ARTICLES

PUBLISHED ONLINE: XX MONTH XXXX | DOI: 10.1038/NGEO1124

Yo-yo subduction recorded by accessory minerals

in the Italian Western Alps
Daniela Rubatto1 *, Daniele Regis2 , Jrg Hermann1 , Katherine Boston1 , Martin Engi2 ,
Marco Beltrando3 and Sarlae R. B. McAlpine1
High-pressure metamorphic rocks form during subduction of Earths crust to mantle depths at convergent plate margins.
Their exposure at the surface of Earth today provides a record of the subduction zone process. In general, such metamorphic
rocks record only a single cycle of subduction and exhumation, yet tectonic models suggest that individual rock units
should undergo multiple subductionexhumation cycles. Here we investigate the microstructure and chemical composition
of metamorphic minerals in high-pressure rocks exposed in the Sesia zone in the Italian Western Alps. We find that the
minerals white mica, garnet, allanite and zircon each exhibit multiple generations of mineral overgrowths. In particular, two
generations of overgrowths in white micas contain high-silicon contents, indicative of formation at high pressures, separated
by overgrowths with low-silicon contents, formed during exhumation to low pressures. Furthermore, the trace-element
signatures of distinct zones within zircon and allanite also reveal two episodes of high-pressure metamorphism, separated
by a period of rapid exhumation. We use uraniumlead dating of zircon and allanite overgrowths to constrain the timing of
this subductionexhumationsubduction cycle to 7965 Myr ago. We conclude that slices of the Sesia zone crust experienced
two cycles of burial to mantle depths in less than 20 million years. The yo-yo subduction occurred during oblique convergence
between the African and European plates, which involved a continental margin.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

22
23
24
25
26
27
28
29

ubduction zones are a unique feature of the Earth and

provide important insights into how plate tectonics works.
High-pressure (HP, eclogite and blueschist facies) rocks have
played a crucial role in characterizing the evolution of subduction
zones1,2 . Forty years of intense research into the petrology of
high-pressure terranes has provided important information on
several aspects of subduction, notably (1) the maximum depth to
which rocks now exposed at the surface have been subducted; and
(2) how fast they went down and how fast they were exhumed38 .
As the petrologic record of metamorphic rocks is intrinsically
discrete and depends on episodic re-crystallization, the detailed
dynamics of subduction zones are only poorly known. On the
basis of pressuretemperature (PT ) trajectories derived from
individual eclogite-facies rock samples, the dominant view is that
subducted rocks now found at the surface undergo a simple
loop from blueschist- to eclogite-facies conditions, followed by
rapid retrograde decompression9 . However, conceptual10,11 and
numerical12 models of subduction suggest that under specific
circumstances, tectonic units may undergo multiple subduction
exhumation cycles. So far, observations on natural-rock samples
from orogenic belts supporting this possibility are sparse.

Subduction in the Western Alps

The Western Alps (Supplementary Fig. S1a) formed as a result
of the convergence between Europe and Adria, a promontory of
Africa. This convergence led to the progressive subduction of units
of the Adria rifted margin, the Tethys ocean and the European
margin between 70 and 35 Myr ago1316 . Existing tectonic models
account for a single burialexhumation cycle for each unit. This
study focuses on rocks from the Sesia zone, which derives from

the thinned Adriatic margin17 . The Sesia zone is a predominantly

continental block (Supplementary Fig. S1) that has been traditionally subdivided into three subunits on the basis of lithostratigraphy
and Alpine metamorphic grade: Eclogitic Micaschist Complex
(EMC), Second DioriticKinzingitic zone and Gneiss Minuti18 .
During the Alpine orogeny, the entire EMC reached eclogite-facies
conditions of 1.52.0 GPa and 550600 C, indicating subduction
to 5070 km depth19,20 . The Second DioriticKinzingitic zone
mainly preserves pre-Alpine amphibolite-facies assemblages with
localized Alpine blueschist re-equilibration. The Gneiss Minuti
recorded mild Alpine metamorphism, generally in greenschist facies, that occasionally reached omphacite, but not jadeite, grade.
Previous analyses of the chronology of eclogites using UPb in
titanite and zircon13,21 and LuHf in garnet14 point to a widely
accepted eclogitic stage at 7065 Myr bp for the EMC.

