1. Introduction
The Variscides in Europe can be followed from the Iberian Peninsula in the west to the Moravo-Silesian unit in the east. Their south-eastward continuation along the margin of the East European Craton is obscured by the overthrusting of the Carpathian orogen (Fig. 1a).
Figure 1. (a) Simplified sketch map of the Variscan belt after Franke et al. (Reference Franke, Dallmeyer, Weber, Dallmeyer, Franke and Weber1995); (b) simplified geological map of the Carpathian Mountains within the Alpine orogenic belt with (c) the Bucovinian, Sub-Bucovinian and Infrabucovinian nappes shown (compilation after Kräutner, Reference Kräutner and Zoubek1988; Vodă, Reference Vodă2002; Matskiv et al. Reference Matskiv, Pukach, Vorobkanych, Pastukhanova and Hnylko2009).
The Carpathian Mountains belt in central Europe extends for more than 1300 km and is composed of the internides, deformed mainly in the Mesozoic and containing crystalline cores of Pre-Alpine (mostly Variscan) basement, and the externides that were deformed during the Cenozoic (Golonka et al. Reference Golonka, Gahagan, Krobicki, Marko, Oszczypko, Ślączka, Golonka and Picha2006; Schmid et al. Reference Schmid, Bernoulli, Fügenschuh, Matenco, Scheffer, Schuster, Tischler and Ustaszewski2008; Golonka et al. Reference Golonka, Gawęda, Waśkowska, Alderon and Elias SA2021 and references therein). The border between them is marked by the Pieniny Klippen Belt (Fig. 1b). The Outer Carpathians consist of a complex nappe system that verges towards the outer parts of the arcuate orogen. The nappe complex formed during Cretaceous to Neogene times due to the collision of several microplates with the European Plate. Sedimentary rocks within the nappes were deposited on the polyphase-deformed pre-Alpine basement representing the attenuated crust of the North European Platform (Ceahlau-Severin Ocean, e.g. Schmid et al. Reference Schmid, Bernoulli, Fügenschuh, Matenco, Scheffer, Schuster, Tischler and Ustaszewski2008; Matenco et al. Reference Matenco, Krezsek, Merten, Schmid, Cloetingh and Andriessen2010). In the Outer Western Carpathians, the nature of this basement is known only from olistoliths and exotic pebbles (Poprawa et al. Reference Poprawa, Malata, Pécskay, Kusiak, Banaś and Paszkowski2006; Gawęda et al. Reference Gawęda, Golonka, Waśkowska, Szopa, Chew, Starzec and Wieczorek2019 and references therein). However, in the Eastern Carpathians, the crystalline basement underlying the Outer Eastern Carpathians sedimentary basins is exposed. Termed the Marmarosh Massif in Ukraine and the Maramuresh Massif in Romania (Fig. 1c), it is composed of Pre-Alpine rocks and is the largest preserved fragment of the Marmarosh Ridge, a ridge of basement that was thrust NE over the Outer Carpathian flysch during Alpine collision (Tkatchuk & Gurżij, Reference Tkatchuk and Gurżij1957; Kräutner, Reference Kräutner and Zoubek1988). The rocks of the Marmarosh/Maramuresh Massif and Marmarosh Klippen Zone just represent part of the Pre-Alpine North European Platform. Reconstructing their geological history not only aids our understanding of the Outer Eastern Carpathians basement but also the geotectonic development of the SW margin of Baltica. The Marmarosh Ridge is believed to represent the eastern prolongation of the Silesian Ridge of the Outer Western Carpathians. The Silesian Ridge separated the Silesian and Magura basins during the late Mesozoic and Paleogene and is inferred to be the source of the coarse detritus deposited in both basins (Săndulescu, Reference Săndulescu, Royden and Horvath F1988; Oszczypko et al. Reference Oszczypko, Oszczypko-Clowes, Golonka and Krobicki2005; Golonka et al. Reference Golonka, Gawęda, Waśkowska, Alderon and Elias SA2021; Bónová et al. Reference Bónová, Bóna, Pańczyk, Kováčik, Mikuš and Laurinc2019).
This study aims to decipher the Pre-Alpine metamorphic history of the Marmarosh/Maramuresh massifs (the Bilyipotik Nappe in Ukraine; the Bretila Group of the Vaser Infrabucovinian Nappe in Romania), where amphibolite-facies pelitic schists with amphibolite intercalations predominate. Multi-tool petrochronology was applied, including U-Pb apatite, rutile and titanite dating for amphibolites and for interleaved metapelites U-Th-Pb CHIME dating of monazite and U-Pb apatite and rutile dating. Geochronological data are coupled with thermodynamic modelling of stable mineral assemblages, which allow us to reconstruct P-T-t paths of the metamorphic units and to refine the geodynamic evolution of the pre-Alpine basement of the Outer Eastern Carpathians.
2. Geological setting and sampling
The crystalline segment, called Marmarosh/Maramuresh Massif, is a constituent of northernmost part of the Outer Eastern Carpathians, located north-east of the Pieniny Klippen Belt (Kräutner & Bindea, Reference Kräutner and Bindea2002; Munteanu & Tatu, Reference Munteanu and Tatu2003). The Pre-Alpine terranes of the Eastern Carpathians were subjected to Aptian-Albian thrusts that generated Alpine-age nappes with metamorphic (crystalline) basements and dominantly Mesozoic sedimentary cover, which Uhlig (Reference Uhlig1907) termed the Crystalline-Mesozoic Zone (CMZ).
In the Ukrainian part of the massif two crystalline units, the Bilyipotik and Dilove nappes, are recognized (Matskiv et al. Reference Matskiv, Pukach, Vorobkanych, Pastukhanova and Hnylko2009) (Figs. 1c, 2a). The Bilyipotik Nappe is composed of amphibolite-facies metamorphic rocks of the Bilyipotik suite (1300 m) which consist of orthogneisses and mica-schists, locally garnet-bearing, intercalated with rare amphibolites and marble lenses. They are structurally overlain by weakly metamorphosed rocks of the Rozyn suite (phyllites, marbled limestones and dolomites up to 600 m thick) and predominantly unmetamorphosed volcanic-sedimentary formations (up to 900 m). The structurally higher Dilove Nappe is composed of greenschist-facies metamorphic rocks, including Neoproterozoic metapelites/meta-tuffs and felsic to intermediate metavolcanics of early Caledonian age, connected to the closure of the easternmost Tornquist Ocean on the margin of Baltica (Munteanu & Tatu, Reference Munteanu and Tatu2003; Oszczypko et al. Reference Oszczypko, Oszczypko-Clowes, Golonka and Krobicki2005; Gawęda et al. Reference Gawęda, Szopa, Golonka, Chew, Stepanyiuk, Belskyy, Waśkowska, Siliauskas and Drakou2022).
Figure 2. The tectonostratigraphy of the metamorphic terranes of the Eastern Carpathians (not to scale). Abbreviations of the metamorphic sequences in the Romanian and Ukrainian part of the Eastern Carpathians: BS – Berlebash Suite; MS – Megura Suite; RS – Rozyn Suite; BpS – Bilyi Potik Suite; Br – Bretila Group; Ng – Negrişoara Group; Rb – Rebra Group; Tg – Tulgheş Group.
In the Romanian part of the Eastern Carpathians, the following Alpine nappes with pre-Alpine basement are identified: the Bucovinian Nappe, Sub-Bucovinian Nappe and several Infrabucovinian Nappes (Săndulescu, Reference Săndulescu1984; Reference Săndulescu1994). The metamorphic basement of the Bucovinian and Sub-Bucovinian nappes is made up of the following pre-Alpine nappe stack from top to bottom (Fig. 2b):
-
• The Rarău Nappe comprised of the amphibolite-facies Bretila Group (garnet ± staurolite metapelitic schists and gneisses, orthogneisses, amphibolites).
-
• The Putna Nappe comprising the greenschist-facies Tulgheş Group (chlorite-sericite-quartz schists ± graphite ± albite ± rutile pseudomorphs on ilmenite, graphitic metacherts, metarhyolites, felsic metatuffs, metagraywackes, rare limestones and metamorphosed mafic rocks).
