2.a. Lower unit of the Tatra mountains

The lower unit is exposed within a tectonic window and attains an exposed thickness of up to 1000 metres. It consists mainly of mica schists with intercalations of quartz-rich metapsammites and subordinate metapelites. Kahan (Reference Kahan1968) interpreted this assemblage as the metamorphosed equivalent of a flysch-type sedimentary succession. Two distinct metamorphic zones have been identified within the lower unit: (a) staurolite–kyanite (St–Ky) zone in the western part and (b) kyanite-fibrolite (Ky–Fib) zone in the east (Figure 1c; Janák, Reference Janák1994). This transition reflects an increase in metamorphic grade up-structural section towards the major Variscan thrust fault, with estimated peak conditions ranging from ∼570°C to 640°C at pressures of 6–7 kbar (Janák, Reference Janák1994).

Geochronological data for the lower unit are limited. Monazite electron microprobe dating conducted by Moussallam et al. (Reference Moussallam, Schneider, Janák, Thöni and Holm2012) on four mica schist samples yielded two age groups: two samples gave consistent ages of ∼370–372 Ma, whereas two others recorded older ages of ∼418 Ma and ∼479 Ma, although with several analyses clustering near 370 Ma. Based on U-Pb dating, Kohút et al. (Reference Kohút, Poller, Gurk and Todt2008, Reference Kohút, Linnemann, Hofmann, Gärtner and Zieger2022) and Soejono et al. (Reference Soejono, Bukovská, Collett, Novotná, Kohút, Janoušek, Zelinková, Jitka Míková, Schulmann, Hora and Veselovský2024) constrained the maximum depositional ages of detrital zircons from the lower and upper units at approximately 524 and 526 Ma.

2.b. Upper unit of the Tatra mountains

The upper unit consists of a volcano-sedimentary succession characteristic of a former passive continental margin and includes paragneisses, orthogneisses, migmatites, amphibolites with lenses of retrogressed eclogites and granitoids. The base of the upper unit (Figure 1c) marks the transition to kyanite zone, marked by the appearance of kyanite in paragneisses and the development of leucosome segregations indicative of incipient migmatization at P–T conditions of up to 10–12 kbar and 700–735°C (Janák et al. Reference Janák, Hurai, Ludhová, O´Brien and Horn1999). Orthogneisses exhibit mylonitic textures with augen-like K-feldspar porphyroclasts (Janák, Reference Janák1994; Poller et al. Reference Poller, Janák, Kohút and Todt2000). Banded amphibolites display alternating mafic and felsic layers (tonalitic to trondhjemitic) on a millimetre to decimetre scale, and eclogite boudins in banded amphibolites are strongly retrogressed (Janák et al. Reference Janák, O´Brien, Hurai and Reutel1996). Eclogite from the Baranec area in the Western Tatra shows minimum peak-pressure conditions of 15–16 kbar at 670–700°C and overprint upper amphibolite to granulite facies conditions at 10–14 kbar and 700–750°C (Janák et al. Reference Janák, O´Brien, Hurai and Reutel1996). At higher structural levels, in the sillimanite zone, migmatization becomes more pervasive. Index minerals include prismatic sillimanite, garnet, K-feldspar and cordierite (Janák et al. Reference Janák, Hurai, Ludhová, O´Brien and Horn1999; Ludhová and Janák, Reference Ludhová and Janák1999). Thermodynamic modelling of sillimanite-bearing paragneisses and migmatites from the High Tatra yielded P–T conditions of partial melting at 7–8 kbar and 760–770°C (Janák et al. Reference Janák, Petrik, Konečný, Kurylo, Kohút and Madaras2022). Amphibolites from the Bystrá area contain garnet but no relics of eclogite-facies metamorphism. The structurally highest levels of the upper unit are intruded by a sheet-like granitoid body (Gorek, Reference Gorek1959), varying in composition from leucogranite to biotite tonalite and amphibole-bearing diorite (Kohút and Janák, Reference Kohút and Janák1994).

