crustal attenuation as a tracer for the emplacement of the...

1614

Crustal attenuation as a tracer for the emplacement of the Beni Bousera ultramafi c massif (Betico-Rifean belt)

Antonio M. Álvarez-Valero1,†, Oliver Jagoutz1, Jessica Stanley1,§, Christian Manthei1, Abdelkader El Maz2, Ali Moukadiri3, and Alison Piasecki1,#

1Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA2Department of Geology, University Moulay Ismail, Meknes, 11201 Beni M’Hamed, Morocco3Department of Geology, University Mohamed Ben Abdellah, 1796 Fès-Atlas, Morocco

ABSTRACT

The study of petrology (fi eldwork, petrog-raphy, and phase diagram modeling) and structural data of the metapelitic granulites and the southern, high-temperature exposed peridotites in the Beni Bousera massif (north-ern Morocco), combined with results from previous regional studies of the Alborán, sug-gest a new emplacement mechanism for the mantle rocks in the Betico-Rifean belt. We document two key metamorphic episodes in the granulites within a temperature window of 710–830 ± 50 °C: (1) An earlier prograde high-pressure period (from 9 ± 1.0 to 12 ± 1.0 kbar) characterized by the assemblage gar-net + biotite + kyanite + K-feldspar + rutile. Pressure differences of ~3 kbar are found over a continuous crustal section of ~1.5 km of exposed granulites that indicate a signifi -cant crustal attenuation during exhumation of the ultramafi c rocks; and (2) a later post-kinematic low-pressure (5 ± 0.8 kbar) sym-plectic assemblage of cordierite + spinel + plagioclase + sillimanite.

At the scale of the entire Betico-Rifean belt, two main contacts are observed as mirror images in both sides of the Alborán Sea: (1) the long axis of the high-temperature duc-tile contact between granulites and peridotites occurs in the west side of the Beni Bousera and Ronda massifs, coupled with (2) the consistent high-angle, east-dipping normal fault in the

east parts of the belt massifs. The integra-tion of the petrologic results with information on the rotation of both contacts reveals duc-tile deformation in the lower crust related to the emplacement of the ultramafi c rocks in the Betico-Rifean belt along deep-reaching normal faults. The presence of the early high-temperature contact suggests that it was origi-nally a shallow, west-dipping detachment fault developed in a back-arc environment of the east-dipping, retreating subduction zone (cur-rent western part of the Gibraltar arc).

This scenario is in concordance with the tectonic evolution in western Italy, where anticlockwise Pleistocene rotations associ-ated with northeast-directed thrusting in the Apennines—and coeval with the southeast-ward motion in the Calabria-Peloritani ter-rane—were triggered by retreat and rollback of the Adriatic-Ionian slab toward the south-east during the northwest-directed subduc-tion beneath the Calabrian arc.

INTRODUCTION

The presence of large coherent bodies of mantle rocks exposed within the continental crust (i.e., orogenic lherzolite massifs) is per-plexing, as mantle rocks have densities 300–500 kg/m3 higher than their host crustal rocks. The emplacement mechanism of these rocks into the crust remains under discussion (e.g., Bodinier and Godard, 2005; Jagoutz et al., 2006; Labrousse et al., 2011). Many large coherent mantle bodies related to the emplacement of ophiolites (e.g., Oman, Himalaya) are gener-ally thrusted, i.e., obducted, onto the continental crust (e.g., Bard, 1983; Hacker et al., 1996). For instance, peridotite types associated with high- and ultra-high-pressure metamorphic terranes are exhumed by the return fl ux of the subducted continental crust and associated emplacement of

mantle rocks into the continental crust at high temperature (e.g., Western Gneiss Region, Nor-way; Dabie Shan, China) (e.g., Labrousse et al., 2011). Other exhumed pieces of lower crust–upper mantle complexes are also described in passive-margin settings (e.g., Alps, west Iberian margin, Red Sea; Fügenschuh et al., 1997; Müntener et al., 2000).

The ultramafi c massifs in the Betico-Rifean belt (e.g., Ronda in southern Spain, and Beni Bousera in northern Morocco) are among the largest exposures of mantle rocks on Earth’s surface, yet the mechanisms leading to their emplacement into the continental crust are still under active discussion, because these pieces of mantle rocks do not conform to the men-tioned emplacement models. Associated basal-tic ophio litic segments near the Betic ultramafi c bodies are either related to small amounts of extension (e.g., Puga et al., 1999) or missing, and the present east-northeastern crustal bound-ary of the peridotitic body is formed by non-metamorphic rocks (e.g., El Maz and Guiraud, 2001). Instead, both the Beni Bousera and Betic ultramafi c peridotites bodies (Fig. 1A) have systematic but highly variable contact relation-ships with their surrounding host. The hinter-land contact (southern and northern contact in Beni Bousera and Ronda, respectively) is a high-tempera ture contact juxtaposing unaltered, strongly sheared, high-temperature peridotite and mostly metapelitic, high-grade, locally mylonitized granulites (so-called kinzingites) that continuously grade—over a distance of 8 km—into lower grade metamorphic rocks (El Maz and Guiraud, 2001). The foreland contact (northern and southern contact of Beni Bousera and Ronda, respectively) is character-ized by seaward-dipping high-angle normal faults that emplaced unmetamorphosed fl ysch-type rocks of the Maghrebian nappes over strongly serpentinized ultramafi c rocks (e.g.,

For permission to copy, contact [email protected]© 2014 Geological Society of America

GSA Bulletin; November/December 2014; v. 126; no. 11/12; p. 1614–1624; doi: 10.1130/B31040.1; 6 fi gures; 1 table; Data Repository item 2014238; published online 30 June 2014.

†Current address: Departamento de Geología, Uni-versidad de Salamanca, 37008 Salamanca, Spain; [email protected].

§Current address: Department of Geological Sci-ences, University of Colorado, Boulder, Colorado 80309, USA

#Current address: Division of Geological and Plan etary Sciences, California Institute of Techno-logy, Pasadena, California 91125, USA

Beni Bousera massif emplacement

Geological Society of America Bulletin, November/December 2014 1615

Chalouan and Michard, 2004). The ultramafi c bodies are considered to be elongated sheets (e.g., Gysi et al., 2011), and the lowermost contact is only exposed in the Betic side. This contact is interpreted to be a late-stage shear zone that resulted in the thrusting of the ultra-mafi c material and surrounding crustal rocks landward (e.g., Tubía et al., 1993).

Geochronological studies on the country rocks in both Beni Bousera (Montel et al., 2000) and Ronda (Sánchez-Rodríguez and Gebauer, 2000), as well as data on pyroxenite dikes in the mantle rocks (ca. 19 Ma: Polvé, 1983; ca. 22 Ma: Reisberg et al., 1989; ca. 25 Ma: Blichert-Toft et al., 1999), indicate that the mantle rocks were exhumed during early Miocene extension, and

subsequently thrusted onto the Iberian (north) and Moroccan (south) margins in the middle Miocene (18–15 Ma; Esteban et al., 2004). Fission-track (in zircon and apatite), U-Pb (in zircon), and Ar-Ar (in hornblende, muscovite, and biotite) ages reveal fast cooling in the inter-val 21.2–20.4 Ma associated with the Miocene exhumation event (Platt et al., 2003). Granitic