-
• The Pietrosu Bistriţei Nappe consisting of the Pietrosu Bistriţei metarhyodacite and the Negrişoara Group (intensely deformed micaschists and gneisses, with amphibolite and carbonate intercalations; Balintoni & Balica, Reference Balintoni and Balica2013).
-
• The Rodna Nappe, consists of the amphibolite-facies Rebra Group, which is comprised predominantly of micaschists and pelitic gneisses with amphibolites, dolomites and limestones (± tremolite ± wollastonite ± diopside, quartzite ± graphite).
The Infrabucovinian Nappe is exposed only in tectonic windows and is composed of metamorphic basement attributed to the Bretila Group and Mesozoic ± Permian sedimentary cover. In this research, we investigated rocks of the Infrabucovinian Bretila Group from the Vaser and Bilyipotik tectonic windows. We sampled garnet-staurolite-mica schists from the Bilyipotik Nappe (SR20Ca and SR20Cb) and two interleaving amphibolites (SR21 and SR22), near a tourist path following the Bilyi Potik stream (Figs. 1c, 3a; Table 1) in the Ukrainian part of the Marmarosh Massif. From the Romanian part of the Maramuresh Massif (Vaser window), garnet-bearing MMS-2 and MMS-3 schists were sampled with two interleaving amphibolites MMA1 (associated with MMS2) and MMA3 (associated with MMS3) outcropping in the Bretila Group, Infrabucovinian Vaser Nappe (Figs. 1c, 2b, 4a, b; Table 1).
Figure 3. Textural relationships in the Marmarosh Massif (Ukraine). (a) Field photograph showing the sampling locality of samples SR20C and SR21, (b) photomicrograph showing snowball garnet (Grt); (c) photomicrograph of staurolite (St) and post-tectonic garnet (Grt) and a foliation defined by muscovite (Ms) and biotite (Bt); (d) striped amphibolite sample SR22 with a foliation defined by plagioclase-quartz segregations; (e) Mg-hornblende (Mg-hbl) overgrowing tschemakitic hornblende (Tsch-Hbl) in sample SR22.
Table 1. GPS co-ordinates of the sampling points
Figure 4. Textural relationships in the Maramuresh Massif (Romania). (a) Field photograph of amphibolite MMA1; (b) field photograph of metapelite MMS2; (c) photomicrograph of garnet-bearing metapelite sample MMS2, (d) photomicrograph of garnet-bearing metapelite sample MMS3 with an S-C fabric; (e) photomicrograph of the amphibolite sample MMA3 with garnet porphyroblasts surrounded by Mg-hornblende. Abbreviations as in Figure 3.
3. Analytical techniques
3. a. Microscopy and electron probe micro-analyses (EPMA)
Petrographic analyses of thin sections were undertaken at the Institute of Earth Sciences in the University of Silesia using Olympus BX-51 and SX-10 microscopes to describe mineral parageneses and to constrain the textural position of the minerals studied, including monazite, apatite, titanite and rutile.
Microprobe analyses of the main rock-forming and accessory minerals were carried out at the Laboratory of Electron Microscopy, Microanalysis and X-Ray Diffraction (Faculty of Geology, University of Warsaw, Poland), using a CAMECA SX-100 electron microprobe. The analytical conditions employed an accelerating voltage of 15 kV, a beam current of 20 nA, counting times of 20–30 s for peak and 10–15 s for background and a beam diameter of 1–5 μm. Reference materials, analytical lines, diffracting crystals, mean detection limits (in wt%) and uncertainties were as follows: rutile – Ti (Kα, PET, 0.03, 0.05), diopside – Mg (Kα, TAP, 0.02, 0.11), Si – (Kα, TAP, 0.02, 0.21), Ca – (Kα, PET, 0.03, 0.16), orthoclase – Al (Kα, TAP, 0.02, 0.08), and K (Kα, PET, 0.03, 0.02), albite – Na (Kα, TAP, 0.01, 0.08), hematite – Fe (Kα, LIF, 0.09, 0.47), rhodonite – Mn (Kα, LIF, 0.03, 0.10), phlogophite – F (Kα, TAP, 0.04, 0.32), tugtupite – Cl (Kα, PET, 0.02, 0.04), Cr2O3 – Cr (Kα, PET, 0.04, 0.01), ZirconED2 – Zr (Lα, PET, 0.01, 0.01), Nb2O3-MAC – Nb (Lα, PET, 0.09, 0.01), V2O5 – V (Kα, LIF, 0.02, 0.01), YPO4 – Y (Lα, TAP, 0.05, 0.05), CeP5O14 – Ce (Lα, LPET, 0.09, 0.02), NdGaO3 – Nd (Lα, LIF, 0.31, 0.24), ThO2 – Th (Mα, LPET, 0.09, 0.09) and UO2 – U (Mα, LPET, 0.16, 0.13).
3. b. Whole-rock chemical analyses
Whole-rock analyses were undertaken by X-ray fluorescence (XRF) at Bureau Veritas Minerals (Canada). Preparation involved lithium borate fusion and dilute digestions, while LOI (loss on ignition) was determined at 1000 °C. Uncertainties for most of elements are 0.01%, except for SiO2 which is 0.1%.
3. c. U-Th-Pb monazite dating
Due to the abundance of small monazite crystals (20 – 40 μm in size) in the metapelitic samples, we undertook in situ U-Th-Pb monazite dating by electron probe micro-analyses (EPMA) to define the age of metamorphism. Monazite chemical dating was undertaken on a CAMECA SXFiveFE electron microprobe equipped with a Schottky emitter at the Laboratory of Electron Microscopy, Microanalysis and X-Ray Diffraction (Faculty of Geology, University of Warsaw, Poland) at 15 kV with an unregulated beam current of 100 nA. Beam current stability within a one hour window was within ±0.5%, except for the last run (for sample MMS2) where beam drift exceeded 1% for 12 min). No statistically significant biases due to beam drift were found.
The analytical setup for monazite dating employs simultaneous measurement of Pb (120 s peak/ 120 s background) on the second WDS spectrometer (LPET), with Th (90 s peak/ 50 s background) and U (120 s peak/120 s background) on fifth spectrometer (LPET). Reference materials, analytical lines, diffracting crystals, mean detection limits (in wt%) and uncertainties are given in Supplementary Table 1. U, Pb, Th measurements were done at differential pulse height analysis modes with narrow energy window which improves signal-to-noise ratio and minimizes higher order interferences, however, does not eliminates peak overlaps completely. The monazite dating setup has 62 peak-overlap correction entries defined, including the most relevant peak overlap corrections for dating – Pb Mα by Y Lγ2, Th Mα by 2nd order Pr, Ce and Sm L lines and U Mα by Pb M2N4. Monazite from Namaqualand (South Africa Republic) was measured as an age standard. Its mean age from 30 EPMA points is 1046 ± 3 Ma, which statistically does not differ from the published 206Pb/238U age of 1046.5 ± 7.1 Ma (LA-ICP-MS, Liu et al. Reference Liu, Wu, Yang and Wilde2012).
3. d. Apatite, rutile and titanite separation and imaging
Apatite, rutile and titanite crystals from four amphibolite and one metapelite samples were separated using standard density separation techniques (crushing, sieving, washing and panning) at the Institute of Geological Sciences at the Polish Academy of Sciences, Kraków, Poland. Mineral phases were hand-picked under a binocular microscope, cast in 25 mm diameter epoxy resin mounts and then ground and polished to expose the grain interiors. Mineral textures were then imaged using back-scattered electron (BSE) and cathodoluminescence (CL) detectors on a FET Philips 30 scanning electron microscope with a 15 kV accelerating voltage and a beam current of 1 nA at the Faculty of Natural Sciences, University of Silesia in Katowice, Poland.