In contrast to the lower unit, the upper unit has been extensively dated. SIMS U–Pb zircon dating of eclogite from Baranec records two metamorphic events: (a) at 367 Ma, interpreted as the minimum age of eclogite-facies metamorphism, and (b) at 349 Ma, recording retrogression under amphibolite-facies conditions (Burda et al. Reference Burda, Klötzli, Majka, Chew, Li, Liu, Gawęda and Wiedenbeck2021). In contrast, Lu–Hf dating of extensively retrogressed eclogite from Bryšno yielded 356 Ma age (Kohút et al. Reference Kohút, Anczkiewicz and Boczkowska2023). Dating of monazite (Th–U–Pb method, Janák et al. Reference Janák, Petrik, Konečný, Kurylo, Kohút and Madaras2022) and garnet (Lu-Hf and Sm-Nd methods, Kohút et al. Reference Kohút, Anczkiewicz and Boczkowska2023) in paragneisses and migmatites from the High Tatra yielded consistent results, constraining high-grade metamorphism and partial melting to 350–345 Ma. Numerous granitoids have been dated by U–Pb zircon geochronology yielding crystallization ages between 350–340 Ma (Poller et al. Reference Poller, Janák, Kohút and Todt2000; Burda et al. Reference Burda, Gawęda and Klötzli2013; Gawęda et al. Reference Gawęda, Burda, Klötzli, Golonka and Szopa2016, Reference Gawęda, Szopa, Chew, O’Sullivan, Burda, Klötzli and Golonka2018; Kohút and Larionov, Reference Kohút and Larionov2021; Broska et al. Reference Broska, Janák, Svojtka, Yi, Konečný, Kubiš and Maraszewska2022; Catlos et al. Reference Catlos, Broska, Kohút, Etzel, Kyle, Stockli, Miggins and Campos2022). ⁴⁰Ar/³⁹Ar dating of biotite from migmatitic garnet-sillimanite gneiss suggests cooling below closure temperatures of 300–350°C at 330–300 Ma, marking the final exhumation of the upper unit (e.g. Janák and Onstott, Reference Janák and Onstott1993; Janák, Reference Janák1994).

3.a. Electron microprobe analysis

Quantitative chemical analyses of garnet, feldspar, biotite and staurolite, as well as garnet X-ray maps, were carried out using a JEOL SuperProbe JXA-8230 at the AGH University of Krakow, Poland. All analyses were performed at an accelerating voltage of 15 kV, 20nA beam current and beam size between 1 and 5μm depending on mineral. Analytical conditions for X-ray maps of garnet were as follows: 15 kV accelerating voltage, 100 nA beam current, 100 ms dwell time and a step size of 2 μm. All EMP data are provided in the supplementary materials.

Zirconium concentrations in rutile were measured using the JEOL JXA-8530F at the Earth Science Institute of the Slovak Academy of Sciences in Banská Bystrica. Analytical conditions included an accelerating voltage of 15 kV, a probe current of 150 nA and a beam diameter of 3–7 μm depending on grain size. Measurement times were 300 seconds on-peak and 150 seconds on-background. High-sensitivity H- and L-type spectrometers were employed for Zr detection. The detection limit for Zr was 17 ppm, with an analytical uncertainty of 6–2% for concentrations between 200 and 600 ppm, and ≤2% for concentrations ≥1000 ppm.

Quantitative chemical characterization of monazite was performed using wavelength-dispersive X-ray spectroscopy (WDS) on electron microprobes at two laboratories: the Polish Geological Institute – National Research Institute in Warsaw (Cameca SX-100) and the Department of Earth Sciences, Uppsala University (JEOL JXA-8530F). Both instruments were used for point analyses and X-ray mapping of monazite grains. Analytical conditions included a beam current of 40 nA, an accelerating voltage of 15 kV and a focused beam with a diameter of 0.5–1 μm. The dwell time was set to 200 ms per point. X-ray maps were acquired with pixel size adapted to the grain dimensions and targeted Y, Th, Nd and Ca distributions to detect compositional zoning. Elemental calibration was performed using the following standards: jadeite and wollastonite for Si, wollastonite for Ca, orthoclase for K and Al, olivine for Fe, crocoite for Pb, arsenopyrite for As and glasses or pure-element phosphates for U, Th and rare earth elements (REE): YPO₄ (Y), SrPO₄ (Sr), GdPO₄ (Gd), DyPO₄ (Dy), SmPO₄ (Sm), NdPO₄ (Nd), PrPO₄ (Pr), CePO₄ (Ce) and LaPO₄ (La). BaSO₄ was used as a standard for S. Data were processed using XMapTools (Lanari et al. Reference Lanari, Vho, Bovay, Airaghi and Centrella2019). All monazite chemistry data can be found in supplementary materials.

