Introduction

After Paleozoic accretionary processes linked to protracted subduction, the basement of the proto-Alps underwent significant deformation and metamorphism during the Variscan Orogeny (e.g., von Raumer et al. 2009, 2013; Siegesmund et al. 2021). Towards the Permian, a transition from Variscan post-collisional processes to continental extension related to Pangea break-up is recorded, giving rise to high-temperature/low-pressure metamorphism coupled with rifting-related magmatism (Marotta and Spalla 2007; Schuster and Stüwe 2008; Quick et al. 2009; Kunz et al. 2018; Manzotti et al. 2018; Yuan et al. 2020). Protracted crustal extension during the Triassic favoured the development of rift basins and associated magmatism (e.g., Doglioni 1987; Satterley 1996; Kurz et al. 1998; Veselá et al. 2008, 2011; Storck et al. 2019; De Min et al. 2020), cultiminating in the final Jurassic opening of the Penninic ocean (Ratschbacher et al. 2004; Gleiβner et al. 2021).

The Variscan to post-Variscan tectonic evolution of the proto-Alps is particularly well-recorded in the Tauern Window (Lammerer et al. 2008; Schmid et al. 2013). Devonian to Carboniferous Variscan tectonometamorphic and magmatic processes were succeeded by Permian–Triassic rifting-related basin development and magmatism (Söllner et al. 1991; Eichhorn et al. 1999, 2000; Cesare et al. 2002; Kebede et al. 2005; Veselá and Lammerer 2008; Veselá et al. 2008, 2011; Siegesmund et al. 2021). Rifting ended in the Late Jurassic, with deposition of platform carbonates and the final separation of the European and Adria plates.

In this contribution, new U–Pb and Lu–Hf detrital zircon data of Permian–Triassic metasedimentary sequences of the Tauern Window are presented. Maximum sedimentation ages and provenance are discussed, providing constraints on Permian–Triassic paleogeography of Alpine basement domains and adjacent crustal blocks. In addition, insights into Variscan to post-Variscan crustal growth and tectonosedimentary evolution of the Tauern Window are evaluated.

Geological setting

The Tauern Window occupies a central position in the orogenic architecture of the Eastern Alps. The lowermost stratigraphic sequences can be found here, including the pre-Alpine European basement and its sedimentary cover, albeit significantly overprinted by metamorphism and deformation (Fig. 1; e.g. Frasl 1958). From a structural perspective, the Tauern Window is formed by a duplex structure in the centre of the Eastern Alps, which underwent up to ca. 35 km of Alpine burial and exhumation caused by N–S crustal shortening and E–W extension during indentation of the Adriatic plate (Selverstone 1985, 1988; Selverstone et al. 19841995; Rosenberg et al. 2007; Ratschbacher et al. 1989; Frisch et al. 2000; Lammerer et al. 2008; Schmid et al. 2013). Basement and cover rocks of European plate provenance (Inner Tauern Window) and oceanic origin (Glockner nappe) are stacked together and subsequently folded (Figs. 2 and 3), favoured by anistropic layering due to granitoid sill and laccolith intrusions (Höck and Miller 1980; Lammerer and Weger 1998).

Fig. 1
figure 1

a Regional sketch map of the western Tauern Window. Red area shows location of map of b. b Detailed geological map of the study area. The A–B profile depicts the geological section of Fig. 2a. Thrusts are indicated with red dotted lines

Fig. 2
figure 2

a Regional geological profile showing units and sampling location. For samples collected to the west and east of the profile, their location is schematically depicted above or below the main profile (e.g., samples SMQZT, MKG). 1: Variscan orthogneisses, 2 and 4: Riffler Basin metasedimentary rocks with intercalation of (3) late Carboniferous metarhyodacite, 5: grey metaquartzites and metaconglomerates, 6: brown sandy marbles, 7: Hochstegen Marble, 8: Kaserer Formation, 9: middle Triassic dolomitic and calcareous marbles, 10: metagabbro, 11: metarhyolite, 12: mixed metasiliciclastic-carbonatic sequences (Seidlwinkel Formation), 13: Bündnerschiefer with intercalation of (14) black schist. b Schematic stratigraphic section showing samples and age constraints. Geochronological data from this work and Veselá et al. (2008, 2011)

