New evidence for an old idea: Geochronological

Lithos 302–303 (2018) 278–297

Contents lists available at ScienceDirect

Lithos journal homepage: www.elsevier.com/locate/lithos

New evidence for an old idea: Geochronological constraints for a paired metamorphic belt in the central European Variscides T.M. Will a,⁎, E. Schmädicke b, X.-X. Ling c, X.-H. Li c, Q.-L. Li c a b c

Institut für Geographie und Geologie der Universität Würzburg, Am Hubland, 97074 Würzburg, Germany GeoZentrum Nordbayern, Universität Erlangen-Nürnberg, Schlossgarten 5a, 91054 Erlangen, Germany State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China

a r t i c l e

i n f o

Article history: Received 1 August 2017 Accepted 8 January 2018 Keywords: Odenwald-Spessart basement Variscan orogeny Paired metamorphic belt U-Pb geochronology Lu-Hf isotope data

a b s t r a c t New geochronological data reveal a prolonged tectonothermal evolution of the Variscan Odenwald-Spessart basement, being part of the Mid-German Crystalline Zone in central Europe. We report the results from (i) secondary ion mass spectrometry (SIMS) U-Pb dating of zircon, rutile and monazite, (ii) SIMS zircon oxygen isotope analyses, (iii) laser ablation-multicollector-inductively coupled plasma mass spectrometry (LA-MCICPMS) zircon Lu-Hf isotope analyses and, (iv) LA-ICPMS zircon and rutile trace element data for a suite of metamorphic rocks (five amphibolite- and eclogite-facies mafic meta-igneous rocks and one granulite-facies paragneiss). The protoliths of the mafic rocks formed from juvenile as well as depleted mantle sources in distinct tectonic environments at different times. Magmatism took place at a divergent oceanic margin (possibly in a back-arc setting) at 460 Ma, in an intraoceanic basin at ca. 445 Ma and at a continental margin at 329 Ma. Regardless of lithology, zircon in eclogite, amphibolite and high-temperature paragneiss provide almost identical Carboniferous ages of 333.7 ± 4.1 Ma (eclogite), 329.1 ± 1.8 to 328.4 ± 8.9 Ma (amphibolite), and 334.0 ± 2.0 Ma (paragneiss), respectively. Rutile yielded ages of 328.6 ± 4.7 and 321.4 ± 7.0 Ma in eclogite and amphibolite, and monazite in high-temperature paragneiss grew at 330.1 ± 2.4 Ma (all ages are quoted at the 2σ level). The data constrain coeval high-pressure eclogite- and high-temperature granulite-facies metamorphism of the Odenwald-Spessart basement at ca. 330 Ma. Amphibolite-facies conditions were attained shortly afterwards. The lower plate eclogite formed in a fossil subduction zone and the upper plate high-temperature, lowpressure rocks are the remains of an eroded Carboniferous magmatic arc. The close proximity of tectonically juxtaposed units of such radically different metamorphic conditions and thermal gradients is characteristic for a paired metamorphic belt sensu Miyashiro (1961). Thus, the Odenwald-Spessart basement represents the first recognised paired metamorphic belt in the European Variscides. © 2018 Elsevier B.V. All rights reserved.

1. Introduction and general tectonic setting of the Mid-German Crystalline Zone For most of the Palaeozoic Gondwana was separated from the northern continents (Laurentia, Baltica, Siberia) by the intervening Iapetus and Rheic oceans. Towards the end of the Ordovician the Iapetus Ocean closed and Laurentia, Baltica and the microcontinent Avalonia, which had separated from the northern margin of West Gondwana in the late Neoproterozoic/early Palaeozoic (e.g., Nance and Linnemann, 2008; Stampfli et al., 2013) amalgamated to form Laurussia. Diachronous collision of Laurussia with Gondwana in the late Palaeozoic led to the closure of the Rheic Ocean and the assembly of the supercontinent Pangaea (e.g., Stampfli et al., 2013). The formation of the Appalachian-Variscan orogen in North America and Europe is related to that late Palaeozoic ⁎ Corresponding author. E-mail address: [email protected] (T.M. Will).

https://doi.org/10.1016/j.lithos.2018.01.008 0024-4937/© 2018 Elsevier B.V. All rights reserved.

collision. The associated suture is referred to as the ‘Rheic suture’ and is thought to run from Mexico to Europe and beyond (e. g., Nance and Linnemann, 2008). In the European section of the orogen the suture extends from the Iberian Peninsula to NW France and Germany and further into SW Poland (e.g., Nance and Linnemann, 2008) and separates basement rocks with West African Craton or NE Gondwana affinities in the south from Baltica-derived terranes in the north (e.g., Henderson et al., 2016; Zeh and Gerdes, 2010). Some authors (e.g., Franke et al., 2017) proposed that the German segment of the Rheic suture is located at the northern outcrop limit of the Northern Phyllite Zone, whereas others (e.g., Linnemann et al., 2004; Oncken, 1997, 2000; Will et al., 2015, 2017; Zeh and Gerdes, 2010) argued that the suture is situated within the Mid-German Crystalline Zone (Fig. 1), the largest part of which is exposed in the Odenwald-Spessart basement (Fig. 2). Regardless of the exact position of the suture there seems to be a general consensus that the Mid-German Crystalline Zone is composed of rocks with different palaeogeographic affinities and thermal histories. Here, we focus on the

T.M. Will et al. / Lithos 302–303 (2018) 278–297

Laurentia

IS

Baltica

Avalonia

TS

VDF

279

100 km

10° E Harz Mts.

RS Leon

RH O

SAX

R

S

MOLD

Kyffhäuser

ADF North Gondwana

Leipzig

51° N

RH

Ruhla

ADF

Gießen

Rhenish Massif

Ardennes

Rhön (xenoliths)

WTF

MGCZ Frankfurt

MF

50° N

Spessart 50° N

Odenwald

SAX

OSZ

Nürnberg

ben

Fig. 2 Gra

Post-Carboniferous Cover Rhenohercynian Zone

hine

Palatinate

Gießen-Harz Nappe

Strasbourg

Upp

er R

49° N

Stuttgart

MOLD Black Forest

Vosges 7° E

8° E

Northern Phyllite Zone Mid-German Crystalline Zone Saxothuringian Zone sensu str. Münchberg Nappe

9° E

Moldanubian Zone

Fig. 1. Location of the Mid-German Crystalline Zone (MGCZ) in Central Europe (modified after Will et al., 2015). RH-Rhenohercynian Zone, SAX-Saxothuringian Zone, MOLD-Moldanubian Zone; only the exposed areas are shown on the map. Additional abbreviations in inset: IS-Iapetus Suture, TS-Tornquist Suture, RS-Rheic Suture; VDF-Variscan deformation front, ADF-Alpine deformation front; MF-Michelbach Fault, OSZ-Otzberg Shear Zone, WTF-West Thuringian Fault; O-Odenwald, S-Spessart, R-Rhön.

tectonothermal evolution of the Odenwald-Spessart basement, whose metamorphic conditions as well as the tectonic settings in which many basement rocks formed are reasonably well known (Will et al., 2015; Will and Schmädicke, 2001, 2003; Zeh and Will, 2010). However, except for a recent study by Will et al. (2017), who determined the age of monazite grains in low- to high-grade schist and gneiss by in-situ U-Th-Pb electron microprobe (EMP) dating, direct geochronological constraints on the age of metamorphism are sparse (see below). Thus, with the aim of better constraining the age and tectonothermal evolution of the Odenwald-Spessart basement and its position within the central European Variscides we present geochronological data of zircon, rutile and monazite from one metasedimentary and five mafic metaigneous samples determined by secondary ionization mass spectrometry (SIMS). In addition, we present zircon Lu-Hf and oxygen isotope data that were collected by SIMS and MC-ICP-MS (multi-collector laser ablation inductively coupled plasma mass spectrometry) techniques and zircon trace element data obtained by LA-ICP-MS. Zircon, rutile and monazite are particularly useful minerals for providing geochronological information on magmatic and metamorphic events because of their high closure temperatures (e.g., Cherniak and Watson, 2000; Heaman and Parrish, 1991; Vry and Baker, 2006) and/or because radiogenic Pb is believed to be retained in the minerals even under granulite-facies metamorphic

conditions (e.g., Cherniak et al., 2004; Copeland et al., 1988; Taylor et al., 2016). Zircon in mafic rocks could be of igneous or metamorphic origin (e.g., Grimes et al., 2007, 2015; Bolhar et al., 2016). Thus, based on textural and geochemical criteria we attempt to clarify whether our zircon data specify an igneous or a metamorphic event. In contrast, rutile and monazite in our samples grew during metamorphism and, hence their (re-)crystallisation ages should closely constrain the time of metamorphism. Rutile, as a typical high-pressure mineral (e.g., KylanderClark et al., 2008; Li et al., 2011) may allow determining the age of high-pressure metamorphism and, monazite is a useful mineral to provide information on the timing of high-temperature metamorphism. 2. Regional geology The geological setting of the Odenwald and Spessart basement areas has been described in detail by several authors (e.g., Nickel, 1975; Krohe, 1992, 1996; Weber, 1995; Stein, 2001; Zeh and Will, 2010; Okrusch et al., 2000, 2011; Will et al., 2017; and references therein) and is only outlined briefly. However, the mafic meta-igneous rocks, which are the main subject of the present investigation, are described in somewhat more detail. The results of previous geochronological studies are summarised in Fig. 3.

280


08° 35’ E

09° 00’ E

N

09° 25’ E

MF

A

Geiselbach

Spessart

SP-68 Alzenau

Mömbris

10 km SP-57 50° 00‘ N

Aschaffenburg

Spessart-Odenwald tectonic window

Elterhof

SP-14 A’ MF

Darmstadt A

GroßUmstadt

A’

Unit I B

Syncline axis

EH-10 EH-19

Anticline axis Foliation trend (in cross-sections only) Monazite age (Will et al., 2017) Ar-Ar white mica age (Will et al., 2017) SIMS U-Pb age dating samples (this study)

Unit IV B’

Gadernheim

Unit II

GH-11

Post-Carboniferous cover

Spessart basement OSZ

Quartzdiorite-Granodiorite Amphibolite

Unit III

Orthogneiss unit Garnet-mica schist and pyllonite (Geiselbach Formation) Garnet-muscovite-biotite schist (Schweinheim Formation) Garnet-staurolite schist (Mömbris Formation) Paragneiss, amphibolite and calcsilicate rock (Alzenau/Elterhof Formation)

49° 30‘ N

Odenwald

Odenwald basement Heidelberg

Granite OSZ

Unit I

B

Unit II

Unit III Unit IV

B’

Flasergranitoid (granite to diorite) Granodiorite

Orthogneiss Amphibolite Eclogite lenses

Diorite

Garnet-biotite schist

Gabbro

HT-Paragneiss

Fig. 2. Geological map of the Spessart-Odenwald basement (modified after Will et al., 2017). OSZ-Otzberg Shear Zone, MF-Michelbach Fault. The Spessart and Odenwald cross-sections are after Weber (1995) and Krohe (1992), respectively. The grey dashed line denotes the approximate extent of the Spessart-Odenwald tectonic window sensu Will et al. (2015).

2.1. Spessart basement The Spessart basement (Fig. 2) is interpreted as a NE-trending asymmetric antiform (Weber, 1995) with its central part consisting of garnet-staurolite schist with rare kyanite and/or sillimanite (Mömbris Formation), micaschist and phyllonite that locally contain pseudomorphs after garnet (Geiselbach Formation), and a well-foliated calcalkaline granitic to granodioritic orthogneiss unit with an upper Silurian to lower Devonian intrusion age of 418–410 ± 18 Ma (single zircon evaporation age; Dombrowski et al., 1995). Garnet- and rarely also garnet-staurolite-bearing biotite-muscovite schist (Schweinheim Formation) is exposed south of the orthogneiss unit. The Alzenau and the Elterhof Formations occur in the northern- and southernmost part of the Spessart basement, respectively, and are composed of variably

deformed granitic gneiss, quartzite, marble, calc-silicate rock, garnetbearing biotite gneiss and minor amphibolite. The Elterhof Formation was intruded by weakly deformed quartzdiorite and granodiorite at ca. 330 Ma (single zircon lead evaporation ages; Anthes and Reischmann, 2001; Siebel et al., 2012). The poorly exposed Michelbach Fault (Behr and Heinrichs, 1987) separates the Alzenau Formation in the north from the central and southern Spessart basement (Fig. 2). The previously held opinion that Alzenau and Elterhof formations once formed a coherent lithological unit (Behr and Heinrichs, 1987; Okrusch et al., 2011) must be abandoned in the light of geochemical data provided by Will et al. (2015). Even though desirable, direct structural evidence for this interpretation cannot be obtained because of the very poor outcrop exposures that prohibit the determination of the kinetics and the geometry of the faults involved. In-situ monazite age dating (Will et al., 2017)


constrains the age of amphibolite-facies metamorphism in the Spessart basement to between 317 and 324 Ma, which is in agreement with KAr, Ar-Ar and Rb-Sr age data that provide evidence that the basement cooled from temperatures in excess of c. 500 °C to below 300 °C between 328 Ma and 311 Ma (e.g., Dombrowski et al., 1995; Lippolt, 1986; Nasir et al., 1991).

locations around the eastern Odenwald orthogneiss core. Dörr et al. (2017) suggested an upper Devonian depositional age of the paragneiss protolith and, Will et al. (2017) determined an in-situ electron microprobe age of 316 Ma for monazite growth. Garnet-amphibolite lenses that are aligned subparallel to the Otzberg Shear Zone occur within the orthogneiss core. In places, this rock type has been recognised as retrogressed eclogite that followed a clockwise P-T path and formed during high-pressure, subduction-related metamorphism that reached peak temperatures of ca. 700 ± 50 °C at minimum pressures of 16–17 kbar (Will and Schmädicke, 2001). In contrast, the granulite-facies paragneiss in the western Odenwald followed a hairpin-like, counterclockwise P-T path at maximum pressures of 4 kbar and temperatures of ca. 680 °C (Will and Schmädicke, 2003). These authors suggested that the high-pressure eclogite and the adjacent, low-pressure/hightemperature granulite that are separated by the Otzberg Shear Zone bear resemblance to a paired metamorphic belt. Scherer et al. (2002) determined Lu-Hf garnet-whole rock ages of 357 ± 7 and 353 ± 11 Ma for two retrogressed eclogite samples, which provide a single isochron age of 357 ± 6 Ma (Fig. 3). Within error this age overlaps with an in-situ EMP monazite age of 349 ± 14 Ma for a granulitic lowpressure paragneiss in the western Odenwald (Will et al., 2017).