3.b. Laser Ablation inductively coupled plasma mass spectrometry (LA-ICP-MS)

In situ Th–U–Pb isotope analysis of monazite was conducted at the Vegacenter, Swedish Museum of Natural History (Stockholm), using a Nu Plasma 3 multi-collector ICP-MS (MC-ICP-MS) coupled with an ESI-New Wave NWR193UC ArF excimer laser ablation system (λ = 193 nm). Ablation was performed for 18 seconds per spot with a repetition rate of 5 Hz, a fluence of ∼3.4 J/cm² and a spot diameter of 8 μm.

Isotope data were calibrated against the 44069 monazite reference material (age = 424 Ma; Aleinikoff et al. Reference Aleinikoff, Schenck, Plank, Srogi, Fanning, Kamo and Bosbyshell2006). Additional reference materials included Manangotry (555 Ma; Horstwood et al. Reference Horstwood, Foster, Parrish, Noble and Nowell2003) and Skalná (311 Ma; Konečný et al. Reference Konečný, Kusiak and Dunkley2018). Age calculations used iterative ²⁰⁷Pb corrections based on the Stacey and Kramers (Reference Stacey and Kramers1975) crustal Pb isotope evolution model. The ²⁰⁶Pb/²³⁸U ages were corrected using ²⁰⁷Pb, whereas ²⁰⁷Pb/²⁰⁶Pb ages were corrected using measured ²⁰⁴Pb. All reported isotope ratios and ages include 2σ analytical uncertainties. All data can be found in supplementary materials.

3.c. Secondary ion mass spectrometry (SIMS)

Secondary ion mass spectrometry (SIMS) U-Pb and oxygen isotope analyses in zircon were performed using a CAMECA IMS1280 large-geometry ion microprobe at the Swedish Museum of Natural History. Separated and hand-picked zircon grains were cast into epoxy mounts together with reference zircon grains and polished to reveal their interiors. Prior to SIMS analysis, a very thin gold coat (<5 nm) was applied, and all grains were imaged using FEI Quanta 650 with an Oxford Instruments Chroma CL2 detector. An additional 25 nm of gold was then added for the SIMS analyses.

Analytical protocols for U-Pb broadly followed those described by Whitehouse and Kamber (Reference Whitehouse and Kamber2005) and references therein. A ca. 10 nA O₂- primary beam with 23 kV incident energy (−13kV primary, +10 kV secondary) was generated using an Oregon Physics H201 RF-plasma source and used to sputter zircon. For this study, the primary beam was operated in critically focused (Gaussian) mode with a small raster retained during analysis to flatten the bottom of the ca. 15 μm analytical crater. Presputtering with a 25 μm raster for 60 seconds, centring of the secondary ion beam in the 3000 μm field aperture (FA), mass calibration optimization and optimization of the secondary beam energy distribution were performed automatically for each run, field aperture and energy adjustment using the ⁹⁰Zr₂ ¹⁶O⁺ species at nominal mass 196 Da. Mass calibration of all peaks in the monocollection sequence was performed at the start of each session; within-run mass calibration optimization scanned only the ⁹⁰Zr₂ ¹⁶O⁺, peak, with small drift corrections applied to all mass stations in each run. A mass resolving power (M/ΔM) of ca. 5400 was used to ensure adequate separation of Pb isotope peaks from nearby HfSi⁺ species. Ion signals were detected using an axial ion-counting electron multiplier. All analyses were run in fully automated chain sequences. Data reduction followed the protocols described by Jeon and Whitehouse (Reference Jeon and Whitehouse2015) and assume a power law relationship between Pb⁺/U⁺ and UO₂ ⁺/U⁺ ratios with an empirically derived slope in order to calculate actual Pb/U ratios based on those in the M257 zircon standard (Nasdala et al. Reference Nasdala, Hofmeister, Norberg, Martinson, Corfu and Dörr2008). U concentrations and Th/U ratio were also referenced to the M257 standard. Common Pb corrections were made only when ²⁰⁴Pb counts statistically exceeded 3x average detector background and assume a ²⁰⁷Pb/²⁰⁶Pb ratio of 0.83 (equivalent to present-day Stacey and Kramers (Reference Stacey and Kramers1975) model terrestrial Pb). Decay constants follow the recommendations of Steiger and Jäger (Reference Steiger and Jager1977). All data can be found in supplementary materials.

