Andes:
Active subduction orogeny: Berlin/Heidelberg, Germany, Springer, p. 459–474.
Schmid, S., 1975, The Glarus overthrust: Field evidence
and mechanical model: Eclogae Geologicae Helvetiae,
v. 68, p. 247–280.
Schmid, S., and Haas, R., 1989, Transition from near-surface thrusting to intrabasement decollement; Schlinig
thrust, Eastern Alps: Tectonics, v. 8, p. 697–718.
Schmid, S., Pfiffner, O. A., Schönborn, G., Froitzheim, N.,
and Kissling, E., 1997, Integrated cross section and
tectonic evolution of the Alps along the the eastern
traverse, in Pfiffner, O. A., Lehner, P., Heitzmann, P.,
Mueller, S., and Steck, A., eds., Deep structure of the
Swiss Alps: Results of NRP20: Basel, Switzerland,
Birkhäuser, p. 289–304.
Schmid, S. M., Aebli, H. R., Heller, F., and Zingg, A.,
1989, The role of the Periadriatic Line in the tectonic
evolution of the Alps, in Coward, M. P., Dietrich, D.,
and Park, R. G., eds., Alpine tectonics: Geological
Society of London, Special Publications, v. 45, p. 153–
171.
Schmid, S. M., Rück, P., and Schreurs, G., 1990, The significance of the Schams nappes for the reconstruction
of the paleotectonic and orogenic evolution of the Penninic zone along the NFP-20 East traverse (Grisons,
eastern Switzerland), in Roure, F., Heitzmann, P., and
Polino, R., eds., Deep structure of the Alps, Société
Géologique de France, Mémoire 156, p. 263–287.
Schoenbohm, l., and Strecker, M. P., 2005, Extension of
the Puna-Altiplano plateau since 1-2 Ma: Lithospheric
delamination, lower crustal flow or gravitational collapse? [abs.]: Geological Society of America, Abstracts
with Programs, v. 37, no. 7, p. 272.
Scholl, D. W., von Huene, R., Vallier, T. L., and Howell, D.
G., 1980, Sedimentary masses and concepts about tectonic processes at underthrust ocean margins: Geology,
v. 8, p. 564–568.
Schreurs, G., 1993, Structural analysis of the Schams
nappes and adjacent tectonic units: Implications for
the orogenic evolution of the Penninic zone in eastern
Switzerland: Bulletin de la Société Géologique de
France, v. 164, p. 415–435.
Schreyer, W., 1995, Ultradeep metamorphic rocks: the retropective viewpoint: Journal of Geophysical Research,
v. 100B, p. 8353–8366.
Schweller, W. J., Kulm, L. D., and Prince, R. A., 1981,
Tectonics, structure and sedimentary framework of the
Peru-Chile Trench: Geological Society of America,
Memoir, v. 154, p. 323–350.
Segawa, J., and Tomoda, Y., 1976, Gravity measurements
near Japan and study of the upper mantle beneath the
oceanic trench-marginal sea transition zones: Ameri-
751
can Geophysical Union, Geophysical Monograph, v.
19, p. 35–52.
Seidel, E., Okrusch, M., Kreuzer, H., Raschka, H., and
Harre, W., 1976, Eo-alpine metamorphism in the
uppermost unit of the Cretan nappe system–petrology
and geochronology. Part 1. The Léndas Area (Asterousia Mountains): Contributions to Mineralogy and
Petrology, v. 57, p. 259–275.
Seidel, E., Okrusch, M., Kreuzer, H., Raschka, H., and
Harre, W., 1981, Eo-alpine metamorphism in the
uppermost unit of the Cretan nappe system—petrology
and geochronology. Part 2. Synopsis of high-temperature metamorphics and associated ophiolites: Contributions to Mineralogy and Petrology, v. 76, p. 351–
361.
Séranne, M., 1992, Late Paaeozoic kinematics of the
Møre-Trøndelag fault zone and adjacent areas, central
Norway: Norsk Geologisk Tidsskrift, v. 72, p. 141–158.
Shervais, J. W., Kimbrough, D. L., Renne, P., Hanan,
B. B., Murchey, B., Snow, C. A., Zoglman Shuman,
M. M., and Beaman, J., 2004, Multi-stage origin of the
Coast Range Ophiolite, California: Implications for the
life cycle of supra-subdution zone ophiolites: International Geology Review, v. 46, p. 289–315.