2.2. Odenwald basement It has long been known (e.g., Nickel, 1975) that the Odenwald consists of two distinct basement units (Bergsträsser and Böllsteiner or western and eastern Odenwald, respectively) that are separated by the NNESSW trending Otzberg Shear Zone (Fig. 2), a steeply westerly dipping ductile to semi-brittle fault and shear zone with a transtensional sinistral strike-slip sense-of-shear (Krohe, 1992, 1996; Stein, 2001; Will, 2001). The western Odenwald basement contains large amounts of granitic to gabbroic calc-alkaline magmatic rocks that were emplaced at 362 Ma in the north (single zircon lead evaporation and average Ar-Ar hornblende and plagioclase ages; Kirsch et al., 1988) and some 30 million years later further south (Siebel et al., 2012). Minor amphibolite-facies metabasic rocks and high-temperature, low-pressure granulite and migmatitic gneiss are also present (e.g., von Raumer, 1973). Zircon fusion ages of 342 to 332 Ma (Todt et al., 1995) overlap with hornblende 40 Ar/39Ar cooling ages (Schubert et al., 2001) and were interpreted by these authors to date the time of metamorphism and partial melting (Fig. 3). The core of the eastern Odenwald is dominated by S-type granitic to granodioritic orthogneisses, which are interpreted to form a NNE-SSW trending antiform (Chatterjee, 1960) and whose protoliths intruded at ca. 410–405 Ma (upper discordia intercept and single zircon lead evaporation ages, respectively; Reischmann et al., 2001). Slivers of variably deformed mafic and ultramafic rocks are present in the eastern Odenwald (Knauer et al., 1974; Nickel, 1975), and a garnet-bearing biotite schist unit, also referred to as schist envelope, crops out at several

2.3. Mafic meta-igneous rocks of the Odenwald-Spessart basement Geochemical and whole rock isotope data presented by Will et al. (2015) demonstrated that the protoliths of amphibolites exposed in the Odenwald-Spessart basement formed in different tectonic settings such as ocean ridge, within-plate and continental margin environments. Amphibolites from the northernmost Spessart and the western Odenwald are tholeiitic and calc-alkaline in composition. Their protoliths were extracted from the shallow, depleted MORB mantle and were interpreted to have formed at a divergent plate margin (Will et al., 2015). Whole rock Nd isotope data indicate that the magmas were juvenile additions to the crust at the end of the Neoproterozoic. The protoliths of the

(10)

Spessart basement

(10) (3)

(1, 2)

(3)

(4)

Eastern Odenwald basement (unit IV)

(9)

(10)

(6)

Western Odenwald basement (unit I)

(7) (8)

Pb-Pb zircon age (Felsic intrusion age) Pb-Pb zircon age (Assumed age of thermal peak) Ar-Ar hornblende crystallisation age in amphibolite (Cooling age)

Western Odenwald basement (unit II)

(10)

Western Odenwald basement (unit III)

Ar-Ar hornblende crystallisation age in gabbro (Cooling age) K-Ar hornblende crystallisation age in amphibolite (Cooling age)

(8) (8) (8) (1)

250

(5) U-Th-Pb monazite age (Age of metamorphism) Ar-Ar white mica crystallization age (Cooling age) Lu-Hf garnet-whole rock age (Minimum age of eclogite-facies metamorphism)

(10)

(10)

(8) (8)

300 Permian

252

281

(9)

350

Carboniferous

299

400 Devonian

359

450 Silurian

418

444

Ordovician

500 Cambrian

485

Fig. 3. Time-space diagram for various units of the Odenwald-Spessart basement (modified after Will et al., 2017). Data sources: 1: Siebel et al. (2012), 2: Anthes and Reischmann (2001), 3: Dombrowski et al. (1995), 4: Nasir et al. (1991) and Lippolt (1986), 5: Reischmann et al. (2001), 6: Scherer et al. (2002), 7: Kirsch et al. (1988), 8: Schubert et al. (2001), 9: Todt et al. (1995), 10: Will et al. (2017). The interpretation of the geochronological data is given in parentheses.

282


metabasic rocks from the southern and central Spessart basement have variable compositions that can either be attributed to formation in an oceanic intraplate or at a continental margin setting. Amphibolite with volcanic arc characteristics formed by recycling of older crustal material or mixing of juvenile mantle-derived material with pre-existing crust in a supra-subduction setting. Rocks with intraplate geochemical signatures originated as ocean island basalts that were derived from an enriched mantle source. These rocks were variably modified by later assimilation of continental crust. The retrogressed eclogites from the eastern Odenwald are tholeiitic metabasalts whose precursor melt was derived from the shallow, depleted MORB mantle and emplaced in a mid-ocean ridge setting.

2.4. Suspected terrane boundary in the Odenwald-Spessart basement The geochemical data summarised above corroborate earlier propositions (e.g., Krohe, 1992; Nickel, 1975) that the Odenwald-Spessart basement is composed of two distinct basement units, which are separated by the Otzberg-Michelbach Fault Zone (Fig. 2). Will et al. (2015) argued that this fault zone is part of the Rheic suture and may separate rocks with Gondwanan (Saxothuringian) and Baltican (Rhenohercynian) affinities from each other. The former constitute the northernmost Spessart and Odenwald basement west of the Otzberg Shear Zone and the latter occur in a lower plate tectonic window that is exposed in the eastern Odenwald and the Spessart basement south of the Michelbach Fault Zone. Based on the lack of Baltica-derived Meso- and Palaeoproterozoic detrital zircon in the eastern Odenwald garnet-biotite schist unit, Dörr et al. (2017) questioned this hypothesis and the interpretation that the eastern Odenwald is part of a lower plate tectonic window in the MidGerman Crystalline Zone. Further provenance studies should help to clarify this matter. Nonetheless, the interpretation by Will et al. (2015) coincides with that of several other authors (e.g., Eckelmann et al., 2014; Franke and Dulce, 2017; Krohe, 1996; Oncken, 1997, 2000). Based on structural and geophysical data Oncken (1997, 2000) argued that the Mid-German Crystalline Zone contains upper plate Saxothuringian rocks with Armorican/Gondwanan affinities and lower plate Rhenohercynian material with Baltica-derived cover, the latter of which was tectonically accreted to the base of the overriding plate by underplating during the late stages of the Variscan orogeny. If true, the Otzberg-Michelbach Fault Zone delineates the western outcrop limit of the Spessart-eastern Odenwald tectonic window (Fig. 2) and would be a segment of a Variscan terrane boundary in the central European Variscides.

3. Sample description and results of previous geochemical studies Three amphibolite-facies (SP-14, SP-57, SP-68), two eclogite-facies (EH-10, EH-19) metabasic rocks and one granulite-facies paragneiss (GH-11) of the Odenwald-Spessart basement are investigated in this study (Table 1). The sample locations are marked in Fig. 2 and, the geochemical and whole rock isotope data of the metabasic rocks (Will et al., 2015) are briefly summarised. Sample SP-14 is a weakly foliated tholeiitic to calcalkaline amphibolite from the Elterhof Formation in the southern Spessart basement and consists of green amphibole, plagioclase, quartz and minor biotite, opaque phases and zircon. The sample is enriched in light rare earth elements, LREE (Fig. 4a) and has a prominent negative Nb-Ta anomaly (Fig. 4b). Based on these and other geochemical features, Will et al. (2015) suggested that the amphibolite protolith formed at an active continental margin where it was overprinted by crustal and/or subduction-derived fluids. The subalkaline to alkaline amphibolite SP-57 was sampled near the contact of the Mömbris with the Geiselbach Formation in the central Spessart basement (Fig. 2). The massive amphibolite consists of green amphibole, plagioclase, epidote, quartz, opaque phases, and minor zircon and rutile. The composition of the rock corresponds to that of a high-Ti ocean basalt, its REE pattern is subparallel to the modern ocean island basalt trend (Fig. 4a) and, the sample lacks a negative Nb-Ta anomaly (Fig. 4b). Such features are typical of rocks that formed in an oceanic within-plate setting. A geochemically enriched mantle source, which is typical for rocks emplaced in such an environment, is also implied by other geochemical features discussed by Will et al. (2015). The composition of sample SP-68 from the Alzenau Formation in the northern Spessart basement corresponds to that of a tholeiitic basalt. The strongly foliated amphibolite consists of green amphibole, plagioclase, opaque phases, minor quartz, and zircon. The flat REE pattern plots close to the modern N-MORB trend (Fig. 4a). Will et al. (2015) proposed that the sample originated at a divergent margin but was subsequently overprinted at an active continental margin by subductionderived fluids. The retrogressed eclogite samples EH-10 and EH-19 were collected at the same locality in the eastern Odenwald (Fig. 2). The rocks contain variable amounts of garnet, omphacite, sodic diopside, calcic amphibole, plagioclase, epidote/zoisite, quartz, rutile, ilmenite, titanite, and zircon. Prominent symplectites of clinopyroxene and plagioclase (Fig. 2 in Will and Schmädicke, 2001) and coronas of amphibole and plagioclase are typical textural features of these rocks. The rocks have a tholeiitic

Table 1 Sample list. Sample Spessart basement Spessart amphibolite-facies metabasite SP14 SP57 SP68 Odenwald basement Eastern Odenwald eclogite-facies metabasite EH-10-II EHU-19 Western Odenwald granulite-facies metapelite GH-11

Composition

Formation/Unit

Geographic location

Geographic coordinates

δ18O zrc

U-Pb zrc

Lu-Hf zrc

Tholeiitic to calcalkaline Subalkaline to alkaline Tholeiitic

Elterhof

Grauberg

X

X

X

Mömbris/ Geiselbach Alzenau

Road Hörstein-Hohl Alzenau train station

49°57′01″N 09°10′49″E 50°02′58″N 09°05′35″E 50°05′19″N 09°03′58″E

X

X

X

X

X

X

Tholeiitic

Unit IV

Eierhöhe

X

X

X

X

Tholeiitic

Unit IV

Eierhöhe

X

X

X

X

Pelitic

Unit II

Gadernheim

X

X

X

49°48′33″N 08°56′19″E 49°48′33″N 08°56′19″E 49°42′45″N 08°44′51″E

U-Pb rt

U-Pb mzt

Teconic setting⁎

Continental arc Oceanic within plate Back-arc basin

X

Mid-ocean ridge Mid-ocean ridge X

Odenwald units according to Krohe (1992). *Inferred tectonic setting of metabasic rocks according to Will et al. (2015). Mineral abbreviations: zrc-zircon, rt-rutile, mzt-monazite.


1000

Rock/Chondrite

SP-14 SP-57 SP-68 EH-10 & -19 100

10

OIB

283

otherwise stated the uncertainties for individual analyses (ratios and ages) are given at the 1σ level in the tables and the concordia ages are reported at the 2σ level. We tried to use only zircon U-Pb data with a concordance (=100 ∗ (206Pb/238U age) / (207Pb/206Pb age)) of 90– 110% for the age calculations. However, this was impossible for sample EH-10 that contains zircon grains with very low U concentrations. In this and a few other cases analyses with a lower concordance had to be used. Cathodoluminescence (CL) images of several zircon grains are shown in Fig. 5 and the results of the zircon dating are summarised in Table 2 and graphically displayed in Fig. 6.