3. e. Apatite, titanite and rutile LA-ICP-MS U-Pb dating
LA-ICP MS U-Pb age and trace element data were acquired using a Photon Machines Analyte Excite 193 nm ArF excimer laser-ablation system with a HelEx 2-volume ablation cell coupled to an Agilent 7900 ICP MS at the Department of Geology, Trinity College Dublin. The instruments were tuned using NIST612 standard glass to yield Th/U ratios of unity and low oxide production rates (ThO+/Th+ typically <0.15%). A repetition rate of 11 Hz and a circular spot of 60 μm (apatite), 47 μm (titanite) and 45 μm (rutile) were employed. A quantity of 0.4 l min−1 He carrier gas was fed into the laser cell, and the aerosol was subsequently mixed with 0.6 l min−1 Ar make-up gas and 11 ml min−1 N2. Each analysis comprised 27 s of ablation and 12 second washout, the latter portions of which were used for the baseline measurement.
The data reduction of the raw U-Th-Pb isotopic data was undertaken using the freeware IOLITE package (Paton et al. Reference Paton, Hellstrom, Paul, Woodhead and Hergt2011) with the ‘VizualAge UcomPbine’ data reduction scheme (Petrus & Kamber, Reference Petrus and Kamber2012; Chew et al. Reference Chew, Petrus and Kamber2014) which can account for the presence of variable common Pb in the primary age reference material. Common Pb in the apatite, titanite and rutile standards was corrected using the 207Pb-based correction method. Conventional sample-standard bracketing was then applied to account for both downhole fractionation and long-term drift in isotopic or elemental ratios by normalizing all ratios to those of the U-Th-Pb reference materials. All apatite, rutile and titanite ages were calculated offline using the Isoplot plug-in for Excel (Ludwig, Reference Ludwig2012).
Twenty-nine isotopes (31P, 35Cl, 43Ca, 51V, 55Mn, 88Sr, 89Y, 90Zr, 139La, 140Ce, 141Pr, 146Nd, 147Sm, 153Eu, 157Gd, 159Tb, 163Dy, 165Ho, 166Er, 169Tm, 172Yb, 175Lu, 204Pb, 206Pb, 207Pb, 208Pb, 232Th and 238U) were acquired for apatite. NIST612 standard glass was used as the apatite trace-element reference material with 43Ca employed as an internal elemental standard, and a crushed aliquot of Durango apatite that has been characterized by solution quadrupole-ICP-MS analyses (Chew et al. Reference Chew, Babechuk, Cogne, Mark, O’Sullivan, Henrichs, Doepke and McKenna2016) was used as the apatite trace-element secondary standard. For U-Pb apatite analyses, the Madagascar natural mineral standard (Thomson et al. Reference Thomson, Gehrels, Ruiz and Buchwaldt2012) was used as the primary age reference material, using a c. 1 cm sized crystal of Madagascar apatite which has yielded a weighted average ID-TIMS concordia age of 473.5 ± 0.7 Ma (unpublished data from the study of Cochrane et al. Reference Cochrane, Spikings, Chew, Wotslaw, Chiaradia, Tyrell, Schaltegger and Van der Lelij2014). McClure Mountain syenite apatite (total Pb/U isochron age of 525.3 ± 1.7 Ma; Krestianinov et al. Reference Krestianinov, Amelin, Neymark and Aleinikoff2021) and Durango apatite (linear 3-D isochron age of 32.683 ± 0.050 Ma; Paul et al. Reference Paul, Spikings and Gaynor2021) were used as secondary standards and yielded 207Pb-corrected ages of 522.9 ± 56.8 Ma (MSWD = 1.8) and 32.49 ± 0.94 Ma (MSWD = 1.4), respectively, using a 207Pb/206Pb value of 0.874 derived from an apatite ID-TIMS total U-Pb isochron (Krestianinov et al. Reference Krestianinov, Amelin, Neymark and Aleinikoff2021) and a 207Pb/206Pb value derived from the Stacey & Kramers (Reference Stacey and Kramers1975) terrestrial Pb evolution model at 32.68 Ma.
For titanite, thirteen isotopes (43Ca, 48Ti, 89Y, 90Zr, 140Ce, 172Yb, 200Hg, 204Pb, 206Pb, 207Pb, 208Pb, 232Th and 238U) were acquired. The primary titanite standard used in this study is OLT1 titanite, which has yielded a TIMS concordia age of 1014.8 ± 2.0 Ma (Kennedy et al. Reference Kennedy, Kamo, Nasdala and Timms2010). BLR titanite Aleinikoff et al. (Reference Aleinikoff, Wintsch, Fanning and Dorais2002), which has yielded an ID-TIMS age of 1048.0 ± 0.7 Ma, was used as the titanite secondary LA-ICP-MS age reference material and yielded weighted average 207Pb-corrected ages of 1013.5 ± 8.9 Ma (MSWD = 0.65) and 1028 ± 14 Ma (MSWD = 1.4).
R10 was employed as the primary standard, and R19 (ID-TIMS date of 489.5 ±0.9 Ma; Zack et al. Reference Zack, Stockli, Luvizotto, Barth, Belousova, Wolfe and Hinton2011) and R13 (U-Pb SHRIMP age of 5045 ± 4 Ma, Schmitt & Zack, Reference Schmitt and Zack2012) were employed as secondary standards and treated as unknowns during data reduction and yielded weighted average 207Pb-corrected ages of 489.4 ± 2.6 Ma and 505 ± 16 Ma, respectively.
3. f. Thermodynamic modelling
Phase diagrams were constructed using the bulk rock chemical compositions of the representative metapelite samples SR20b and MMS3 (Table 2). For this, THERIAK DOMINO software package (dataset tcds62cid; de Capitani & Brown, Reference de Capitani and Brown1987; de Capitani & Petrakakis, Reference de Capitani and Petrakakis2010) was used in the MnNCKFMASHT system under fully water-saturated conditions with the internally consistent dataset of Holland and Powell (Reference Holland and Powell2011). The following phases and a-X models were implemented: clinopyroxene and melt (Green et al. Reference Green, White, Diener, Powell, Holland and Palin2016), biotite, cordierite, garnet, ilmenite, staurolite, spinel and orthopyroxene (White et al. Reference White, Powell, Holland, Johnson and Green2014a; Reference White, Powell and Johnson2014b), chlorite (White et al. Reference White, Powell, Holland, Johnson and Green2014a; Reference White, Powell and Johnson2014b; Powell et al. Reference Powell, White, Green, Holland and Diener2014), epidote (Holland & Powell, Reference Holland and Powell2011), feldspar (Holland & Powell, Reference Holland and Powell2003) and white mica (Coggon & Holland, Reference Coggon and Holland2002). Quartz, titanite, rutile, sillimanite, andalusite, kyanite and water were treated as pure phases Isopleths for garnet and plagioclase endmembers and Si in white mica were plotted. Additionally, we apply the Ti-in-biotite geothermometer after Henry et al. (Reference Henry, Guidotti and Thomson2005) and the Zr-in-rutile geothermometer (Zack et al. Reference Zack, Moraes and Kronz2004) for additional geothermometry constraints, while the amphibole-plagioclase NaSi-CaAl geobarometric approach (Molina et al. Reference Molina, Moreno, Castro, Rodriges and Fershtater2015) and Ti-in-calcium amphibole thermometry (Liao et al. Reference Liao, Wei and Rehman2021), coupled with the classical Blundy & Holland (Reference Blundy and Holland1990) thermobarometer, were applied to amphibolites. The revised nomenclature of amphiboles (Leake et al. Reference Leake, Wooley, Birch, Burke, Ferraris, Grice, Hawthorne, Kisch, Krivovichev, Schumacher, Stephenson and Whittaker2004) was applied.
Table 2. Chemical composition of the whole-rock metapelitic samples from the Marmarosh Massif (Ukraine/Romania)
4. Results
4. a. Petrography and mineral chemistry
4. a.1. Bilyipotik Nappe (Ukraine) – metapelites
The metapelite sample (SR20c) is characterized by the presence of garnet and staurolite porphyroblasts, containing mineral inclusions, and a biotite + muscovite + quartz + plagioclase matrix (Figs. 3b, c). Accessory phases are ilmenite, rutile, Ca (F, OH)-apatite, monazite-Ce and zircon.