Fig. 3
figure 3

a Type locality of the Kaserer Formation at the Kleiner Kaserer (3093 masl). 1: granitic basement of the Tux gneiss nappe, 2: Middle Triassic (?) dolomitic marbles, 3: Early Jurassic black schists, 4: Hochstegen Marble, 5: folded Kaserer Formation, 6: Bündnerschiefer calc-schists. View from helicopter to the north. b Kaserer Formation underlying tightly folded Middle Triassic dolomites of the Schöberspitz. View from the Kleiner Kaserer summit towards the west

The post-Variscan intramontane basins were filled with coarse- to fine-grained clastic sediments and acid volcanic rocks (Eichhorn et al. 2000; Veselá et al. 2008, 2011). During the post-rift subsidence, the Late Jurassic Hochstegen limestone covered the entire region (Kiessling 1992). A comparable situation can be found in the northern foothills of the Alps, where it is documented in deep boreholes (Lemcke 1988). In this context, the tectonostratigraphic position of rocks exposed in the hanging wall of the Hochstegen limestone, namely the Kaserer Formation (metaquartzites, metarkoses, schists, breccias and thin metadolomite lenses or layers) and equivalent sequences, is controversial. In short, there are two main proposals:

  1. 1.

    The Kaserer metasedimentary rocks comformably overlie the Hochstegen limestone, implying a sedimentary contact and therefore a Cretaceous age of the former (Frisch 1974; Rockenschaub et al. 2003; Schmid et al. 2013).

  2. 2.

    The Kaserer metasedimentary rocks are overthrust as a nappe system. In this case, rocks may be older, being originally deposited further south but still within the European margin of the Penninic ocean. In this case, the Kaserer Formation would be somehow related to the Permian–Triassic Haselgebirge, Verrucano and Buntsandstein formations and even to the continental to shallow marine Bellerophon and Werfen sequences of the Northern Calcareous Alps and the Southern Alps (Fenti and Friz 1973; Tollmann 1977; Baggio et al. 1982).

Materials and methods

Samples were collected for coupled U–Pb and Lu–Hf LA-ICP-MS zircon analysis, comprising mainly metasedimentary sequences from the basal Glockner Nappe (BSF, FMA, SMQZT, TXQZT, MSTG), the Kaserer Nappe (MSBG, KAS), the Riffler basin (MKG) and one metaigneous rock (TXRH), which marks together with metagabbro lenses the thrust plane of the Glockner Nappe. Sample locations are shown in Fig. 2, whereas sampling site coordinates are detailed in Supplementary Material 1. Approximately 3–5 kg for each sample were crushed and disintegrated using disc mill to obtain a grain fraction between ca. 50–500 μm. The size fraction < 315 μm was processed using the Wilfley table to concentrate the heavy minerals. The heavy fraction of the samples was loaded to sodium polytungstate and subsequently to diiodomethane heavy liquids, followed by removal of the magnetic minerals using the Frantz isodynamic separator. Zircons were arranged in rows, cast into epoxy resin discs having a diameter of 2.54 cm, and polished to reveal grain centres. Prior to analysis, cathodoluminescence and transmitted light images were obtained so that the best sites for analysis could be chosen. Zircon analyses were performed using a Neptune multicollector inductively coupled plasma mass spectrometer (ICP-MS) and an Analyte G2 excimer laser ablation (LA) system at the University of São Paulo (Brazil). Analytical results of U–Pb and Lu–Hf data are presented in Supplementary Materials 2 and 3, respectively, whereas details on analytical procedures are included in Supplementary Material 4.

An overview of the studied samples is presented in Table 1. Sample MKG corresponds to a metaconglomerate to meta-arenite of the Riffler Basin, which is tectonically the lowermost unit. For the metasedimentary Kaserer Formation, sample KAS was collected; it corresponds to a brownish calcareous metaquartzite, intercalated with a graphitic phyllite. MSBG comprises a graphitic micaschist of the Seidlwinkl Formation, collected from the footwall the Glockner Nappe thrust, which is marked by the mylonitic orthogneiss of sample TXRH. Graphitic micaschists (MSTG) and chlorite-muscovite-bearing metaquartzites (TXQTZ) were sampled in the hanging wall of the Glockner Nappe thrust (Fig. 2). Sample SMQZT (Schmirntal Quartzite) comprises a white metaquartzite, whereas FMA is a fine-grained phengitic meta-arkose, so far attributed to the Wustkogel Formation (Thiele 1976) but more likely corresponding to the Seidlwinkel Formation (see “Age and provenance”). Finally, a black sandy phyllite with calcareous layers (BSF) was sampled close to the base of the Bündnerschiefer unit of the Glockner Nappe.