Shutian, S., Zengqiu, Z., Hanwen, Z., Zhendong, Y., Hongfei, Z., and Li, Z., 2005, Tectonic evolution of the
Dabie-Sulu UHP and HP metamorphic belts, eastcentral China: Structural record in UHP rocks: International Geology Review, v. 47, p. 1207–1221.
Silver, P. G., and Russo, R. M., 1995, The origin of the
Andean Orogen since the mid-Tertiary [abs.], in IUGG
XX1 General Assembly. Boulder, CO, p. A441.
Skår, Ø., and Pedersen, R. B., 2003, Relations between
granitoid magmatism and migmatization: U-Pb
geochronological evidence from the Western Gneiss
Complex, Norway: Journal of the Geological Society, v.
160, p. 935–946.
Smellie, J. L., and Stone, P., 2001, Geochemical characteristics and geotectonic setting of early Ordovician
lavas in the Ballantrae complex, SW Scotland: Transactions of the Royal Society of Edinburgh, Earth
Sciences, v. 91, p. 539–555.
Smith, A. G., 2006, Tethyan ophiolite emplacement,
Africa to Europe motions, and Atlantic spreading, in
Robertson, A. H. F., and Mountrakis, D., eds., Tectonic
development of the Eastern Mediterranean region:
Geological Society, Special Publication 260, p. 11–34.
Smith, D. C., 1980, A tectonic mélange of foreign eclogites
and ultramafites in the Basal Gneiss Region, West
Norway: Nature, v. 287, p. 366–367.
Smith, D. C., 1984, Coesite in clinopyroxene in the Caledonides and its implication for geodynamics: Nature,
v. 310, p. 641–644.
Smithson, S. B., Ramberg, I. B., and Grönlie, G., 1974,
Gravity interpretation of the Jotun nappe of the Norwegian Caledonides: Tectonophysics, v. 22, p. 359–366.
752
MILES F. OSMASTON
Sokol, A. G., and Pal’yanov, Y. N., 2008, Diamond formation in the system MgO–SiO2–H2O–C at 7.5 GPa and
1600°C: Contributions to Mineralogy and Petrology, v.
155, no. 1, p. 33–43 [doi: 10.1007/s00410-007-02219].
Somoza, R., 1998, Updated Nazca (Farallon)–South America relative motions during the last 40 My: Implications for mountain building in the central Andean
region: Journal of South American Earth Sciences, v.
11, no. 3, p. 211–215.
Sperner, B., Lorenz, K., Bonjer, K., Hettel, S., Müller, B.,
and Wenzel, F., 2001, Slab break-off—abrupt cut or
gradual detachment? New insights from the Vrancea
Region (SE Carpathians, Romania): Terra Nova, v. 13,
no. 3, p. 172–179.
Spjeldnaes, N., 1985, Biostratigraphy of the Scandinavian
Caledonides, in Gee, D. G., and Sturt, B. A., eds., The
Caledonide Orogen—Scandinavia and related areas:
Chichester, UK, Wiley, p. 317–329.
Stampfli, G. M., and Marthaler, M., 1990, Divergent and
convergent margins in the northwestern Alps: Confrontation to actualistic models: Geodynamica Acta
(Paris), v. 4, no. 3, p. 159–184.
Staub, R., 1937, Gedanken zum Bau der Westalpen zwischen Bernina und Mittlemeer. II Teil: Vierteljahrsschrift der Naturforschenden Gesellschaft (Zurich), v.
82, p. 197–336.
Stauder, W., 1975, Subduction of the Nazca plate under
Peru as evidenced by focal mechanisms and by seismicity: Journal of Geophysical Research, v. 80, p.
1053–1064.
Stearns, D. W., 1978, Faulting and forced folding in the
Rocky Mountains foreland, in Matthews, V., III, ed.,
Laramide folding associated with basement block
faulting in the western United States: Geological Society of America Memoir 151, p. 1–37.
Steck, A., 1980, Deux directions principales de flux synmétamorphiques dans les Alpes centrales: Bulletin de
la Société vaudoise de Science naturelle, v. 75, p. 141–
149.
Stephens, M. B., (co-ordinator and compiler), 1986,
Stratabound sulphide deposits in the AppalachianCaledonian orogen: Map 1649: Geological Survey of
Canada, Scale 1:5,000,000.