Garnet crystals of 1 to 4 mm in size (Fig. 3b) are zoned with respect to Fe, Mg, Ca and slightly Mn (Figs. 5a-d, f), and they are classified as almandine, with subordinate substitutions of grossular, pyrope and spessartine. The garnet core composition is Alm65-60Gros24-20Py6.5-3.0Spess8-15, while the rim composition is Alm80-79Gros12.0-9.5Py10.3-7.0Spess0.7-2.0 (Supplementary Table 2a). They contain abundant inclusions of ilmenite, plagioclase, quartz, zircon and monazite while rutile intergrown with ilmenite is found in the rims also. Inclusions define the snowball structure of the garnet (Fig. 3b). Plagioclase, both in the matrix and as inclusions in garnet, is oligoclase (An22-27). Biotite crystals are often chloritized. Unaltered biotite exhibits beige-brick-red pleochroism, with #mg in the range of 0.47–0.48 and Ti in the range of 0.19–0.20 a.p.f.u. (Table 3). Ilmenite (Ilm), mostly intergrown with rutile (Rt), occurs both as inclusions in garnet porphyroblasts and as a component of the matrix (Figs. 5a, e). All ilmenite crystals exhibit Mn substitution (Table 5).
Figure 5. BSE images from metapelite sample SR20C of the Marmarosh Massif (Ukraine): (a) portion of a garnet porphyroblast with inclusions defined by ilmenite+rutile (Ilm, Rt), monazite (Mnz) and quartz (Qtz); (b – d) chemical distribution of Ca, Mg and Fe in same area as a); (e) intergrowths of rutile (Rt) and ilmenite (Ilm) on the rim of a garnet porphyroblast; (f) chemical transect along the garnet porphyroblast in (a); (g) monazite Mnz1 and zircon from metapelite sample SR20C; (h) monazite Mnz2 rimmed by REE-epidote from sample SR20C.
Table 3. Chemical compositions, crystal-chemical formulae and calculated crystallization temperatures (Henry et al., Reference Henry, Guidotti and Thomson2005) of representative biotite crystals from metapelites
Abbreviations: b.d.l. – below detection limit; #mg = Mg/(Mg+Fe); T H = temperature of crystallization (after Henry et al., Reference Henry, Guidotti and Thomson2005).
Staurolite porphyroblasts, up to 1 mm in size, are parallel to the foliation defined by mica group minerals and mineral lineation (Fig. 3c). A slight chemical zonation due to Mg-Fe exchange is observed, but, in general, the staurolite is the Fe-variety (Supplementary Table 3). It is associated with small garnet crystals (Grt 3 – Alm75-73Gros13-6Py13-11Spess2-10; Fig. 3c; Supplementary Table 2a). Muscovite exhibits relatively small phengite contents (Si = 6.097–6.114 a.p.f.u. coupled with Mg = 0.075–0.091 a.p.f.u.) and elevated Na contents (0.376–0.576 a.p.f.u.; Table 4). Chlorite with an Fe/(Fe+Mg) ratio in the range of 0.54–0.61 (classified as ripidolite – brunsvigite; Table 6) replaces primary biotite. Among accessory phases, zircon crystals are rare. Monazite crystals from c. 20 μm to 80 μm in length are present both as inclusions in garnet and in the matrix (Figs. 5e, g, h). Most of them are unaltered, only locally overgrown by REE-rich epidote (Fig. 5h; Supplementary Table 1). Apatite crystals are quite common but show intensive recrystallization.
Table 4. Chemical compositions and crystal-chemical formulae of representative muscovite crystals from metapelites
Abbreviations: b.d.l. – below detection limit; #mg = Mg/(Mg+Fe).
Table 5. Chemical compositions and crystal-chemical formulae of representative ilmenite, rutile and titanite crystals from metapelites and amphibolites
Table 6. Chemical compositions and crystal-chemical formulae of representative chlorite crystals from metapelites
Abbreviations: b.d.l. – below detection limit. Temperature is based on three geothermometers: T1: Cathelineau & Nieva (Reference Cathelineau and Nieva1985), T2: Jowett (Reference Jowett1991), and T3: Kranidiotis & MacLean (Reference Kranidiotis and MacLean1987).
Four main mineral assemblages can be defined, related to kinematic history of the rock. Remanence of assemblage 1 (prograde) is defined by garnet core (Grt 1) and inclusions hosted within it (plagioclase, quartz, ilmenite, zircon and monazite). Assemblage 2 (peak of metamorphism) is defined by garnet rims (Grt 2) with inclusions of plagioclase, quartz and rutile in it and muscovite in the matrix. Assemblage 3 (post-kinematic) is represented by staurolite, containing inclusions of ilmenite, biotite and quartz and overgrown by biotite with similar composition and muscovite, as well as small elongated garnet (Grt 3; Fig. 3c). Assemblage 4 (retrograde) is composed of secondary minerals: post-biotite chlorite, RE-epidote, apatite and carbonate minerals (Figs. 3b, c; 5h), and which continued to grow during decompression. Shearing was probably continuous as evidence by inclusion trails in garnet porphyroblasts and S-C structures in the rock (Figs. 3b, c).
4. a.2. Bilyipotik Nappe (Ukraine) – amphibolites
The amphibolites (SR21 & SR22) show a strong metamorphic foliation and lineation, highlighted by dark and leucocratic layers (Fig. 3d). Amphibole group minerals, being the major mineral components, are classified as tschermakitic hornblende – alumino-tschermakitic hornblende – magnesio-hornblende (Supplementary Table 4a) and coexist with plagioclase (An35-27), accessory ilmenite, (Ca,F)-apatite and zircon. Tschermakitic- and alumino-tschermakitic hornblende forms porphyroblasts, together with andesine plagioclase, while magnesio-hornblende, equilibrated with oligoclase and ilmenite, postdates it and defines the foliation and lineation (Fig. 3e). Accessories are represented by rare zircon and abundant Ca-F apatite. Ilmenite (Fe1.00-0.93Mn0.08-0.02TiO3; Table 5) is replaced by titanite and is locally subidioblastic (Table 5). Rare secondary actinolite and actinolitic hornblende, together with common epidote and chlorite, replace amphibole and plagioclase. The rock is locally cut by carbonate veinlets up to 0.5 mm thick.
4. a.3. Bretila group, vaser infrabucovinian Nappe (Romania) – metapelites
The biotite-garnet schists (MMS2 & MMS3) from the Bretila Group are medium-grained, with garnet porphyroblasts ranging from 0.1 to 3.0 mm in size (Figs. 4b-d).
Garnets are classified as almandine and are zoned with respect to Fe, Mn, Mg and Ca. In the metapelite sample, MMS-2 garnets (0.1–0.5 mm in size) are almost homogeneous, with core composition of Alm73-72Gros7Py12-11Spess6-5, and rims have a composition of Alm75-74Gros7Py13-12Spess6-7. In sample MMS-3, the garnet (1.5–3 mm in diameter) has cores compositions of Alm70-66Gros19-10Py11-8Spess11-6, while the rim composition is Alm65-63Gros22-18Py14-8Spess7-2 (Supplementary Table 2b). The garnets contain inclusions of quartz, biotite, oligoclase, ilmenite and rutile (Figs. 4c, d). Mineral inclusions and minerals in the matrix do not differ in chemical composition. Biotite shows a characteristic beige-brown pleochroic scheme (Fig. 4c), with #mg in the range of 0.42–0.46 and Ti in the range of 0.18–0.20 a.p.f.u. (Table 3). In muscovite, defining the S-C fabric (Fig. 4d) elevated phengite and elevated Na contents were noted (0.21–0.25 a.p.f.u.; Table 4). Ilmenite exhibits Mn substitution and is locally replaced by titanite (Table 5). Accessory phases are monazite, partly replaced by epidote+apatite coronas, rare zircon, long-prismatic rutile and apatite of yellow to green CL (Figs. 6a-c, e). Rutile is characterized by Zr contents ranging from 210–381 ppm, Cr from 55.6–1494 ppm, Nb from 1455–3850 ppm and Hf from 1.73–10.2 ppm. (Supplementary Table 5).