Table 1 Maximum sedimentation and metamorphism ages

Results

U–Pb LA-ICP-MS

Maximum sedimentation ages for metasedimentary rocks were obtained following three approaches, i.e., youngest detrital grain (YDG), youngest grain cluster at 1σ (YGC1) and youngest grain cluster at 2σ (YGC2) (Table 1; Dickinson and Gehrels 2009; Coutts et al. 2019). In all cases, calculations were made considering 206Pb/238U ages, since the youngest zircons of all samples yield Late Paleozoic to Triassic ages. All calculations of the youngest cluster weighted mean age were based on n = 2 and n = 3 for ages overlapping at 1σ and 2σ, respectively. In some samples, scarce zircons yielded anomalously young ages (< 210 Ma) which seem to record Pb loss and are thus not considered for estimation of sedimentation ages (Supplementary Material 2). Results of detrital zircon age spectra are presented in Fig. 4 and Table 1.

Fig. 4
figure 4

Kernel density estimate curves and histograms plotted using DensityPlotter (Vermeesch 2012) for U–Pb detrital zircon data results. Plots only include data with ± 10% discordance. Bin and band width = 20 myr

Metasedimentary samples commonly show euhedral to subhedral and fragmented zircons, though subordinate subrounded to ovoid crystals are present as well. Internal structures mainly correspond to oscillatory and minor sector zoning, whereas thin homogeneous dark or bright rims are common.

Sample KAS shows a well-defined Guadalupian maximum sedimentation age, based on all parameters (YDG = 267 ± 3 Ma, YGC1 = 268 ± 4 Ma, YGC2 = 269 ± 3 Ma). In terms of provenance, Paleozoic ages are clearly dominant, with main peaks at ca. 480–440 and 320–280 Ma. Scarce Cryogenian and Stenian crystals are also present.

The youngest single grain of sample MSBG yields a late Triassic age of 236 ± 3 Ma and scattered Permian crystals are also present. However, estimations of YGC1 and YGC2 are much older, providing early to middle Devonian ages. The main detrital zircon population corresponds to ages of ca. 455–435 Ma. A minor peak of 580–600 Ma is also recorded, together with subordinate Paleoproterozoic, Stenian and Tonian contributions.

A late Triassic maximum sedimentation age is clear for MSTG, as indicated by the YDG at 207 ± 3 Ma, and slightly older ages of 224 ± 4 and 225 ± 10 Ma of YGC1 and YGC2, respectively. Provenance is dominated by ages of ca. 320–280 Ma and a second peak at ca. 460–440 Ma. A minor group at ca. 560–570 Ma and a few Tonian crystals are present as well.

In the case of sample TXQZTG, a middle to early Triassic age maximum sedimentation age of ca. 250–243 Ma is obtained by all calculations. Two main peaks mainly at ca. 460–430 and 280–260 Ma are observed, though a minor group at ca. 610–580 Ma and three Neoarchean-Paleoproterozoic grains are also recorded.

For sample SMQZT, all parameters point to a comparable late Triassic maximum sedimentation age of ca. 235–232 Ma. Provenance is clearly dominated by early Permian ages, mainly of ca. 290–280 Ma, though a minor contribution of Silurian ages between ca. 440 and 420 Ma is present as well. A Paleoproterozoic and an Ediacaran crystal are also recorded.

Middle to late Triassic maximum sedimentation ages, mainly at ca. 240–233 Ma, are also documented by all three calculations for sample FMA, which is dominated by euhedral to subhedral zircons with oscillatory zoning and dark rims. The main peak corresponds to Ordovician to early Siluarian ages of ca. 480–420 Ma, followed by Late Carboniferous-Permian crystals of ca. 320–260 Ma. A minor contribution of Neoproterozoic zircons is also observed.