Stephens, M. B., Furnes, H., Robins, B., and Sturt, B. A.,
1985, Igneous activity within the Scandinavian Caledonides, in Gee, D. G., and Sturt, B. A., eds., The
Caledonide Orogen—Scandinavia and related areas:
Chichester, UK, Wiley, p. 623–656.
Strachan, R. A., Holdsworth, R. E., Friderichsen, J. D.,
and Jepson, H. T., 1992, Regional Caledonian structure within an oblique convergence zone, Dronning
Louise Land, NE Greenland: Journal of the Geological
Society, London, v. 149, p. 359–379.
Suo, S., Zhong, Z., and You, Z., 1999, Location of Triassic
suture between collided Sino-Korean and Yangtze cra-
tons in Dabie-Sulu Region, China: Journal of China
University of Geosciences, v. 10, p. 281–286.
Suo, S., Zhong, Z., Zhou, H., and You, Z., 2002, Triassic
collisional structures and post-collisional deformation
of Bixiling UHP rock stack: Insights for tectonic evolution of UHP metamorphic belt in Dabie Massif, Central China: Journal of China University of Geosciences,
v. 13, p. 1–13.
Tarif, P. A., Wilkens, R. H., Cheng, C. H., and Paillet,
F. L., 1988, Laboratory studies of the acoustic properties of samples from Salton Sea Scientific Drilling
Project and their relation to microstructure and field
measurements: Journal of Geophysical Research, v.
93, p. 13,057–13,067.
Terabayashi, M., Okamoto, K., Yamamoto, H., Kaneko, Y.,
Ota, T., Maruyama, S., Katayama, I., Komiya, T.,
Ishikawa, A., Anma, R., Ozawa, H., Windley, B. F.,
and Liou, J. G., 2005, Accretionary complex origin of
the mafic-ultramafic bodies of the Sanbagawa Belt,
Central Shikoku, Japan: International Geology
Review, v. 47, p. 1058–1073.
Terry, M. P., Robinson, P., Hamilton, M. A., and Jerçinovic, M. J., 2000, Monazite geochronology of UHP
and HP metamorphism, deformation, and exhumation,
Nørdoyane, Western Gneiss Region, Norway: American Mineralogist, v. 85, p. 1651–1664.
Thöny, W., Ortner, H., and Scholger, R., 2006, Paleomagnetic evidence for large en-bloc rotations in the Eastern Alps during Neogene orogeny: Tectonophysics, v.
414, p. 169–189.
Thornburg, T. M., and Kulm, L. D., 1981, Sedimentary
basins of the Peru continental margin: Structure,
stratigraphy and Cenozoic tectonics from 6° to 16°S
Latitude: Geological Society of America Memoir, v.
154, p. 393–422.
Tollmann, A., 1977, Geologie von Österreich. Band 1, Die
Zentralalpen: Vienna, Austria, Deuticke, 766 p.
Tollmann, A., 1978, Plattentectonische Fragen in den
Östalpen: Mitteilungen österreichischen Geologischen
Gesellschaft, v. 69 (for 1976), p. 291–351.
Tollmann, A., 1980, Geology and tectonics of the eastern
alps (Middle Sector), Guidebook G-05 for Excursion
035A, in International Geological Congress, 26th,
Paris, Abhandlungen der Geologischen Bundesanstalt, v. 34, p. 197–255.
Tollmann, A., 1986, Geologie von Österreich. Band III.
Gesamtübersicht: Vienna, Austria, Franz Deuticke,
718 p.
Trommsdorff, V., 1990, Metamorphism and tectonics in
the Central Alps: The Alpine lithospheric mélange of
Cima Lunga and Adula: Memoria della Società Geologia Italiana, v. 45, p. 39–49.
Trommsdorff, V., and Nievergelt, P., 1983, The Bregaglia
(Bergell) Iorio intrusive and its field relations: Memoria della Società Geologia Italiana, v. 26, p. 55–68.
BASAL SUBDUCTION TECTONIC EROSION AND UHP BELT CONSTRUCTION
Trümpy, R., 1960, Paleotectonic evolution of the Central
and Western Alps: Bulletin of the Geological Society of
America, v. 71, p. 843–908.
Trümpy, R., 1973, The timing of events in the Central
Alps, in de Jong, K. A., and Scholten, R., eds., Gravity
and tectonics: New York, NY, John Wiley, p. 229–251.