3.d. Sampling strategy

Fifteen metapelite samples were collected from the St–Ky and Ky-Fib metamorphic zones of the lower unit in the Western Tatra Mountains. Sampling was carried out along a north–south transect in the Jalovecká Valley and near the structural boundary between the Lower and upper units in the Žiarska Valley (see Figure 1c, Table 1).

Table 1.

Sample location and mineral assemblages of the studied samples

Seven samples were selected for detailed monazite analysis: three from Jalovecká Valley (LZ2204, ZT206 and ZT219) and four from Žiarska Valley (LZ2202, ZT96, MB2217B and MB2218). Structural measurements indicate that samples from Jalovecká Valley and LZ2202, ZT96 display a penetrative foliation dipping 45–70° to the NW–NNW. In contrast, MB2217B and MB2218, located in the upper Žiarska Valley, show a superimposed foliation parallel to SE–NW-trending fold axes. In addition, three samples of two-mica granite were collected in the upper unit, from the outcrops on Baníkov, Príslop and Plačlivé peaks (Figure 1c). The sampled rocks are coarse to medium grained granodiorites to monzogranites and are composed of quartz, plagioclase, K-feldspar, biotite and muscovite. A more detailed petrography of this granite has been described by Poller et al. (Reference Poller, Janák, Kohút and Todt2000) and is not repeated here. Zircon grains were extracted from crushed and sieved rock using standard mineral separation techniques (washing, panning, magnetic separation and heavy liquids). High-quality, inclusion-free and transparent zircons were hand-picked under a binocular microscope. Selected grains were mounted in 1-inch epoxy resin discs along with zircon reference materials (Qinghu and 91500) and subsequently polished for isotopic analysis.

4.a. Staurolite–Kyanite zone

Mica schist samples from the St–Ky zone contain the following major minerals: quartz, plagioclase, biotite, muscovite, kyanite, sillimanite, staurolite, garnet and Fe–Ti oxides. Chlorite, chloritoid and margarite have been interpreted as retrograde phases, partly replacing garnet, staurolite and kyanite (Janák et al. Reference Janák, Kahan and Jančula1988). Accessory phases include zircon, monazite, apatite, tourmaline, rutile and ilmenite. Garnet occurs as euhedral to subhedral porphyroblasts, commonly rimmed by kyanite, plagioclase or staurolite (Figure 2a). Atoll-textured garnets are common, with biotite and chlorite infilling the central core (Figure 2b). Most garnets display inclusion-rich cores and inclusion-free rims (Figure 2a,b,e). Staurolite forms elongated, often poikiloblastic crystals incorporating inclusions of quartz and biotite (Figure 2a,c). Kyanite is present as elongated, bladed grains that coexist with staurolite. The mineral assemblage Grt–Qz–Pl–Fsp–Bt–Ms–Ky–St–Sill–Rt is interpreted to represent peak-pressure metamorphic conditions. Foliation is defined by alternating quartz- and mica-rich layers (Figure 2d,f), with quartz elongation parallel to the alignment of white mica, biotite and kyanite. In high-strain samples (e.g. LZ2204), quartz exhibits undulose extinction and dynamic recrystallization features such as bulging and subgrain rotation (Figure 2f). In contrast, quartz inclusions within garnet and kyanite remain strain-free. Feldspar frequently shows bent twin lamellae, indicative of deformation under ductile conditions, and often contains abundant inclusions in high-strain samples. Biotite is locally replaced by chlorite, while feldspar exhibits variable degrees of sericitization, likely resulting from late-stage fluid infiltration. Euhedral chloritoid occurs in some samples within chlorite-rich domains.

Figure 2.