Figure 6. BSE and CL images from metapelites and amphibolites of the Maramuresh Massif (Romania): (a) monazite Mnz1 (with secondary epidote – Ep) and zircon (Zr) inclusions in quartz from schist sample MMS2; (b) deformed monazite (Mnz1) from the MMS2 schist; (c) aggregates of rutile needles (Rt2), zircon (Zr) and monazite (Mnz) in quartz from the MMS3 schist sample; (d) ilmenite (Ilm) replacement by titanite (Ttn) in amphibolite sample MMA1; (e) Cl image of apatite from the MMS3 schist; (f) intergrowths of rutile (Rt) and ilmenite (Ilm), replaced by titanite (Ttn) in amphibolite sample MMA3; (g) associations of the two generations of rutile: primary Rt 1 and secondary Rt 2, locally overgrowing titanite (Ttn) and ilmenite from the MMA1 amphibolite; (h) replacement of titanite by secondary rutile in the MMA3 amphibolite.
Remnants of the prograde mineral assemblage (Assemblage 1) are represented by garnet cores and mantles (Grt1) and inclusions (biotite, plagioclase, ilmenite? and rutile). The peak assemblage (Assemblage 2) is Grt2 + Ms + Pl + Bt + Rt1 + Qtz. The post-kinematic mineral assemblage (Assemblage 3) is defined by biotite, muscovite, ilmenite, monazite, titanite and apatite. The final late retrograde assemblage (Assemblage 4) is formed by secondary minerals: chlorite, rare earth-epidote + apatite and rutile (Rt2). All mineral assemblages grew during continuous ductile shearing, as evidenced by the inclusion trails in garnets and associated S-C fabric (Figs. 4c, d).
4. a.4 Bretila group, Vaser infrabucovinian Nappe (Romania) – Amphibolites
The amphibolite interleaved with the metapelites exhibits a strong foliation and mineral lineation (Fig. 4a). In the striped amphibolite sample MMA1, the main mineral phases (amphiboles) can be classified as Mg-hornblende (porphyroblasts) to tschermakitic hornblende and ferrian tschermakitic hornblende, which defines the foliation and postdates Mg-hornblende. Minor actinolite and actinolitic hornblende rims were found around Mg-hornblende (Supplementary Table 4b). Plagioclase is replaced by sericite + albite. Accessories are represented by abundant rutile locally replaced by ilmenite and titanite, while ilmenite is replaced by titanite (Fig. 6d).
In the garnet-bearing amphibolite sample MMA3, magnesio-hornblende to tschermakitic hornblende (Supplementary Table 4b) and oligoclase-andesine (An26-33) are the main components, enclosing accessory almandine-grossulare-rich garnet crystals (Alm54-51Gros52-23Py11-7Spess11-2; Supplementary Table 1c; Fig. 4e). Mg-hornblende, actinolitic hornblende, Fe-actinolite and actinolite together with oligoclase-albite, RE-rich epidote and (Ca,F)apatite (Fig. 6e) are also present as the retrogressive assemblage.
In both amphibolite samples, two types of rutile are present: dark-brown, short prismatic crystals (Rt1), from 50–100 μm in length, locally replaced by ilmenite and titanite (Figs. 6f-g) and 2) brick-red, prismatic (Rt2) with aspect ratios between 2 and 3:1, which overgrows titanite and ilmenite (Fig. 6h). In Rt1 from sample MMA1, Zr contents range from 171–2850 ppm, Cr ranges from 13–123 ppm and Nb ranges from 631–922 ppm, while Hf ranges from 5.3–76 ppm. Rt2 from the same sample is characterized by Zr ranging from 173–980 ppm, Cr ranging from 32.8–142.3 ppm, Nb ranging from 660–1720 ppm, while Hf ranges from 7.32–29.1 ppm. Rt1 from sample MMA3 show Zr contents from 155.6–3000 ppm, Cr from 149.2–1970 ppm, Nb from 1180–887 ppm and Hf from 6.48–106 ppm (Supplementary Table 5).
4. b. Pressure–temperature estimations
4. b.1 Bilyipotik Nappe (Ukraine) – metapelites
The modelled metamorphic conditions are based on the whole-rock composition (Table 2) and major rock-forming minerals. Since matrix forming minerals re-equilibrated during the later stages and earlier assemblages could only be inferred from mineral inclusions in refractory phases like garnet, the pressure and temperature estimations here are largely based on the measured garnet compositions in sample SR20c. The modelled garnet compositional isopleths, corresponding to the core composition, define the first metamorphic episode (M1), with a mineral assemblage: Bt + Ph + Msc + Pl + Chl + Grt1 (core) + Ilm + Qtz at 560–630 MPa and 515–535 °C. The modelled isopleths corresponding to garnet rim compositions intercept at much higher pressure and similar temperature of 900–1180 MPa and 590–620 °C with a metamorphic assemblage of Pl + Ph + Msc + Bt + Grt2 (rim) + Rt + Qtz (Fig. 7a), defining the M2 episode. The M1 and M2 episodes mark the prograde metamorphism. Garnet isopleths corresponding to composition of the small, elongated post-kinematic garnets (Grt3), intercept at P-T conditions of 500–690 MPa and 550–610 °C, within staurolite field, with an assemblage of Grt + Pl + Msc + Bt + St + Ilm + Qtz (defining the M3 episode; Fig. 7a). Alternatively, staurolite porphyroblasts can be interpreted as a part of peak assemblage. However, the presence of ilmenite inclusions in staurolite (Fig. 3c) suggests that this mineral grew during uplift. These temperature estimates are in agreement with the Henry et al. (Reference Henry, Guidotti and Thomson2005) Ti-in-biotite geothermometry, yielding temperatures of 576–581 °C (Table 2; Fig. 7a). As biotite is stable at all computed P-T conditions, most probably it re-equilibrated throughout the P-T loop. Rare allanite + apatite + carbonate coronas around monazite in metapelites (Fig. 5h) mark retrogression and oxidized fluid influx (M4; Fig. 7a). Chlorite geothermometry (Cathelineau & Nieva, Reference Cathelineau and Nieva1985; Jowett, Reference Jowett1991; Kranidiotis & MacLean Reference Kranidiotis and MacLean1987) implies two phases of late chlorite growth – a higher temperature event at 350–400 °C (in agreement with M4 episode) and a subsequent event of low-temperature chlorite crystallization at 245–300 °C (Table 6).
Figure 7. P-T paths of the metamorphic evolution of the two metapelite rocks from the Marmarosh/Maramuresh Massif with the stability fields of the main mineral assemblages indicated. (a) Ukrainian part; (b) Romanian part. See the main text for explanations. Abbreviations according to Whitney & Evans (Reference Whitney and Evans2010) are amph – amphibole, bt – biotite, chl – chlorite, grt – garnet, ky – kyanite; pl – plagioclase, kf – K-feldspar, ms – muscovite, ph – phengite, qtz – quartz, rt – rutile, sill – sillimanite, ilm – ilmenite, ttn –titanite; st – staurolite, zo – zoisite.
4. b.2. Bilyipotik nappe (Ukraine) – amphibolites
In the amphibolites, P-T conditions of 880–1030 MPa and 573–620 °C for tschermakitic to alumino-tschermakitic hornblende coexisting with andesine plagioclase and 610–700 MPa and 610–630 °C for magnesio-hornblende and coexisting plagioclase pairs are determined based on amphibole-plagioclase geobarometry (Molina et al. Reference Molina, Moreno, Castro, Rodriges and Fershtater2015), Ti-in-calcium amphibole geothermometry (Liao et al. Reference Liao, Wei and Rehman2021), coupled with the Blundy & Holland (Reference Blundy and Holland1990) thermobarometer. These P-T constraints are compatible with a clockwise P-T loop for intercalated metapelites, with ilmenite crystallization at the expense of rutile on the exhumation path in the amphibolites. Decreasing temperature on the retrograde path was associated with an increase in oxygen fugacity and titanite crystallization at the expense of ilmenite/rutile in the amphibolites. That is supported by the low-temperature actinolite growth (423–508 °C, based on Ti-in-calcium amphibole geothermometry by Liao et al. (Reference Liao, Wei and Rehman2021) at low pressure at c. 340 MPa (Blundy & Holland Reference Blundy and Holland1990) (Supplementary Table 4a).