Trümpy, R., 1975, Penninic-Austroalpine boundary in the
Swiss Alps: A presumed former continental margin
and its problems: American Journal of Science, v.
275A, p. 209–238.
Trümpy, R., 1980, An outline of the geology of Switzerland: Basel, Switzerland, Wepf & Co., 104 p.
Trümpy, R., 1982, Alpine paeogeography: A reappraisal,
in Hsu, K.J., ed., Mountain building processes: London, UK, Academic Press, p. 149–156.
Trümpy, R., 1988, A possible Jurassic-Cretaceous transform system in the Alps and Carpathians, in Clark, S.
P., Jr., Burchfiel, B. C., and Suppe, J., eds., Processes
in continental lithosphere deformation. Geological
Society of America Special Paper, v. 218, p. 93–109.
Trümpy, R., 1997, Switzerland, in Moores, E. M., and Fairbridge, R. W., eds., Encyclopaedia of European and
Asian Geology: London, UK, Chapman and Hall, p.
704–710.
Trümpy, R., 2001, Why plate tectonics was not invented in
the Alps: International Journal of Earth Science, v. 90,
p. 477–483.
Tsujimori, T., Sisson, V. B., Liou, J. G., Harlow, G. E., and
Sorensen, S. S., 2006, Very-low-temperature record of
the subduction process: A review of world-wide lawsonite eclogites, in Sajeev, K., and Santosh, M., eds.,
Extreme crustal metamorphism and related crust-mantle processes: Lithos, v. 92, nos. 3–4, p. 609–624 [doi:
10.1016/j.lithos.2006.03.054].
Usui, T., Nakamura, E., Kobayashi, K., Maruyama, S., and
Helmstaedt, H., 2003, Fate of the subducted Farallon
plate inferred from eclogite xenoliths in the Colorado
Plateau: Geology, v. 31, no. 7, p. 589–592.
Uyeda, S., 1983, Comparative subductology: Episodes, v.
6, no. 2, p. 19–24.
Uyeda, S., and Kanamori, H., 1979, Back-arc opening and
the mode of subduction: Journal of Geophysical
Research, v. 84, p. 1049–1061.
Vallier, T. L., Scholl, D. W., Fischer, M. A., and Bruns, T.,
1994, Geologic framework of the Aleutian arc, Alaska,
in Plafker, G., and Berg, H. C., eds., The geology of
Alaska; The Geology of North America, (DNAG) vol.
G.1: Boulder, CO, Geological Society of America, p.
367–388.
Vialon, P., Rochette, P., and Ménard, G., 1989, Indentation and rotation in the western Alpine arc, in Coward,
M. P., Dietrich, D., and Park, R. G., eds., Alpine tectonics: Geological Society of London, Special Publications, v. 45, p. 329–338.
Viganò, A., Martin, S., and Ranalli, G., 2007, Shallow
crustal seismicity, magnitudes and focal mechanisms
753
in the Giudicarie belt (eastern Alps, Italy), in XXIV
IUGG 2007: Perugia, Italy [http://www.iugg2007perugia
.it/webbook/], IASPEI Session SW 002 Geophysical
studies of active faults. abstr. No 6525.
Vigny, C., Simons, W. J. F., Abu, S., Bamphenyu, R., Satirapod, C., Choosakul, N., Subarya, C., Socquet, A.,
Omar, K., Abidin, H. Z., and Ambrosius, B. A. C.,
2005, Insight into the Sumatra–Andaman earthquake
from GPS measurements in southeast Asia: Nature, v.
436, p. 201–206 [doi:10.1038/nature03937].
von Blanckenburg, F., Kagami, H., Deutsch, A., Wiedenbeck, M., Oberli, M., Meier, M., and Fischer, H., 1998,
The origin of Alpine plutons along the Periadriatic
Lineament: Schweizerische Mineralogische und Petrographische Mitteilungen, v. 78, p. 55–66.
von Huene, R., Ranero, C. R., and Vannucchi, P., 2004,
Generic model of subduction erosion: Geology, v. 32,
no. 10, p. 913–916.
von Quadt, A., Peytcheva, I., Bolz, V., and Heinrich, C. A.,
2005, Cretaceous magmatism and Cu-Au mineralization in the region of the Apuseni–Banat–Timok–
Sredno-gorie belt—constrains from U-Pb zircons and
Re-Os molybdenite dating: Geochemica et Cosmochemica Acta, v. 98 (10S), p. A568.