Evidence of polyphase eclogite-facies conditions

31
32
33
34
35
36
37
38
39
40
41
42
43
44

45

Detailed fieldwork and investigation of various samples of the

southern part of the EMC provide new evidence for a much
more complex evolution of eclogite-facies metamorphism. Here we
report the findings of key samples.
Micaschists (samples SL21 and SL22) from Brosso, Val Chiusella
(Supplementary Fig. S1), were sampled in the area of the Sesia zone
where the oldest evidence of Alpine deformation is preserved22 .
The foliation (S1) is marked by the typical HP assemblage
phengite, paragonite, and glaucophane or jadeite-rich omphacite
that coexist with garnet, rutile and accessory apatite, zircon
and allanite. Zr-in-rutile and garnetphengite thermobarometry
analyses combined with the Si content in phengite indicate
temperatures of 570580 C and pressures of 2.1 GPa (details in

1 Research

School of Earth Sciences, The Australian National University, Mills Road, Canberra 0200, Australia, 2 Institute of Geological Sciences, University
of Bern, Baltzerstrasse 3, CH-3012 Bern, Switzerland, 3 Department of Mineralogical and Petrological Sciences, University of Torino, via Valperga Caluso 35,
I-10125 Torino, Italy. *e-mail: Daniela.rubatto@anu.edu.au.
NATURE GEOSCIENCE | ADVANCE ONLINE PUBLICATION | www.nature.com/naturegeoscience

30

46
47
48
49
50
51
52
53
54
55
56
57
58

NATURE GEOSCIENCE DOI: 10.1038/NGEO1124

ARTICLES
a

a SL1

Z3

Z2

SL1

Z3

Z1

Omp

Ab

Ab

D
Z3

SL1

d
Ab

SL21

B24

All core

Z1

B24

Jd

Ep rim
10 m

All mantle

500 m
2 LP
Phe

Phe2 LP

Phe3
HP2

Phe1
HP1

10,000
1,000
Zircon/chondrite

S3

100

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

Supplementary Note). Sample SL21 is important because the age of

high-pressure metamorphism is significantly older than recognized
before in the Sesia zone (see below).
A phengite-bearing quartzite with abundant accessory allanite/epidote from Cima Bonze, Scalaro valley (B24) contains structural evidence for a complex HP evolution. The white micas are
aligned along two foliations that can be regionally correlated to S2
and S3. The white micas have three distinct compositions (Fig. 1b,c
and Supplementary Fig. S4): relict cores of high-Si phengite (Phe1)
are rimmed by low-Si phengite (Phe2 aligned along S2), which in
turn is rimmed by a second generation of high-Si phengite (Phe3)
that defines the main foliation (S3). In the same quartzite, rare
earth element (REE)-rich allanite cores have mantles of REE-poor
allanite and external rims of epidote (Fig. 2e,f and Supplementary
Fig. S5). These observations provide evidence for a succession
from high-pressure (S1) to low-pressure (S2), followed by a new
high-pressure stage (S3).
An eclogitic micaschist (SL1) from Quincinetto in the Aosta
valley contains jadeite, phengite, quartz and minor amounts
of garnet, omphacite, rutile, and accessory allanite, apatite and
zircon. This sample also has petrographic evidence for two stages
of high-pressure metamorphism (Fig. 1a). Eclogite-facies jadeitic
pyroxene (HP1 stage) is replaced by albite (LP stage), which is
2

Z3: 7762 Myr

0.01

Figure 1 | Mineral textural relationships showing the sequence HPLPHP.

a, Backscatter electron micrograph of jadeite (Jd) breaking down to albite
(Ab), which in turn is overgrown by omphacite (Omp). Sample SL21.
b, Sketch of the textural relationships between different mica generations in
sample B24. c, Photomicrograph of zoned phengite grains (crossed
polarizer).

Z2: 73.7 0.9 Myr

10

0.10

200 m

Z1: 78.6 0.9 Myr

SL21: 76.8 0.9 Myr

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Figure 2 | Accessory-mineral texture and composition.

ac, Cathodoluminescence images of zircon grains that preserve three
distinct metamorphic rims (R1, R2 and R3) in sample SL1.
d, Cathodoluminescence image of a zircon grain with a large metamorphic
rim in sample SL21. e,f, Backscatter electron micrographs of zoned allanite
in quartzite B24. In af, circles indicate the location of the SHRIMP analysis
(diameter 25 m). g, The REE composition and age of the different zircon
domains. Average ages are reported at the 95% confidence level.

then partially replaced by omphacite (HP2 stage). See details in

Supplementary Note. Furthermore, the sample contains garnet
with two distinct growth zones (Supplementary Fig. S3). Jadeite
inclusions in garnet indicate a first HP assemblage, whereas
omphacite coexists with the garnet rim and constitute the HP2
assemblage. Conditions of 580 C and 2.2 GPa are indicated
for the HP1 stage, as thermodynamic models predict mineral
compositions matching those in the assemblage garnetphengite
jadeiteparagonitequartz. Thermobarometry using the garnet rim
in equilibrium with phengite and omphacite, and Zr-in-rutile
thermometry give 560 C and 2.0 GPa for the HP2 stage. The
stability of sodic plagioclase during the LP stage in between the two
HP assemblages is constrained to pressures lower than 1.4 GPa for
the temperature of 550 C.