N-MORB

4.1. Amphibolite SP-14

Lu

Yb

Er

Tm

Ho

Tb

Dy

Eu

Gd

Sm

Pr

Nd

La

1

Ce

(a)

Rock/Primitive mantle

1000 SP-14 SP-57 SP-68 EH-10 & -19

100

10

Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Sm Zr Hf Eu i T Gd Tb Dy Y Ho Er Tm Yb Lu

(b) 1

Fig. 4. REE and trace element patterns of amphibolites from the Odenwald-Spessart basement. The normalisation values and the values for the ocean island basalt (OIB) and ‘normal’ mid-ocean ridge basalt (N-MORB) reference curves are from Sun and McDonough (1989).

composition and high TiO2 contents. The REE patterns are generally unfractionated and are subparallel to the modern N-MORB trend at elevated REE concentrations (Fig. 4a). Sample EH-10 has an εNd(360 Ma) of 7.8. Will et al. (2015) proposed that the eclogite precursor rocks formed in a divergent setting, most likely in a shallow magma chamber below a mid-ocean ridge and were juvenile additions to the crust from the depleted MORB mantle. Subsequent subduction, concomitant highpressure metamorphism and assimilation of continental crust variably modified the sample compositions. Sample GH-11 from the western Odenwald is a granulite-facies quartz-garnet-sillimanite-bearing paragneiss that, in addition, contains cordierite, plagioclase, biotite, opaque phases, zircon, and monazite. No leucosomes or other features indicative of partial melting are present. 4. Results The analytical procedures and techniques used to obtain the isotope and trace element data are described in Appendix A. The results are summarised in the electronic Supplement Tables S1 to S6 (Table S1: zircon U-Pb and O isotope data; Tables S2 and S3: rutile and monazite U-Pb isotope data; Table S4: zircon Lu-Hf isotope data; Tables S5 and S6: zircon and rutile trace element data, including the calculated apparent zircon and rutile crystallisation temperatures). Unless

More than a 1000 sub- to euhedral zircon grains were separated from this sample. The grains are up to 200 μm long and most grains have a weak luminescence and appear grey to dark in CL imaging but irregular patches with higher CL response occur as well (Fig. 5a). A strong oscillatory zoning pattern is present in many grains (Fig. 5a). Twenty-seven spot analyses provide a concordia age of 329.1 ± 1.8 (2σ) Ma (Fig. 6a). All grains used for the age determination have a Th/U ratio well above 0.4 with a median of 0.74 (Fig. 6b, Table 2). The 30 zircon grains analysed for δ18O yield values ranging from 5.94 to 6.75‰ with a median value of 6.4 (Fig. 7a, Table 2), which is above the average uncontaminated mantle δ18O value of 5.3 ± 0.6‰ (2σ) reported by Valley et al. (1998). The εHf initials and Hf model ages (Fig. 8, Table 2) of the zircon grains range between −1.61 and 1.81 (median of 0.62) and from 1291 to 1106 Ma (median of 1171 Ma). The chondrite-normalised zircon rare Earth element (REE) patterns have distinct positive Ce and negative Eu anomalies. The REE patterns are characterised by low concentrations of LREE and a steep slope towards higher concentrations of HREE (Fig. 9a), which is typical for igneous zircons (Hoskin and Schaltegger, 2003). There are no differences between core and rim analyses with respect to U-Pb age, Hf and O isotopes and trace element patterns. Titanium-in-zircon thermometry yields apparent crystallisation temperatures ranging from 758 to 830 °C (n = 6), with a median of 809 °C. 4.2. Amphibolite SP-57 Some 50 zircon grains could be separated from sample SP-57. The grains are generally rounded to anhedral but a few 70–80 μm long subhedral zircons occur. Oscillatory zoning is preserved in many prismatic grains (Fig. 5b). Most grains appear grey in CL imaging but a few grains have bright rims and/or dark cores. Individual core and rim ages of the same grain could not be obtained. The majority of the grains yield Phanerozoic ages but a few Mesoproterozoic and one Mesoarchaean age were also determined. The REE patterns are typical for igneous zircon with high HREE concentration except for one grain that has distinctly lower HREE (Fig. 9b). The εHf initials of the 15 grains analysed are very variable and range from −15.4 to +11.3 (Fig. 8). The zircon δ18O values are also very variable and lie between 4.4 and 10.8‰ (Fig. 7b). Four grains in sample SP-57 provide a concordia age of 444.9 ± 6.4 (2σ) Ma (Fig. 6c). These grains are characterised by oscillatory zoning (Fig. 5b) and Th/U ratios of 0.12 to 0.26 (median of 0.16; Fig. 6d). In addition, they have high δ18O values of 7.87 to 9.39‰ and εHf initials of −15.0 to +3.3 (Figs. 7b, 8a). Titanium-in-zircon thermometry yields temperatures of 752 to 937 °C, with the lower and higher temperatures having been obtained at zircon rim and core, respectively. Rutile is generally an- to subhedral and up to 200 μm long. Twelve grains with a uranium content of 0.35 to 33 ppm (mean of 14.7) were analysed by SIMS. Data regression of eleven analyses (excluding one with very low U concentration and resulting large errors, Table S2) yields a lower concordia intercept age of 321.4 ± 7.0 Ma (2σ) (Fig. 10a). A weighted average 206Pb/238U of 321.7 ± 9.3 Ma (95% confidence) is consistent with this age. Zirconium-in-rutile thermometry (n = 5) provided a very narrow temperature range of 736–751 °C for

284


Sample SP-14

Sample SP-57

SP-14#8

SP-14#2

332 Ma (0.84)

327 Ma (0.78)

SP57#17

435 Ma (0.12)

SP57#8

439 Ma (0.26) SP57#10

SP57#20

(b) SP-14#16

460 Ma (0.19)

448 Ma (0.14)

Sample SP-68

SP-68#2

SP-68#1

329 Ma (0.78) SP-14#22

333 Ma (0.01)

SP-68#20 328 Ma (0.74)

325 Ma (0.66)

Sample EH-10

470 Ma (0.56)

335 Ma (0.04)

(c)

SP-68#19

SP-68#18

467 Ma

450 Ma (0.50)

GH-11#6

EH-10#9

452 Ma (0.42) SP-68#14

447 Ma (0.47)

GH-11#11

341 Ma (0.14)

EH-10#16 EH-10#2

(d)

329 Ma (0.23)

SP-68#17

SP-68#23 468 Ma (0.65)

Sample GH-11

318 Ma (0.10)

314 Ma (0.04)

EH-10#19

EH-10#28

329 Ma (0.21)

SP-68#13

SP-14#24 338 Ma (0.02)

(a)

SP-68#4

331 Ma (0.41) 337 Ma (0.02)

100 μm

GH-11#12

(e)

338 Ma (0.23)

339 Ma (0.38) GH-11#28

331 Ma (0.32)

Fig. 5. Cathodoluminescence (CL) images of zircon grains from amphibolite, eclogite and paragneiss from the Odenwald-Spessart basement. The SIMS U-Pb measurement spots are indicated by the circles. The age and the Th/U ratio of each measurement are given. Scale bar = 100 μm.

rutile equilibration, which corresponds closely to the temperature obtained for zircon rim growth by Ti-in-zircon thermometry (see above). 4.3. Amphibolite SP-68 Sixty sub-to euhedral zircon grains were separated from this sample. Most grains are prismatic and generally shorter than 50–60 μm (Fig. 5c) even though a few grains are up to 100 μm in length. Rounded grains are rare. Many grains are zoned (oscillatory or sector zoning) and appear

grey to black in CL imaging. Most of the twenty grains analysed by SIMS yield Phanerozoic ages but some Proterozoic grains were also found. Prismatic grains with strong oscillatory zoning are, in places, well rounded at their terminations and have high Th/U ratios well above 0.47 with a median of 0.6 (Fig. 6h). Eight of these grains provide a concordia age of 459.7 ± 4.7 (2σ) Ma (Fig. 6g) and have δ18O values ranging from 4.73 to 6.02‰, with a median of 5.36‰ (Fig. 7c, Table 2). These zircons have εHf initials of 2.33 to 13.08 (median of 10.67) and Hf model ages of 592 to 715 Ma, with a median of 712 Ma (Fig. 8, Table 2).

Table 2 SIMS U-Pb zircon, rutile and monazite age summary and interpretation of Odenwald-Spessart basement rocks. All age uncertainties are given at the 2σ or 95% confidence level; the latter are italizised. Th/U, δ18O, εHf and TDM, Hf values in parentheses are median values. Lithology

n (nAge)a tZircon (Ma)b

SP-57

SP-68

EH10

EH19

Amphibolite

Amphibolite

Retrogressed eclogite

Retrogressed eclogite

Granulitic paragneiss

30 (27) 329.1 ± 1.8 (27)

28 (4) 444.9 ± 6.4 (4)

20 (12) a: 328.4 ± 8.9 (4) b: 459.7 ± 4.7 (8)

48 (20) 333.7 ± 4.1 (20)

59 (0) No reliable agec

30 (24) 334.0 ± 2.0 (24)

328.6 ± 4.7 (27)

327.3 ± 3.0 (33)

tRutile (Ma)b tMonazite (Ma)b Th/U of the nAge zircons

0.41–0.88 (0.74)

0.12–0.26 (0.16)

δ18O of the nAge zircons

6.01–6.75 (6.41)

7.87–9.39 (8.74)

Range of all (n = 30) δ18O zircon analyses nLu/Hfd Range of εHfte

5.94–6.75 15 −1.61 to 1.81 (0.62)

4.35–10.78 15 −15.35 to 11.27

TDM, Hf (Ma)e

1106–1291 (1171)

769–3589

tZircon (this study) εNdtf TDM, Nd (Ma)*/** Interpretation

329 1.02 1008* Magmatism & metamorphism

Comments Tectonic setting (after Will et al., 2015)

Crustal input in cont. volcanic arc Continental volc. arc

445 −8.44 1876** Zrc: Magmatism Rt: Metamorphism Palaeoprot. source Intraoceanic/OIB

321.4 ± 7.0 (11) a: 0.01–0.23 (0.03) b: 0.47–0.90 (0.60) a: 8.35–8.95 (8.68) b: 4.73–6.02 (5.36) 4.73–10.12 14 −18.83 to 13.23 (all data) b: 2.33 to 13.08 (10.67) 592–2840 (all data) b: 592–715 (712) 328/460 5.04/5.13 673**/774** a: Metamorphism b: Magmatism Juvenile Neoproteroz. source Back-arc basin

GH-11

0.01–0.33 (0.04)

330.1 ± 2.4 (29) 0.03–0.46 (0.34)

6.31–6.94 (6.60)

10.04–11.31

6.22–6.94 14 9.61–11.44 (10.23)

6.28–8.81 15 5.64–14.16 (9.97)

9.98–11.31 13 −7.77 to −6.03 (−6.95)

609–686 (646)

551–884 (660)

1542–1635 (1600)

Not determined

Not determined

Rt: Metamorphism

Zrc & mzt: Metamorphism

Juvenile Neoproteroz. source Mid-ocean ridge

Recycled Continental margin

334 7.82 452** Zrc & rt: Metamorphism Juvenile Neoproteroz. source Mid-ocean ridge


SP-14 Amphibolite

*TDM 1-stage and **TDM 2-stage whole rock Nd model ages according to Goldstein et al. (1984) and using the zircon ages obtained in the current study. a n: total number of U-Pb zircon analyses. nAge: number of analyses used for age determinations. b Number of analyses used for age determinations in parentheses. For further details see text. c No reliable age because of very low U concentrations in zircon. d Number of zircon Lu/Hf analyses. e εHft and TDM, Hf calculated using the measured U-Pb age of each zircon grain. f Using Nd data presented by Will et al. (2015).

285

286


Several grains (top row in Fig. 5c) are multi-faceted with welldeveloped sector zoning and (with one exception) very low Th/U ratios of b0.04 (Fig. 6f). These zircons have high δ18O values of 8.35 to 8.95‰ (Fig. 7c). Four of these grains were analysed and provided a concordia age of 328.4 ± 8.9 (95% confidence) Ma (Fig. 6e). Titanium-in-zircon thermometry yields temperatures between 741 and 886 °C (n = 5). The zircons have low LREE and high HREE concentrations and the chondrite-normalised REE patterns have distinct positive Ce and negative Eu anomalies (Fig. 9c).

0.6 330

0.052 0.050 0.048 0.34

Concordia age = 329.1 ± 1.8 Ma (2σ, decay const. errors included) MSWD (of concordance) = 0.00012 Probability (of concordance) = 0.99

310

0.36

0.38 0.40 207Pb/235U

0.076

0.2

(b) 0.0 320

340

0.8

470

SP-57 Amphibolite 0.6

450

0.072

330

Age (Ma)

Th/U

206Pb/238U

0.42

0.4

SP-57 Amphibolite, n = 4 (Zircon)

0.074

0.070 430

0.4

0.068 0.066

0.2


410

(c)

(d)

0.064 0.48 0.50 0.52 0.54 0.56 0.58 0.60 207Pb/235U

0.0 300

0.057

1.0 SP-68 Amphibolite, n = 4 (Zircon)

0.8 0.6 0.4

320

0.049

(e)

Concordia age = 328.4 ± 8.9 Ma (95% confidence, decay const. errors included) MSWD (of concordance) = 1.9 Probability (of concordance) = 0.16

0.2

(f)

0.047 0.32 0.34 0.36 0.38 0.40 0.42 0.44 207Pb/235U

0.0 300

0.080

1.0

0.078

600

SP-68 Amphibolite

330

0.051

500

Th/U

0.053

400

Age (Ma)

350

0.055 206Pb/238U

SP-14 Amphibolite

0.8

Th/U

206Pb/238U

350

0.054

(a)

400

500

Age (Ma) SP-68 Amphibolite

SP-68 Amphibolite, n = 8 (Zircon) 0.8

480

0.076 0.6

0.074

Th/U

206Pb/238U

Several hundred zircon grains could be separated from samples EH10 and EH-19. The zircon grains are highly variable in shape and appearance. A few grains are short-prismatic to isometric and have oscillatory and, locally, fir-tree zoning but most zircons have highly variable crystal faces. The latter have very complex internal features and are characterised by heterogeneous, patchy structure and/or zonation in CL image (Fig. 5d). Some grains have bright luminescent

1.0

SP-14 Amphibolite, n = 27 (Zircon)

0.056

4.4. Retrogressed eclogites EH-10 and EH-19

460

0.4 0.072 440

0.070

(g) 0.068 0.48

0.52


0.56 0.60 207Pb/235U

0.64

0.2

(h) 0.0 300

400

500

Age (Ma)

Fig. 6. SIMS zircon U-Pb concordia (left column) and age vs. Th/U diagrams (right column) of Odenwald-Spessart basement samples. The samples represented by the filled symbols in the left column were used for the age calculations.