Sample BSF yields a very robust middle Permian maximum sedimentation age of ca. 268–266 Ma, indicated by comparable YDG, YGC1 and YGC2 values. Provenance is dominated by an Ediacaran peak at ca. 610–580 Ma, followed by a second Ordovician to early Silurian group at ca. 470–435 Ma. In addition, subordinate contributions of ages of ca. 330–310 and 720–700 Ma, together with scattered Archean to Tonian crystals, are documented as well.

Though only scarce zircons were obtained from sample MKG, they provide a late Permian (YGC2) to middle-late Triassic (YGC1 and YDG) maximum sedimentation age. In a similar way, provenance is restricted to Permian–Triassic crystals, mainly yielding ages of ca. 260–240 Ma.

Finally, euhedral to subhedral zircons of sample TXRH, characterized by oscillatory zoning, yield a concordia age of 267 ± 2 Ma for the timing of crystallization of the felsic orthogneiss protolith (Fig. 5). In addition, the sample shows significant inheritance of xenocrysts, mainly of ca. 325–270 Ma, 460–420 and 620–580 Ma, with further scattered grains yielding Paleoproterozoic, Tonian, Cryogenian and Paleozoic ages (Supplementary Material 2).

Fig. 5
figure 5

U–Pb concordia age for the orthogneiss of sample TXRH

Lu–Hf isotopes

Coeval detrital zircons of different stratigraphic units and samples show similar Lu–Hf signature, indicating that no major differences are recorded among them. Such similarities are also observed for xenocrysts of the orthogneiss sample TXRH (Fig. 6). Paleoproterozoic and Stenian-early Tonian zircons yield dominantly subchondritic compositions, with εHf below ca. − 4. Though scarce, scattered values between ca. + 6 and − 27 are recorded by late Tonian to Cryogenian crystals. Ediacaran to early Cambrian zircons yield variable compositions between ca. + 7 and − 27 as well. In contrast, the composition of Ordovician to Silurian crystals is almost restricted to subchondritic values between ca. − 4 and − 28.

Fig. 6
figure 6

U–Pb vs. εHf data of studied units. The lower diagram shows a detail of age distribution for zircons < 700 Ma. Data were recalculated considering a constant decay λ 176Lu = 1.867 × 10−11 year−1 (Söderlund et al. 2004) and CHUR values of 176Hf/177Hf = 0.282785 and 176Lu/177Hf = 0.0336 (Bouvier et al. 2008). The red line indicates a crustal array with 176Lu/177Hf = 0.015, typical for a crustal reservoir (e.g., Vervoort and Blichert-Toft 1999)

Carboniferous to very early Permian zircons are dominated by subchondritic compositions, mainly between − 3 and − 11. In contrast, zircons with crystallization ages between the middle Cisuralian (ca. 290 Ma) and the Middle Triassic show a broad compositional spectrum, with εHf between + 4 and − 24. They define an apparent bimodal distribution, with a first group dominated by subchondritic values below ca. − 9 and a second cluster with slightly subchondritic to suprachondritic values between ca. + 4 and − 5. The latter group is mainly recorded by zircons of sample TXQZTG and magmatic crystals of sample TXRH.

Discussion

Age and provenance

Despite minor differences, all studied metasedimentary rocks yielded similar results, with maximum sedimentation ages between the late Permian and the Triassic (Table 1). The Guadalupian maximum sedimentation age of sample KAS is in line with correlations of the Kaserer Formation upper section with Anisian rocks of the Wolfendorn area (Frisch 1975; Veselá et al. 2008). Consequently, a late Permian to early Triassic deposition age is inferred. Even though a Cretaceous age for the Kaserer Formation cannot be completely ruled out, there is no evidence for the suspected Cretaceous age. In contrast to previous proposals (e.g., Frisch 1975; Thiele 1976; Rockenschaub et al. 2003; Schmid et al. 2013), a tectonic contact between the top of the Upper Jurassic Hochstegen Marble and the Kaserer Formation is assumed (Fig. 3), further supported by field evidence, such as different folding styles and local sheared gypsum lenses at the contact (e.g., Steinernes Lamm crest), and a conformal contact with the Middle Triassic carbonates at the Schöberspitz and other locations. In addition, the Late Triassic maximum deposition age for the overlying MSBG sample (Fig. 2) reinforces this interpretation.