Ichendirene

~~

~~~~

~~~~~~~~~~

~~~~

~~~~

~~~~

~~~~

~~

~~~~

~~

~~

~~~~

~~~~

~~~~

~~

~~~

~

~~

~~~

~~

~

~

~~

~~

~~

~

~~

~~

~ ~

~~

~~

~~

~

~

~

~ ~~

~ ~ ~~~

~ ~~

~ ~~

~~~

~~

~~~ ~

~

~

~~ ~~~~

~~

~~~

~~~~

~~

~ ~

~~

~~

~

~

~~

~~~

~

~~

~

~~

~~

~~~

~

~

~

~~

~

~

~

~

~ ~~

~~~~

~~~ ~

~~~ ~~~~

~~

~

~

~~~

~~

~~

~~

~

~~

~

~~

~~~~ ~~ ~~ ~ ~~

~~

~~~~ ~ ~ ~

~~~~ ~~

~~~~ ~~ ~~

~~~ ~

~~~

~

~~~~

~ ~~~ ~~ ~~

~~~

~

~

AA′

Alborán Sea

A

Tendmane

IberianForeland

Betic UM exposures

African Foreland

Alborán SeaTell

38°N

36°N

4°W 0°W34°N

BeniBousera

100 Km

N

Amtar

BB-11-23

27A34A

BB-11-17

Fig.6B

Yahia Aarab Contacts5° - 20°21°-40°41°-65°65°-90°

xsection

n.faultinferred

Foliationattitude

Rock UnitsBeni Bousera UltramaficsSheared peridotiteGranuliteFilali gneiss and micaschist Gomarides shales

~~~~

Rif

+

N

G

G

P

P

C

Chmaala beach (Fig. 1B)

4 Km1 3

4°48′0″W4°50′0″W4°52′0″W4°54′0″W4°56′0″W35°

19′N

35°14′N

A A′

BB-11-56 (Fig.2A)BB-11-49 (Fig.2B,C)BB-11-2

NW SE

1.4 km exposed granulitic rocksB

9 kbar 12 kbar10 kbarBB-11-33 (Fig.2F)

BB-11-3010 kbar

Figure 1. (A) Simplifi ed geologic map of the Beni Bousera peridotite massif (50 m contour interval). UM—ultramafi c; n. fault—normal fault. (B) Schematic Chmaala beach cross-section on the granulites. Pressure value uncertainties are ±1 kbar (2σ). Note in the granu-lites the shift in foliation dip and thickness toward the peridotite contact, and the different lenses and inclusions of mafi c material, quartz veins, and leucocratic dikes. (C) Stereographic projections of foliation (poles, lower hemisphere; black dots for granulites [G] and black squares for peridotites [P]) and lineation (white circles for granulites and white squares for peridotites). Gray areas (folia-tion) and white areas of dashed contour (lineation) are from Afi ri et al. (2011).

Álvarez-Valero et al.

1616 Geological Society of America Bulletin, November/December 2014

dikes formed by crustal melting are commonly found at the contact between the peridotites and granulites, and have been also dated through U-Pb sensitive high-resolution ion microprobe analysis in zircons within the range of 22–19 Ma (Esteban et al., 2011). An emplacement age of 21.8 ± 0.5 Ma has also been determined using the same technique on surrounding zircon-bear-ing chlorite schists (Esteban et al., 2007). The mechanism and the driving forces for the exhu-mation of the mantle rocks and the surrounding crustal rocks are under active debate: e.g., Obata (1980) proposed a model of emplacement based on rapid cooling of garnet lherzolites along the crustal contact, and slower cooling of spinel and plagio clase lherzolite at shallower levels. Platt et al. (2003) suggested that ductile normal faults reaching down into the lithospheric mantle (and subsequent thrusting of the external thrust belts) led to the emplacement of upper mantle rocks. Tubía et al. (2004) proposed that a nar-row mantle diapir was responsible for peridotite exhumation. Afi ri et al. (2011) described the exhumation by a model of lithospheric thinning in the footwall of an extensional shear zone. Mazzoli and Martín-Algarra (2011) showed that Miocene emplacement of the Ronda peridotites within the Betic Cordillera may be interpreted in terms of deformation partitioning associated with oblique convergence during continental subduction and subsequent exhumation, involv-ing the coeval activity of kinematically linked systems of reverse, strike-slip, and normal-sense shear zones.

Contrary to the well-studied but rarely exposed high-temperature contact of the ultramafi c Ronda bodies, the contact relationships in Beni Bousera are well exposed along the coast, in the Filali area (Ichendirene, Fig. 1A), and very locally at the southeast part of the massif (e.g., sample BB-11-17 in Fig. 1A), but have not been studied with modern petrologic or thermodynamic tools, including phase diagram modeling. The combi-nation of our results with the geological record and previous petrologic, structural, and geo-chronologic studies on the Betic Cordillera—Ronda (Sierra Bermeja and Sierra Alpujata) and Carratraca massifs (e.g., Lundeen, 1978; Tubía and Cuevas, 1986; Argles et al., 1999; Tubía et al., 2004; Garrido et al., 2011)—allows us to propose an alternative geodynamic scenario for the exhumation of the ultramafi c massifs that resolves the long-lasting discussion of the emplacement of these massifs.

STRUCTURE

At Beni Bousera, a high-temperature (high-T) package of granulitic rocks (~1.5 km in lateral extent exposed material) rims the massif along

the western to the southeastern contact (Figs. 1A, 1B). The northwesternmost part of the massif, along the beach cliffs, offers spectacu-lar exposure of a continuous section from the perido tite contact along the entire granulite out-crop (Figs. 1A, 1B) to weakly metamorphosed rocks of the Filali series. The alternation of leuco cratic quartz-feldspar– and melanocratic garnet-biotite–dominated layers defi nes compo-sitional banding up to tens of centimeters thick in the granulites. Local shear bands as well as the elongation of quartz grains, as lineation features in the granulites, show shear sense to southeast (Fig. 1C). Locally along the profi le, the foliation is isoclinal to asymmetrically folded (Fig. 2A), with a change in dip along the section from ~30° toward the northwest in the most distal parts from the contact, to around 70° approaching the contact (see also Afi ri et al., 2011). Locally deformed igneous felsic segregations (plagio-clase + quartz + K-feldspar + turmaline crystals; Fig. 2B), and centimeter-scale ultramafi c frag-ments (Figs. 2C, 2D, 2E), occur within the gran-ulites, indicating that granite emplacement and peridotite exhumation were related. Approach-ing the peridotite contact within the granulites, garnet size gets progressively larger (from ~0.5 to 1 mm diameter), local S-C fabrics indicating top-to-the-northwest sense of shear are pre-served (Fig. 2A), the relative amount of leuco-cratic material increases, and foliation banding is progressively thicker. Locally, the foliation is also sheared or boudinaged, with melt accu-mulating in the boudin necks. The presence of a mylonitic layer (Fig. 2F) at the contact between the granulites and the peridotites indicates a general increase in strain toward the contact (Fig. 1A).

Approaching the granulite contact within the peridotites, foliation is defi ned by elongate orthopyroxene and olivine crystals. In addition, the presence of boudinage and elongate garnet within layers of pyroxenite (generally parallel to the main foliation) may document a high-T deformation episode. This foliation is nearly parallel to the one observed in the granulites (Afi ri et al., 2011). In summary, the structural observations in Beni Bousera indicate that the Moroccan contact of the Betico-Rifean belt is a high-T extensional shear zone, consistent with those in the Betic side of the Alborán Sea (e.g., Platt et al., 2003).

PETROGRAPHY AND PHASE DIAGRAM MODELING

We sampled 59 representative crustal rocks—and studied their respective thin sections and selected 14 for chemical analysis—within the granulite unit (including felsic and mafi c inclu-