287

0.060

EH-10 Retrogressed eclogite, 0.058 n = 20 (Zircon)

0.3

EH-10 Retrogressed eclogite

360

Th/U

206Pb/238U

0.056 0.054 0.052 320

0.1

0.050 Concordia age = 333.7 ± 4.1 Ma (95% confidence, decay const. errors included) MSWD (of concordance) = 0.012 Probability (of concordance) = 0.91

0.048

(i) 0.046 0

0.2

0.4

0.6

0.8

1.0

1.2

320

340 360 Age (Ma)

380

GH-11 Granulite-facies paragneiss

0.4

0.056

Th/U

0.3

340

0.054

1.4

(j) 0.0 300

0.5

GH-11 Granulite-facies paragneiss, n = 24 360 (Zircon)

0.058

206Pb/238U

0.2

0.2

0.052 320

0.050 (k) 0.048 0.34

0.36

Concordia age = 334 ± 2.0 Ma (2σ, decay const. errors included) MSWD (of concordance) = 0.30 Probability (of concordance) = 0.58

0.38 0.40 0.42 207Pb/235U

0.44

0.1 (l) 0.0 320

330

340

350 Age (Ma)

360

370

Fig. 6 (continued).

cores and dark rims or vice versa or show almost featureless bright or, in many places, dark CL textures. The latter may be due to metamictization of the U-rich parts of the zircon. Several grains have lobate, convoluted, or schlieren-like textures or a combination of all or some of these features (Fig. 5d). The zircon grains have very low U and Th concentrations of typically b5 and 1 ppm, respectively, near-zero values of Pb and U/Yb ratios of b0.1. The common Pb-uncorrected data of 20 zircon grains in sample EH-10 provide a concordia age of 333.7 ± 4.1 (95% confidence) Ma (Fig. 6i). On a Tera-Wasserburg diagram (not shown) these data define a linear array providing an identical lower-intercept age of 334.8 ± 4.0 (95% confidence) Ma and a Y-intercept of common 207 Pb/206Pb composition of 0.84 ± 0.02. With this composition common Pb correction was performed using the 207Pb-based method. The 207-corrected data yield a weighted average 206Pb/238U age of 334.5 ± 5.1 Ma (95% confidence). Zircon grains in this sample have rather uniform δ18O values of 6.22 to 6.94‰ (median of 6.60‰), whereas those in sample EH-19 range from 6.28 to 8.81‰, with a median of 6.69‰ (Fig. 7d, e; Table 2). The zircons in samples EH10 and EH-19 have highly depleted εHf initials (median of 10.2 and 10.0, respectively) and late Neoproterozoic Hf model ages of 686 to 609 Ma (median of 646 Ma) and 884 to 551 Ma (median of 660 Ma), respectively. Titanium-in-zircon thermometry could only be performed on two grains yielding temperatures of c. 800 °C. The LREE concentrations of the zircon grains in the retrogressed eclogite are very low and mostly below the detection limit of the LA-ICP-MS. In addition, the grains have distinctly lower HREE concentrations than the amphibolite samples (Fig. 9d, e). Rutile is sub- to anhedral and up to 180 μm long. The uranium content of rutile (Table S6) in sample EH-10 ranges from 0.07 to 50 ppm (median of 1.7 ppm) and that of sample EH-19 varies between 1.2 and 152 ppm (median of 32 ppm). The two samples have indistinguishable lower concordia intercept ages of 328.6 ± 4.7 Ma (n = 27) and 327.3 ± 3.0 Ma (n = 33) at the 95% confidence level, respectively (Fig. 10b, c). Zirconium-in-rutile thermometry provides temperatures

ranging from 598 to 713 °C and 591 to 722 °C, respectively. Combining both samples yields a median rutile equilibration temperature of 657 °C (n = 14).

4.5. Granulitic paragneiss GH-11 More than 2000, generally prismatic and up to c. 140 μm long zircon grains were separated from this sample. Many zircons have a welldeveloped sector zoning but fir-tree zoning is also present in some grains (Fig. 5e). The Th/U ratios are (with one exception) higher than 0.2 resulting in a median of 0.34 (Fig. 6l). Twenty-four spot analyses provide a U-Pb concordia age of 334.0 ± 2.0 (2σ) Ma (Fig. 5k). The δ18O values of these grains vary between 10.04 and 11.31‰ (Fig. 7f) and the εHf zircon initials are enriched and range from −7.77 to −6.03 (median: −6.95). Hafnium model ages vary between 1635 and 1542 Ma, with a median of 1600 Ma (Fig. 8, Table 2). Titanium-in-zircon thermometry yields very uniform temperatures between 739 and 795 °C (n = 6; median of 770 °C). The zircon REE concentrations in the paragneiss are lower than in the mafic rocks analysed; this is especially true for the HREE, which are two to three orders of magnitude lower than in the mafic samples. In addition, there is no marked Ce anomaly but a distinct negative Eu spike (Fig. 9f). Over a 1000 monazite grains were extracted from the sample. The grains are sub- to anhedral and up to c. 100 μm long. In places they have rounded grain boundaries and some grains have inclusions that are too small to be identified with the optical microscope. The U contents of thirty monazite grains analysed ranges from 0.05 to 1.23 wt% (mean of 0.26 wt%), the Th content is 4.2–6.1 wt% (mean of 5.3 wt%) and the Th/U ratios vary between 3 and 119. On a Tera-Wasserberg plot the common Pb- and 230Th-uncorrected data give a lower intercept age of 330.1 ± 2.4 (2σ) Ma (Fig. 10d). After 207-based common Pb and 230 Th corrections (Ludwig, 2001), identical weighted average 206 Pb/238U and208Pb/232Th ages of 330.3 ± 1.9 (2σ) Ma and 330.9 ± 1.4 (2σ) Ma are obtained.

288


11

7.0 SP-14 Amphibolite

SP-57 Amphibolite

10 9

δ18O

δ18O

6.5

6.0

5

(a) 330

7 6

Zircon Th/U > 0.4

5.5 320

Zircon Th/U < 0.3

8

4

340

300

400

Age (Ma)

600

7.0 SP-68 Amphibolite

6.8

Zircon Th/U mostly < 0.1

EH-10 Retrogressed eclogite

6.6

7 Zircon Th/U > 0.4

6

δ18O

8

δ18O

500

Age (Ma)

10 9

(b)

SP-57 Amphibolite

Zircon Th/U mostly < 0.1

6.4 6.2

5 (c) 4 300

400

(d) 6.0 300

500

320

Age (Ma)

340

360

380

Age (Ma)

7.5

11.5 EH-19 Retrogressed eclogite

GH-11 High-T paragneiss 11.0

δ18O

δ18O

7.0 10.5

6.5 Zircon Th/U < 0.2

Zircon Th/U > 0.2

10.0

(e) 6.0 300

350

400

450

500

Age (Ma)

(f) 9.5 320

330

340

350

360

370

Age (Ma)

Fig. 7. SIMS zircon U-Pb age versus δ18O relationship of Odenwald-Spessart basement samples. The grey bands give the sample ages including their 2σ uncertainties. The average uncontaminated mantle has a δ18O value of 5.3 ± 0.6 (2σ) ‰ (Valley et al., 1998).

5. Discussion 5.1. Zircon origin and significance of the mineral ages 5.1.1. Amphibolite SP-14 The oscillatory zoning (Fig. 5a), the high Th/U ratios (Fig. 6b) and the REE patterns (Fig. 9a) of the zircons provide consistent evidence that they crystallised from a melt and are of igneous origin. The broad zonation of most grains is thought to be typical for zircon growth in a plutonic environment (Corfu et al., 2003). The rather uniform εHf initials and the narrow range of Hf model ages (Table 2) point to a common source region of the protolithic melts (Fig. 7), with the near-zero εHf initials either indicating recycling of older crustal material or mixing of juvenile, mantle-derived material with pre-existing continental crust. These data fit in well with those presented by Will et al. (2015), which indicated that amphibolite SP-14 formed at an active continental edge. The concordia age of 329.1 ± 1.8 Ma (Fig. 6a) is interpreted to date the time of arc magmatism. In addition, the sample experienced

amphibolite-facies metamorphism but metamorphic overgrowth rims are not visible on the zircon grains. Emplacement into the continental crust must therefore have occurred coevally with or only slightly after arc magmatism. 5.1.2. Amphibolite SP-57 Zircon grains in this sample have very variable shapes, U-Pb spot ages, U and Th concentrations, Th/U ratios and Hf isotope values (Fig. 6d; Table 2), all of which may point to a detrital origin of many grains. However all grains, except for grain #12, have REE patterns (Fig. 9b) that are typical for zircons that grew from a melt. The zircon Hf and high δ18O values data indicate interaction of the protolithic melt with supracrustal rocks and/or fluids. This supports an earlier suggestion (Will et al., 2015) that the precursor of this sample formed in a within-plate oceanic setting but was subsequently modified by crustal assimilation at a continental margin. The significance of the Ordovician/Silurian zircon age of ca. 445 Ma (Fig. 6c) is unclear. The grains used to constrain this age may be


15

289

13

10

11

δ18O

εHft

5 0 -5

9 7

-10 5

-15 (a) -20 300

400

500

600

Zircon age (Ma)

400

500

600

Zircon age (Ma) SP-14 SP-57 SP-68 EH-10 EH-19 GH-11

13 11

δ18O

(b) 3 300

9 7 5 (c) 3 -20

-10

0

10

20

εHft Fig. 8. Zircon age-Hf-O relationships of Odenwald-Spessart basement samples. The shaded bands in (b) and (c) show the average mantle zircon δ18O value of 5.3 ± 0.6 (2σ) ‰ (Valley et al., 1998).

detrital (see above) but have a distinct magmatic oscillatory zoning, which is typical for igneous zircon (Fig. 5b). It is conceivable that the protolithic melt formed close to the Ordovician-Silurian boundary but subsequent interaction with continental crust caused assimilation of detrital zircons. We regard the U-Pb rutile age of 321.4 ± 7.0 Ma to date the time of amphibolite-facies metamorphism experienced by the sample.

5.1.3. Amphibolite SP-68 Prismatic zircon grains with strong oscillatory zoning and Th/U ratios that are typical of magmatic zircon (Fig. 6h) provide a concordia age of ca. 460 Ma (Fig. 6g), which is interpreted to constrain the time of formation of the amphibolite protolith. The very high εHf initials of these zircons (Table 2) point to formation in the highly depleted MORB mantle, which is further supported by their near-uncontaminated δ18O mantle values (Fig. 7c) and whole rock Nd isotope data (Will et al., 2015). Some grains that have morphological features (top row in Fig. 5c) and Th/U ratios (Table 2) that are characteristic of zircon that experienced high-grade metamorphism (e.g., Hoskin and Black, 2000; Kröner et al., 2006) provide a concordia age of ca. 328 Ma. This, and the high δ18O values of the zircon grains (Fig. 7c), which are typical of crustal recycling and/or contamination by upper crustal material (e.g., Valley et al., 1998) are taken as evidence that this sample experienced amphibolite-facies metamorphism in a supra-subduction setting at that time. Thus, the amphibolite protolith formed at ca. 460 Ma in the shallow MORB mantle, probably in a marginal basin (Will et al., 2015) and was metamorphosed during its accretion to a continental margin some 130 million years later. This time span corresponds closely to the average lifetime of an ocean basin and is in the same order as the age of the present-day South Atlantic Ocean (e.g., Will and Frimmel, 2018).

