The metamorphic footprint of western Laurentia preserved in subducted rocks from southern Australia

Polymetamorphic metapelitic rocks in central‐west Tasmania, southern Australia, contain high‐pressure mineral assemblages that formed during Cambrian‐aged subduction and relict garnet with published Lu–Hf ages of c. 1285–1240 Ma. These garnet ages, along with published detrital zircon data from throughout western Tasmania and western North America, have been used to propose the presence of Mesoproterozoic Laurentian crust in western Tasmania. In this study, we combine zircon petrochronology with compositional information from the inclusion assemblages in relict garnet to extract Mesoproterozoic pressure–temperature data from subduction‐overprinted rocks, which effectively constitute an interpreted remnant of Laurentian crust now residing in central‐west Tasmania. The new data suggest Mesoproterozoic metamorphism involved two stages. The first event is recorded by c. 1480–1235 Ma zircon that formed in a garnet‐absent, plagioclase‐present, high‐thermal gradient environment at pressures no greater than ~5–5.5 kbar. The second event recorded by c. 1285–1240 Ma relict garnet was characterized by the development of a moderate‐pressure kyanite–plagioclase–biotite‐bearing mineral assemblage, which formed at ~8.5 kbar and ~590–680°C. These pressure–temperature constraints are attributed to extension within a deep basin system associated with the cryptic East Kootenay Orogeny in North America, which coincides with the final stages of c. 1450–1370 Ma upper Belt‐Purcell Basin sedimentation. Taking into account new detrital zircon U–Pb–Hf isotopic data from central‐west Tasmania in this study and existing zircon provenance data from throughout western Tasmania and the Belt‐Purcell Basin, our results strengthen the hypothesis of a Laurentian footprint that potentially encompasses much of western Tasmania and relates to both Nuna and Rodinian tectonism.


| INTRODUCTION
Reconstructing the tectonic framework of terranes requires a considered evaluation of the pressures, temperatures, and time at which mineral assemblages develop, achieved in part through the petrological investigation of metamorphic rocks (e.g., Brown, 2014;Spear et al., 2017).A continuing impediment to extracting useful thermobarometric information from crustal rocks is the potential for polymetamorphism and the development of superimposed mineral assemblages.Such assemblages have been documented in the Rayner Complex and Prydz Bay region of East Antarctica (e.g., Kelsey et al., 2007;Morrissey et al., 2016), the Ford Ranges in West Antarctica (e.g., Korhonen et al., 2012;Yakymchuk et al., 2015), the Arunta region in central Australia (e.g., Morrissey et al., 2011;Scrimgeour et al., 2005), the central Lewisian Complex in northwest Scotland (e.g., Zirkler et al., 2012), the western Bohemian Massif (e.g., Peřestý et al., 2017), the western Altai Range (e.g., Nakano et al., 2015), the Greater Himalayan Sequence (e.g., Johnson et al., 2021), and the central Appalachian orogen (e.g., Broadwell et al., 2019), with these representing just a few examples.Placing pressure-temperature (P-T) constraints on older metamorphic events can be challenging where rocks experience high degrees of mineralogical overprinting from later crustal reworking, but such constraints are necessary to place a given terrane into a more complete tectonic framework.Accessing reliable P-T information associated with older crustal processes is also necessary in the context of informing palaeogeographical models.Although isotopic information from detrital zircon has been used extensively to address questions concerning the ancestry of present-day geological terranes (e.g., Cawood et al., 2007;Gehrels, 2014), correlations of metamorphic events and associated P-T conditions between crustal domains provide a greater degree of confidence to such models (e.g., Cutts et al., 2013;Goodge & Fanning, 2016;Liu et al., 2022;Morrissey et al., 2019).