sions) surrounding the peridotites, and focus on the petrologic features in the northern part of the massif where a detailed east-west sequence from the lherzolites to the crustal material of granu-lites, gneisses, and schists is exposed (Figs. 1A, 1B). The granulitic rocks are mainly metapelites interlayered with minor mafi c granulites (see also Loomis, 1972; Kornprobst, 1974), which are mineralogical and chemically similar to the samples in the Betic massifs (e.g., Kornprobst, 1974; Tubía et al., 1997; Argles et al., 1999; Haissen et al., 2004). We focus on microtextural analysis combined with thermodynamic mod-eling to constrain pressure-temperature (P-T) conditions and regional evolution. Pressure and temperature were constrained from the best fi t between observed and calculated mineral data by utilizing Perple_X (Connolly, 2005; includ-ing the thermodynamic database of Holland and Powell, 1998, with updates). Details on bulk-rock determination and composition used for modeling are in the GSA Data Repository1.

Phase diagram modeling, as P-T pseudo-sections, is advantageous over conventional thermo barometers (Fig. 3; Table 1) in constrain-ing metamorphic P-T conditions, as it allows the observed assemblage to be quantitatively constrained for a specifi ed rock composition (e.g., Powell et al., 1998; White et al., 2002) or microdomain composition (e.g., Álvarez-Valero and Kriegsman, 2007, 2010; Álvarez-Valero and Waters, 2010), and is not dependent on establishing original mineral compositions. The direct comparison of mineral modes and com-position with calculated phase relationships in the corresponding pseudosection permits the deduction of P-T conditions by the interpreta-tion of the textural evolution of the rock.

Results of Modeling

Our results indicate the preservation of two dif-ferent metamorphic mineral assemblages in the granulites: (1) A higher-P assemblage character-ized by the melting reaction biotite + kyanite + quartz → garnet + melt + rutile, which, on aver-age, constrains T to ~750 ± 50 °C and indicates a regional pressure gradient, progressively moving from lower P in the west (9 ± 1.0 kbar, 2σ), to higher P in the easternmost samples (12 ± 1.0 kbar, 2σ) along the Chmaala beach (Table 1; Figs. 1B, 3A). This pressure difference is simi-larly observed along the northeast-southwest Ichendirene section, over a distance of ~1.5 km of exposed granulites . (2) A lower-P, higher-T

1GSA Data Repository item 2014238, determin-ing bulk composition for phase diagram modeling and composition of modeled material, is available at http:// www .geosociety .org /pubs /ft2014 .htm or by request to editing@ geosociety .org.



50cmB

D

E

400μm

SpSc

A

C

NW SE NW

F

200μm

N S

SE

NW SE

NW SE

NW SE

Figure 2. (A) Ductile deforma-tion features in the granulites. Inset is a detail of S-C fabric. The sense of shear is top to the northwest. (B) Felsic vein defor-mation. Inset shows detail of one vein showing turmaline crystals. (C) Elongate orthopy-roxene and olivine crystals near the contact between the ultramafics and granulites. (D) Ultramafi c (dashed lines) and mafi c fragments included in the host granulites. (E) Elon-gate garnet crystals within pyroxenites enclaves (white triangles) in the granulites; (F) Microscopic general view of the mylonite layer at the base of the granulites.



4750

5500

6250

P(ba

r)

M Crd Grt Sil Ru

M Crd Grt Sil Ilm

M Crd Grt Sil Qtz Ru

M Crd Grt Sil Pl Ilm

M Crd Grt Sil Ru Ilm

M Crd Spl Grt Ilm

M Crd Spl Grt Sil Ilm

M Crd Spl Grt Sil Ru IlmM Crd Spl Grt Sil Ru

910870830T (°C)

Bt M Grt Ky Pl QtzRu

Bt M Grt Ky Pl Kfs Qtz Ru

Bt M Grt Ky Kfs Qtz Ru

Bt M Grt Kfs Qtz Ru

Bt M Grt Pl Kfs Qtz Ru

Bt M Grt Qtz Ru

10000

10800

11600

12400

13200

14000

T(°C)

P(ba

r)

860820780740

4000

7000

A

Grt

Bt

Ky

Ky

Qtz

Pl

Kfs

Grt

Qtz Pl

M

Ky

KyKyQtz

Bt

Spl-Crd

Spl-Crd

Spl-Crd

KyCrd-M

Pl

Gph

M Crd Grt Ilm

NCKFMASHT

NCKFMASHT B

M Crd Grt PlIlm

Bt

Crd

XFe:(Fe/Fe+Mg)

Grt

86074010

14

T(°C)P(

Kba

r)

9507504

7

T(°C)

Spl

XFe:(Fe/Fe+Mg)

Grt

Vol %melt

2

39

0.67

0.79

0.26

0.30

0.81

0.71

0.26

0.68

0.62

0.18

49

0.18

6

22

9

20

32

Vol %melt

BtGrt

12

P(K

bar)

Figure 3. Pressure-temperature (P-T ) pseudosections (sample BB-11-56) highlighting the area corresponding to the best estimate (gray ellipse) between modeled and observed phase proportions (isomodes) and compositions (mode isopleths) of: (A) the higher-P assemblage (garnet [Grt] + kyanite [Ky] + biotite [Bt] + rutile [Ru] + melt [M]); and (B) lower-P assemblage (symplectites after Grt and sillimanite (Sil) of spinel [Spl] + cordierite [Crd] + melt). Darker gray fi eld shading indicates higher variance. Modeling was done in the system NCKFMASHT (Na2O-CaO-K2O-FeO-MgO-Al2O3-SiO2-H2O-TiO2: 1.43, 2.43, 2.69, 8.14, 6.27, 15.59, 60.68, 2.52, 0.50 mol, respectively, for A; and 0.83, 1.28, 1.42, 16.16, 6.51, 24.19, 47.74, 1.54, 0.33 mol, respectively, for B), and involved the following phases: garnet, cordierite, aluminous spinel, biotite, sillimanite, kyanite, plagioclase (Pl), K-feldspar (Kfs), ilmenite (Ilm), rutile, quartz (Qtz) and silicate liquid. The model for silicate melt, as well as activity-composition models for garnet are taken from White et al. (2007); for spinel and hydrous cor-dierite from Holland and Powell (1998); and for feldspars from Fuhrman and Lindsley (1988). The activity-composition model for biotite is from Tajcmanová et al. (2009). The model biotite is saturated in Ti by incorporating enough TiO2 to allow excess ilmenite. Textural images are examples of the two studied assemblages (higher and lower P; mineral abbreviations as in the pseudosection).



symplectic mineral assemblage defi ned by the reaction garnet + kyanite/sillimanite + biotite + quartz → spinel + cordierite + melt at ~850 °C and 5 ± 0.8 (2σ) kbar (Table 1; Fig. 3B). These values are analogous to those obtained from metapelitic granulites in the Betic Cordillera (slightly lower P for the higher-P event; Table 1) by Argles et al. (1999). Thus, both crustal sec-tions (Rif and Betic) match a structural attenua-tion related to high-T decompression (up to 2–4 kbar in the Betic side) at the base of the crust. These granulites represent an example of the so-called near-isothermal decompression paths described by Harley (1989).

Glass inclusions (quenched melt) of rhyolitic composition (typical after partial melting of a metapelitic crust) are found in both the higher- and lower-P assemblages in Grt, Ky, Pl, Zr, and Crd crystals (Fig. 4A; Data Repository [see footnote 1]). Based on observed textures, the recrystallized quartz grains were formed earlier than the lower-P symplectic coronas after Grt

A

B

Figure 4. (A) Scanning electron microscope (back-scattered electrons) view of a melt in-clusion (M.I.) in a zircon (Zr) crystal within the higher-pressure metamorphic assem-blage (electron microprobe analysis [EMP] analysis [wt%]: CaO, 1.74; Al2O3, 16.44; MnO, 0.02; Na2O, 1.74; TiO2, 0.05; SiO2, 71.86; FeO, 0.40; MgO, 0.08; K2O, 6.25; total , 98.57). (B) Plane-polarized light micros-copy view of post-kinematic quartz ribbons after symplectite formation. Mineral abbre-viations as in Figure 3.

TAB

LE 1

. TH

ER

MO

BA

RO

ME

TR

IC R

ES

ULT

S IN

TH

E P

ELI

TIC

GR

AN

ULI

TE

S O

F B

EN

I BO

US

ER

A

Geo

ther

mob

arom

eter

)rabk,C°(

egalbmessa

erusserp-rewoL

)rabk,C°(

egalbmessa

erusserp-re hgiH

Con

vent

iona

l(1)

Ber

man

(19

88)(2

)H

olla

nd a

nd P

owel

l (1

998)

(3)

Con

vent

iona

l(1)

Ber

man

(19

88)(2

)H

olla

nd a

nd P

owel

l (1

998)

(3)

Sam

ple

Geo

grap

hic

area

(F

ig.1

)G

eogr

aphi

c co

ordi

nate

s(N

, W)

BB

-11-

17A

mta

r35

°12′

31.5

20′′

4°51′0

8.12

0′′

860

± 7

0,

9.0

± 0

.880

0 ±

50,

9.

5 ±

0.5

750

± 5

0,

8.5

± 0

.865

0 ±

30,

5.

5 ±

0.5

–78

0 ±

50,

4.

5 ±

0.5

BB

-11-

23Ic

hend

irene

35°1

7′09

.200′′

4°55′4