Accessory-phase record of multiple eclogite-facies stages

Zircon from micaschists SL1 and SL21, and allanite from sample
B24 were investigated by cathodoluminescence (Fig. 2ad) and
backscatter electron microscopy (Fig. 2e,f), respectively. Imaging
revealed complex growth zones that were characterized using Th,
UPb ages, trace elements and mineral inclusions (Supplementary
Table S4S7 and Figs S6 and S7).
Zircons from micaschists SL1 and SL21 contain rounded cores of
variable age that are interpreted to represent detrital zircon grains
of the original sedimentary rock. In sample SL1, three distinct

NATURE GEOSCIENCE | ADVANCE ONLINE PUBLICATION | www.nature.com/naturegeoscience

24
25
26
27
28
29
30
31
32
33
34
35
36
37

38
39
40
41
42
43
44
45
46
47

NATURE GEOSCIENCE DOI: 10.1038/NGEO1124

a

ARTICLES
b

S1, HP1
S2
S3, HP2
eclogite
low P
eclogiteblueschist
78.5 0.9 HP zircon
Include Jd Phe Rt
73.7 0.9 LP zircon
Micaschist SL1
7762 HP zircon
Include Omp Phe Rt
Micaschist SL21 76.8 0.9 HP zircon
75.6 0.8
allanite core

Quartzite B24

Age (Myr)

rth

75

65

A
e
pin
Al

69.8 0.8 allanite rim

70

ria

Ibe

?
Africa

66 1 Myr titanite
80

At
lan

tic

65 3 Myr
HP zircon
76 1 Myr LP zircon
69 3 Myr garnet

Previous works
Ref. 13
Ref. 14
Ref. 21

Europe

No

60

th
Te

Sesia
Adria

ys

67.7
79.1 73.6
83.0

Continental plate
Oceanic plate
Thrust
Africa movement vector

Figure 3 | Age and palaeogeographic setting for the yo-yo subduction of the Sesia zone. a, Summary of ages for the Sesia high-pressure unit from this
study and previous works. b, Palaeogeographic setting of the Western Tethys at 83 Myr BP (mainly based on ref. 42). The position of the Sesia zone17
(Se) and the vector of relative motion between the African and European plates, with interposed Alkapeca microcontinent (A), in the Cretaceous period15
are shown. Brown shading indicates the distal Adriatic continental margin. Possible segmentation is shown in the Adriatic margin. The comparison of time
constraints from a and b suggests that the yo-yo subduction occurred during highly oblique convergence.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

metamorphic zircon domains are present, the oldest of which grew

directly on some detrital cores. This texturally older zircon rim
(Z1 in Fig. 2ac) shows dark cathodoluminescence emission and
weak zoning. Sensitive high-resolution ion micro probe (SHRIMP)
dating of this domain yields a 206 Pb/238 U age of 78.50.9 Myr (95%
confidence level). Inclusions consist of phengite, rutile and jadeite
and provide evidence that this zircon domain formed during the
jadeiteeclogite stage (HP1). Ti-in-zircon thermometry23 indicates
temperatures of 560620 C. Zircon1 REE patterns show no
significant Eu anomaly and a relatively steep heavy-REE pattern
(Fig. 2g). The lack of a negative Eu anomaly indicates the absence
of coexisting plagioclase24 , and thus zircon growth at high pressure,
beyond the stability of feldspar. The steep heavy-REE pattern of
zircon1 suggests that garnet, which readily sequesters heavy-REE,
was not a major constituent of the assemblage.
This early HP stage is also recorded in the zircons of micaschist
SL21, which have a single generation of metamorphic rims on
detrital cores (Fig. 2d). The rims yield an age of 76.8 0.9 Myr, in
line with the observation that an early Alpine foliation is dominant
in this area22 . The REE pattern of these zircons is characterized by a
flat heavy-REE and only a weak negative Eu anomaly (Eu/Eu 0.5,
Fig. 2g), suggesting growth at high pressure in the presence of garnet
and lack of plagioclase24 . Ti contents are low (mainly <2 ppm),
indicating a T of 560620 C.
In micaschist SL1, a second, younger zircon domain generally
surrounds the detrital cores in place of the older zircon1. Zircon2
(Z2 in Fig. 2ac) overgrowths have intermediate cathodoluminescence emission, exhibit no zoning and yield an age of 73.70.9 Myr.
The REE pattern shows a marked negative Eu anomaly and a
steep heavy-REE pattern (Fig. 2g), indicating growth within the
stability field of feldspar and in the absence of garnet, and therefore at lower pressures than the 78.5 Myr stage. Ti-in-zircon
thermometry yields temperatures of 610660 C, but the buffering
assemblage is uncertain.
The texturally youngest zircon domain in micaschist SL1 (Z3
in Fig. 2ac) occurs either as rims around zircon1 domains or
directly on detrital cores. The age of this third overgrowth is
not well constrained, with apparent 206 Pb/238 U ages scattering
between 77 and 62 Myr. This age scatter could be partly due to
physical mixing with older domains, given the particularly small
size of these overgrowths, or to growth over an extended period
of time. Phengite, rutile and omphacite inclusions were found
within this domain, providing evidence that the third overgrowth