5.1.4. Retrogressed eclogites EH-10 and EH-19 The highly depleted zircon εHf and whole rock εNd initials (Table 2) indicate that the eclogite precursor formed from a juvenile source in a divergent setting at a mid-ocean ridge. However, the origin of the zircon in the retrogressed eclogite is not clear. Zircons have very low U concentrations and U/Yb ratios (Table S1), which is thought to be typical of magmatic zircon that formed in a midocean ridge setting (e.g., Grimes et al., 2007, 2015). This is consistent with the interpretation of the Hf and Nd isotope data. In contrast, the very low Th/U ratios are characteristic of metamorphic zircon but these ratios must be regarded with caution because of the low element concentrations. Moreover, many zircon grains have highly irregular grain boundaries and convoluted or schlieren-like textures (Fig. 5d), which are generally ascribed to pervasive fluid alteration and/or associated re-equilibration of zircon grains in aqueous fluids and/or melts (e.g., Geisler et al., 2007; Taylor et al., 2016). Thus, the low U, Th and Pb concentrations could have been caused by fluid infiltration, which is in agreement with the elevated zircon δ18O values (Fig. 7d, e). Hence it remains unclear whether initial zircon formation occurred in a magmatic or a metamorphic environment. However, for reasons outlined below, we argue that the geochronological information provided by the zircon grains constrains the age of eclogite-facies metamorphism closely. The zircon REE concentrations are low (Fig. 9d, e) and may indicate that the mineral crystallised either contemporaneously with another phase that preferentially incorporates HREE or, alternatively, that zircon growth post-dates the formation of such a phase. The retrogressed eclogite contain ubiquitous garnet, which is the most likely cause for the reduced HREE concentrations in zircon (e.g., Whitehouse and Platt, 2003). If true, this implies that garnet crystallised either coevally with or prior to zircon. Given that garnet grew during eclogite-facies metamorphism (Will and Schmädicke, 2001) this further implies that the U-Pb zircon age

290


10000

10000

SP-57

SP-14 1000

1000

100

100

10

10

Grain #12-12 (rim)

Grain #12-1 (core)

1

1

(b)

(a)

Lu

Yb

Er

Tm

Ho

Tb

Dy

Gd Gd

Eu

Gd

Eu Eu

Sm Sm Sm

Pr

Nd Nd Nd

La

Lu

Yb

Er

Tm

Ho

Tb

Dy

Eu

Gd

Sm

Pr

Nd

La

Ce

Ce

0.1

0.1

10000

10000

SP-68

EH-10

1000

1000

100

100

10

10

1

1

c. 330 Ma old grains c. 460 Ma old grains

(c)

(d)

0.1

Tb

Dy

Ho

Er

Tm

Yb

Lu

Tb

Dy

Ho

Er

Tm

Yb

Lu

Pr

La

Ce

Lu

Yb

Tm

Er

Ho

Tb

Dy

Gd

Eu

Sm

Pr

Nd

La

Ce

0.1

10000

10000

EH-19

GH-11

1000

1000

100

100

10

10

1

1

(e)

(f) Pr

La

Lu

Yb

Tm

Er

Ho

Tb

Dy

Gd

Eu

Sm

Pr

Nd

La

Ce

Ce

0.1

0.1

Fig. 9. REE data of zircon grains from amphibolite (a–c), eclogite (d–e) and paragneiss (f) samples of the Odenwald-Spessart basement.

of sample EH-10 closely constrains the time of metamorphism. This is strongly supported by the U-Pb rutile ages of 328.6 ± 4.7 and 327.3 ± 3.0 Ma for samples EH-10 and EH-19 (Fig. 10b, c), which are very close to the U-Pb zircon age of 333.7 ± 4.1 Ma for sample EH-10 (Fig. 6i). As rutile is a mineral that typically grows during high-grade metamorphism it should provide a robust estimate for the time of metamorphism. Thus, we conclude that our zircon and rutile U-Pb analyses constrain the time of eclogite-facies metamorphism to ca. 330 Ma. 5.1.5. Granulitic paragneiss GH-11 The zircon morphology and the Th/U ratios point to zircon formation by crystallisation from a magmatic liquid, whereas the high δ18O and strongly enriched εHf initials of the zircon indicate substantial crustal recycling and/or contamination. Zircon with such

low HREE concentrations as determined in sample GH-11 was interpreted to have formed by sub-solidus growth in the presence of garnet (Hoskin and Schaltegger, 2003). However, this interpretation is in contrast to the zircon morphology (Fig. 5e) and the high Th/U ratio, which are better explained by a magmatic origin of the mineral. It is thus conceivable that the zircon grains formed in a melt at 334.0 ± 2.0 Ma (zircon U/Pb age) but their HREE concentrations were modified during slightly later granulite-facies metamorphism at 330.1 ± 2.4 Ma (monazite U/Pb age). Alternatively, if the melt was anatectic and garnet was a residual phase it is also plausible that the melt was originally low in HREE and the zircon age is close to the time of peak metamorphism. The zircon Hf isotope data (Table 2) indicate an Amazonian or southern Baltica provenance (e.g., Henderson et al., 2016; Sagawe et al., 2016), the latter of which is our preferred source area.


SP-57 Amphibolite, n = 11 (Rutile)

0.060

0.28

360

207Pb/206Pb

206Pb/238U

0.052

320

0.048 Concordia age = 321.4 ± 7.0 Ma (2σ, decay const. errors included) MSWD (of concordance) = 0.032 Probability (of concordance) = 0.86

280

0.044

(b) (a)

0.040 0.26

0.34 0.38 0.42 207Pb/235U

0.20 0.16 0.12 0.08

0.46

13

0.20


0.12 0.08

(c)

380

340

16

18 20 238U/206Pb

15

320

17 19 21 238U/206Pb

23

25

GH-11 Granulite-facies paragneiss, n = 29 (Monazite) Intercept at 330.1 ± 2.4 Ma (95% confidence) MSWD = 5.2

0.16 0.12 0.08 410

300

390

370

350

330

310

(d)

0.04 14

400

(b)

0.04

EH-19 Retrogressed eclogite, n = 33 (Rutile)

0.16


0.24

207Pb/206Pb

207Pb/206Pb

0.20

0.30

EH-10 Retrogressed eclogite, n = 27 (Rutile)

0.32

0.056

291

22

0.04 14.5 15.5 16.5 17.5 18.5 19.5 20.5 238U/206Pb

Fig. 10. SIMS rutile (a–c) and monazite (d) U-Pb concordia ages of Odenwald-Spessart basement samples.

5.2. Comparison with other age data from the Odenwald-Spessart basement In a recent study Will et al. (2017) provided information on the age of metamorphism in the Odenwald-Spessart basement by white mica 39 Ar/40Ar data and in-situ electron microprobe dating of monazite (Fig. 3). Our new SIMS U-Pb zircon, rutile and monazite data expand and refine this geochronological data set. Regardless of rock type, zircon in eclogite, amphibolite and high-temperature paragneiss provide almost identical Carboniferous ages of 333.7 ± 4.1 Ma (eclogite), 329.1 ± 1.8 to 328.4 ± 8.9 Ma (amphibolite), and 334.0 ± 2.0 Ma (HT paragneiss), respectively. Rutile in eclogite and amphibolite yield ages between 328.6 ± 4.7 and 321.4 ± 7.0 Ma, and monazite in HT paragneiss grew at 330.1 ± 2.4 Ma. Thus, the new data demonstrate that the OdenwaldSpessart basement experienced high-pressure and high-temperature metamorphism contemporaneously at ca. 330 and amphibolite-facies conditions shortly afterwards. Two amphibolite samples from the Spessart basement provide evidence for an Ordovician tectonothermal event. Sample SP-68 yields a well-constrained concordia age of 459.7 ± 4.7 Ma, which we interpret to constrain the time of protolith formation in a back-arc basin (Will et al., 2015; Figs. 11a & 12a). Magmatic rocks of that age have not yet been described from the Odenwald-Spessart basement (Fig. 3). A somewhat younger but less well-constrained upper Ordovician age of some 445 Ma was obtained for sample SP-57, whose precursor rock is interpreted to have formed in an oceanic within-plate setting (Will et al., 2015; Figs. 11b & 12b). The Silurian in-situ monazite EMP age of 430 ± 33 Ma for a granulitic paragneiss from the western Odenwald (Will et al., 2017) could not be duplicated in the present study even though sample GH-11 was collected at the same locality as the sample investigated in the previous EMP study. This is unfortunate because it

would have been desirable to verify this rather unusual age of a western Odenwald basement sample. Nevertheless, the older monazite EMP age should be real (even though not very well constrained) as it is based on in-situ monazite analyses, which excludes any contamination by foreign material. If true, this age as well as the suspected Baltican provenance of the zircon grains (Fig. 12b; see above) would support the exotic nature of the high-temperature western Odenwald paragneiss as suggested by Will et al. (2017). Our new SIMS U-Pb zircon (333.7 ± 4.1 Ma) and rutile (328.6 ± 4.7 Ma and 327.3 ± 3.0 Ma) ages of the eclogite samples EH-10 and EH-19 are ca. 30 million years younger than a garnet-whole rock Lu-Hf age of 357 ± 6 Ma for the time of eclogite-facies metamorphism of sample EH-10 (Scherer et al., 2002). In this context it has to be emphasised (i) that the Lu-Hf garnet-whole rock isochron is a two-point isochron and, (ii) that it is strictly necessary that the whole rock chemistry did not change since the time of garnet growth. However, the samples are retrogressed and it is plausible that the eclogite Lu/Hf ratio was altered during that process; for example, a slight increase in Lu/Hf will invariably lead to a steeper isochron. Hence, it is likely that the retrograde overprint of the eclogite was not isochemical, and the Lu-Hf garnet-whole rock age will overestimate the age of eclogite-facies metamorphism, which is better constrained by our SIMS U-Pb zircon age data (K. Mezger, pers. comm., 2017). Furthermore, as argued above the inferred crystallisation order of garnet and zircon (garnet must have formed prior to or coevally with zircon) and the zircon and rutile U-Pb data clearly indicate that metamorphism occurred at ca. 330 Ma. It cannot be entirely ruled out that the samples experienced a polymetamorphic history with an older eclogite-facies event and a younger thermal event at lower pressure-temperature conditions. However, this is unlikely for several reasons. (1) The closure temperature of U-Pb in rutile is estimated to be ca. 620 ± 20 °C (Cherniak, 2000; Kooijman

292


465 Ma Siberia

Laurentia cean

Iapetus O

Baltica

Avalonia Rheic Ocean

Sample SP-68 IB+AM

Laurentia

Iapetu

s Oce

an

Gondwana

(a)

Siberia

440 Ma

Avalonia Baltica

Sample SP-57

Rheic Ocean

(b)

IB+AM

Siberia

Gondwana

Laurussia

Continents

Avalonia IB+AM

340-330 Ma Palaeotethys

Sample SP-14

Oceans Gondwana

Spreading centers Subduction zones (c) Fig. 11. Simplified palaeogeographic sketches from the Middle Ordovician to the Carboniferous (modified after Catalán et al., 2009; and Cocks and Torsvik, 2006). Solid blue lines are spreading ridges, and red lines are subduction zones with barbs on their downward sides. IB + AM: Iberia and the Armorican microcontinent, which includes the Armorican and Bohemian Massifs and parts of the Saxothuringian Zone in central Europe. The inferred locations of formation of the magmatic protoliths of some of our samples are indicated.

et al., 2010; Vry and Baker, 2006). Hence, our SIMS U-Pb rutile ages document a metamorphic event above this temperature, which is in the same order or only slightly lower than the estimated eclogitefacies peak temperatures of 700 ± 50 °C (Will and Schmädicke, 2001) and/or the median Zr-in-rutile temperature of 657 °C obtained in the present study. Any (hypothetical) older metamorphic event would have been largely obliterated at that time. (2) The retrogressed eclogite preserve very delicate clinopyroxene-plagioclase symplectites and amphibole-plagioclase coronas around garnet that formed at only somewhat lower pressures but identical temperature than the eclogite-facies peak assemblage (Will and Schmädicke, 2001). Invariably, such textures are very unstable and would have been erased by any tectonothermal event that succeeded their formation. Thus, we conclude that the eastern Odenwald basement experienced highpressure metamorphism at ca. 330 Ma. In summary, there is compelling evidence from the new SIMS U-Pb zircon, rutile and monazite age data, previous Ar-Ar and in-situ monazite ages and several other geochronological studies from the OdenwaldSpessart basement (Fig. 3) that eclogite-, amphibolite- and granulitefacies metamorphic peak conditions in the Odenwald-Spessart basement were reached in a narrow time interval between 334 and 321 Ma, with the majority of our new SIMS data providing ages of ca. 330 Ma.

5.3. Regional geological implications Will et al. (2015) advocated the idea that amphibolite north of the Michelbach Fault and west of the Otzberg Shear Zone (Fig. 2) belong to the same tectonostratigraphic unit and formed in a marginal basin. Our new geochronological data now define an age of 460 Ma for this event (Fig. 12a), which is contemporaneous with back-arc spreading and the opening of the eastern branch of the Rheic Ocean as suggested by von Raumer and Stampfli (2008) and Murphy et al. (2010). It has been suggested (e.g., Eckelmann et al., 2014; Franke et al., 2017; Linnemann et al., 2007; Oncken, 1997; Zeh and Will, 2010) that rocks exposed in the eastern Odenwald and south of the Michelbach Fault in the Spessart basement (Fig. 2) formed in a pre-Variscan magmatic arc during N- to NW-directed subduction of the Rheic Ocean underneath the Avalonian-Baltican margin prior to the lower Devonian. Geochemical (Will et al., 2015) and geochronological investigations (this study) support this idea and provide evidence that some of the Spessart amphibolite protoliths formed close to the Ordovician/Silurian boundary (Figs. 11b & 12b). Several authors (e.g., Franke et al., 2017; Oncken, 1997; Stampfli et al., 2013; von Raumer and Stampfli, 2008; Zeh and Gerdes, 2010) proposed that the subduction direction of the Rheic Ocean changed from NW to SE in the Devonian, which led to upper Devonian to Carboniferous continental arc magmatism (Figs. 11c & 12c, d). As shown by previous (Fig. 3) and our new SIMS age data this resulted in mafic magmatism between ca. 360 and 330 Ma and somewhat younger felsic intrusions between ca. 340 and 330 Ma in most of the OdenwaldSpessart basement area. Collision between Baltica/Avalonia- and Gondwana-derived terranes (Figs. 11c & 12e) led to N- to NW-directed tectonic stacking and, eclogite- and granulite-facies metamorphism at ca. 330 Ma. Exhumation of these rocks must have occurred within a few million years after they had reached peak metamorphic conditions as indicated by slightly younger white mica cooling ages of ca. 324 to 322 Ma from the Odenwald-Spessart basement (Will et al., 2017). The Mid-German Crystalline Zone is considered as a tectonic stack of metamorphic rocks with, (i) significantly different temperature– pressure gradients (dT/dP) and, (ii) whose precursor rocks formed in different tectonic settings and different palaeogeographic locations (e.g., Franke et al., 2017). Mafic and felsic upper Ordovician and Silurian magmatic rocks formed at or close to the northern margin of the Rheic Ocean, whereas upper Devonian to lower Carboniferous felsic and mafic rocks formed in a continental magmatic arc setting that was most likely located at or close to the Gondwanan margin of the Rheic Ocean (Zeh and Gerdes, 2010; Zeh and Will, 2010). Nonetheless, the numbers and palaeogeographic positions of the (micro) continents involved in the formation of the Mid-German Crystalline Zone and the central European Variscides, the locations and widths of the intervening ocean basins as well as the distribution of the Variscan sutures are still a matter of debate (e.g., von Raumer and Stampfli, 2008; Kroner and Romer, 2013; Stampfli et al., 2013; von Raumer et al., 2015; Will et al., 2015; Franke et al., 2017; and references therein).