6.09

0′′

750

± 8

0,

9.2

± 0

.575

0 ±

50,

8.

0 ±

0.5

710

± 5

0,

10.0

± 0

.875

0 ±

80,

4.

8 ±

0.5

750

± 5

0,

5.8

± 0

.573

0 ±

50,

5.

0 ±

0.5

34A

Iche

ndire

ne35

°16 ′

08.2

66′′

4°55′0

4.07

7′′

690

± 3

0,

8.3

± 0

.680

0 ±

70,

8.

0 ±

0.5

–69

0 ±

30,

5.

5 ±

0.3

800

± 7

0,

5.5

± 0

.5–

27A

Iche

ndire

ne35

°15′

14.1

34′′

4°54′5

3.97

0′′

––

–65

0 ±

50,

4.

5 ±

0.6

650

± 5

0,

4.2

± 1

.0–

BB

-11-

2C

hmaa

la-b

each

35°1

9′32

.540′′

4°56′1

6.98

0′′

650

± 5

0,

7.5

± 0

.872

0 ±

70,

9.

0 ±

1.0

750

± 5

0,

9.0

± 0

.865

0 ±

50,

5.

2 ±

0.5

720

± 7

0,

4.5

± 1

.085

0 ±

50,

5.

0 ±

0.5

BB

-11-

30C

hmaa

la-b

each

35°1

9′25

.553′′

4°55′5

5.28

2′′

650

± 5

0,

7.0

± 0

.5–

––

750

± 5

0,

5.0

± 0

.5–

BB

-11-

49C

hmaa

la-b

each

35°1

9′24

.143′′

4°55′4

9.80

3′′

700

± 5

0,

7.5

± 0

.580

0 ±

50,

9.

0 ±

1.0

740

± 5

0,

10.0

± 0

.8–

–85

0 ±

50,

4.

5 ±

0.5

BB

-11-

56C

hmaa

la-b

each

35°1

9′18

.428′′

4°55′4

8.49

3′′

750

± 8

0,

10.0

± 0

.580

0 ±

50,

8.

5 ±

1.0

710

± 5

0,

12 ±

0.8

850

± 8

0,

6.0

± 0

.5–

830

± 5

0,

5.8

± 0

.5

BB

-11-

33 (

myl

onite

)C

hmaa

la-b

each

35°1

9′11

.580′′

4°55′4

5.17

5′′

750

± 5

0,

11.0

± 0

.575

0 ±

50,

20

.0 ±

1.0

–80

0 ±

50,

4.

5 ±

0.5

750

± 5

0,

5.0

± 1

.0–

Car

ratr

aca

(Ron

da m

assi

f, B

etic

sid

e; A

rgle

s et

al.,

199

9)75

0–80

0,

12–1

475

0–80

0,

4N

ote:

Tab

le p

rese

nts

a su

mm

ary

of th

e pr

essu

re-t

empe

ratu

re r

esul

ts o

btai

ned

in th

e pe

litic

gra

nulit

es o

f Ben

i Bou

sera

by

diffe

rent

app

roac

hes:

(1)

Con

vent

iona

l geo

ther

mob

arom

eter

s th

erm

omet

ers:

[g

arne

t-bi

otite

] Fer

ry a

nd S

pear

(19

78),

Hol

daw

ay (

2000

); [g

arne

t-co

rdie

rite]

Tho

mps

on (

1976

), H

olda

way

and

Lee

(19

77);

[tw

o-fe

ldsp

ar] P

erch

uk e

t al.

(198

9); b

arom

eter

s: [g

arne

t-al

umin

osili

cate

-sili

ca-

plag

iocl

ase]

Koz

iol a

nd N

ewto

n (1

988)

, Gan

guly

and

Sax

ena

(198

4); [

garn

et-c

ordi

erite

] Tho

mps

on (

1976

); [s

pine

l-cor

dier

ite] N

icho

ls e

t al.

(199

2); [

garn

et-r

utile

-ilm

enite

-pla

gioc

lase

-sili

ca] B

ohle

n an

d Li

otta

(1

986)

; [ga

rnet

-pla

gioc

lase

-bio

tite-

silic

a (ir

on)]

Hoi

sch

(199

0); (

2) T

herm

odyn

amic

dat

abas

e of

Ber

man

, 198

8 (T

WE

EQ

U);

(3)

Hol

land

and

Pow

ell,

1998

(P

erpl

e_X

; Con

nolly

, 200

5).



and Ky (Fig. 4B). In addition, minor low-grade retrograde phases like Ms and Chl locally occur within the symplectites.

Compositional garnet profi les range from slightly bell shaped (Fig. 5A) with garnet cores containing 27–29 wt% FeO, to fl at with Fe-increasing, Mg-decreasing sharp rims (Fig. 5B). The latter show smooth profi les in MgO, CaO, and MnO that may indicate relicts of growth zoning as well as chemical homogenization by diffusion at high-grade metamorphism. They normally show sharp transition to rims (up to 33 wt% FeO).

P-T Constraints in the Peridotites

In both the spinel peridotites and garnet pyroxenites (i.e., Ariegite subfacies) we utilized the conventional two-pyroxene thermometer of Putirka (2008) and the orthopyroxene and garnet barometer of Nickel and Green (1985). Results in the samples near the contact with the granu-lites indicate P of 16 ± 1 (2σ) kbar, and typically record a T range of 900–1000 °C that is compa-rable with the T values in the center of the mas-sif and is in line with results obtained by Gysi et al. (2011). Rare garnet peridotites, in which garnet is in contact with olivine, reveal P and T values of 18.7 ± 1.2 (2σ) kbar at 982 ± 70 °C.

DISCUSSION

Our petrologic results in the lower crust rocks of the Beni Bousera massif indicate that, before shearing and exhumation of the perido-tites, the granulites equilibrated at >30 km depth (prekinematic minimum conditions at > 9–12 kbar and ~800 °C). This is in concordance with the barometric estimates on the peridotites (local garnet-olivine equilibria) that yield depths of at least ~65 km (see also Gysi et al., 2011; and for the Ronda peridotites, Garrido et al., 2011).

The available fi eld and textural data, such as the presence of ultramafi c elongate fragments entrained in the granulites, indicate that ductile deformation on the preserved fabric in the gran-ulites is associated with the exhumation of the peridotite massif (e.g., Figs. 1, 2). The structural continuity between the granulite and the ultra-mafi cs, i.e., the consistent direction of the linea-tion and similar dip and strike of the foliation (Fig. 1C; Afi ri et al., 2011), permits the use of kinematic criteria in the granulites to constrain the kinematics of the crust-mantle contact. This implies that the peridotite foliation, at least the one close to the contact, developed coevally with the granulite foliation. The deformed igne-ous felsic segregations (Fig. 2B) and centimeter-scale ultramafi c fragments (Figs. 2C, 2D, 2E) in the granulites indicate that granite emplace-

ment and peridotite exhumation were related by a decompression partial melting. In addition, the generally observed top-to-the-northwest sense of shearing suggests that this contact is a kilometer-scale ductile extensional shear zone that may favor the transport and exhumation of the mantle material (Fig. 6; see also Afi ri et al., 2011). The associated crustal ductile thinning, evidenced by our thermobarometric estimates of the high-P assemblage, implies attenuation of up to ~8 km within the current 1.5 km of exposed granulitic crust. These data are in agreement with the high attenuation of the lower crust in the Betic massif (Argles et al., 1999). The esti-mates of this high-P episode are assumed to be minimum, as the best fi t between modeled and observed modes and volumes show isopleths nearly isothermally parallel (Fig. 3A).

Flat profi les of elements suggest chemical equilibration of minerals, which is commonly the result of prolonged residence at high-T con-ditions. The preservation of Grt cores with Fe, Mg, Mn, and Ca bell-shaped profi les (Fig. 5A) indicates that diffusive chemical re-equilibration during high-T conditions did not completely obliterate the original growth zonation. Instead, the cases with chemical species showing fl at core profi les (Fig. 5B) indicates a long-enough residence of Grt at high-T to homogenize.

The chemical diffusion experiments car-ried out in garnet by Chakraborty and Ganguly (1991) represent an optimal tool to constrain

the duration of particular geological processes. They show the necessary time to homogenize a chemical element (fl at profi le) in a garnet crystal as a function of temperature and its radius length:

log a (mm) = log(Dt)1/2 (cm) + 1.1505, (1)

where a is the radius of a spherical garnet crystal, and Dt is the diffusion coeffi cient at a constant P value of 5 kbar. According to their results, we quantitatively constrain in the granulites how long the ductile crustal thinning episode was. The sharp chemical changes at the garnet rims of Figure 5B, which are up to 0.1 mm, formed as a result of a residence period of <0.01 m.y. dur-ing uplifting that yielded to a fast exhumation rate (see also Chakraborty and Ganguly, 1991, their fi gure 13). This, coupled to the presence of ductile foliation in the structures, reveals that ductile thinning is in accordance to the Miocene exhumation model of Argles et al. (1999) on the Betic side of the Alborán Sea.