formed during the omphaciteeclogite stage (HP2). Ti-in-zircon

thermometry returns temperatures of 560660 C. REE patterns
show a weak Eu anomaly (Eu/Eu = 0.240.69), again suggesting
the absence of feldspar. The heavy-REEs vary significantly from
steep to flat patterns (Fig. 2g). The progressive heavy-REE depletion
in Z3 overgrowths is interpreted as the consequence of concomitant
garnet crystallization.
Impure quartzite B24 contains two generations of allanite that
were dated by ion microprobe analysis using the ThPb system. The
epidote rim was too rich in common Pb to yield any significant
age information. The REE-rich allanite cores yield consistent
208
Pb/232 Th ages with an average of 75.6 0.8 Myr. Abundant
mica inclusions in these domains are identical in composition to
the texturally oldest phengite preserved in the sample (Phe1 in
Fig. 1b). The REE-poor allanite mantle is systematically younger, at
69.80.8 Myr, and contains inclusions of low-pressure mica (Phe2
in Fig. 1b and Supplementary Fig. S4). The two allanite populations
are thus indicative of a HPLP cycle that pre-dates the 65 Myr bp
HP stage13,14,21 , which is probably represented by the texturally late
HP foliation (S3 marked by Phe3 in Fig. 1b) preserved in the sample.

Two subductionexhumation cycles in 20 Myr

45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63

64

The combined investigation of structures, petrology and

geochronology using major and accessory phases revealed two
stages of eclogite-facies metamorphism, one at 7975 Myr bp and
one at 7065 Myr bp, separated by a low-pressure stage (Fig. 3a).
Despite extensive investigation of the high-pressure Sesia zone
and, to the best of our knowledge, in other eclogite-facies terranes
worldwideno double loop has been demonstrated and timed
so far. In the Sesia zone, the two stages produced similar mineral
assemblages with comparable mineral composition. Two stages
of eclogite-facies metamorphism within single samples (SL21 and
B24) have been distinguished unambiguously by geochronology
and by linking the growth of the accessory minerals with the main
mineral assemblages and the foliations. In all three samples, the
older eclogite metamorphic stage is preserved in zircon or allanite
that contains HP minerals inclusions such as jadeite (SL21) or
high-Si phengite (B24). These samples were targeted because they
best record the earlier stages of metamorphism. Although the
younger eclogite stage could not be dated precisely here, that stage
has been widely documented by previous geochronological13,14,21
and petrological studies19,20,25 . A low-pressure stage, with zircon
and allanite ages of 7075 Myr, separates the two high-pressure

NATURE GEOSCIENCE | ADVANCE ONLINE PUBLICATION | www.nature.com/naturegeoscience

44

65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85

NATURE GEOSCIENCE DOI: 10.1038/NGEO1124

ARTICLES

19

stages (Fig. 3a). Jadeite breakdown textures and trace-element

signatures of zircon in micaschist SL1 indicate that plagioclase
(probably albite-rich) was stable, and hence metamorphic pressure
was at least 0.5 GPa lower than during the first eclogite-facies
stage. The inclusions of low-Si phengite in the REE-poor allanite
of quartzite B24 are another indicator of low-pressure conditions.
A low-pressure metamorphic stage at 7570 Myr bp is also
supported by a metamorphic vein found in the Aosta valley,
which contains zircon formed at low pressure at 76 1 Myr bp
(ref. 13). Blueschist-facies shear zones have been dated at 64 Myr
(ref. 22), suggesting a short residence time at depth for the
second cycle as well. Therefore, our study shows that a rock unit
experienced two subductionexhumation cycles within less than
20 Myr. The slight overlapping of high-pressure and low-pressure
ages between samples of the Sesia zone (Fig. 3a) could be due to
actual diachronous metamorphism across the zone. Alternatively,
some bias was introduced by different techniques, as for example
dating of prograde garnet14 , or formation of HP zircon and allanite
by different reactions.