On the other hand, metasedimentary rocks of the Glockner Nappe, which are spatially associated with tectonic slices of Cambrian metagabbros and Permian orthogneisses at the base (Fig. 2b, Veselá et al. 2008), yielded middle to late Triassic maximum sedimentation ages (samples MSTG, TXQZT, SMQZT, FMA). This may indicate that, though these rocks may be part of the Triassic clastic sedimentary sequences of the area, they may represent slightly younger deposits than those of the Kaserer Formation sensu strictu (sample KAS). Though sample FMA correspond to sequences previously attributed to the Permian–Triassic Wustkogel Formation of the central Tauern Window (Thiele 1976), the maximum deposition age indicates a more likely affinity with the Seidlwinkel Formation. Furthermore, results are comparable to maximum sedimentation ages of sample MSBG in the footwall of the Glockner Nappe thrust. Alternatively, this sample may be correlated with Late Triassic rocks of the Pfiffkar Formation (Favaro and Schuster 2012).

Despite being scarce, similar results were documented for the meta-arenite of sample MKG (Riffler Basin). The existence of an intercalation of a late Carboniferous metarhyodazite at the base of the Riffler Basin (Fig. 2b, Veselá et al. 2011) implies a protracted sedimentation age for the latter, extending up to the middle to late Triassic.

The obtained late Permian–Triassic maximum deposition ages are also in agreement with the age of the Hochstegen Marble, which overlains post-Variscan metasedimentary sequences of the Tauern Window and yields a late Jurassic Oxfordian age according with fossil content (Schönlaub et al. 1975; Kiessling 1992). Consequently, the studied sequences represent protracted siliciclastic, mainly continental sedimentation, which were afterwards succeeded by a progressive Jurassic marine transgression recorded by platform deposits of the Hochstegen Marble (Veselá and Lammerer 2008; Veselá et al. 2008).

Surprisingly, sample BSF of the Bündnerschiefer Basin shows only middle Permian maximum sedimentation ages (Table 1), though this unit has been classically assigned to the Jurassic-Cretaceous (e.g., Steinmann 1994; Garofalo 2012). The wide spread of zircon ages may be explained by the existence of a relatively open and deep Penninic Ocean basin (Alpine Tethys), in contrast to underlying sequences (Fig. 2b), thus suggesting a generally low topography and a consequent enlarged catchment area.

In a similar way to maximum deposition ages, provenance is also comparable for most sequences, with main late Carboniferous-Permian and Ordovician-early Silurian peaks. A subordinate, yet relevant Ediacaran contribution is present as well, together with minor input from Archean, Paleoproterozoic, Stenian-Tonian and Cryogenian sources.

Late Carboniferous-Permian ages are attributed to Variscan and post-Variscan magmatism, which is well-documented in the Tauern Window (Söllner et al. 1991; Eichhorn et al. 1999, 2000; Cesare et al. 2002; Kebede et al. 2005; Veselá et al. 2008, 2011). On the other hand, Ordovician-early Silurian zircons possibly derived from Cenerian magmatism (e.g., Zurbriggen 2015, 2017; Oriolo et al. 2021), which is, however, absent in the Tauern Window, despite scarce coeval detrital zircons documented by Kebede et al. (2005). This may imply a more likely derivation from an adjacent tectonostratigraphic domain, where such ages are well-documented (see “Permian–Triassic paleogeography of the Tauern Window”).

On the other hand, Ediacaran ages seem to represent a clear Pan-African contribution, being nearly coeval with peak collisional to post-collisional magmatism at ca. 610–580 Ma (e.g., Abdelsalam et al. 2002; Liégeois et al. 2003, 2013; Oriolo et al. 2017). A minor input from Cadomian magmatic sources ubiquitously recorded in pre-Variscan European basement inliers may be feasible as well, particularly for ages between ca. 550–530 Ma (Oriolo et al. 2021 and references therein). Pre-Ediacaran zircons between the Archean and the Cryogenian might have derived from Precambrian sources associated with the northern African crustal blocks, such as the Saharan and Tuareg Shield metacratons (Abdelsalam et al. 2002; Liégeois et al. 2003, 2013). Nevertheless, this may not necessarily imply a derivation from primary Precambrian sources (see “Permian–Triassic paleogeography of the Tauern Window”).