Geodynamic and Regional Implications— A Consistent Emplacement Model of the Ultramafi c Massifs within the General Evolution of the Alborán Domain

Our observations from the Beni Bousera massif reveal striking similarities with the ultramafi c exposures in southern Spain (e.g.,

BB-11-56

BB-11-2

BB-11-17

BB-11-23

0.80.4

0.80.410.4

0.80.4

30

5

5

5

5

distance (mm) rim-core-rim distance (mm) rim-core-rim

FeOMgOMnOCaO

30

30

30

A B

wt%

oxi

des

wt%

oxi

des

Figure 5. Example of garnet zoning profi les in the granulites. (A) Bell shape with non-homogenized core. (B) Homogenized core with sharp transition to the rim.



the Ronda massif) within the Betico-Rifean system. In all massifs, the high-T contact is observed (i.e., the granulites around perido-tites) consistently only on one side of the mas-sifs (i.e., north in Spain, south in Morocco), whereas the opposite contact, showing low-T, steeper normal faults, is generally considered

to be younger than the emplacement (Platt et al., 1995, for Ronda). Similarly, along the contacts of the Beni Bousera massif with the fl ysch-type rocks, the shear sense indicates normal movement. Figure 1A shows the elon-gated geometry (northwest-southeast trend) of the main peridotite body, of which the lower

contact (at the bottom) corresponds to the northeastern boundary.

The emplacement of these ultramafi c mas-sifs has been discussed for decades. The pres-ence of the high-T contact only on one side of the system does not favor a model of ultra-mafi c emplacement in the crust by a hot diapir

IberianForeland

Betic UM

exposures

Rif African Foreland

Alborán Sea

TellBB

200 Km

N

Grt pyroxenites

A

B

SW NE

1

2

2, 3

1 Beginning of both westward lithosphere ejection from the Africa-Iberia collision zone, and foundering of the retreating subduction boundary due to the presence of thick continental crust at N and S. Oligocene.

3

Granulites

12

2Mylonites

websterites

litho

sphe

re

asthenosphere

detail of the crustal attenuation

W Eretreating subductionzone

Granada

Fes

Almería

Melilla

Oujda

Deformed Peridotite

Continental Lithosphere

Oceanic LithosphereSebtides/Alpujarrides

30

60

90

Continuing westward escape of retreating subduction system. Early to Middle Mioceneand Middle to Late Miocene, respectively.

12

8

4800700

P (k

bar)

T (ºC)

Grt+Ky+BtKfs+Ru+M

Grt+Sil+Spl+Crd+M

30

60

90

km

Ky

Sil

10 Km

Figure 6. (A) Geographic loca-tion of the Beni Bousera (BB) and Betic ultramafi c (UM) mas-sif exposures within the Betico-Rifean belt. Dashed arrows represent the paleomagnetic reconstruction/rotation of the high-temperature contact at the Beni Bousera and Ronda massifs (anticlockwise and clockwise, respectively) due to retreating subduction. Num-bers in black squares show chronologic collision evolution (white arrow shows collision direction; gray arrows indicate the motion of the retreating subduction boundary). Dashed detachment exposes the man-tle rocks in the footwall and granulites in the hanging wall. (B) Detail of the subduction evolution, current disposition of peridotites and granulites in the Rif side, and pressure-tempera-ture–related summary of the two key episodes documented in the granulites. Cross-section location is shown in Figure 1A. Mineral abbreviations as in Figure 3.



(Loomis, 1972; Obata, 1980, with a normal fault at its top; Kornprobst and Vielzeuff, 1984, with peridotitic mantle core complexes through extensional detachments; Doblas and Oyarzun, 1989; Tubía et al., 2004, for the Betic mas-sifs), nor injection of mantle material within the lower crust along thrust faults during com-pression (Reuber et al., 1982, for the Ronda massif). Alternatively, the exhumation process may have dominantly occurred along deep-reaching normal faults (El Maz and Guiraud, 2001; for the Betic massifs: Argles et al., 1999; Platt et al., 2003; Mazzoli and Martín-Algarra, 2011). Our fi eld and petrological data support these latter models.

It is generally accepted that the Betico-Rifean peridotite bodies were linked in space and time (e.g., Azañón and Crespo-Blanc, 2000; Esteban et al., 2004; Platt et al., 2005; Vergés and Fernández, 2012). Therefore any emplacement mechanism needs to explain the consistent asymmetry of structures observed on both Alborán margins. Paleomagnetic data indicate that the ultramafi c massifs in Morocco and Iberia systematically rotated anticlockwise and clockwise, respectively, up to 74° during their emplacement (Platzman, 1992; Feinberg et al., 1996; Villasante-Marcos, 2003). Reori-enting the different massifs to their original orientation (by rotating the high-T granulitic contact/foliation back; Fig. 6) results in: (1) a north-south orientation of the long axes of the peridotite massifs subparallel to the east-dipping retreating subduction zone (Royden, 1993); (2) the high-T ductile contact, between the granulites and peridotites, then located on the western side of the peridotite massifs; and (3) the low-T eastern contact as a high-angle east-dipping normal fault.

These observations lead us to propose a model of exhumation and emplacement for the ultramafi c rocks by an initial exhumation along an east-dipping detachment system that was driven by extension behind the retreating west-dipping subduction system (Fig. 6). The kinematic features in concert with the rotation kinematic observations of Platt et al. (1995) indicate that the high-T contact was originally a shallow west-dipping detachment fault. This is consistent with our petrological results that reveal a signifi cant attenuation of the lower crust during exhumation of the ultramafi c massifs. We propose that the west-dipping detachment system formed in a back-arc environment of the east-dipping, retreating subduction zone (pres-ently the western part of the Gibraltar arc). The retreat resulted in extension of the upper plate, i.e., the Alborán domain, ultimately exhuming the mantle material. As such, much of the ini-tial exhumation of the mantle rocks resembles

the exhumation mechanism described for an oceanic-continental transition zone such as the peridotites in the Alps (Manatschal and Müntener , 2009) and also in the Iberia-New-foundland conjugated margin system (Jagoutz et al., 2007; Péron-Pinvidic et al., 2007).

We speculate that the high-angle normal fault that consistently occurs to the paleo-east of these massifs is a conjugated fault system (north-south–trending transtensional set; Balanyá and García-Dueñas, 1987; García-Dueñas et al., 1992) related to the major detachment system. This, coupled with the footwall decompression indicated by our petrologic evidence, may have led to exhumation of the ultramafi c bodies.