The geological histories of crust at subduction margins can provide insights into palaeogeographical links between now disparate terranes (e.g., Brown, Morrissey, et al., 2021;Wan et al., 2015).However, the development of mineralogical overprints due to subduction presents unique challenges for unravelling older P-T histories.Specifically, subduction environments are commonly hydrous, which promotes mineralogical reactivity during metamorphism, typically resulting in the partial to complete obliteration of precursor minerals (e.g., Brown et al., 2020;Proyer, 2003;Schorn, 2018;Young & Kylander-Clark, 2015).The potential to extract thermobarometric information from subduction-related polymetamorphic rocks is arguably increased if the rocks contain garnet, given garnet and the inclusions which it hosts commonly represent the sole recorder of P-T information associated with the growth of the relict mineral assemblage (e.g., Cutts et al., 2010;Dutch et al., 2010;Thiessen et al., 2019).
In central-west Tasmania, southern Australia, Mesoproterozoic-aged (c. 1285-1240 Ma) relict garnet occurs within metapelitic schists that were extensively overprinted during Cambrian-aged subduction and high-pressure metamorphism (Brown et al., 2022).The Mesoproterozoic ages for garnet-grade metamorphism in central-west Tasmania, combined with the comparable detrital zircon records from the Rocky Cape Group in northwest Tasmania and the Belt-Purcell Basin in western North America (Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015), are argued to represent compelling evidence for the preservation of Laurentian crust in southern Australia (Figure 1; Brown et al., 2022).If the Mesoproterozoic Tasmania-Laurentia connection hypothesis is correct, an obvious expectation is that these regions also preserve analogous metamorphic histories.However, despite the widespread record of Mesoproterozoic amphibolite to granulite facies metamorphism in western North America (e.g., Doughty & Chamberlain, 1996;Johnson, 2015;McFarlane, 2015), there is little quantitative P-T data from age-equivalent rocks in western Tasmania.Quantifying the P-T conditions of Mesoproterozoic metamorphism in western Tasmania and constraining the protolith age and affinity of the polymetamorphic rocks which record this early tectonic history offer a novel test of the Tasmania-Laurentia connection hypothesis.
Here, we use mineral equilibria modelling and compositional information from the inclusion assemblages of c. 1285-1240 Ma garnet porphyroblasts to extract information from highly overprinted subducted rocks to quantify the P-T conditions on Mesoproterozoic metamorphism in western Tasmania.Additionally, new petrochronological zircon U-Pb and rare earth element (REE) data are used to identify a second, early Mesoproterozoic metamorphic event that predates garnet in these rocks.Detrital zircon U-Pb-Hf isotopic data are also used to constrain the protolith ages of the metasedimentary rocks and demonstrate the central-west Tasmanian system has a heritage equivalent to the upper Belt-Purcell Basin-not unlike the Rocky Cape Group in northwest Tasmania (Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015).Based on these results, we hypothesize that an extensive proportion of western Tasmania shares a common Mesoproterozoic tectonic history with North America.
F I G U R E 1 (a) Schematic illustration of the Cambro-Ordovician East Gondwana margin showing the relative positions of Australia (AUS), Tasmania (TAS), and East Antarctica (EA).The Western Tasmanian Terrane microcontinent sits outboard of the margin with its eastern margin subducting beneath an island arc.(b) Relative positions of Australia, Tasmania, East Antarctica, and Laurentia (LA) in a Nuna to Rodinia framework (c.1.7-1.1 Ga).In (b), the bold dashed boundary indicates the extent of the Mawson Continent (MC), while the dashed line signifies its inferred extent.The Rocky Cape Group (RCG) and Belt-Purcell Basin (BP) are also shown.The schematic in (b) is modified after Goodge and Fanning (2016) and Goodge et al. (2017).(c) Simplified map of the Mesoproterozoic to early Palaeozoic geology of western Tasmania highlighting the locations of (1) the Rocky Cape Group, (2) the Franklin Metamorphic Complex, and (3) the Clark Group in Tasmania's Jubilee region.Inset shows the location of the Franklin Metamorphic Complex and the extent of the field area.Geology of the Franklin Metamorphic Complex and locations of the investigated samples are shown in a simplified map modified after Palmeri et al. (2009).Sample coordinates: À42.123472 , 145.855753 (FMC-1b), À42.123560 , 145.855377 (FMC-2c), and À42.122078 , 145.87123 (FMC-17).