5.3.1. A Variscan paired metamorphic belt in the Odenwald basement Eclogite-facies metabasic rocks in the Odenwald basement were first recognised by Will and Schmädicke (2001). After realising that these high-pressure rocks are tectonically separated from hightemperature rocks with an entirely different metamorphic history, Will and Schmädicke (2003) suggested that both units constitute a paired metamorphic belt. However, age data of these units did not exist at that time to test his idea. The results of the present SIMS study close that gap and leave no doubt that eclogite- and granulite-facies metamorphism in the eastern and western Odenwald occurred coevally.


ca. 480-460 Ma

East Avalonia

Baltica Iapetus

(a)

293

Incipient Rheic Ocean periGondwana

Deposition of paragneiss precursor GH-11?

Back-arc Marginal basin basalt SP-68

ca. 440-410 Ma Silurian Arc

(b)

Rheic Ocean PeriGondwana

Baltica/ Avalonia Silurian HT metamorphism of GH-11?

Silurian HT event

Oceanic intraplate basalt SP-57

ca. 410-390 Ma Nth Avalonia

(c)

Baltica/ Avalonia

Sth Avalonia PeriGondwana

? Opening of RH basin

Slab roll-back and break-off (?), opening of marginal basin

ca. 370-350 Ma Closing Narrowing RH basin Rheic Ocean

(d)

Frankenstein Gabbro Intrusion PeriGondwana

Baltica/ Avalonia

?

ca. 340-320 Ma Arc basalt SP-14

(e)

Baltica/ Avalonia

Carboniferous Arc Carboniferous HT metamorphism of sample GH-11 20

Carboniferous HT event in peri-Gondwanan upper plate, HP eclogite-facies metamorphism in lower plate and subsequent juxtapositon of both units

(f) P (kbar)

Carboniferous HP metamorphism of mid-ocean ridge basalts EH-10 & -19

High-pressure metamorphism in upper plate (Eastern Odenwald) zircon: 333.7 ± 4.1 Ma rutile: 328.6 ± 4.7 Ma

16

12

8

High-temperature metamorphism in lower plate (Western Odenwald) zircon: 334.0 ± 2.0 Ma monazite: 330.1 ± 2.4 Ma

4 550

600

650

700

750

800

T(°C)

Fig. 12. Simplified model for the development of the Variscan orogen in Central Europe (modified after Zeh and Will, 2010). The inferred locations of our samples are indicated. Note that the sequence of events depicted in this cartoon is far from certain as there are several contrasting models about the tectonic evolution of the European Variscan Belt (e.g., von Raumer and Stampfli, 2008; Catalán et al., 2009; Kroner and Romer, 2013; Stampfli et al., 2013; Eckelmann et al., 2014; Franke et al., 2017; and many more).

The close proximity of tectonostratigraphic units with such radically different metamorphic conditions and thermal gradients that, however, were reached at the same time are characteristic of a classical paired metamorphic belt sensu Miyashiro (1961). However, we are neither implying that the units in the Odenwald-Spessart basement resulted from subduction of a wide Pacific-type ocean basin nor that their formation is analogous to the evolution of the circum-Pacific metamorphic belts that are in parts characterised by successive metamorphic facies series, including blueschists, which are so far unknown in the Mid-German

Crystalline Zone. Instead, following Brown (2010) we use the idiom ‘paired metamorphic belt’ to express that two tectonically juxtaposed belts of similar age (in the case of the Odenwald basement they are of the same age) but of very different type of metamorphism and thermal gradients occur next to each other. However, inherent in our usage of this term is the implication that there is a relationship between tectonic setting and thermal regime such as, for example, a high dT/dP gradient in an arc/back-arc setting and a low dT/dP gradient in a subduction zone environment.

294


The western Odenwald upper plate is characterised by migmatite and high-grade granulitic rocks that are characterised by high dT/dP gradients and which followed counter-clockwise, hairpin like PT paths (Will and Schmädicke, 2003; Fig. 12f). The mafic oceanic rocks in the western Odenwald basement (at least those in unit II sensu Krohe, 1992) and those north of the Michelbach Fault in the Spessart basement (Fig. 2) formed in a pre-Variscan, rheologically weak but thermally hot back-arc setting (Will et al., 2015). Carboniferous shallow crustal level granites are ubiquitous in the western Odenwald basement and testify to the high dT/dP gradient attained in this area at that time. In contrast, the lower plate eastern Odenwald and Spessart basement consists of rocks recording low to intermediate dT/dP gradients, containing oceanic crustal rocks that formed initially in the shallow, depleted mid-ocean ridge mantle (Will et al., 2015). These rocks followed a clockwise PT path during their evolution (Fig. 12f). Metamorphic peaks were reached coevally in the lower (eclogite) and upper plate (granulite) units as demonstrated in the present study for the very first time. These observations can be explained by terrane amalgamation along the Otzberg Shear Zone in response to oblique subduction of the Rheic Ocean underneath the peri-Gondwanan continental margin at ca. 330 Ma or shortly afterwards (Fig. 12e). This idea is consistent with the different crustal structure across the Otzberg Fault, the age of metamorphism in both units and the geochemical signatures preserved in the oceanic mafic rocks on both sides of the fault zone. We cannot prove whether the high and low dT/dP belts formed in-situ or whether one of them or even both travelled far. However, given that eclogite- and granulitefacies peak conditions were reached contemporaneously at 330 Ma it is unlikely that the units had travelled far given that exhumation of the juxtaposed metamorphic pile was already underway at 324 Ma as indicated by white mica Ar-Ar cooling ages (Will et al., 2017). The interpretation of the two units as a paired metamorphic belt is a consequence of the combined geological, geochemical and geochronological data. To our knowledge this is the first documented occurrence of a paired metamorphic belt in the Variscides and also one of the very few already recognised in the northern hemisphere. 6. Conclusions and geodynamic implications From our new SIMS U-Pb zircon, rutile and monazite data, zircon Lu-Hf and oxygen isotope data and zircon and rutile trace element data together with whole rock isotope and geochronological data presented by Will et al. (2015, 2017) several new conclusions can be drawn, some of which are illustrated in a simplified tectonic sketch (Fig. 12). 1. Regardless of lithology and peak metamorphic conditions the time of metamorphism recorded in the Odenwald-Spessart basement samples ranges between 334 and 321 Ma, with most U-Pb age data clustering around 330 Ma. 2. Emplacement of the magmatic protoliths of the amphibolites occurred at ca. 460, 445 and 329 Ma. 3. The 460 Ma old magmatic protolith was most likely emplaced in a marginal basin, which is thought to be the early Rheic Ocean. Subsequent opening of this ocean basin and its final closure towards the end of the Variscan orogeny led to crustal accretion and amphibolitefacies metamorphism at ca. 330 Ma. Thus, the zircon dating provides evidence that the ocean basin existed for at least 130 million years. This timespan would have been sufficient for the Rheic Ocean to reach the size of a mature ocean basin. 4. A segment of the Rheic suture is present in the Odenwald-Spessart basement and its position should correspond to the combined Otzberg-Michelbach Fault Zone. This fault zone separates highpressure eclogite in the eastern Odenwald from high-temperature, low-pressure granulite-facies paragneiss in the western Odenwald basement. The eclogite formed in a fossil subduction zone and the coeval low-pressure rocks in the upper plate are the remains of a

deeply eroded magmatic arc. The U-Pb zircon, rutile and monazite data leave no doubt that these radically different metamorphic conditions were reached contemporaneously at ca. 330 Ma, which strongly supports the earlier hypothesis of the Odenwald-Spessart basement as a remnant of a paired metamorphic belt. 5. The Mid-German Crystalline Zone is a tectonic stack of metamorphic rocks, which protoliths formed in different tectonic settings and different palaeogeographic locations at different times. The protoliths of mafic meta-igneous rocks formed from juvenile as well as depleted mantle source regions at mid-ocean ridge, intraoceanic island and active continental margin settings during the Palaeozoic. Metamorphism and juxtaposition of different crustal blocks occurred towards the end of the Carboniferous during the final assembly of the Variscan orogen. 6. High-pressure eclogite-facies lower plate rocks of the OdenwaldSpessart basement experienced a clockwise PT path and are tectonically separated from upper plate high-temperature granulite-facies rocks that followed a counter-clockwise PT trajectory. Metamorphism occurred contemporaneously in both units. As a consequence the Odenwald basement is interpreted as a paired metamorphic belt. This is the first recognition of a paired metamorphic belt in the European Variscides and one of the few in the northern hemisphere. Supplementary data to this article can be found online at https://doi. org/10.1016/j.lithos.2018.01.008. Acknowledgements E.S and T.M.W. thank the Chinese Academy of Sciences for financial and technical support during the course of this study. H. Brätz (GeoZentrum Nordbayern, University of Erlangen, Germany) is thanked for performing the LA-ICP-MS zircon and rutile trace element analyses. This work was supported by Chinese State Key Research and Development Program (2016YFE0203000). We thank J. Wang, H. Ma, Q. Yuan, Y. Liu, G. Tang, J. Li and Y.-H. Yang for their technical support during the course of this study. K. Mezger (Universität Bern, Switzerland) is thanked for his comments regarding the interpretation of the Scherer et al. (2002) Lu/Hf age of sample EH-10. Two unknown reviewers are thanked for their very constructive comments and criticisms. Appendix A. Analytical methods Zircon, rutile and monazite were separated by standard density and magnetic separation techniques and hand-picking under a binocular microscope. The minerals, together with zircon standards (Plešovice, Qinghu and Penglai), rutile standards (DXK and JDX) and monazite standards (RW-1 and 44,069), respectively, were mounted in epoxy mounts that were then polished to expose the crystal interiors for analysis. Transmitted and reflected light as well as cathodoluminescence (CL) imaging was used to document the internal structures of the minerals, including mineral inclusions. The mount was vacuumcoated with high purity gold to reach b20 Ω resistance prior to SIMS analysis. A.1. Secondary ion mass spectrometry (SIMS) zircon, rutile and monazite age dating Measurements of zircon, rutile, and monazite U-Th-Pb isotopes were conducted using a Cameca IMS-1280HR SIMS at the Institute of Geology and Geophysics, Chinese Academy of Sciences (IGG-CAS). Zircon was analysed using the operating and data processing procedures described by Li et al. (2009). The primary O− 2 ion beam has an intensity of ca. 8 nA and a spot size of about 20 × 30 μm. In the secondary ion beam optics, a mass resolution of ca. 7000 (at 50% peak height) was used to separate Pb+ peaks from isobaric interferences. An electron multiplier was used to measure secondary ion beam