(a-f) Photomicrographs and BSE images of samples from the St-Ky zone. (a) Staurolite and kyanite with garnet inclusions. (b) Atoll garnet with central part composed of biotite and chlorite. (c) Euhedral grains of staurolite surrounded by chloritized biotite. (d) Rounded and slightly elongated grain of monazite in matrix. (e) Garnet in matrix with numerous inclusions in the central part and inclusion-free rim. (f) Dynamically recrystallized quartz grains displaying subgrain rotation and undulose extinction. (g-l) Photomicrographs of samples from the Ky-Fib zone. (g, h) Kyanite partially replaced by sericite and sillimanite. (i) Pleochroic halos around monazite in biotite. (j) Garnet with numerous inclusions, including monazite. The garnet is slightly elongated parallel to the foliation. (k) BSE image of elongated garnet grain in matrix, with Bt and Ms inclusion. (l) Dynamically recrystallized quartz grains.

4.b. Kyanite–fibrolite zone

Metapelitic samples from the Ky-Fib zone are composed of quartz, plagioclase, biotite, muscovite, garnet, kyanite, fibrolitic sillimanite and Fe–Ti oxides. Accessory minerals include zircon, monazite, apatite and tourmaline. The transition into this zone is marked by the disappearance of staurolite and the abundance of kyanite, which occurs as elongate grains aligned parallel to foliation. Kyanite is commonly replaced by fibrolitic sillimanite along grain boundaries (Figure 2g,h). Sillimanite occurs exclusively as fibrolite, either forming thin layers along foliation planes or as intergrowths with muscovite.

Garnet forms euhedral to subhedral porphyroblasts (Figure 2j,k) with inclusion zoning and local signs of resorption. Some grains contain fractures partially filled with chlorite (Figure 2k) and host larger inclusions of quartz, feldspar or biotite. As in the Ky–St zone, the foliation is defined by aligned micas, biotite, kyanite and elongate quartz (Figure 2g,i,l). Evidence for dynamic recrystallization of quartz, such as bulging grain boundaries, is observed in sample LZ2202 (Figure 2l).

Rutile, titanite and ilmenite are present in all samples. Rutile is locally overgrown or replaced by ilmenite. The mineral assemblage Grt–Qz–Pl–Fsp–Bt–Ms–Ky–Sill–Rt is interpreted to represent the peak-pressure metamorphic assemblage.

5.a. Staurolite–kyanite zone

Sample ZT206, which preserves a representative assemblage of the St–Ky zone, was selected for P–T estimates. Garnet is observed in two main textural positions: (i) as inclusions within staurolite and kyanite (Figure 2a) and (ii) as isolated porphyroblasts in the matrix (Figure 2e). All garnet grains display homogeneous compositions, with near-flat chemical profiles. Representative garnet compositions are: X_Prp = 0.08–0.11, X_Alm = 0.77–0.81, X_Grs = 0.04–0.07, and X_Sps = 0.05–0.08 (Fig. S1a–f, Supplementary Table S1). Plagioclase is sodic, with compositions Ab_65–75An_25–34Or_0–1 (Supplementary Table S2). Staurolite exhibits low X_Mg values between 0.14 and 0.17 (X_Mg = Mg/[Mg + Fe²⁺], Supplementary Table S5). Matrix biotite has X_Mg values between 0.46 and 0.48, with TiO₂ contents ranging from 0.09 to 0.11 apfu (Supplementary Table S4). White mica is muscovite, with Si contents from 3.02 to 3.11 apfu and Na = 0.14–0.20 apfu (Supplementary Table S3).

5.b. Kyanite–fibrolite zone

Sample ZT96, exhibiting the least retrogression, was chosen as representative for P–T estimation in the Ky–Fib zone. Garnet is compositionally homogeneous with the following ranges: X_Prp = 0.06–0.14, X_Alm = 0.65–0.68, X_Grs = 0.03–0.04, and X_Sps = 0.16–0.24 (Fig. S1g-j, Supplementary Table S1). Zoning is observed, with increasing spessartine and decreasing pyrope contents toward the rim (Fig. S1d). Plagioclase is rich in albite with compositions Ab_59–75An_16–40Or_0–3 (Supplementary Table S2). Biotite in the matrix has X_Mg = 0.52–0.56 and Ti = 0.002–0.118 apfu (Supplementary Table S4). Inclusions of biotite in garnet show slightly higher X_Mg values of 0.54–0.61 and similarly low Ti contents (0.002–0.118 apfu). White mica compositions range from 3.02 to 3.10 apfu Si (Supplementary Table S3).