| BACKGROUND
The Western Tasmanian Terrane includes southern Australia's western Tasmania and King Island and contains Mesoproterozoic strata unconformably overlain by Neoproterozoic strata and mafic volcano-sedimentary sequences (Figure 2; Berry et al., 2008).The Proterozoic strata of western Tasmania were metamorphosed and deformed during the middle to late Cambrian Tyennan Orogeny-an expression of broader subduction orogenesis along the early Palaeozoic East Gondwana margin (Figure 1a; Brown, Hand, & Morrissey, 2021;Meffre et al., 2000;Mulder et al., 2016).The Tyennan Orogeny involved the collision of the east-facing passive margin of the Western Tasmanian Terrane with an intraoceanic island arc, resulting in c. 515-500 Ma high-pressure metamorphism, emplacement of an ophiolite onto the continental margin, and ultimately the distribution of several metamorphic complexes across western Tasmania (Figure 2; Berry & Crawford, 1988;Berry et al., 2007;Brown, Hand, & Morrissey, 2021;Mulder et al., 2016;Palmeri et al., 2009;Turner & Bottrill, 2001).As a consequence of Tyennan orogenic and post-orogenic processes (the latter involving rapid exhumation of the high-pressure products and the development of syn-sedimentary normal faults and felsic extrusive magmatism; Mortensen et al., 2015;Noll & Hall, 2005) and subsequent overprinting during the Devonian Tabberabberan Orogeny, much of the Proterozoic history recorded in western Tasmania has been obscured (Berry et al., 2007;Brown et al., 2022).
The Proterozoic rocks of the Western Tasmanian Terrane record basin development and continental rifting associated with the breakup of supercontinent Nuna and the opening of the Pacific Ocean following Rodinia assembly (Berry et al., 2008;Halpin et al., 2014;Mulder et al., 2018Mulder et al., , 2020)).The lower middle Rocky Cape Group exposed in northwest Tasmania represents Tasmania's oldest basin system, preserving Mesoproterozoic (c.1450-1300 Ma), generally subgreenschist facies quartz arenite, siltstone, and mudstone (Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015).Lithologically, similar sequences are widely exposed in the Tyennan Region of central-west Tasmania and on King Island to the northwest of Tasmania (Figure 2; Black et al., 2004;Halpin et al., 2014).The age, stratigraphy, and detrital zircon provenance of the lower middle Rocky Cape Group are strikingly similar to the Lemhi and Missoula Groups in North America's upper Belt-Purcell Basin and the Marqueñas Formation in New Mexico (Daniel et al., 2013;Jones et al., 2011;Stewart et al., 2010), supporting a connection between the Western Tasmanian Terrane and western Laurentia within supercontinent Nuna (Figure 1b; Halpin et al., 2014;Morrissey et al., 2019;Mulder, Halpin, & Daczko, 2015).Disconformably overlying, the lower middle Rocky Cape Group are the late Mesoproterozoic strata of the upper Rocky Cape Group, which correlate with sedimentary rocks in the Jubilee region of southern Tasmania (Clark Group) and possibly late Mesoproterozoic strata in the foreland of the Grenville orogen in southwest North America-suggesting proximity between western Tasmania and southwest Laurentia within Rodinia (Mulder et al., 2018).
Despite detrital zircon dates suggesting a Laurentian provenance for northwest Tasmania, evidence for Mesoproterozoic-aged metamorphism in the Western Tasmanian Terrane is sparse compared with the evidence recorded in the potentially correlative rocks of the Belt-Purcell Basin.In the Surprise Bay Formation on King Island, monazite geochronology indicates metapelitic schists record 1287 ± 18 Ma greenschist to amphibolite facies metamorphism with conditions of $3 kbar and $500 C and associated folding (Berry et al., 2005;Calver et al., 2014).Low-pressure conditions are consistent with the garnet-andalusite-bearing assemblages, which are interpreted to have developed during the primary phase of deformation (Berry et al., 2005).The Mesoproterozoic metamorphism and deformation on King Island are apparently not recorded in northwest Tasmania where the Rocky Cape Group is generally subgreenschist facies and only preserves a cryptic history of Mesoproterozoic low-temperature fluid flow, which is recorded by authigenic monazite dates between c. 1345 and 1085 Ma (Halpin et al., 2014).A comparable record of Mesoproterozoic fluid flow is also widely documented in authigenic and low-temperature mineral assemblages throughout the Belt-Purcell Basin strata of North America (Aleinikoff, Lack, et al., 2012;Halpin et al., 2014).