Subsequently, ongoing westward retreat of the subduction system (García-Dueñas et al., 1992; Royden, 1993), coupled by through-trench suction forces (Chemenda et al., 1995) onto the hanging wall, pulled the exhumed mantle rocks, the granulites, and the Internal units of the Alborán domain to the west and ulti-mately thrusted them onto the Rifean and Betic margins. The high angle between the retreating slab and the original margin boundaries resulted in anticlockwise and clockwise rotation of the Rifean and Betic ultramafi c rocks, respectively, during thrusting of the units on land. This tec-tonic scenario is consistent with our petrologic results, i.e., the signifi cant heating during the Miocene decompression/exhumation episode (Fig. 3) that is in accordance with a west-dip-ping subduction offshore in the Atlantic Ocean. Indeed, melts parental to Cr-diopside pyroxenite present in the Beni Bousera ultramafi c mas-sif show geochemical features of a subduction process similar to that of subduction-related volcanic rocks found in the Alborán Sea (Gysi et al., 2011). Additionally, geophysical stud-ies support this idea: for instance, Gutscher et al. (2002) and Spakman and Wortel (2004) recently described evidence of an active sub-duction area with thrusts and earthquakes up to 600 km depth. Our results are also compatible with the conclusions of Mazzoli and Martín-Algarra (2011), Vergés and Fernández (2012), and Mazzoli et al. (2013). They showed that Miocene emplacement of the Ronda perido-tites may be interpreted in terms of deformation partitioning (i.e., orogen-parallel strike-slip and orogen-perpendicular thrusting) associated with oblique convergence during continental subduc-tion and subsequent exhumation involving the coeval activity of kinematically linked systems of reverse, strike-slip, and and normal-sense shear zones. Top-to-the-hinterland shear along the contact between the Ronda peridotites and overlying crustal rocks is consistent with extru-sion of the subcontinental mantle rocks within the framework of continental subduction.

CONCLUSIONS

We summarise the main concluding remarks in: (1) The petrologic and structural data pre-sented reveal that the ductile deformation in the granulites is directly related to the exhumation of the Beni Bousera massif along deep reaching normal faults; (2) Results of phase diagram mod-elling in the granulites evidence crustal atten-uation of up to ~8 km; (3) The compositional zoning patterns of the granulites garnet-rims indicate a fast exhumation rate (<0.01 m.y.) during the Miocene uplifting; (4) The opposite contacts within the Betico-Rifean belt, i.e., the high-T contact between granulites and perido-tites (north in Spain, south in Morocco) and the low-T normal faults, occur similarly on one side of the ultramafi c massifs; (5) The anticlockwise and clockwise rotation of the high-T contact in the Rifean and Betic ultramafi c rocks, respec-tively, indicates that it was originally a shal-low, west-dipping detachment fault that may have formed in a back-arc environment of the east-dipping, retreating subduction zone (cur-rent western part of the Gibraltar arc); (6) Our geodynamic scenario, as mechanism of the ultramafi cs emplacement at surface, is coherent with the proposed tectonic Pleistocene evolution in western Italy during the northwest-directed subduction beneath the Calabrian Arc (e.g., Royden et al., 1987; Mattei et al., 2004; Ascione et al., 2012), where rotations, northeast-directed thrusts, as well as slab’s motion, retreat and roll-back were also identifi ed.

ACKNOWLEDGMENTS

This work was supported by an U.S. National Sci-ence Foundation grant (EAR 0910644) through O.J. The constructive and in-depth revisions of the Asso-ciate Editor Stefano Mazzoli and the anonymous re-viewers largely helped to improve the paper. We also appreciate the manuscript handling by C. Koeberl . A.M.A-V. also thanks the assistance of the Spanish research program Ramón y Cajal. This paper is a personal tribute of A.M.A.-V. to the memory of Prof. Víctor García-Dueñas, one of the most important con-tributors to the knowledge of the Betico-Rifean belt, and a special master (Life and Science).

REFERENCES CITED

Afi ri, A., Gueydan, F., Pitra, P., Essaifi , A., and Précigout, J., 2011, Oligo-Miocene exhumation of the Beni-Bousera peridotite through a lithosphere-scale extensional shear zone: Geodinamica Acta, v. 24, p. 49–60, doi: 10 .3166 /ga .24 .49 -60 .

Álvarez-Valero, A.M., and Kriegsman, L.M., 2007, Crustal thinning and mafi c underplating beneath the Neo-gene Volcanic Province (Betic Cordillera, SE Spain): Evidence from crustal xenoliths: Terra Nova, v. 19, p. 266–271, doi: 10 .1111 /j .1365 -3121 .2007 .00745 .x .

Álvarez-Valero, A.M., and Kriegsman, L.M., 2010, Chemi-cal, petrological and mass balance constraints on the textural evolution of pelitic enclaves: Lithos, v. 116, p. 300–309, doi: 10 .1016 /j .lithos .2009 .09 .009 .



Álvarez-Valero, A.M., and Waters, D.J., 2010, Partially melted xenoliths as a window into sub-volcanic processes : Evidence from the Neogene magmatic province of the Betic Cordillera, SE Spain: Journal of Petrology, v. 51, p. 973–991, doi: 10 .1093 /petrology /egq007 .

Argles, T.W., Platt, J.P., and Waters, D.J., 1999, Attenuation and excision of a crustal section during extensional exhumation: The Carratraca massif, southern Spain: Journal of the Geological Society, v. 156, p. 149–162, doi: 10 .1144 /gsjgs .156 .1 .0149 .

Ascione, A., Ciarcia, S., Di Donato, V., Mazzoli, S., and Vitale, S., 2012, The Pliocene-Quaternary wedge-top basins of southern Italy: An expression of propagat-ing lateral slab tear beneath the Apennines: Basin Research, v. 24, p. 456–474, doi: 10 .1111 /j .1365 -2117 .2011 .00534 .x .

Azañón, J.M., and Crespo-Blanc, A., 2000, Exhumation dur-ing a continental collision inferred from the tectono-metamorphic evolution of the Alpujarride Complex in the central Betics (Alboran Domain, SE Spain): Tec-tonics, v. 19, p. 549–565, doi: 10 .1029 /2000TC900005 .

Balanyá, J.C., and García-Dueñas, V., 1987, Les directions structurales dans le Domaine d’Alboran de part et d’autre du Détroit de Gibraltar: Comptes Rendus de l’Académie des Sciences, Ser. II, v. 304, p. 929–933.

Bard, J.P., 1983, Metamorphism of an obducted island arc: Example of the Kohistan Sequence (Pakistan) in the Himalayan collided range: Earth and Planetary Science Letters, v. 65, p. 133–144, doi: 10 .1016 /0012 -821X (83)90195 -4 .

Berman, R.G., 1988, Internally-consistent thermodynamic data for minerals in the system Na2O-K2O-CaO-MgO-FeO-Fe2O3-Al2O3-SiO2-TiO2-H2O-CO2: Journal of Petrology, v. 29, p. 445–522, doi: 10 .1093 /petrology /29 .2 .445 .

Blichert-Toft, J., Albarède, F., and Kornprobst, J., 1999, Lu-Hf isotope systematics of garnet pyroxenites from Beni Bousera, Morocco: Implications for basalt origin: Science, v. 283, p. 1303–1306, doi: 10 .1126 /science .283 .5406 .1303 .

Bodinier, J.L., and Godard, M., 2005, Orogenic, ophiolitic, and abyssal peridotites, in Carlson, R., ed., Treatise on Geochemistry, Volume 2: The Mantle and Core: Am-sterdam, Elsevier, p. 103–170.

Bohlen, S.R., and Liotta, J.J., 1986, A barometer for garnet amphibolites and garnet granulites: Journal of Petrol-ogy, v. 27, p. 1025–1034, doi: 10 .1093 /petrology /27 .5 .1025 .

Chakraborty, S., and Ganguly, J., 1991, Compositional zon-ing and cation diffusion in aluminosilicate garnets, in Ganguly, J., ed., Diffusion, Atomic Ordering, and Mass Transport: Selected Problems in Geochemistry (Ad-vances in Physical Geochemistry, Volume 8): Berlin, Springer, p. 120–170.

Chalouan, A., and Michard, A., 2004, The Alpine Rif belt (Morocco): A case of mountain building in a subduc-tion-subduction-transform fault triple junction: Pure and Applied Geophysics, v. 161, p. 489–519, doi: 10 .1007 /s00024 -003 -2460 -7 .

Chemenda, A.I., Mattauer, M., Malavieille, J., and Bokun, A.N., 1995, A mechanism for syn-collisional rock exhumation and associated normal faulting: Results from physical modelling: Earth and Planetary Science Letters, v. 132, p. 225–232, doi: 10 .1016 /0012 -821X (95)00042 -B .

Connolly, J.A.D., 2005, Computation of phase equilibria by linear programming: A tool for geodynamic modeling and its application to subduction zone decarbonation: Earth and Planetary Science Letters, v. 236, p. 524–541, doi: 10 .1016 /j .epsl .2005 .04 .033 .

El Maz, A., and Guiraud, F., 2001, Example of low-variance parageneses in the metapelite of Filali (inner Rif, Morocco ): Description, interpretation and geody-namic implication: Bulletin de la Société Géologique de France, v. 172, p. 469–485, doi: 10 .2113 /172 .4 .469 .