6.a. Staurolite–Kyanite zone

The peak metamorphic conditions for sample ZT206 were determined using compositional isopleths of garnet (X_Grs = 0.05–0.07; X_Sps = 0.05–0.07) and staurolite (X_Mg = 0.14–0.17). The intersection of these isopleths defines a P–T field of 6–8 kbar and 600–640°C, within the stability fields of garnet + muscovite + biotite + staurolite ± kyanite + plagioclase + rutile + quartz (Figure 3a), which is consistent with petrographic observations described above.

Rutile in sample ZT206 is found both in the matrix, associated with quartz and plagioclase, and as inclusions within garnet, staurolite and kyanite. Zirconium concentrations (15 analyses) range from 155 to 326 ppm, with the lowest values in staurolite-hosted rutile and the highest in kyanite-hosted rutile. The average Zr content (239 ppm) yields a temperature of 618°C at 7 kbar, while the maximum value (326 ppm) corresponds to 643°C (Figure 4a). A pressure of 7 kbar was chosen based on the results of phase equilibrium modelling.

Figure 4.

Results of Zr-in-rutile thermometry.

6.b. Kyanite–fibrolite zone

P–T conditions for sample ZT96 were estimated from compositional isopleths of garnet (X_Grs = 0.03–0.04; X_Alm = 0.65–0.68) and white mica (Si = 3.02–3.06 apfu). These isopleths intersect within the stability field of biotite + garnet + kyanite + muscovite + plagioclase + quartz + rutile, defining P–T conditions of 6.5–8.5 kbar and 650–690°C (Figure 3b).

Rutile occurs both in the matrix (associated with muscovite, quartz and plagioclase) and as inclusions in kyanite. Zr concentrations (10 analyses) range from 274 to 370 ppm, with negligible differences between inclusion- and matrix-hosted rutile. The average Zr content (329 ppm) yields a temperature of 644°C at 7 kbar, and the maximum value (370 ppm) corresponds to 654°C (Figure 4b).

In summary, the combined results indicate peak metamorphic conditions of approximately 600–640°C at 6–8 kbar for the St–Ky zone, and 640–655°C at 6.5–8.5 kbar for the structurally higher Ky-Fib zone. These are consistent with previous P–T estimates based on conventional geothermobarometry (∼570–640°C and 6–7 kbar; Janák, Reference Janák1994).

7.a. Staurolite–kyanite zone

Monazite grains occur predominantly in the matrix (Figures 2d and 5a,b) and less frequently as inclusions in garnet, staurolite, kyanite or Fe–Ti oxides (Figures 5c–e). Grains are typically rounded to slightly elongated, with long axes ranging from ∼20 to 150 μm (average ∼50 μm). Inclusions within garnet are generally smaller (up to 8 μm; Figure 5e). Lower-strain samples (ZT219 and ZT206) contain larger monazite grains compared to the more deformed sample LZ2204. Irregular and jagged grain morphologies are occasionally observed (Figure 5c,d). Matrix-hosted monazite commonly occurs adjacent to biotite, muscovite and quartz, but is also found as inclusions in porphyroblastic feldspar and Fe-oxides. Allanite–apatite coronas around monazite are frequent in LZ2204 and rare in ZT206 and ZT219 (Figure 5f).

Figure 5.

BSE images of monazite grains in different mineral assemblages. The left panel (a-f) shows monazite from the St-Ky zone of the lower unit, while the right panel (g-l) shows monazite from the Ky-Fib zone. (a, b) Rounded monazite grains in matrix. (c) Monazite in kyanite. (d) Monazite in staurolite. (e) Monazite in garnet. (f) Matrix monazite surrounded by coronas of allanite and apatite. (g) Irregularly shaped monazite in muscovite. (h, i) Rounded monazite in matrix. (j) Monazite in kyanite. (k) Monazite at the boundary between biotite and FeOx. (l) Garnet with monazite inclusion, garnet also visible in Figure 2h.