Permian–Triassic paleogeography of the Tauern Window

Provenance of the studied late Permian–Triassic metasedimentary sequences provide valuable constraints to reconstruct the paleogeographic position of the Tauern Window during the Variscan to post-Variscan evolution. In the first place, the dominance of late Carboniferous-Permian and Ordovician-early Silurian ages together with the relative scarcity of Precambrian zircons seems to indicate a relatively long-distance from the Gondwana mainland (i.e., northern Africa). Consequently, the most likely source could be attributed to Paleozoic intrusions and their metasedimentary wall rocks, related to the protracted Paleozoic orogenic evolution of the pre-Alpine basement (e.g., Schulz et al. 2008; Siegesmund et al. 2018, 2021; Oriolo et al. 2021). This implies that sediments of Permian–Triassic basins partially resulted from recycling of older (meta) sedimentary sequences.

Ediacaran to Early Paleozoic paleogeographic reconstructions suggested a position of the Tauern Window close to the Bohemian Massif, immediately west/northwest of the Austroalpine and South Alpine units and relatively far away from Gondwana mainland (Ratschbacher and Frisch 1993; Eichhorn et al. 2001; Siegesmund et al. 2021). A roughly comparable paleogeography was also suggested for the Variscan evolution (von Raumer 1998; Eichhorn et al. 2000; Veselá et al. 2011), indicating that the Tauern Window might have been placed between the Austroalpine and South Alpine units, External Alpine units and the Bohemian Massif during the Permian–Triassic, representing one of the easternmost domains of the Penninic units (Fig. 7). Similarities in the provenance of coeval sequences in adjacent Alpine domains further reinforce this model (Beltrán-Triviño et al. 2016; Wizevich et al. 2019).

Fig. 7
figure 7

Sketch showing the position of the Alpine basement domains during the late Permian–Triassic (modified after von Raumer 1998; Haas et al. 2020). AGM Aar and Gotthard Massifs, AM Argentera Massif, BPM Belledonne and Pelvoux Massifs, BSTW basement south of the Tauern Window, CA Carnic Alps, DMM Dora Maira Massif, EGZ Eastern Greywacke Zone, GPU Gran Paradiso Unit, GU Gleinalm Unit, IVZ Ivrea-Verbano Zone, LU Ligurian Unit, MBARM Mont Blanc and Aiguilles Rouges Massifs, MD Moldanubian Domain, ÖU Ötztal Unit, SBU St. Bernhard Unit, SLZ Sesia–Lanzo Zone, SN Silvretta Nappe, SSU Schladming-Seckau Unit, VU Veitsch Unit, WU Wechsel Unit

Such a paleogeographic position explains the significant recycling of older Paleozoic sequences together with the input of coeval magmatic sources, particularly of those related to the Cenerian orogeny. Though Variscan and even relics of Ediacaran-Cambrian magmatism are well-documented in the Tauern Window (Veselá et al. 2008), Ordovician-early Silurian magmatism is nearly absent, being essentially restricted to amphibolites yielding a U–Pb zircon crystallization ages of ca. 486–482 Ma (von Quadt 1992; Eichhorn et al. 2001) and detrital zircon ages from the Biotitporphyroblastenschiefer (ca. 471–411 Ma) and Felbertal gneisses (ca. 447–405 Ma) (Kebede et al. 2005). However, dominantly felsic peraluminous magmatism and crustal anatexis at ca. 490–450 Ma are well-recorded in the Moldanubian Domain and allochthonous units of the Saxo-Thuringian Domain of the Bohemian Massif (Friedl et al. 2004; Teipel et al. 2004; Sagawe et al. 2016; Koglin et al. 2018; Soejono et al. 2019). Comparable felsic peraluminous magmatism and migmatites yielding ages of ca. 490–440 Ma are also widespread in Austroalpine, South Alpine and Penninic Units (Klötzli-Chowanetz et al. 1997; Meli and Klötzli 2001; Schulz and Bombach 2003; Schulz et al. 2004, 2008; Siegesmund et al. 2007, 2021; Thöny et al. 2008; Bussien et al. 2011; Rode et al. 2012; Cavargna-Sani et al. 2014; Bergomi et al. 2018; Arboit et al. 2019). Furthermore, the peraluminous geochemical signature of the magmatism and the evidence of crustal anatexis explains the dominant subchondritic Lu–Hf composition of Ordovician to early Silurian zircons (Fig. 6).