intensities by peak jumping mode. Analyses of the Plešovice zircon standard (206Pb/238U age = 337 Ma; Sláma et al., 2008) were interspersed with unknown grains. In order to monitor the uncertainties related to the Plešovice standard, a second zircon standard Qinghu (X.H. Li et al., 2013) was alternately analysed as an unknown together with the zircon grains. A long-term uncertainty of 1.5% (1σ RSD) for 206 Pb/238U measurements of the standard zircons was propagated through the unknowns (Q.L. Li et al., 2010). The rutile analytical procedures were described in detail by Li et al. (2011) and are only briefly summarised here. An O− 2 primary ion beam was used with an intensity of ~15 nA and an ellipsoidal spot size of ca. 20 × 30 μm. Pb/U ratios were calibrated against a DXK rutile standard (1782.6 ± 2.8 Ma; Li et al., 2013b) and were monitored using a JDX rutile standard (518 ± 4 Ma; Li et al., 2013b). The 207Pb-based common Pb correction method is used. For more details see Li et al. (2011). The monazite analytical procedures during this study follow Li et al. (2013a). U-Pb ratios and absolute abundances were determined relative to the monazite standard RW-1 (Ling et al., 2017), analyses of which were interspersed with those of the unknown grains. The O− 2 primary ion beam was accelerated at 13 kV, with an intensity of 1.5 nA and an analytical spot size of 10 × 15 μm. The 207Pb-based common Pb correction method is used. Data reduction was carried out using the Isoplot/Ex v. 2.49 program (Ludwig, 2001). SIMS zircon, rutile, and monazite U–Pb data are presented in the electronic Supplement Tables S1, S2, and S3, respectively. Uncertainties of individual analyses are reported at a 1σ level and, except where noted otherwise, the calculated Concordia or intercept ages are quoted at the 95% confidence level, including uncertainties in calibration against the mineral standards. A.2. SIMS zircon oxygen isotopic analysis Zircon oxygen isotopic compositions were measured by using a Cameca IMS-1280 SIMS at IGG-CAS, following standard procedures (Tang et al., 2015; X.H. Li et al., 2010). The focused Cs+ primary ion beam has an intensity of 2 nA, and the analytical spot size is typically 20 μm in diameter. Oxygen isotopes were measured in multi-collector mode using two off-axis Faraday cups. Uncertainties on single analyses are usually around 0.3‰ (2σ SE). The instrumental mass fractionation factor (IMF) was corrected using the zircon standard Penglai with a δ18O value of 5.3‰ (X.H. Li et al., 2010). A second zircon standard Qinghu was analysed as an unknown to ascertain the veracity of the IMF. Twenty measurements of Qinghu zircon standard during the course of this study yielded a weighted mean of δ18O = 5.4 ± 0.4‰ (2σ SD), the zircon oxygen isotopic data are given together with the U-Pb data in electronic Supplement Table S1. A.3. Multi-collector ICP-MS zircon Hf analysis In-situ zircon Hf isotopic analysis was carried out using a Neptune multi-collector ICP-MS, equipped with a 193 nm laser at IGG-CAS; the analytical procedures were similar to those described by Wu et al. (2006). Where possible, the zircon Lu–Hf isotope composition was acquired at the same spot as in the SIMS U-Pb dating, with a spot size of 65 μm and an 8 Hz pulse frequency. Analyses were carried out with beam diameters of 44 μm, 10 Hz repetition rates, laser beam energy density of 10 J/cm2 and 26 s of ablation time. Measured 176Hf/177Hf ratios were normalised to 179Hf/177Hf = 0.7325. Zircon standards Penglai and Mud Tank were analysed alternately with the unknowns to evaluate the analytical accuracy and precision. During the course of this study, we obtained 176Hf/177Hf ratio of 0.282909 ± 0.000032 (2σ SD, n = 15) for Penglai zircon and 0.282494 ± 0.000024 (2σ SD, n = 7) for Mud Tank zircon, in good agreement within the errors and the recommended value of X.H. Li et al. (2010) and Woodhead and Hergt (2005). Zircon Lu-Hf isotopic data are given in electronic Supplement Table S4.

295

Initial epsilon Hafnium (εHfi) values of zircon grains can provide information on the crust from which they were derived, with positive Hf initials generally pointing to a juvenile (mantle-derived) source, whereas negative initials are commonly interpreted to indicate derivation from, or mixing with, pre-existing continental crust (e.g., Vervoort, 2015). Hf (and Nd) model ages provide information on the time elapsed since melt separated from a depleted mantle reservoir (e.g., Patchett, 1992; Hawkesworth et al., 2017) and can be regarded as the minimum age of reworked crust and/or of the source of the parental magma. It should be noted however that model ages do not reflect ‘true’ protolith ages but may place some constraints on the timing and nature of crustforming events (Vervoort, 2015). A.4. Trace element analysis Trace element concentrations in zircon and rutile were determined at the GeoZentrum Nordbayern (University of Erlangen, Germany) by LA-ICP-MS using a 193 nm UP193-FX Excimer laser and an Agilent 7500c mass spectrometer. Single spot zircon and rutile analyses were performed by ablating sample spots of 15 to 35 μm in diameter at 17 Hz with an energy density of 2.99 and 2.85 J/cm2 per pulse, respectively. Signal integration times were 20 s for the Ar background and 23 and 25 s for the zircon and rutile ablation intervals. The NIST SRM 612 50 ppm glass standard (Pearce et al., 1997) was used for external calibration. Data reduction was performed using ideal Si and Ti concentrations for zircon and rutile as internal standards. The software program GLITTER v. 4.4.4 (van Achterbergh et al., 2000) was used for data reduction. Zircon and rutile trace element data and apparent zircon and rutile crystallisation temperatures are presented in electronic Supplement Tables S5 and S6. The zircon and rutile crystallisation temperatures were calculated using the Titanium-in-zircon and Zirconium-in-rutile thermometers of Watson et al. (2006) without the pressured correction suggested by Fu et al. (2008) and assuming unit activities of TiO2 and SiO2 (Ferry and Watson, 2007). Uncertainties in temperatures are estimated to be smaller than ~10 °C for the temperatures range obtained in this study (Watson et al., 2006). References van Achterbergh, E., Ryan, C.G., Griffin, W.L., 2000. GLITTER Version 4.4.4. On-line Interactive Data Reduction for LA-ICPMS. Macquarie Research Ltd. Anthes, G., Reischmann, T., 2001. Timing of granitoid magmatism in the eastern MidGerman crystalline rise. Journal of Geodynamics 31, 119–143. Behr, H.J., Heinrichs, T., 1987. Geological interpretation of DEKORP 2-S: a deep seismic reflection profile across the Saxothuringian and possible implications for the late Variscan structural evolution of Central Europe. Tectonophysics 142, 173–202. Bolhar, R., Ring, U., Ireland, T.R., 2016. Zircon in amphibolites from Naxos, Aegean Sea, Greece: origin, significance and tectonic setting. Journal of Metamorphic Geology 35, 413–434. Brown, M., 2010. Paired metamorphic belts revisited. Gondwana Research 18, 46–59. Catalán, J.R.M., Arenas, R., Abati, J., et al., 2009. A rootless suture and the loss of the roots of a mountain chain: the Variscan belt of NW Iberia. Comptes Rendus Geoscience 341, 114–126. Chatterjee, N.D., 1960. Geologische Untersuchungen im Kristallin des Böllsteiner Odenwaldes. Neues Jahrbuch für Geologie und Paläontologie Abhandlungen 37, 223–256. Cherniak, D.J., 2000. Pb diffusion in rutile. Contributions to Mineralogy and Petrology 139, 198–207. Cherniak, D.J., Watson, E.B., 2000. Pb diffusion in zircon. Chemical Geology 172, 5–24. Cherniak, D.J., Watson, E.B., Grove, M., Harrison, T.M., 2004. Pb diffusion in monazite: a combined RBS/SIMS study. Geochimica et Cosmochimica Acta 68, 829–840. Cocks, L.R.M., Torsvik, T.H., 2006. European geography in a global context from the Vendian to the end of the palaeozoic. In: Gee, D.G., Stephenson, R.A. (Eds.), European Lithosphere Dynamics. Geological Society of London Memoirs 32, pp. 83–95. Copeland, P., Parrish, R.R., Harrison, T.M., 1988. Identification of inherited radiogenic Pb in monazite and its implications for U-Pb systematics. Nature 333, 760–763. Corfu, F., Hanchar, M., Hoskin, P.W.O., Kinny, P., 2003. Atlas of zircon textures. In: Hanchar, J.M., Hoskin, P.W.O. (Eds.), Zircon. Reviews in Mineralogy and Geochemistry 53, pp. 469–500. Dombrowski, A., Henjes-Kunst, F., Höhndorf, A., Kröner, A., Okrusch, M., Richter, P., 1995. Orthogneisses in the Spessart Crystalline Complex, Northwest Bavaria: witnesses of Silurian granitoid magmatism at an active continental margin. Geologische Rundschau 84, 399–411.

296


Dörr, W., Zulauf, G., Gerdes, A., Loeckle, F., 2017. Provenance of Upper Devonian clastic (meta) sediments of the Böllstein Odenwald (Mid-German-Crystalline-Zone, Variscides). International Journal of Earth Sciences 106, 2927–2943. Eckelmann, K., Nesbor, H.-D., Königshof, P., Linnemann, U., Hofmann, M., Lange, J.M., Sagawe, A., 2014. Plate interactions of Laurussia and Gondwana during the formation of Pangaea — constraints from U–Pb LA–SF–ICP–MS detrital zircon ages of Devonian and Early Carboniferous siliciclastics of the Rhenohercynian zone, Central European Variscides. Gondwana Research 25, 1484–1500. Ferry, J.M., Watson, E.B., 2007. New thermodynamic models and revised calibrations for the Ti-in-zircon and Zr-in-rutile thermometers. Contributions to Mineralogy and Petrology 154, 429–437. Franke, W., Dulce, J.-C., 2017. Back to sender: tectonic accretion and recycling of Balticaderived clastic sediments in the Rheno-Hercynian Variscides. International Journal of Earth Sciences 106, 377–386. Franke, W., Cocks, L.R.M., Torsvik, T.H., 2017. The Paleozoic Variscan oceans revisited. Gondwana Research 48, 257–284. Fu, B., Page, F.Z., Cavosie, A.J., Fournelle, J., Kita, N.T., Lackey, J.S., Wilde, S.A., Valley, J.W., 2008. Ti-in-zircon thermometry: applications and limitations. Contributions to Mineralogy and Petrology 156, 197–215. Geisler, T., Schaltegger, U., Tomaschek, F., 2007. Re-equilibration of zircon in aqueous fluids and melts. Elements 3, 43–50. Goldstein, S.L., O'Nions, R.K., Hamilton, P.J., 1984. A Sm-Nd study of atmospheric dusts and particulates from major river systems. Earth and Planetary Science Letters 70, 221–236. Grimes, C.B., John, B.E., Kelemen, P.B., Mazdab, F.K., Wooden, J.L., Cheadle, M.J., Hanghøj, K., Schwartz, J.J., 2007. Trace element chemistry of zircon from oceanic crust: a method for distinguishing detrital zircon provenance. Geology 35, 643–646. Grimes, C.B., Wooden, J.L., Cheadle, M.J., John, B.E., 2015. “Fingerprinting” tectonomagmatic provenance using trace elements in igneous zircon. Contributions to Mineralogy and Petrology 170, 46. Hawkesworth, C.J., Cawood, P.A., Dhuime, B., Kemp, T.I.S., 2017. Earth's continental lithosphere through time. Annual Review of Earth and Planetary Sciences 45, 169–198. Heaman, L.M., Parrish, R.R., 1991. U-Pb geochronology on accessory minerals. In: Heaman, L.M., Ludden, J.N. (Eds.), Applications of Radiogenic Isotope Systems to Problems in Geology. Mineralogical Association of Canada, pp. 59–102. Henderson, B.J., Collins, W.J., Murphy, J.B., Guiterrez-Alonso, G., Hand, M., 2016. Gondwanan basement terranes of the Variscan-Appalachian orogen: Baltican, Saharan and West African isotopic fingerprints in Avalonia, Iberia and the Armorican Terranes. Tectonophysics 681, 278–304. Hoskin, P.W.O., Black, L.P., 2000. Metamorphic zircon formation by solid-state recrystallization of protolith igneous zircon. Journal of Metamorphic Geology 18, 423–439. Hoskin, P.W.O., Schaltegger, U., 2003. The composition of zircon and igneous and metamorphic petrogenesis. In: Hanchar, J.M., Hoskin, P.W.O. (Eds.), Zircon. Reviews in Mineralogy and Geochemistry 53, pp. 27–62. Kirsch, H., Kober, B., Lippolt, H.J., 1988. Age of intrusion and rapid cooling of the Frankenstein gabbro (Odenwald, SW Germany) evidenced by 40Ar/39Ar and single zircon 207 Pb/206Pb measurements. Geologische Rundschau 77, 693–711. Knauer, E., Okrusch, M., Richter, P., Schmidt, K., Schubert, W., 1974. Die metamorphe Basit-Ultrabasit-Assoziation in der Böllsteiner Gneiskuppel, Odenwald. Neues Jahrbuch für Mineralogie Abhandlungen 122, 186–228. Kooijman, E., Mezger, K., Berndt, J., 2010. Constraints on the U–Pb systematics of metamorphic rutile from in-situ LA-ICP-MS analysis. Earth and Planetary Science Letters 293, 321–330. Krohe, A., 1992. Structural evolution of intermediate-crustal rocks in a strike-slip and extensional setting (Variscan Odenwald, SW Germany): differential upward transport of metamorphic complexes and changing deformation mechanisms. Tectonophysics 206, 357–386. Krohe, A., 1996. Variscan tectonics of central Europe: postaccretionary intraplate deformation of weak continental lithosphere. Tectonics 15, 1364–1388. Kroner, U., Romer, R., 2013. Two plates - many subduction zones: the Variscan orogeny reconsidered. Gondwana Research 24, 298–329. Kröner, A., Wilde, S.A., Zhao, G.C., O'Brien, P.J., Sun, M., Liu, D.Y., Wan, Y.S., Liu, S.W., Guo, J.H., 2006. Zircon geochronology and metamorphic evolution of mafic dykes in the Hengshan Complex of northern China: evidence for late Palaeoproterozoic extension and subsequent high-pressure metamorphism in the North China Craton. Precambrian Research 146, 45–67. Kylander-Clark, A.R.C., Hacker, B.R., Mattinson, J.M., 2008. Slow exhumation of UHP terranes: titanite and rutile ages of the Western Gneiss Region, Norway. Earth and Planetary Science Letters 272, 531–540. Li, X.H., Liu, Y., Li, Q.L., Guo, C.H., Chamberlain, K.R., 2009. Precise determination of Phanerozoic zircon Pb/Pb age by multi-collector SIMS without external standardization. Geochemistry, Geophysics, Geosystems 10, Q04010. https://doi.org/10.1029/ 2009GC002400. Li, Q.L., Li, X.H., Liu, Y., Tang, G.Q., Yang, J.H., Zhu, W.G., 2010a. Precise U-Pb and Pb-Pb dating of Phanerozoic baddeleyite by SIMS with oxygen flooding technique. Journal of Analytical Atomic Spectrometry 25, 1107–1113. Li, X.H., Long, W.G., Li, Q.L., Liu, Y., Zheng, Y.F., Yang, Y.H., Chamberlain, K.R., Wan, D.F., Guo, C.H., Wang, X.C., Tao, H., 2010b. Penglai zircon megacryst: a potential new working reference for microbeam analysis of Hf-O isotopes and U-Pb age. Geostandards and Geoanalytical Research 34, 117–134. Li, Q.L., Lin, W., Su, W., Li, X.H., Shi, Y.H., Liu, Y., Tang, G.Q., 2011. SIMS U–Pb rutile age of low-temperature eclogites from southwestern Chinese Tianshan, NW China. Lithos 122, 76–86. Li, Q.L., Li, X.H., Lan, Z., Guo, C., Yang, Y., Liu, Y., Tang, G., 2013a. Monazite and xenotime U– Th–Pb geochronology by ion microprobe: dating highly fractionated granites at