Wagreich, M., 1995, Subduction tectonic erosion and Late
Cretaceous subsidence along the northern Austroalpine margin (Eastern Alps, Austria): Tectonophysics,
v. 242, p. 63–78.
Wakabayashi, J., and Dumitru, T. A., 2007, 40Ar/39Ar ages
from coherent, high-pressure metamorphic rocks of the
Franciscan Complex, California: Revisiting the timing
of the world’s type subduction complex: International
Geology Review, v. 49, p. 873–906.
Wang, Y., 2006, The onset of the Tan-Lu fault movement
in eastern China: Constraints from zircon (SHRIMP)
and 40Ar/39Ar dating: Terra Nova, v. 18, no. 6, p. 423–
431 [doi: 10.1111/j.1365-3121.2006.00708.x].
Wang, Q., Shi, Y., Lin, W., and Guo, J., 2008, Exhumation
of the Dabie UHP Terrane, China: International Geology Review, v. 50, p. 15–31.
Wang, X., Liou, J.-G., and Mao, H. K., 1989, Coesite-bearing eclogites from the Dabie Mountains in central
China: Geology, v. 17, p. 1085–1088.
Watson, J. V., 1984, The ending of the Caledonian orogeny
in Scotland: Journal of the Geological Society, London,
v. 141, p. 193–214.
Watts, A. B., 2001, Isostasy and flexure of the lithosphere:
Cambridge, UK, Cambridge University Press, 458 p.
Webb, L. E., Leech, M. L., and Yang, T. N., 2006, 40Ar/
39Ar thermochronology of the Sulu terrane: Late Triassic exhumation of high- and ultrahigh-pressure rocks
and implications for Mesozoic tectonics in East Asia,
in Hacker, B. R., McClelland, W. C., and Liou, J. G.,
eds., Ultrahigh pressure metamorphism: Deep continental subduction: Geological Society of America Special Paper 403, p. 77–92 doi: 10.1130/2006.2403(04).
754
MILES F. OSMASTON
Wenzel, F., and Sandmeier, K.-J., 1992, Geophysical evidence for fluids in the crust beneath the Black Forest,
SW Germany: Earth Science Reviews, v. 32, p. 61–75.
Wood, B. L., and Yuen, D. A., 1983, The role of lithospheric phase transitions on seafloor flattening at old
ages: Earth and Planetary Science Letters, v. 66, p.
303–314.
Xu, Z., Zeng, L., Liu, F., Yang, J., Zhang, Z., McWilliams,
M., and Liou, J. G., 2006, Polyphase subduction and
exhumation of the Sulu high-pressure and ultrahighpressure metamorphic terrane, in Hacker, B. R.,
McClelland, W. C., and Liou, J. G., eds., Ultrahigh
pressure metamorphism: Deep continental subduction: Geological Society of America Special Paper 403,
p. 93–113 [doi: 10.1130/2006.2403(05)].
Ye, K., Cong, B., and Ye, D., 2000, The possible subduction of continental material to depths exceeding 200
km: Nature, London, v. 407, p. 734–736.
Yuan, X., Sobolev, S. V., Kind, R., Oncken, O., Bock, G.,
Asch, G., Schurr, B., Graeber, F., Rudloff, A., Hanka,
W., Wylegalla, K., Tibi, R., Haberland, C., Rietbrock,
A., Giese, P., Wigger, P., Röwer, P., Zandt, G., Beck, S.,
Wallace, T., Pardo, M., and Comte, D., 2000, Subduction and collision processes in the Central Andes
constrained by converted seismic phases: Nature, v.
408, p. 958–961.
Zheng, Y.-F., Zhou, J.-B., Wu, Y.-B., and Xie, Z., 2005,
Low-grade metamorphic rocks in the Dabie-Sulu
orogenic belt: A passive-margin accretionary wedge
deformed during continent subduction: International
Geology Review, v. 47, no. 8, p. 851–871.
Zimmermann, R., Hammerschmidt, K., and Franz, C.,
1994, Eocene high-pressure metamorphism in the
Penninic units of the Tauern Window (Eastern Alps):
evidence from 40 Ar/ 39 Ar dating and petrological
evidence: Contributions to Mineralogy and Petrology,
v. 117, p. 175–186.