4. b.3. Bretila group, vaser infrabucovinian nappe (Romania) – metapelites
Similarly to sample SR 20c, sample MMS3 thermodynamic modelling was based on the whole-rock composition (Table 2), garnet compositional zoning patterns and rock-forming mineral assemblages. The modelled Grt isopleths show a prograde path from 455–620 MPa and 545–555 °C through 670–745 MPa and 540–560 °C (M1 - M2 – Grt core) to 910–965 MPa and 645–660 °C (M3 – Grt rim; Fig. 7b). The modelled assemblages change from Grt + Msc + Pl + Bt + St + Ilm + Qtz (M1) to Grt + Msc + Pl + Bt + Rt (M2) and Grt + Ms + Pl + Bt + Rt + Qtz + melt (M3), describing a clockwise P-T loop of increasing pressure at almost constant temperature, culminating in a low degree of partial melting at the inferred peak temperature (Field 13, Fig. 7b), probably obscured by the retrogression. The inclusions of ilmenite and rutile in garnet core and rim correspond to the modelled P-T path starting in ilmenite field going well into the rutile field. The modelled assemblage M3 corresponds with the matrix assemblage observed in the sample (Assemblage 2, MMS3). Temperature estimates based on the Henry et al. (Reference Henry, Guidotti and Thomson2005) geothermometer show values of 530–580 °C (Table 2) and are interpreted as recording retrogression linked to shearing.
Zr-in-rutile geothermometry in sample MMS2 (Zack et al. Reference Zack, Moraes and Kronz2004) yields two temperature ranges: a higher temperature one at 654 °C (R1, mean of 18 analyses) and a lower one at 569 °C (R2, mean of 9 analyses), which can be linked to retrogression in the sample (Supplementary Table 5). Temperature estimates employing Ti-in-biotite geothermometry (Henry et al. Reference Henry, Guidotti and Thomson2005) are 551–590 °C (Table 2), which correlates with the temperature range of the R2 rutile population and are similar to the MMS3 sample.
4. b.4. Bretila group, Vaser infrabucovinian Nappe (Romania) – amphibolites
The lack of unaltered plagioclase precludes the use of Pl-Amph geothermobarometry in the MMA1 amphibolite. However, Ti-in-calcium amphibole thermometry implies a temperature range of 688–640 °C (Supplementary Table 4b), and Zr-in-rutile geothermometry from the same rock sample yields a temperature at 672 °C (mean of 23 analyses). While this is in agreement with peak temperature estimates for the MMS2 schist (670–690 °C), the stability field of rutile at that temperature range is above 1000 MPa. The P-T estimates in amphibolite sample MMA3 for coexisting tschermakitic hornblende and andesine plagioclase are c. 800–840 MPa and 748–756 °C, while P-T estimates for coexisting Mg-hornblende and oligoclase yielded 450–500 MPa and 654–670 °C, within the ilmenite stability field. The Zr-in-rutile geothermometer yielded a mean value of 670 °C (Rt2), while a small rutile population (Rt1; 5 grains) yielded a mean temperature of c. 900 °C. As there is no other proof for high-temperature metamorphism in Bretila Group, these results remain uncertain.
The presence of low-grade (retrograde) amphiboles (actinolitic hornblende, Fe-actinolite and actinolite) and oligoclase-albite lie on the cooling path with a pressure of 366–480 MPa and a temperature range 317–619 °C as defined by Ti-in amphibole geothermometry (Liao et al. Reference Liao, Wei and Rehman2021) and Blundy & Holland (Reference Blundy and Holland1990) thermobarometer (Supplementary Table 4b).
4. c. Geochronology
4. c.1. Monazite geochemistry and dating
Chemical EPMA dating of monazite crystals was undertaken on the metapelite sample SR20C (Bilyipotik Nappe). Chemically homogeneous monazite crystals ranging in size between 15 and 50 μm in size were dated, including inclusions in garnet and in the matrix (Fig. 4a, b). Ce and La are the most abundant REE (23.1–29.5 wt% Ce2O3 and 9.7–13.7 wt % La2O3), while ThO2 and UO2 contents are 1.0–3.1 wt% and 0.5–1.15 wt%, respectively (Supplementary Table 1). On a Th versus Pb plot, two main generations of monazite were distinguished, and only one of them intersects the origin (Fig. 8a–c). Thirty-five point analyses of this population yielded an average age of 351 ± 14 Ma (MSWD = 0.28). Twenty-four point analyses yielded an average age of 287 ± 17 Ma (MSWD = 0.29), two points represent inherited pre-Variscan ages (c. 403 and 422 Ma), one point yielded an age of 70 Ma ± 106 Ma, two point analyses yielded early Alpine ages (171 and 172 Ma – rims of monazite from matrix biotite) and one point analyses was rejected due to disequilibrium (Supplementary Table 1; Figs. 8a and b).
Figure 8. Geochronology of metamorphic rocks from the Marmarosh Massif (Ukraine): (a) and (b) weighted mean CHIME age of monazite Mnz 1 and Mnz 2 from sample SR20C; (c) ThO2 vs PbO plot showing two populations of monazite in the SR20C schist; (d) and (e) U-Pb LA-ICP-MS Tera-Wasserburg apatite concordias.
Fifty-nine (59) chemically homogeneous Ce and La rich monazite crystals from the metapelite sample MMS2 (Bretila Group) were analysed (La2O3 = 13.6–15.7 wt %, Ce2O3 = 27.6–29.8 wt %, ThO2 = 7.4–3.6 wt % and UO2 = 0.61–1.06 wt %). Fifty-seven point analyses yielded an average age of 327.6 ± 8.5 Ma (MSWD = 0.35; Fig. 9a). Two-point analyses that yielded lower U contents (UO2 = 0.45–0.48 wt %) yielded ages of 238 Ma and 259 Ma (Supplementary Table 1).
Figure 9. Geochronology of metamorphic rocks from the Maramuresh Massif (Romania): (a) ThO2 vs PbO plot and weighted mean CHIME age of monazite Mnz 1 from the MMS3 schist sample; b-h) U-Pb LA-ICP-MS Tera-Wasserburg concordias for apatite, titanite and two rutile generations from amphibolites and metapelites.
4. c.2. Apatite geochemistry and U-Pb dating
Bilyipotik Unit
Short prismatic apatite crystals from the Bilyipotik Nappe, 70–150 μm in length, are fluorapatite in composition. A total of 64 spot analyses were undertaken on 53 crystals from the SR21 sample, while 86 spot analyses were undertaken on sample SR22. On the support vector machine (SVM) apatite classification diagrams (Sr/Y vs LREE [La, Ce, Pr, Nd]; O‘Sullivan et al. Reference O’Sullivan, Chew, Kenny, Henrichs and Mulligan2020) the apatite analyses from the SR21 and SR22 amphibolites plot in low- and medium-grade metamorphic (LM) and I-type granitoids and mafic igneous (IM) fields. Two analyses from the SR21 amphibolite plot in the high-grade metamorphic (HM) field (Fig. 10). Th/U ranges from 0.003–1.402 in SR21 to from 0.012–2.322 in SR22 sample, while Sr/Mn is mostly above 0.5 (Supplementary Table 6). Despite the small differences in chemistry, apatite from both samples yielded very similar U-Pb ages. Excluding one core, the 52 grains of metamorphic apatite from SR 21 sample yielded a Tera-Wasserburg lower intercept age of 317.3 ± 5.5 Ma (MSWD = 2.3; Fig. 8d), while all 86 analyses from the SR22 sample yielded a Tera-Wasserburg lower intercept age of 319.0 ± 2.5 Ma (MSWD = 0.88; Fig 8e; Supplementary Table 6).
Figure 10. Trace element geochemistry of apatite from the amphibolite samples on a support vector machine (SVM) classification diagrams (Sr/Y vs LREE [La, Ce, Pr, Nd]) after O‘Sullivan et al. (Reference O’Sullivan, Chew, Kenny, Henrichs and Mulligan2020). ALK = alkaline-rich igneous; HM = high-grade metamorphic; IM = I-type granitoids and mafic igneous; LM = low- and medium-grade metamorphic; S = S-type granites; UM = ultramafic igneous.