On the other hand, late Stenian to Tonian zircons, which represent a minor age group in Permian–Triassic sequences, are absent in pre-Variscan metasedimentary rocks of the Tauern Window (Siegesmund et al. 2021). This age group, which is also lacking in Early Paleozoic metasedimentary rocks of the Bohemian Massif, is well documented in coeval Austroalpine and South Alpine sequences (Heinrichs et al. 2012; Siegesmund et al. 2018, 2021). Considering the input of Early Paleozoic magmatism as well, the Austroalpine and South Alpine basement might have thus represented a relatively major source for detritus in Variscan and post-Variscan basins of the Tauern Window.

The narrow age range of the investigated samples and their structural position at the base of the Bündnerschiefer (Penninic nappes) indicate that they formed in a transitional area between the European continental margin and the future Penninic ocean. Deep drilling for the Brenner base tunnel encountered horizons of evaporitic and shallow marine Permian–Triassic strata at several tectonic levels (Brandner 2008; Töchterle et al. 2011). These tectonically weak units served as detachement horizons during Alpine compression and possibly allowed thrusting of the Kaserer Formation over the Hochstegen Marble. Consequently, Permian–Triassic siliciclastic deposits mixed with shallow marine carbonate and evaporitic horizons, Variscan gneiss relics and pre-Variscan mafic and ultramafic basement blocks (metagabbros, amphibolites, serpentinites) may represent a tectonic mélange zone, which originated on a distal and hyperextended continental margin that was imbricated during the Alpine Orogeny.

Pre-Variscan to Variscan tectonosedimentary and magmatic evolution

The earliest record of Variscan magmatic activity in the Tauern Window corresponds to arc granitoids yielding U–Pb SHRIMP zircon ages of ca. 374 Ma (Eichhorn et al. 2000). Subduction culminated at ca. 345–340 Ma during the Variscan collision, triggering crustal anatexis and syncollisional felsic peraluminous intrusions (Finger et al. 1997; Eichhorn et al. 2000), associated with siliciclastic sedimentation (Kebede et al. 2005; Lerchbaumer et al. 2010). Widespread post-collisional magmatism is documented at ca. 340–290 Ma, including the intrusion of durbachitic plutons, which further reinforces the correlation of the Tauern Window with adjacent Alpine basement domains and the Bohemian Massif (e.g., Finger et al. 1997; Veselá et al. 2011; von Raumer et al. 2013; Janoušek et al. 2020). The source of these post-collisional magmas, attributed to a mixing of metasomatized mantle-derived magmas and crustal components, can satisfactorily explain the subchondritic εHf values obtained for Carboniferous to early Permian detrital zircons of the studied sequences (Fig. 8).

Fig. 8
figure 8

a Compilation of U–Pb vs. εHf data of detrital zircons of Permian–Triassic Alpine sequences (Tauern Window magmatic and detrital zircons: this work; Austroalpine and South Alpine detrital zircons: Beltrán-Triviño et al. 2016; Austroalpine magmatic zircons: Yuan et al. 2020). b Schematic late to post-Variscan tectonic evolution of the Alpine Units. The star shows the approximate position of the Tauern Window. Studied sequences are indicated

The change in the isotopic signature of early Permian (< ca. 290 Ma) to middle Triassic detrital zircons, showing a broader spectrum of suprachondritic to subchondritic compositions (Fig. 8a), can be linked to the transition from the Variscan post-collisional stage to a generalized intracontinental extension (Fig. 8b). Protracted Permian to Triassic crustal extension is further supported by the progressive transition from closed intermontane to open sedimentary conditions recorded by geochemical and δ11B isotopic data of the Permian–Triassic metasedimentary cover of the Western Tauern Window (Franz et al. 2021).

The tectonosedimentary evolution of the studied sequences is comparable to that of late Carboniferous to Triassic deposits of the Pfitsch-Mörchner Basin, characterized by spatially restricted depocenters controlled by horst and graben structures that are overlapped by the Hochstegen Marble (Veselá and Lammerer 2008; Veselá et al. 2008). Furthermore, coeval South Alpine and Austroalpine units show also a similar history of continental rifting and resulting oceanic basin development, with dominance of Permian–Triassic continental deposits and volcanic and volcaniclastic rocks, succeeded by marine Jurassic deposits (e.g., Bertotti et al. 1993; Cassinis et al. 2012; Beltrán-Triviño et al. 2016).