20

Implications for subduction-zone models

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65

Our case study provides new evidence for short-term, multiple

eclogite-facies metamorphism within a single unit, indicating
fast burial and exhumation cycles of rocks along convergent
plate margins. Pressure cycles of low amplitude have previously
been detected2629 . The timescale of <20 Myr for the double-loop
metamorphism is in agreement with exhumation of high-pressure
rocks at plate tectonic rates of several centimetres per year68 . The
timescale and tectonic implications of such yo-yo subduction are
significantly different from subduction of high-grade metamorphic
rocks formed during a previous orogeny30 .
Burialexhumation cycles of the type presented here may
in principle result from orogen-scale switches from shortening
to extension11,31 . Alternatively, such cycles may result from
Franciscan-type subduction channel dynamics12,32 , as recently
documented in a serpentinites mlange in Cuba33 . However, models
of rapid burial and exhumation within a thin and rapidly shearing
layer of unconsolidated sediments32 or hanging-wall serpentinites12
are not applicable to the Sesia zone, which largely consists of
laterally continuous slivers of Palaeozoic continental basement18 .
Palaeogeographic and plate kinematic studies provide good
constraints on the large-scale geodynamic setting in which such a
complex evolution took place. In the Jurassic period, the Sesia zone
was part of the distal rifted margin of the Adriatic continental plate17
(Fig. 3b). During the early stages of convergence, in the Cretaceous
period, this margin was oriented at a high angle to the direction
of relative motion between the African and European plates15
(Fig. 3b). At the onset of convergence, deformation localized within
the distal Adriatic continental margin, before involving the Western
Tethys slab sensu stricto16 . In this setting characterized by oblique
subduction, small changes in the relative direction of plate motion
and along-strike variations in the plate margin geometry may
have pronounced effects on the force balance acting on subducted
rocks. This may lead to episodic switches from transpressional to
transtensional deformation. Large, coherent slivers of subducted
material, once detached from the downgoing slab, may reside
along the interface with the upper plate and undergo episodic
exhumation in response to variations in boundary forces. Renewed
mechanical coupling with the subducting plate may lead to
further tectonic burial.
In any case, the first finding of rocks experiencing two
loops within less than 20 Myr of eclogite-facies metamorphism
provides evidence that the pressuretemperaturetime trajectories
of eclogite-facies rocks can be much more complex than previously
thought and testifies to the highly dynamic character of subduction
zones and plate margins at large.
4

Methods

66
67

Imaging. Accessory zircon and allanite were retrieved by mineral separation.

Grains were mounted in epoxy resin and polished down to expose the
near-equatorial section. Imaging of grain sections was carried out at the Electron
Microscopy Unit at The Australian National University (ANU) in Canberra.
Cathodoluminescence investigation of zircon employed a Hitachi S2250N
scanning electron microscope (SEM) supplied with an ellipsoidal mirror for
cathodoluminescence. Operating conditions for the SEM were 15 kV, 60 A and
a 20 mm working distance. Backscattered electron micrographs of allanite were
obtained with a Cambridge S360 SEM using a voltage of 20 kV, current of 2 nA
and a working distance of 1720 mm.

68

SHRIMP. Zircon and allanite were analysed for U, Th and Pb in the epoxy
mount using the SHRIMP (reverse geometry) at the ANU in Canberra. For
zircon, instrumental conditions and data acquisition were generally as described
previously34 . The data were collected in sets of six scans throughout the masses
and a reference zircon (transmission electron microscopy35 ) was analysed each
fourth analysis. The analyses (Supplementary Tables S4 and S5) were corrected
for common Pb on the basis of the measured 207 Pb/206 Pb ratios as described
previously34 . The common Pb composition was assumed to be that predicted by
the model in ref. 36. For allanite, instrumental conditions and data acquisition were
as described previously37 , with isotope data collected from sets of six scans through
the masses. The measured 208 Pb/232 Th ratio was corrected using the allanite
standard TARA (417 Myr; ref. 37). A secondary standard (Bona, 208 Pb/232 Th age of
30.1 Myr; ref. 38) returned ages within 1% of the nominal value. All analyses were
corrected for common Pb on the basis of measured 207 Pb/206 Pb and 208 Pb/206 Pb
(ref. 37). Common Pb measured in allanite is essentially inherent; therefore, an
estimate of initial Pb composition at the time of crystallization from an evolving
model Pb composition was assumed36 . This assumption was justified by applying
free regressions to the uncorrected data: for the allanite rims the TeraWasserburg
initial 207 Pb/206 Pb intercepts of 0.831 0.027 (2 ) is within error of the model Pb
compositions at 70 Myr bp. For both minerals, data evaluation and age calculation
were done using the software Squid and Isoplot/Ex (ref. 39), respectively. Average
ages are quoted at the 95% confidence level.

69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99

Trace elements. Trace elements of zircon and rutile (Supplementary Table S6)
were analysed by laser-ablation inductively coupled plasma mass spectrometry at
the Research School of Earth Sciences (ANU) using a pulsed 193 nm ArF excimer
laser with 100 mJ output energy at a repetition rate of 5 Hz (ref. 40) coupled to
an Agilent 7,500 quadrupole inductively coupled plasma mass spectrometer. A
synthetic glass (NIST 612 for zircon and NIST 610 for rutile) was used for external
calibration, and reference values were taken from ref. 41. Internal standards
were SiO2 determined either with the SEM-EDS (allanite) or stoichiometric
SiO2 (zircon), and a TiO2 of 98% for rutile. A BCR-2G glass was used as a
secondary standard.