Bretila Unit
Thirteen short prismatic fluorapatite crystals (50–100 μm in length) from the MMA1 amphibolite were analysed, yielding a Tera-Wasserburg lower intercept age of 280.6 ± 32 Ma (MSWD = 0.71; Fig 9b; Supplementary Table 5). On the SVM apatite classification diagrams (Sr/Y vs LREE [La, Ce, Pr, Nd]; O‘Sullivan et al. Reference O’Sullivan, Chew, Kenny, Henrichs and Mulligan2020), the apatites plot in the low- and medium-grade metamorphic (LM) and high-grade metamorphic (HM) fields (Fig. 10). The Sr/Mn and Sr/Y ratios are below 0.5 and 1, respectively (Supplementary Table 6).
Forty-eight short-prismatic fluorapatite crystals from the amphibolite sample MMA3 (60–120 μm in length) were analysed. All points plot exclusively in the LM field on the SVM apatite classification diagram (Fig. 10) (O‘Sullivan et al. Reference O’Sullivan, Chew, Kenny, Henrichs and Mulligan2020), yielding a Sr/Mn ratio above 0.5 and a Tera-Wasserburg lower intercept age of 280.8 ± 27 Ma (MSWD = 0.41; Fig. 9c; Supplementary Table 5).
4. c.3. Titanite geochemistry and U-Pb dating (Bretila group)
Ten elongated titanite crystals from the amphibolite sample MMA3 (70–110 μm in length) were analysed. The aluminium content ranges from 6220–15390 ppm, yttrium ranges from 72.5–1220 ppm and correlates positively with ΣREE, ranging from 51.37–2297.40 ppm (r2 = 0.935). Uranium prevails over thorium (Th/U = 0.09–0.22; Supplementary Table 5). All ten analyses yielded an imprecise Tera Wasserburg age of 322 ± 25 Ma (MSWD = 4.0; Fig. 9d; Supplementary Table 5).
4. c.4. Rutile geochemistry and U-Pb dating (Bretila group)
In both amphibolite samples, two populations of rutile were analysed: Rt1 and Rt2. In sample MMA1 25, spot analyses on 35 crystals of first-generation rutile (Rt1) were undertaken. They yielded a Tera-Wasserburg lower-intercept age of 341 ± 18 Ma (MSWD = 1.45; Fig. 9e; Supplementary Table 5). Twenty-eight brick-red and elongated rutile crystals (Rt 2) were analysed. They yielded a lower intercept age of 276 ± 15 Ma (MSWD = 2.3; Fig. 9f; Supplementary Table 5).
Five short prismatic grains of rutile from amphibolite MMA3 yielded a Tera Wasserburg lower intercept age of 354 ± 21 Ma (MSWD = 0.55; Fig. 9g; Supplementary Table 5).
Twenty-seven dark-red and prismatic rutile crystals from the metapelite sample MMS2, ranging from 80 to 200 μm in length, were selected for analysis. From these crystals, 21 yielded a Tera Wasserburg lower intercept age of 278 ± 14 Ma (MSWD = 2.2; Fig. 9h; Supplementary Table 5), while six analyses do not lie on the discordia intercept.
5. Discussion
5. a. Linking the P-T paths with geochronology
5. a.1. Bilyipotik Nappe (Ukraine)
Monazite is thermally resistant mineral, with a Pb closure temperature exceeding 900 °C (Crowley & Ghent, Reference Crowley and Ghent1999; Cherniak et al. Reference Cherniak, Watson, Grove and Harrison2004), but it can crystallize over a wide range of temperatures and in magmatic, metamorphic and hydrothermal environments (Spear, Reference Spear2010). The older monazite (Mz1), found as inclusions in garnets (both in the cores and rims), yields an age of 351 ± 14 Ma (Fig. 8a) and constrains garnet crystallization to Meso-Variscan M1 – M2 metamorphism (Figs. 7a; 10a). The presence of rutile, intergrown with ilmenite, shows out that the sample passed through the stability fields of both phases during the metamorphic evolution of the rock complex (Fig. 11a).
Figure 11. A synthesis of the metamorphic paths from the northern (Ukrainian) (a) and southern (Romanian) (b) parts of the Marmarosh/Maramuresh Massif. See the main text for explanations. Red stability fields after Spear (Reference Spear2010).
The younger, second-generation monazite (Mz2; 287 ± 17 Ma – M4 episode; Figs. 8b, 11a) can be linked to post-Variscan (Permian) shearing, either within the monazite stability field or at the transition between the monazite and allanite stability fields on the retrograde path. Small monazite crystals, locally overgrown by allanite – REE-epidote coronas, imply oxidized fluid ingress, below 400–450 °C and c. 400 MPa (Finger et al. Reference Finger, Broska, Roberts and Schermaier1998) and are linked to the higher temperature chlorite population crystallizing at temperatures of 350–400 °C (M5; Fig. 11a). This episode was, however, intensive enough to partially mobilize Pb in monazite, which would explaining why PbO versus ThO2 line does not meet the origin on a Th versus Pb plot (Fig. 8c).
Apatite has a lower closure temperature window for the U-Pb system (350–550 °C, Schoene & Bowring, Reference Schoene and Bowring2007) and yields reasonably precise age data (317.3 ± 5.5 Ma and 319.0 ± 2.5 Ma for the M3 episode; Fig. 11a), which constrains the history of the cooling path (Chew & Spikings, Reference Chew and Spikings2015; Chew et al. Reference Chew, Sylvester and Tubrett2011). The apatite affinity on the SVM biplot suggests hydrothermal/low-grade metamorphic resetting of mafic igneous apatite (particularly Sr/Mn ratios) that underwent high-grade (amphibolite facies) metamorphism (Fig. 10). As shown by the replacement of intergrown ilmenite and rutile by titanite in the amphibolites, oxidizing conditions are again inferred for the M4 episode (Fig. 11a). A few rims of actinolite/ actinolitic hornblende yielded temperatures of 423 °C, 481 °C and 508 °C, with pressures at 340 MPa and below (Supplementary Table 4a), all located below the closure temperature of apatite, supporting the retrogression path described above (Fig. 11a).
5. a.2. Bretila group (Romania)
Four U-Pb mineral system pairs were used to date the metamorphic episodes. The oldest (Variscan) episode was defined by the Rt1 population from the MMA1 and MMA3 amphibolites and yielded Tera-Wasserburg U-Pb lower intercept ages of 341 ± 18 Ma and 354 ± 21 Ma, respectively (Figs. 9e, g; Supplementary Table 6) and defining the prograde mineral assemblage (M1 – M2 metamorphic episodes; Fig. 11b). These ages overlap with the U-Th-Pb ages of texturally early monazite from the Bilyipotik Nappe, Ukraine (Fig. 8a).
Titanite from the MMA3 amphibolite yielded an imprecise Tera-Wasserburg U-Pb lower intercept age of c. 322 Ma, in agreement with the U-Th-Pb monazite CHIME age of 327.6 ± 8.5 Ma (Figs. 9a & 9d) from the interleaved metapelite sample MMS2, possibly marking cooling into the titanite stability field (M4) and slightly older than the U-Pb apatite ages of c. 317 and 319 Ma from the Bilyipotik Nappe.