Xihuashan tungsten mine, SE China. Contributions to Mineralogy and Petrology 166, 65–80. Li, Q.L., Yang, Y.N., Shi, Y.H., et al., 2013b. Eclogite rutile U-Pb dating: constraint for formation and evolution of continental collisional orogen. Chinese Science Bulletin 58, 2279–2284. Li, X.H., Tang, G.Q., Gong, B., Yang, Y.H., Hou, K.J., Hu, Z.C., Li, Q.L., Liu, Y., Li, W.X., 2013c. Qinghu zircon: a working reference for microbeam analysis of U-Pb age and Hf and O isotopes. Chinese Science Bulletin 58, 4647–4654. Ling, X.X., Huyskens, M.H., Li, Q.L., Yin, Q.Z., Werner, R., Liu, Y., Tang, G.Q., Yang, Y.N., Li, X.H., 2017. Monazite RW-1: a homogeneous natural reference material for SIMS UPb and Th-Pb isotopic analysis. Mineralogy and Petrology 111, 163–172. Linnemann, U., McNaughton, N.J., Romer, R.L., Gehmlich, M., Drost, K., Tonk, C., 2004. West African provenance for Saxo–Thuringia (Bohemian Massif): did Armorica ever leave pre-Pangean Gondwana? — U/Pb-SHRIMP zircon evidence and the Ndisotopic record. International Journal of Earth Sciences 93, 683–705. Linnemann, U., Gerdes, A., Drost, K., Buschmann, B., 2007. The continuum between Cadomian orogenesis and opening of the Rheic Ocean: constraints from LA-ICP-MS U-Pb zircon dating and the analysis of plate-tectonic setting (Saxo-Thuringian zone, NE Bohemian Massif, Germany). In: Linnemann, U., Nance, R.D., Kraft, P., Zulauf, G. (Eds.), The Evolution of the Rheic Ocean: From Avalonian-Cadomian Active Margin to Alleghenian-Variscan Collision. Geological Society of America Special Paper 423, pp. 61–96. Lippolt, H.J., 1986. Nachweis altpaläozoischer Primäralter (Rb-Sr) and karbonischer Abkühlalter (K-Ar) der Muscovit-Biotit-Gneise des Spessarts und der Biotit-Gneise des Böllsteiner Odenwaldes. Geologische Rundschau 75, 569–583. Ludwig, K.R., 2001. Users Manual for Isoplot/Ex Rev. 2.49. Berkeley Geochronology Center Special Publication 1a (54 pp.). Miyashiro, A., 1961. Evolution of metamorphic belts. Journal of Petrology 2, 277–311. Murphy, J.B., Keppie, J.D., Nance, R.D., Dostal, J., 2010. Comparative evolution of the Iapetus and Rheic Oceans: a North American perspective. Gondwana Research 17, 482–499. Nance, R.D., Linnemann, U., 2008. The Rheic Ocean: origin, evolution, and significance. GSA Today 18, 4–12. Nasir, S., Okrusch, M., Kreuzer, H., Lenz, H., Höhndorf, A., 1991. Geochronology of the Spessart Crystalline Complex, Mid-German crystalline rise. Mineralogy and Petrology 44, 39–55. Nickel, E., 1975. Geologische Position und Petrogenese des kristallinen Odenwaldes. Aufschluss Sonderband 27, 1–25. Okrusch, M., Schubert, W., Stähle, V., 2000. The Odenwald, Germany: Variscan metamorphic evolution and igneous events. Beiheft zum European Journal of Mineralogy 12, 45–89. Okrusch, M., Geyer, G., Lorenz, J., 2011. Spessart. Geologische Entwicklung und Struktur, Gesteine und Minerale. Sammlung geologischer Führer Band 106. Gebrüder Bornträger, Stuttgart (368 pp.). Oncken, O., 1997. Evidence for precollisional subduction erosion in ancient collisional belts: the case of the Mid-European Variscides. Geology 26, 1075–1078. Oncken, O., 2000. Transformation of a magmatic arc and an orogenic root during oblique collision and its consequences for the evolution of the European Variscides (Mid-German Crystalline Rise). Geologische Rundschau 86, 2–20. Patchett, P.J., 1992. Isotopic studies of Proterozoic crustal growth and evolution. In: Condie, K.C. (Ed.), Proterozoic crustal evolution. Developments in Precambrian Geology vol. 10, pp. 481–508. Pearce, N.J.G., Perkins, W.T., Westgate, J.A., Gorton, M.P., Jackson, S.E., Neal, C.R., Chenery, S.P., 1997. A compilation of new and published major and trace element data for NIST SRM 610 and NIST SRM 612 glass reference materials. Geostandards Newsletter 21, 115–144. von Raumer, J.F., 1973. Die mineralfazielle Stellung der Metapelite und Metagrauwacken zwischen Heppenheim und Reichelsheim (Odenwald). Neues Jahrbuch für Mineralogie Abhandlungen 118, 313–336. von Raumer, J.F., Stampfli, G.M., 2008. The birth of the Rheic Ocean — early Palaeozoic subsidence patterns and subsequent tectonic plate scenarios. Tectonophysics 461, 9–20. von Raumer, J.F., Stampfli, G.M., Arenas, R., Sánchez Martínez, S., 2015. Ediacaran to Cambrian oceanic rocks of the Gondwana margin and their tectonic interpretation. International Journal of Earth Sciences 104, 1107–1121. Reischmann, T., Anthes, G., Jaeckel, P., Altenberger, U., 2001. Age and origin of the Böllsteiner Odenwald. Mineralogy and Petrology 72, 29–44. Sagawe, A., Gärtner, A., Linnemann, U., Hofmann, M., Gerdes, A., 2016. Exotic crustal components at the northern margin of the Bohemian Massif - implications from U-Th-Pb and Hf isotopes of zircon from the Saxonian Granulite Massif. Tectonophysics 681, 234–249. Scherer, E.E., Mezger, K., Münker, C., 2002. Lu-Hf ages of high-pressure metamorphism in the Variscan fold belt of southern Germany. Geochimica et Cosmochimica Acta. Supplement. 66, A677. Schubert, W., Lippolt, H.J., Schwarz, W., 2001. Early to Middle Carboniferous hornblende 40 Ar/39Ar ages of amphibolites and gabbros from the Bergsträsser Odenwald. Mineralogy and Petrology 72, 113–132. Siebel, W., Eroglu, S., Shang, C.K., Rohrmüller, J., 2012. Zircon geochronology, elemental and Sr-Nd isotope geochemistry of two Variscan granitoids from the OdenwaldSpessart Crystalline Complex (Mid-German crystalline rise). Mineralogy and Petrology 105, 187–200. Sláma, J., Košler, J., Condon, D.J., et al., 2008. Plešovice zircon - a new natural reference material for U–Pb and Hf isotopic microanalysis. Chemical Geology 249, 1–35. Stampfli, G.M., Hochard, C., Vérard, C., Wilhem, C., von Raumer, J., 2013. The formation of Pangea. Tectonophysics 593, 1–19. Stein, E., 2001. The geology of the Odenwald Crystalline Complex. Mineralogy and Petrology 72, 7–28.

T.M. Will et al. / Lithos 302–303 (2018) 278–297 Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean Basins. Geological Society of London Special Publication 42, pp. 313–345. Tang, G.Q., Li, X.H., Li, Q.L., Liu, Y., Ling, X.X., Yin, Q.Z., 2015. Deciphering the physical mechanism of the topography effect for oxygen isotope measurements using a Cameca IMS-1280 SIMS. Journal of Analytical Atomic Spectrometry 30, 950–956. Taylor, R.J.M., Kirkland, C.L., Clark, C., 2016. Accessories after the facts: constraining the timing, duration and conditions of high-temperature metamorphic processes. Lithos 264, 239–257. Todt, W.A., Altenberger, U., von Raumer, J.F., 1995. U-Pb data on zircons for the thermal peak of metamorphism in the Variscan Odenwald, Germany. Geologische Rundschau 84, 466–472. Valley, J.W., Kinny, P.D., Schulze, D.J., Spicuzza, M.J., 1998. Zircon megacrysts from kimberlite: oxygen isotope variability among mantle melts. Contributions to Mineralogy and Petrology 133, 1–11. Vervoort, J., 2015. Lu-Hf dating: the Lu-Hf isotope system. In: Rink, W.J., Thompson, J.W. (Eds.), Encyclopedia of Scientific Dating Methods. Springer Science and Business Media, Dordrecht, pp. 379–390. Vry, J.K., Baker, J.A., 2006. LA-MC-ICPMS Pb–Pb dating of rutile from slowly cooled granulites: confirmation of the high closure temperature for Pb diffusion in rutile. Geochimica et Cosmochimica Acta 70, 1807–1820. Watson, E.B., Wark, D.A., Thomas, J.B., 2006. Crystallization thermometers for zircon and rutile. Contributions to Mineralogy and Petrology 116, 413–433. Weber, K., 1995. Structure of the Spessart Crystalline Complex. In: Dallmeyer, R.D., Franke, W., Weber, K. (Eds.), Pre-Permian Geology of Central and Eastern Europe. Springer, Berlin-Heidelberg-New York, pp. 167–173. Whitehouse, M.J., Platt, J.P., 2003. Dating high-grade metamorphism-constraints from rare-earth elements in zircon and garnet. Contributions to Mineralogy and Petrology 145, 61–74.

297

Will, T.M., 2001. Paleostress-tensor analysis of late deformation events in the Odenwald Crystalline Complex and comparison with other units of the Mid-German Crystalline Rise, Germany. Mineralogy and Petrology 72, 229–247. Will, T.M., Frimmel, H.E., 2018. Where does a continent prefer to break up? Some lessons from the South Atlantic margins. Gondwana Research 53, 9–19. Will, T.M., Schmädicke, E., 2001. A first report of retrogressed eclogites in the Odenwald Crystalline Complex: evidence for high-pressure metamorphism in the MidGerman Crystalline Rise, Germany. Lithos 59, 109–125. Will, T.M., Schmädicke, E., 2003. Isobaric cooling and anti-clockwise P-T paths in the Variscan Odenwald Crystalline Complex. Journal of Metamorphic Geology 21, 469–480. Will, T.M., Lee, S.-H., Schmädicke, E., Frimmel, H.E., Okrusch, M., 2015. Variscan terrane boundaries in the Odenwald-Spessart basement, Mid-German Crystalline one: new evidence from ocean ridge, intraplate and arc-derived metabasaltic rocks. Lithos 220-223, 23–42. Will, T.M., Schulz, B., Schmädicke, E., 2017. The timing of metamorphism in the Odenwald-Spessart basement, Mid-German Crystalline Zone. International Journal of Earth Sciences 106, 1631–1649. Woodhead, J.D., Hergt, J.M., 2005. Preliminary appraisal of seven natural zircon reference materials for in-situ Hf isotope determination. Geostandards and Geoanalytical Research 29, 183–195. Wu, F.Y., Yang, Y.H., Xie, L.W., Yang, J.H., Xu, P., 2006. Hf isotopic compositions of the standard zircons in U–Pb geochronology. Chemical Geology 234, 105–126. Zeh, A., Gerdes, A., 2010. Baltica- and Gondwana-derived sediments in the Mid-German Crystalline Rise (Central Europe): implications for the closure of the Rheic ocean. Gondwana Research 17, 254–263. Zeh, A., Will, T.M., 2010. The Mid-German Crystalline Rise. In: Linnemann, U., Romer, R.L. (Eds.), Pre-Mesozoic Geology of Saxo-Thuringia–From the Cadomian Active Margin to the Variscan Orogen. Schweizerbart, Stuttgart, pp. 195–220.