100

Element maps. Element maps for garnet (Supplementary Fig. S3) were produced
using an SX100 electron microprobe housed at the Research School of Earth
Sciences (ANU) using a beam current of 40 nA and a dwell time of 100 ms. Mineral
composition data (Supplementary Tables S1 and S2) were acquired using a JEOL
6400 SEM (Electron Microscopy Unit, ANU) with EDS operating at 15 kV and
a focused beam of 1 nA.
Quantitative EMP analyses for major elements of phengite and allanite on
sample B24 (Supplementary Table S3) were obtained on a JEOL JXA8200 at
the University of Bern.

110

Received 19 August 2010; accepted 25 February 2011;

published online XX Month XXXX

References
1. Ernst, W. G. Tectonic history of subduction zones inferred from retrograde
blueschist PT paths. Geology 16, 10811084 (1988).
2. Godard, G. Eclogites and their geodynamic interpretation: A history. J. Geodyn.
32, 165203 (2001).
3. Chopin, C. Coesite and pure pyrope in high grade blueschists of the western
Alps: A first record and some consequences. Contrib. Mineral. Petrol. 86,
107118 (1984).
4. Smith, D. C. Coesite in clinopyroxene in the Caledonides and its implications
for geodynamics. Nature 310, 641644 (1984).
5. Sobolev, N. & Shatsky, V. Diamond inclusions in garnets from metamorphic
rocks: A new environment for diamond formation. Nature 343,
742746 (1990).
6. Rubatto, D. & Hermann, J. Exhumation as fast as subduction? Geology 29,
36 (2001).
7. Baldwin, S. L. et al. Pliocene eclogite exhumation at plate tectonic rates in
eastern Papua New Guinea. Nature 431, 263267 (2004).
8. Parrish, R. R., Gough, S. J., Searle, M. P. & Waters, D. J. Plate velocity
exhumation of ultra-high pressure eclogites in the Pakistan Himalaya. Geology
34, 989992 (2006).

NATURE GEOSCIENCE | ADVANCE ONLINE PUBLICATION | www.nature.com/naturegeoscience

101
102
103
104
105
106
107
108
109

111
112
113
114
115
116
117
118

119
120

121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140

NATURE GEOSCIENCE DOI: 10.1038/NGEO1124

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56

9. Agard, P., Yamato, P., Jolivet, L. & Burov, E. Exhumation of oceanic blueschists
and eclogites in subduction zones: Timing and mechanisms. Earth-Sci. Rev. 92,
5379 (2009).
10. Cloos, M. Flow melanges: Numerical modelling and geologic constraints on
their origin in the Franciscan subduction complex, California. Geol. Soc. Am.
Bull. 93, 330345 (1982).
11. Brueckner, H. K. Dunk, dunkless and re-dunk tectonics: A model for
metamorphism, lack of metamorphism, and repeated metamorphism of
HP/UHP terranes. Int. Geol. Rev. 48, 978995 (2006).
12. Gerya, T., Stockhert, B. & Perchuk, A. Exhumation of high-pressure
metamorphic rocks in a subduction channel: A numerical simulation. Tectonics
21, T10566 (2002).
13. Rubatto, D., Gebauer, D. & Compagnoni, R. Dating of eclogite-facies zircons:
The age of Alpine metamorphism in the SesiaLanzo Zone (Western Alps).
Earth Planet. Sci. Lett. 167, 141158 (1999).
14. Duchne, S. et al. The LuHf dating of garnets and the ages of the Alpine
high-pressure metamorphism. Nature 387, 586589 (1997).
15. Rosenbaum, G., Lister, G. S. & Duboz, C. Relative motions of Africa, Iberia
and Europe during Alpine orogeny. Tectonophysics 359, 117129 (2002).
16. Beltrando, M., Compagnoni, R. & Lombardo, B. (Ultra-) High-pressure
metamorphism and orogenesis: An Alpine perspective. Gondwana Res. 18,
147166 (2010).
17. Froitzheim, N. & Manatschal, G. Kinematics of Jurassic rifting, mantle
exhumation, and passive-margin formation in the Austroalpine and Penninic
nappes (eastern Switzerland). Geol. Soc. Am. Bull. 108, 11201133 (1996).
18. Compagnoni, R. The SesiaLanzo Zone: High pressurelow temperature
metamorphism in the Austroalpine continental margin. Rend. Soc. Ital.
Mineral. Petrol. 33, 335375 (1977).
19. Pognante, U. Tectonic implications of lawsonite formation in the Sesia zone
(Western Alps). Tectonophysics 162, 219227 (1989).
20. Konrad-Schmolke, M., Babist, J., Handy, M. R. & OBrien, P. J. The
physico-chemical properties of a subducted slab from garnet zonation patterns
(Sesia Zone, Western Alps). J. Petrol. 47, 21232148 (2006).
21. Inger, S., Ramsbotham, W., Cliff, R. A. & Rex, D. C. Metamorphic evolution
of the SesiaLanzo Zone, Western Alps: Time constraints from multi-system
geochronology. Contrib. Mineral. Petrol. 126, 152168 (1996).
22. Babist, J., Handy, M. R., Konrad-Schmolke, M. & Hammerschmidt, K.
Precollisional, multistage exhumation of subducted continental crust: The
Sesia Zone, western Alps. Tectonics 25, 125 (2006).
23. Watson, E. B., Wark, D. A. & Thomas, J. B. Crystallization thermometers for
zircon and rutile. Contrib. Mineral. Petrol. 151, 413433 (2006).
24. Rubatto, D. Zircon trace element geochemistry: Distribution coefficients
and the link between UPb ages and metamorphism. Chem. Geol. 184,
123138 (2002).
25. Konrad-Schmolke, M., OBrien, P. J. & Zack, T. Fluid migration above a
subducted slabconstraints on amount, pathways and major element mobility
from partially overprinted eclogite-facies rocks (Sesia Zone, Western Alps).
J. Petrol. (2011, in the press).
26. Beltrando, M., Hermann, J., Lister, G. & Compagnoni, R. On the evolution of
orogens: Pressure cycles and deformation mode switches. Earth Planet. Sci. Lett.
256, 372388 (2007).
27. Harris, C. R., Hoisch, T. D. & Wells, M. L. Construction of a composite
pressuretemperature path: Revealing the synorogenic burial and exhumation
history of the Sevier hinterland, USA. J. Metamorphic Geol. 25, 915934 (2007).
28. Umhoefer, P. J. et al. Yo-yo tectonics in a wrench zone, Central Anatolian fault
zone, Turkey. Spec. Pap. Geol. Soc. Am. 3557 (2007).