Lithospheric thinning coupled with magmatic underplating might have triggered rifting-related magmatism and high-temperature/low-pressure metamorphism (Marotta and Spalla 2007; Schuster and Stüwe 2008; Kunz et al. 2018), thus explaining the observed Hf isotopic evolution as the result of mixing of crustal and mantellic sources, which is further supported by Permian magmatism of the Tauern Window and adjacent regions (e.g., Fig. 5; Eichhorn et al. 1999, 2000; Veselá et al. 2011; Yuan et al. 2020). A comparable evolution was also documented for Permian–Triassic volcano-sedimentary sequences of the western South Alpine and Austroalpine Units, based on U–Pb and Lu–Hf detrital zircon data and mineral chemistry of detrital feldspar, spinel and garnet (Beltrán-Triviño et al. 2016). Consequently, late Permian–Triassic metasedimentary rocks of the Tauern Window monitor the progressive transition from Variscan collisional/post-collisional tectonomagmatic processes to intracontinental rifting, which culminated with the opening of the Penninic ocean towards the Jurassic (Fig. 8b; e.g., Ratschbacher et al. 2004; Gleiβner et al. 2021).

The presence of scarce zircons yielding anomalously young late Triassic to Jurassic ages (< 210 Ma) may indicate the existence of a post-depositional hydrothermal overprint. Pre-Alpine hydrothermal activity linked to tourmalinite veins was documented by Franz et al. (2021) in the Tauern Window. Though the hydrothermal zircon overprint may be related to these mineralizations, it is unclear whether the obtained zircon ages provide a meaningful temporal constraint for this event.

The Jurassic protolith of the Hochstegen marble was deposited on a marine platform on the subsiding shelf and remains the youngest proven sediment of the Tauern Window. As a post-rift sedimentary sequence, it covers Late Variscan granites gneiss nappes on elevated Variscan highs as well as post-Variscan metasedimentary sequences (Veselá et al. 2008).

Conclusions

Detrital zircon data from metasedimentary rocks of the western Tauern Window yielded maximum deposition ages between the late Permian and the Triassic, indicating protracted sedimentation and magmatism between the Late Paleozoic and the Mesozoic. From a paleogeographical perspective, results indicate that the Tauern Window might have been placed close to further Alpine basement units (South Alpine, Austroalpine and External Massifs) and the Bohemian Massif during the Permian–Triassic, representing one of the easternmost domains of the Penninic Units.

Protracted clastic sedimentation in the Riffler Basin extended between the Late Carboniferous to the Early to Middle Jurassic, and were subsequently covered by Late Jurassic marine deposits of the Hochstegen Marble that covered the entire region. On the other hand, the Kaserer Formation, which is part of the Kaserer nappe, was overthrust by the Hochstegen Marble. Deposition of the Kaserer Formation took place between the late Permian and the Early Triassic, as constrained by Late Permian maximum sedimentation ages and overlying Anisian carbonates and Late Triassic metasiliciclastic sequences of the Seidlwinkel Formation.

In the Glockner Nappe, the lowermost section is characterized by slices of metagabbros and felsic orthogneisses. In this block, metaquarzites and metaconglomerates exposed between the Tuxer Joch to Hintertux, so far attributed to the Wustkogl Formation of the Central Tauern Window, are in fact Late Triassic and seem thus to correlate with sequences of the Seidlwinkel Formation from the Groβglockner area.

The studied sequences, which show similarities with coeval facies of the Germanic Basin, document the transition between Late Paleozoic Variscan post-collisional processes to intracontinental extension. This change possibly occurred at ca. 290, as suggested by variable suprachondritic to subchondritic Lu–Hf compositions of detrital zircons, which contrast with the dominantly subchondritic fingerprint of older zircon crystals. Lithospheric thinning coupled with magmatic underplating might have triggered rifting-related magmatism and high-temperature/low-pressure metamorphism, thus explaining the observed Hf isotopic evolution as the result of mixing of crustal and mantellic sources, which is further supported by Permian magmatism of the Tauern Window and adjacent regions. Progressive intracontinental rifting culminated with the final opening of the Penninic ocean by the Jurassic.