Esteban, J.J., Sánchez-Rodríguez, L., Seward, D., Cuevas, J., and Tubía, J.M., 2004, Late thermal history of the Ronda area, southern Spain: Tectonophysics, v. 389, p. 81–92, doi: 10 .1016 /j .tecto .2004 .07 .050 .

Esteban, J.J., Cuevas, J., Tubía, J.M., Liati, A., Seward, D., and Gebauer, D., 2007, Timing and origin of zircon-

bearing chlorite schists in the Ronda peridotites (Betic Cordilleras, Southern Spain): Lithos, v. 99, p. 121–135.

Esteban, J.J., Cuevas, J., Tubía, J.M., Sergeev, S., and Larionov , A., 2011, A revised Aquitanian age for the em-placement of the Ronda peridotites (Betic Cor di lleras, southern Spain): Geological Magazine, v. 148, p. 183–187, doi: 10 .1017 /S0016756810000737 .

Feinberg, H., Saddiqi, O., and Michard, A., 1996, New constraints on the bending of the Gibraltar arc from palaeomagnetism of the Ronda peridotites (Betic Cor-dilleras, Spain), in Morris, A., and Tarling, D.H., eds., Palaeomagnetism and Tectonics of the Mediterranean Region: Geological Society [London] Special Publica-tion 105, p. 43–52.

Ferry, J.M., and Spear, F.S., 1978, Experimental calibration of the partitioning of Fe and Mg between biotite and garnet: Contributions to Mineralogy and Petrology, v. 66, p. 113–117, doi: 10 .1007 /BF00372150 .

Fügenschuh, B., Seward, D., and Mancktelow, N., 1997, Ex-humation in a convergent orogen: The western Tauern Window: Terra Nova, v. 9, p. 213–217, doi: 10 .1046 /j .1365 -3121 .1997 .d01 -33 .x .

Fuhrman, M.L., and Lindsley, D.H., 1988, Ternary-feldspar modeling and thermometry: The American Mineralo-gist, v. 73, p. 201–215.

Ganguly, J., and Saxena, S.K., 1984, Mixing properties of aluminosilicate garnets: Constraints from natural and experimental data, and applications to geothermo-barometry : The American Mineralogist, v. 69, p. 88–97.

García-Dueñas, V., Balanyá, J.C., and Martínez-Martínez, J.M., 1992, Miocene extensional detachments in the outcropping basement of the northern Alboran Basin (Betics) and their tectonic implications: Geo-Marine Letters, v. 12, p. 88–95.

Garrido, C.J., Gueydan, F., Booth-Rea, G., Precigout, J., Hidas, K., Padron-Navarta, J.A., and Marchesi, C., 2011, Garnet lherzolite and garnet-spinel mylonite in the Ronda peridotite: Vestiges of Oligocene backarc mantle lithospheric extension in the western Medi-terranean: Geology, v. 39, p. 927–930, doi: 10 .1130 /G31760 .1 .

Gutscher, M.A., Malod, J., Rehault, J.P., Contrucci, I., Klinghoefer, F., Mendes-Victor, L., and Spakman, W., 2002, Evidence for active subduction beneath Gibral-tar: Geology, v. 30, p. 1071–1075, doi: 10 .1130 /0091 -7613 (2002)030 <1071: EFASBG>2 .0 .CO;2 .

Gysi, A., Jagoutz, O., Schmidt, M.W., and Targuisti, K., 2011, Petrogenesis of pyroxenites and melt infi ltrations in the ultramafi c complex of Beni Bousera, northern Morocco: Journal of Petrology, v. 52, p. 1679–1735, doi: 10 .1093 /petrology /egr026 .

Hacker, B.R., Mosenfelder, J.L., and Gnos, E., 1996, Rapid emplacement of the Oman ophiolite: Thermal and geochronologic constraints: Tectonics, v. 15, p. 1230–1247, doi: 10 .1029 /96TC01973 .

Haissen, F., García-Casco, A., Torres-Roldan, R., and Aghzer, A., 2004, Decompression reactions and P-T conditions in high-pressure granulites from Casares–Los Reales units of the Betic-Rif belt (S Spain and N Morocco): Journal of African Earth Sciences, v. 39, p. 375–383, doi: 10 .1016 /j .jafrearsci .2004 .07 .030 .

Harley, S.L., 1989, The origin of granutites: A metamorphic perspective: Geological Magazine, v. 126, p. 215–247.

Hoisch, T.D., 1990, Empirical calibration of six geobarom-eters for the mineral assemblage quartz + muscovite + biotite + plagioclase + garnet: Contributions to Miner-alogy and Petrology, v. 104, p. 225–234, doi: 10 .1007 /BF00306445 .

Holdaway, M.J., 2000, Application of new experimental and garnet Margules data to the garnet-biotite geother-mometer: The American Mineralogist, v. 85, p. 881–892.

Holdaway, M.J., and Lee, S.M., 1977, Fe-Mg cordierite sta-bility in high-grade pelitic rocks based on experimen-tal, theoretical and natural observations: Contributions to Mineralogy and Petrology, v. 63, p. 175–198, doi: 10 .1007 /BF00398778 .

Holland, T.J.B., and Powell, R., 1998, An internally consis-tent thermodynamic dataset for phases of petrologi-cal interest: Journal of Metamorphic Geology, v. 16, p. 309–343, doi: 10 .1111 /j .1525 -1314 .1998 .00140 .x .

Jagoutz, O., Müntener, O., Burg, J.P., Ulmer, P., and Jagoutz, E., 2006, Lower continental crust formation through fo-

cused fl ow in km-scale melt conduits: The zoned ultra-mafi c bodies of the Chilas Complex in the Kohistan island arc (NW Pakistan): Earth and Planetary Science Letters, v. 242, p. 320–342, doi: 10 .1016 /j .epsl .2005 .12 .005 .

Jagoutz, O., Müntener, O., Manatschal, G., Péron-Pinvidic, G., Rubatto, D., Villa, I.M., and Turrin, B.D., 2007, The rift-to-drift transition in the North Atlantic: A stut-tering start of the MORB machine?: Geology, v. 35, p. 1087–1090, doi: 10 .1130 /G23613A .1 .

Kornprobst, J., 1974, Contribution à l’étude pétrographique et structurale de la Zone interne du Rif (Maroc septen-trional): Notes et Mémoires Service géologique Maroc, v. 251, p. 1–256.

Kornprobst, J., and Vielzeuf, D., 1984. Transcurrent crustal thinning: A mechanism for the uplift of deep continen-tal crust–upper mantle associations, in Kronprobst, J., ed., Kimberlites II: The Mantle and Crust-Mantle Re-lationships (Developments in Petrology, Volume 11B): Amsterdam, Elsevier, p. 347–359.

Koziol, A.M., and Newton, R.C., 1988, Redetermination of the anorthite breakdown reaction and improvement of the plagioclase-garnet-Al2SiO5-quartz geobarometer: The American Mineralogist, v. 73, p. 216–223.

Labrousse, L., Prouteau, G., and Ganzhorn, A.C., 2011, Continental exhumation triggered by partial melting at ultrahigh pressure: Geology, v. 39, p. 1171–1174, doi: 10 .1130 /G32316 .1 .

Loomis, T.H., 1972, Diapiric emplacement of the Ronda high-temperature ultramafi c intrusion, southern Spain: Geological Society of America Bulletin, v. 83, p. 2475–2496, doi: 10 .1130 /0016 -7606 (1972)83 [2475: DEO-TRH]2 .0 .CO;2 .

Lundeen, M.T., 1978, Emplacement of the Ronda peridotite, Sierra Bermeja, Spain: Geological Society of America Bulletin, v. 89, p. 172–180, doi: 10 .1130 /0016 -7606 (1978)89 <172: EOTRPS>2 .0 .CO;2 .

Manatschal, G., and Müntener, O., 2009, A type sequence across an ancient magma poor ocean–continent transi-tion: The example of the western Alpine Tethys ophio-lites: Tectonophysics, v. 473, p. 4–19, doi: 10 .1016 /j .tecto .2008 .07 .021 .

Mattei, M., Pretocelli, V., Lacava, D., and Schiattarella, M., 2004, Geodynamic implication of Pleistocene ultrarapid vertical-axis rotation in the Southern Apen-nines, Italy: Geology, v. 32, p. 789–792, doi: 10 .1130 /G20552 .1 .

Mazzoli, S., and Martín-Algarra, A., 2011, Deformation partitioning during transpressional emplacement of a ‘mantle extrusion wedge’: The Ronda peridotites, western Betic Cordillera, Spain: Journal of the Geo-logical Society, v. 168, p. 373–382, doi: 10 .1144 /0016 -76492010 -126 .

Mazzoli, S., Martín-Algarra, A., Reddy, S.M., López Sán-chez-Vizcaíno, V., Fedele, L., and Noviello, A., 2013, The evolution of the footwall to the Ronda subconti-nental mantle peridotites: Insights from the Nieves Unit (western Betic Cordillera): Journal of the Geological Society, v. 170, p. 385–402, doi: 10 .1144 /jgs2012 -105 .