ARTICLES
29. Kabir, M. F. & Takasu, A. Evidence for multiple burialpartial exhumation
cycles from the Onodani eclogites in the Sambagawa metamorphic belt, central
Shikoku, Japan. J. Metamorphic Geol. 28, 873893 (2010).
30. Herwartz, D. et al. Tracing two orogenic cycles in one eclogite sample by LuHf
garnet chronometry. Nature Geosci. 4, 178183 (2011, on line).
31. Cloos, M. & Shreve, R. L. Subduction channel model of prism accretion,
melange formation, sediment subduction, and subduction erosion at
convergent plate margins: 1. Background and description. Pure Appl. Gephys.
128, 455500 (1988).
32. Lister, G. S., Forster, M. A. & Rawling, T. J. in Continental Reactivation and
Reworking, Vol. 184 (eds Miller, J. A., Holdsworth, R. E., Buick, I. S. & Hand,
M.) 89113 (Geological Society of London, 2001).
33. Blanco-Quintero, I. F., Garcia-Casco, A. & Gerya, T. V. Tectonic blocks in
serpentinite mlange (eastern Cuba) reveal large-scale convective flow of the
subduction channel. Geology 39, 7982 (2011).
34. Williams, I. S. in Application of Microanalytical Techniques to Understanding
Mineralizing Processes (eds McKibben, M. A., Shanks, W. C. III & Ridley, W. I.)
135 (Reviews in Economic Geology, Society of Economic Geologists, Vol. 7,
1998).
35. Black, L. P. et al. TEMORA 1: A new zircon standard for Phanerozoic UPb
geochronology. Chem. Geol. 200, 155170 (2003).
36. Stacey, J. S. & Kramers, J. D. Approximation of terrestrial lead evolution by a
two-stage model. Earth Planet. Sci. Lett. 26, 207221 (1975).
37. Gregory, C. et al. Allanite micro-geochronology: A LAICPMS and SHRIMP
UThPb study. Chem. Geol. 245, 162182 (2007).
38. von Blanckenburg, F. Combined high-precision chronometry and geochemical
tracing using accessory minerals: Applied to the Central-Alpine Bergell
intrusion (central Europe). Chem. Geol. 100, 1940 (1992).
39. Ludwig, K. R. (Berkeley Geochronological Centre Spec. Pub. 2003), pp. 70.
40. Eggins, S. M., Rudnick, R. L. & McDonough, W. F. The composition of
peridotites and their minerals: A laser ablation ICP-MS study. Earth Planet. Sci.
Lett. 154, 5371 (1998).
41. Pearce, N. J. G. et al. A compilation of new and published major and trace
element data for NIST SRM 610 and NIST SRM 612 glass reference materials.
Geostand. Newsl. 21, 115144 (1997).
42. Michard, A. et al. How does the Alpine belt end between Spain and Morocco?
Bull. Soc. Ge. Fr. 173, 315 (2002).