7.b. Kyanite–fibrolite zone

Monazite in the Ky-Fib zone is predominantly matrix-hosted (Figure 5g–i), with rare inclusions in garnet and kyanite (Figure 5j,l). Monazite grains are rounded to slightly elongated, ranging from ∼15 to 100 μm (average ∼40 μm). Inclusions within garnet are up to 10 μm in size. As in the Ky–St zone, grain morphologies are variable and include concave or irregular forms (Figure 5g). No clear relationship was observed between the monazite texture (shape, position) and metamorphic grade. The only consistent trend is a decrease in grain size with increasing strain.

7.c. Monazite composition and zoning

Compositional X-ray maps for Y, Nd, Ca and Th were acquired using EMP to evaluate internal zoning patterns in monazite. Representative maps for all dated samples are shown in Figure 6, with compositional trends summarized in Figure 7. Variations in element distribution reflect controls from local bulk chemistry, mineral assemblage, P–T conditions and redox state (e.g. Catlos, Reference Catlos2013).

Figure 6.

High-contrast BSE images, Th, Y and Ca elemental maps of representative monazite grains from the lower unit. (a–c) Staurolite-Kyanite zone (St-Ky): Monazite grains exhibit pronounced zonation in Y and Th, with Y typically increasing from core to rim, while Th zonation is patchy. LA-ICP-MS (yellow circles) and EMPA (pink dots) analysis spots are marked. (d–g) Kyanite-Fibrolite zone (Ky-Fib): Monazite grains display similar zonation patterns, with Y-enriched rims and patchy Th distribution. Ca zonation is irregular, with Ca-depleted cores and Ca-rich rims in some grains. Analysis spots are indicated.

Figure 7.

(a, b) Y vs Th/U diagrams. Orange and blue circles refer to the St-Ky zone and Ky-Fib zone, respectively. The background shading indicates whether the analytical spot was located in the core (yellow) or rim (grey) of the monazite. Variation in Y content between the rim and core is visible but there is no correlation with the Th/U content. (c) Monazite compositional vectors showing dominant huttonite and cheralite substitutions.

8.a. Staurolite–kyanite zone

Monazite in samples from the St–Ky zone yield U–Pb ages ranging from 315 to 338 Ma (Figure 8; Supplementary Table 6). For sample ZT219, the Concordia age of 318.2 ± 5.7 Ma (n = 7; MSWD = 3.6) can be calculated. Sample ZT206 yields no obvious Concordia age. Instead, the monazite age data show rather large spread from 315 to 340 Ma. These results are interpreted to reflect the timing of monazite growth during peak to post-peak metamorphic conditions, likely influenced by fluid infiltration. Textural evidence, such as allanite–apatite intergrowths and coronas around monazite and chemical maps (Figure 6) revealing irregular Y and Th zoning in these monazites, are indicative of compositional modifications during fluid infiltration.

Figure 8.

Results of U–Pb monazite geochronology, presented in Wetherill Concordia diagrams. All uncertainties are reported at the 2σ level. The diagrams in the upper row represent monazites from the St–Ky zone, whereas those in the two lower rows represent monazites from the Ky-Fib zone.

8.b. Kyanite–fibrolite zone

Monazite grains from the Ky-Fib zone are consistently older, ranging from 342 to 333 Ma (Figure 8; Supplementary Table 6). Sample LZ2202 produced a Concordia age of 336.8 ± 1.9 Ma (n = 10; MSWD = 1.2), and sample MB2218 yielded a similar age of 332.8 ± 1.9 Ma (n = 12; MSWD = 1.7), both indicating robust and well-constrained age populations. In contrast, sample ZT96 yielded an older Concordia age of 342.3 ± 2.8 Ma (n = 6), but with a higher MSWD of 11, suggesting significant scatter likely due to Pb loss or post-crystallization disturbance. Sample MB2217B gave an age of 335.3 ± 4.0 Ma (MSWD = 5.2), also reflecting some degree of discordance. These results indicate that monazite growth in the Ky-Fib zone predates that in the underlying St–Ky zone, consistent with the observed inverted metamorphic gradient.