Montel, J.M., Kornprobst, J., and Vielzeuf, D., 2000, Pres-ervation of old U-Th-Pb ages in shielded monazite: Example from the Beni Bousera Hercynian kinzigites (Morocco): Journal of Metamorphic Geology, v. 18, p. 335–342, doi: 10 .1046 /j .1525 -1314 .2000 .00261 .x .

Müntener, O., Hermann, J., and Trommsdorff, V., 2000, Cooling history and exhumation of lower crustal granulite and upper mantle: Journal of Petrology, v. 41, p. 175–200, doi: 10 .1093 /petrology /41 .2 .175 .

Nichols, G.T., Berry, R.F., and Green, D.H., 1992, Internally consistent gahnitic spinel-cordierite-garnet equilibria in the FMASH Zn system: Geothermobarometry and applications: Contributions to Mineralogy and Petrol-ogy, v. 111, p. 362–377, doi: 10 .1007 /BF00311197 .

Nickel, K.G., and Green, D.H., 1985, Empirical geother-mobarometry for garnet peridotites and implications for the nature of the lithosphere, kimberlites and dia-monds: Earth and Planetary Science Letters, v. 73, p. 158–170, doi: 10 .1016 /0012 -821X (85)90043 -3 .

Obata, M., 1980, The Ronda peridotite—Garnetlherzolite, spinel-lherzolite, and plagioclase lherzolite facies and the P-T trajectories of a high-temperature mantle intru-sion: Journal of Petrology, v. 21, p. 533–572.



Perchuk, L.L., Gerya, T.V., and Nozhkin, A.D., 1989, Petrol-ogy and retrogression in granulites of the Kanskiy Formation, Yenisey Range, Eastern Siberia: Journal of Metamorphic Geology, v. 7, p. 599–617, doi: 10 .1111 /j .1525 -1314 .1989 .tb00621 .x .

Péron-Pinvidic, G., Manatschal, G., Minshull, T.A., and Sawyer, D., 2007, The tectono-sedimentary and morpho-tectonic evolution recorded in the deep Iberia-Newfoundland margins: Evidence for a complex breakup history: Tectonics, v. 26, TC2011, doi: 10 .1029 /2006TC001970 .

Platt, J.P., Allerton, S., Kirker, A., and Platzman, E., 1995, Origin of the western Subbetic arc (South Spain): Palaeo magnetic and structural evidence: Journal of Structural Geology, v. 17, p. 765–775, doi: 10 .1016 /0191 -8141 (94)00110 -L .

Platt, J.P., Argles, T.W., Carter, A., Kelley, S.P., Whitehouse, M.J., and Lonergan, L., 2003, Exhumation of the Ronda peridotite and its crustal envelope: Constraints from thermal modeling a P-T-time array: Journal of the Geological Society, v. 160, p. 655–677.

Platt, J.P., Kelley, S.P., Carter, A., and Orozco, M., 2005, Timing of tectonic events in the Alpujárride Complex, Betic Cordillera, S Spain: Journal of the Geological Society, v. 162, p. 451–462.

Platzman, E.S., 1992, Paleomagnetic rotations and kine-matics of the Gibraltar Arc: Geology, v. 20, p. 311–314, doi: 10 .1130 /0091 -7613 (1992)020 <0311: PRATKO>2 .3 .CO;2 .

Polvé, M., 1983, Les isotopes du Nd et du Sr dans les lherzo-lites orogéniques: Contribution à la détermination de la structure et de la dynamique du manteau supérieur [Ph.D. thesis]: Paris, Université Paris Diderot–Paris 7, 361 p.

Powell, R., Holland, T.J.B., and Worley, B., 1998, Calculating phase diagrams involving solid solutions via non-linear equations, with examples using THERMOCALC : Journal of Metamorphic Geology, v. 16, p. 577–588, doi: 10 .1111 /j .1525 -1314 .1998 .00157 .x .

Puga, E., Nieto, J.M., Díaz de Federico, A., Bodinier, J.L., and Morten, L., 1999, Petrology and metamorphic evolution of ultramafi c rocks and dolerite dykes of the Betic Ophiolitic Association (Mulhacén Complex, SE Spain): Evidence of eo-Alpine subduction following an

ocean-fl oor metasomatic process: Lithos, v. 49, p. 23–56, doi: 10 .1016 /S0024 -4937 (99)00035 -3 .

Putirka, K., 2008, Thermometers and barometers for vol-canic systems, in Putirka, K.D., and Tepley, F.J., III, eds., Minerals, Inclusions and Volcanic Processes: Mineralogical Society of America Reviews in Mineral-ogy and Geochemistry 69, p. 61–120.

Reisberg, L., Zindler, A., and Jagoutz, E., 1989, Further Sr and Nd isotopic results from peridotites of the Ronda Ultramafi c Complex: Earth and Planetary Science Letters, v. 96, p. 161–180, doi: 10 .1016 /0012 -821X (89)90130 -1 .

Reuber, I., Michard, A., Chalovan, A., Juteau, T., and Jermoumi , B., 1982, Structure and emplacement of the Alpine type peridotites from Beni Bousera, Rif, Morocco : A polyphase tectonic interpretation: Tecto-nophysics, v. 82, p. 231–251, doi: 10 .1016 /0040 -1951 (82)90047 -6 .

Royden, L.H., 1993, Evolution of retreating subduction boundaries formed during continental collision: Tec-tonics, v. 12, p. 629–638, doi: 10 .1029 /92TC02641 .

Royden, L.H., Patacca, E., and Scandone, P., 1987, Segmen-tation and confi guration of subducted lithosphere in Italy: An important control on thrust-belt and foredeep-basin evolution: Geology, v. 15, p. 714–717, doi: 10 .1130 /0091 -7613 (1987)15 <714: SACOSL>2 .0 .CO;2 .

Sánchez-Rodríguez, L., and Gebauer, D., 2000, Mesozoic formation of pyroxenites and gabbros in the Ronda area (southern Spain), followed by early Miocene subduction metamorphism and emplacement into the middle crust: U-Pb sensitive high-resolution ion micro-probe dating of zircon: Tectonophysics, v. 316, p. 19–44, doi: 10 .1016 /S0040 -1951 (99)00256 -5 .

Spakman, W., and Wortel, M.J.R., 2004, A tomographic view on western Mediterranean geodynamics, in Cavazza W., Roure, F., Spakman, W., Stampfl i, G.M., and Ziegler, P., eds., The TRANSMED Atlas: The Mediterranean Region from Crust to Mantle: Berlin, Springer Verlag, p. 31–52.

Tajcmanová, L., Connolly, J.A.D., and Cesare, B., 2009, A thermodynamic model for titanium and ferric iron solution in biotite: Journal of Metamorphic Geology, v. 27, p. 153–165, doi: 10 .1111 /j .1525 -1314 .2009 .00812 .x .

Thompson, A.B., 1976, Mineral reactions in pelitic rocks: I. Prediction of P-T-X (Fe-Mg) phase relations: Ameri-can Journal of Science, v. 276, p. 401–424, doi: 10 .2475 /ajs .276 .4 .401 .

Tubía, J.M., and Cuevas, J., 1986, High-temperature em-placement of the Los Reales peridotite nappe (Betic Cordillera, Spain): Journal of Structural Geology, v. 8, p. 473–482, doi: 10 .1016 /0191 -8141 (86)90064 -7 .

Tubía, J.M., Cuevas, J., and Gil Ibarguchi, J.I., 1997, Sequen-tial development of the metamorphic aureole beneath the Ronda peridotites and its bearing on the tectonic evolution of the Betic Cordillera: Tectonophysics, v. 279, p. 227–252, doi: 10 .1016 /S0040 -1951 (97)00124 -8 .

Tubía, J.M., Cuevas, J., and Esteban, J.J., 2004, Tectonic evidence in the Ronda peridotites, Spain, for mantle diapirism related to delamination: Geology, v. 32, p. 941–944, doi: 10 .1130 /G20869 .1 .

Vergés, J., and Fernández, M., 2012, Tethys–Atlantic inter-action along the Iberia–Africa plate boundary: The Betic–Rif orogenic system: Tectonophysics, v. 579, p. 144–172, doi: 10 .1016 /j .tecto .2012 .08 .032 .

Villasante-Marcos, V., 2003, Palaeomagnetic study of the Ronda peridotites (Betic Cordillera, southern Spain): Tectonophysics, v. 377, p. 119–141, doi: 10 .1016 /j .tecto .2003 .08 .023 .

White, R.W., Powell, R., and Clarke, G.L., 2002, The in-terpretation of reaction textures in Ferich metapelitic granulites of the Musgrave Block, central Australia: Constraints from mineral equilibria calculations in the system K2O–FeO–MgO–Al2O3–SiO2–H2O–TiO2–Fe2O3: Journal of Metamorphic Geology, v. 20, p. 41–55, doi: 10 .1046 /j .0263 -4929 .2001 .00349 .x .

White, R.W., Powell, R., and Holland, T.J.B., 2007, Progress relating to calculation of partial melting equilibria for metapelites: Journal of Metamorphic Geology, v. 25, p. 511–527, doi: 10 .1111 /j .1525 -1314 .2007 .00711 .x .

SCIENCE EDITOR: CHRISTIAN KOEBERL

ASSOCIATE EDITOR: STEFANO MAZZOLI

MANUSCRIPT RECEIVED 28 NOVEMBER 2013REVISED MANUSCRIPT RECEIVED 23 FEBRUARY 2014MANUSCRIPT ACCEPTED 20 MAY 2014

Printed in the USA

crustal attenuation as a tracer for the emplacement of the...

Documents