57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93

Acknowledgements

94

We thank I. Campbell and H. Brueckner for constructive comments. This research has
been supported by the Australian Research Council (DP0556700) and the Swiss National
Science Foundation (project 200020_126946).

96

Author contributions

98

D. Rubatto and K.B. contributed the geochronological and chemical data; D. Regis
and M.E. provided data for sample B24 and regional knowledge; J.H. contributed to
petrological investigation of samples SL2122 and SL1; M.B. provided the tectonic
model; S.R.B.M. contributed to the investigation of sample SL21. D. Rubatto and J.H.
wrote the paper.

97

99
100
101
102
103

Additional information

104

The authors declare no competing financial interests. Supplementary information

accompanies this paper on www.nature.com/naturegeoscience. Reprints and permissions
information is available online at http://npg.nature.com/reprintsandpermissions.
Correspondence and requests for materials should be addressed to D. Rubatto

NATURE GEOSCIENCE | ADVANCE ONLINE PUBLICATION | www.nature.com/naturegeoscience

95

105
106
107
108

Page 1
Query 1: Line no. 1
Please note that title has been edited as per
journal style. OK?
Query 2: Line no. 1
Please note that the first paragraph has been
edited according to style.
Query 3: Line no. 16
Hyphen added, to give blueschist- here. OK?
Query 4: Line no. 27
According to style, Ma has been changed to
Myr throughout, and where appropriate, BP or
ago has been added. Please check throughout.
Query 5: Line no. 42
Text reworded to Previous analyses of the
chronology of eclogites here. OK?
Query 6: Line no. 57
analyses added here. OK?
Page 2
Query 7: Line no. 5
Please provide text, to be added to the caption,
to define Phe1, Phe2, Phe3, HP1, HP2
and S3 in figure 1 (and, to follow style, then
amend main text to avoid unnecessary use of
acronyms/abbreviations).
Query 8: Line no. 24
Please provide text, to be added to the caption,
to define Ep, D, Z1, Z2, Z3 in figure 2 (and,
to follow style, then amend main text to avoid
unnecessary use of acronyms/abbreviations).
Query 9: Line no. 28
Should the text here be, for example, and
together they constitute or whereas the coexistence
of omphacite and the garnet rim constitutes. or
should constitute be constitutes?
Page 3
Query 10: Line no. 2
Text formatted as eclogiteblueschist in figure
3a, with an en rule (). OK? (See query above for
further information about en rule/hyphen.)
Query 11: Line no. 2
In figure 3a, should Myr be added for all
numbers in the upper part of the figure, as has been
done in the lower part of the figure?

Query 12: Line no. 2

Can (Se) be deleted from figure 3s caption, as
such text does not appear in the figure?
Query 13: Line no. 8
Text formatted as jadeiteeclogite, where the en
rule () represents a conjunction or preposition. OK
(or is jadeite acting as an adjective modifying
eclogite, in which case the hyphen will be
restored)?
Query 14: Line no. 44
Text formatted as omphaciteeclogite here,
with an en rule. OK? (See query above for further
information about en rule/hyphen.)
Query 15: Line no. 46
Eu assumed to represent europium here. OK?
(Please provide text to clarify what the asterisk
represents.)
Page 4
Query 16: Line no. 63
PTtime changed to pressuretemperature
time here. OK?
Query 17: Line no. 66
Heading Methods added here. OK?
Query 18: Line no. 79
Text amended to the SHRIMP (reverse
geometry) here. OK?
Query 19: Line no. 82
TEM expanded to transmission electron
microscopy here. Please check, and please check the
placement of this text within the sentence.
Query 20: Line no. 86
Author names have been removed throughout
this paragraph, according to style. Please check.
Query 21: Line no. 96
Should the text here be intercept ... is or
intercepts ... are?
Query 22: Line no. 107
Please check the sentence Internal standards
were ... for rutile. as the meaning seems unclear.
(Also, please give EDS in full, used here and once
belowSEM is already defined.)
Query 23: Line no. 116
Please provide text to define EMP, and should
this sentence run on from the paragraph above, is
it a separate paragraph, or is it a separate section
requiring a new heading?

Page 5
Query 24: Line no. 48
Any update for ref. 25?
Query 25: Line no. 56
Please provide volume number for ref. 28.
Query 26: Line no. 61
Please provide volume and page/article number
for ref. 30.
Query 27: Line no. 85
Please provide book title for ref. 39.
Query 28: Line no. 103
In the last line of author contributions, D.R.
changed to D. Rubatto. OK?