The Rt2 rutile from the MMS2 schist sample yielded a Tera-Wasserburg lower intercept age of 278 ± 14 Ma; Fig. 9h). Low-temperature Rt2 rutile, replacing titanite in the MMA1 amphibolite sample (Fig. 9f) yielded a similar Tera-Wasserburg lower intercept age of 278 ± 14 (Fig. 9f), in agreement with an imprecise apatite Tera-Wasserburg lower intercept age of 280.6 ± 32 Ma from the same sample (Fig. 9b). A similar Tera-Wasserburg age of 280.8 ± 27 Ma was determined from the MMA3 amphibolite sample, with these c. 280 Ma ages marking M5 shearing and retrogression (Fig. 11b). The apatite crystals from the MMA1 amphibolite plot mostly in HM field (Fig. 10) and yielded the lowest Sr/Mn and Sr/Y ratios, atypical of a metabasite protolith (Supplementary Table 6). The MMA3 amphibolite apatite plot mostly in LM field (Fig. 10) with Sr/Mn and Sr/Y ratios mainly below 0.5 and 1, respectively. In both amphibolite samples fluid-assisted recrystallization of apatite crystals is assumed to take place during low-grade metamorphism, likely related to shearing. This is in agreement with the presence of late, post-kinematic Mg-hornblende and secondary actinolite-group minerals, that crystallized in a P-T range of 366–428 MPa and 525–620 °C (Mg-hbl and Act-Hbl) and 317–456 °C (Supplementary Table 4b; Fig. 11b). As in the case of the Bilyi Potik suite, these P-T estimates can be related to titanite and apatite/Rt2 growth and mark the retrogression path related to decompression and shearing. In the entire massif, late muscovite is post-kinematic, with low phengite contents and represents low pressure (< 200 MPa) retrogression (M6 in Ukraine – Fig. 11a and M5 in Romania – Fig. 11b).
5. b. Geodynamic context and P-T-t evolution
The Marmarosh/Maramuresh Massif is interpreted as a fragment of the Paleozoic North European Platform that rifted off in the Mesozoic (Schmid et al. Reference Schmid, Bernoulli, Fügenschuh, Matenco, Scheffer, Schuster, Tischler and Ustaszewski2008; Matenco et al. Reference Matenco, Krezsek, Merten, Schmid, Cloetingh and Andriessen2010). A Laurussian affinity for the massifs is supported by the presence of Caledonian volcanic arc rocks, likely representing the eastern branch of the European Caledonides (Munteanu & Tatu Reference Munteanu and Tatu2003; Gawęda et al. Reference Gawęda, Szopa, Golonka, Chew, Stepanyiuk, Belskyy, Waśkowska, Siliauskas and Drakou2022).
The P-T-t data, reported here, place the Bilyipotik/Infrabucovinian nappes of the Marmarosh/Maramuresh Massif within the European Variscides. They were subjected to epidote-amphibolite to amphibolite facies Mid-Variscan (c. 350 Ma) metamorphism (M1 – M2 in Romania or M1 – M4 in Ukraine). In the northern part (Bilyipotik Nappe), the pressure increase to 900–1200 MPa (M3; Fig. 11a) can be interpreted as a result of nappe stacking. In the Infrabucovinian Nappes (Bretila Group) to the south, the maximum pressure did not exceed 1000 MPa (Figs. 11a, b). The Bilyipotik Variscan nappe was likely structurally deeper during Variscan nappe stacking. Based on an assumed westward vergence of the pre-Alpine thrusts Săndulescu (Reference Săndulescu1984) and Krautner & Bindea (Reference Kräutner and Bindea2002) considered that the Bretila Group of the Infrabucovinian Nappes was the structurally highest of pre-Alpine (i.e. Variscan) units, which also supports the lowermost position of the northern (Ukrainian) Bilyipotik nappe. A similar Variscan P-T scenario has been documented in the West Sudetes (Leszczyniec Complex) and is interpreted as recording the transition from oceanic to continental subduction following consumption of the Saxoturingian Ocean (Młynarska et al. Reference Młynarska, Barnes, Zack, Majka and Mazur2024). The similarity in age and P-T conditions of peak metamorphism represents another link between Eastern Carpathian pre-Alpine basement with the Central European Variscides.
Peak metamorphism in the Marmarosh/Maramuresh Massif is followed by Late Carboniferous exhumation and shearing at c. 317–330 Ma (Figs. 11a, b). Post-Variscan shearing in the Central European Variscides was likely related to orogen-parallel extension and exhumation of basement blocks (Franke, Reference Franke2014 and references therein; Franke & Stein, Reference Franke, Stein, Franke, Haak, Oncken and Tanner2000; Franke & Żelaźniewicz, Reference Franke and Żelaźniewicz2023). This Permian shearing event is more pronounced in the southern (Romanian) parts of the massif, with a c. 280–290 Ma event recorded by U-Pb apatite and rutile ages (second-generation rutile) and supported by the petrological evidence for retrogression both in metapelites and amphibolites. To the south, Permian and later retrogression is found in the Rodna Nappe (Figs 1c & 2b) (Ar-Ar ages of 200–280 Ma; Culshaw et al. Reference Culshaw, Mosonyi and Reynolds2012) interpreted as post-Variscan – pre-Alpine rifting (future Penninic Ocean; Plašienka, Reference Plašienka2012 and references therein). Permian tectonism (c. 260–275 Ma) has also been documented in the Central Western Carpathians both by chemical monazite dating and U-Pb apatite dating (Finger et al. Reference Finger, Broska, Haunschmid, Hrasko, Kohut, Krenn, Petrik, Riegler and Uher2003; Gawęda et al. Reference Gawęda, Szopa, Chew, O‘Sullivan, Burda, Klötzli and Golonka2018) and is associated with extensional tectonics and a high mantle heat flux (Broska & Uher, Reference Broska and Uher2001; Finger et al. Reference Finger, Broska, Haunschmid, Hrasko, Kohut, Krenn, Petrik, Riegler and Uher2003 and references therein).
The P-T conditions of the Marmarosh/Maramuresh Massif are therefore similar to the pre-Mesozoic crystalline basement of the Alps (Schulz, Reference Schulz2021), recording the same Variscan and post-Variscan collision-related crustal thickening, exhumation and Permian shearing. The principal difference is the lack of Meso-Variscan granitoids and HP/HT rocks in the Marmarosh/Maramuresh Massif, that is typical of the pre-Mesozoic basement of the Alps (von Raumer et al. Reference von Raumer, Bussy, Schaltegger, Schulz and Stampfli2013 and references therein). The metamorphic basement of the Marmarosh/Maramuresh Massif thus likely formed part of the external zone of Variscan orogen, most probably overthrust onto the Baltica/Laurussia margin (see Mazur et al. Reference Mazur, Aleksandrowski, Gągała, Krzywiec, Żaba, Gaidzik and Sikora2020 for discussion). The pre-Alpine crystalline fragments/massifs of the Alps and Outer Carpathians likely represent the same microcontinent, dismembered during Alpine orogenesis.
6. Conclusions
1. Metapelite-amphibolite successions from the Marmarosh/Maramuresh Massif show similar P-T-t histories as shown by U-Th-Pb dating of monazite, apatite, rutile and titanite and P-T modelling and enables detailed correlation between the northern Bilyipotik Nappe (Ukraine) and the southern Bretila Group (Romania).
2. The peak of prograde metamorphism is dated at c. 350 Ma and records Meso-Variscan nappe stacking during accretion to Laurussia. The P-T conditions of Meso-Variscan metamorphism vary between 900–1180 MPa and 515–535 °C for the Bilyipotik Nappe in the north and 910–965 MPa and 645–660 °C for the Bretila Group in the south.
3. Retrogression during late-Variscan (c. 317 Ma–330 Ma) to post-Variscan (c. 280–290 Ma) times was associated with shearing, likely linked to extension predating the opening of the Penninic Ocean.
4. These data allow linking of the Marmarosh/Maramuresh Massif to the Central European Variscides, with the Marmarosh/Maramuresh likely representing the external zone of the Variscan orogen. We infer that all crystalline, Pre-Mesozoic fragments in the Alps and Carpathians represent the remnants of the same microcontinent, dismembered during Alpine orogenesis.
Supplementary material
supplementary material for this article can be found at https://doi.org/10.1017/S0016756824000542
Acknowledgements
This study was financially supported by the internal University of Silesia project entitled: Linking East with the West – geological history of the Marmarosh/Maramuresh Massif (Ukraine/Romania) given to A.G. D.C. acknowledges past and present support from Science Foundation Ireland (SFI) through research grants 12/IP/1663, 13/RC/2092 and 13/RC/2092_P2 (iCRAG Research Centre) and 15/IA/3024. iCRAG is funded under the SFI Research Centres Programme. Detailed comments of the editor Dr Simon Schorn, reviewer Alexandre Peillod and an anonymous reviewer led to a clearer presentation of the paper and are gratefully acknowledged.
Competing interests
The authors declare none.
Open access