A recent advance in understanding the Mesoproterozoic tectonic history of western Tasmania is the recognition of c. 1285-1240 Ma garnet-grade metamorphism in polymetamorphic metapelitic schists from the Cambrian Franklin Metamorphic Complex in central-west Tasmania (Figure 1c; Brown et al., 2022).These rocks contain an early generation of monazite that grew in a garnet-free environment between c. 1420 and 1280 Ma and may record an additional cryptic Mesoproterozoic metamorphic event (Brown et al., 2022).These metamorphic events correlate temporally with metamorphism, magmatism, and deformation in western North America, which is broadly referred to as the East Kootenay Orogeny (e.g., Nesheim et al., 2012;Zirakparvar et al., 2010).However, there are no constraints on the P-T conditions for the enigmatic Mesoproterozoic metamorphic events in western Tasmania that could provide additional support for the presence of western Laurentian crust in southern Australia.
F I G U R E 2 Mesoproterozoic to early Palaeozoic geology of the Western Tasmanian Terrane and King Island.Hill shaded relief map is publicly available and sourced from Esri.Regional geology is from Mineral Resources Tasmania.

| SAMPLE INFORMATION
Two metapelitic samples were collected from the Franklin Metamorphic Complex to investigate the Mesoproterozoic P-T conditions of central-west Tasmania (samples FMC-1b and FMC-2c).Additionally, a quartzite sample was collected for detrital zircon geochronology to constrain the protolith age of the metapelitic rocks in the region (sample FMC-17).The two metapelites were dated by Brown et al. (2022) using in situ garnet Lu-Hf geochronology.

| FMC-2c
Sample FMC-2c is a garnet-kyanite-muscovite-biotiteplagioclase-bearing mylonite.The strong schistose fabric is dominantly composed of fine-to medium-grained muscovite and quartz (0.1-0.6 mm; Figure 3b).The fabric wraps medium-to coarse-grained garnet porphyroblasts (0.5-2.5 mm; Figure 3h) which contain inclusions of apatite and zircon.Quartz inclusions armoured in the garnet porphyroblasts are also common (Figure 3i).The rims of porphyroblastic garnet are typically poikiloblastic overgrowths, containing abundant fine-grained acicular rutile (Figure 3h,k).The fabric also contains elongate, fine-grained garnet (<0.2 mm; Figure 3h-k).Finegrained garnet contains abundant inclusions of needlelike rutile and quartz (Figure 3k).At the margin of porphyroblastic garnet, fine-grained garnet is often observed (Figure 3h,k).Additionally, fine-grained biotite is commonly situated at the margin of both fine-grained garnet and fabric-defining muscovite (Figure 3h,j,k).Fine-grained needle-like kyanite grains (0.05-0.3 mm) also define the fabric and commonly occur as monomineralic foliated domains (Figure 3h,j).These domains may be bordered by fine-grained plagioclase grains (0.05-0.1 mm; Figure 3h,j).Fine-grained kyanite needles are observed at the margin of fine-grained garnet and they are also included within fine-grained plagioclase grains.Fine-grained rutile is preserved in alignment with the fabric, and the rest of the matrix contains minor apatite, graphite, monazite, and zircon.Monazite is also included within porphyroblastic garnet.

| Existing geochronology and interpretation of superimposed assemblages
Metapelitic samples FMC-1b and FMC-2c enclose metreto decimetre-scale mafic eclogites which were subducted in the Cambrian (e.g., Brown, Hand, & Morrissey, 2021)  hypothesis are Cambrian ages for monazite occurring parallel to the fabric in the metapelites (Chmielowski & Berry, 2012) and monazite included within fine-grained, fabric-parallel garnet (Brown et al., 2022).In sample FMC-1b, Neoproterozoic-to Cambrian-aged fine-to medium-grained garnet is located within a weak to moderate fabric and contains inclusions of rutile.Similarly, fine-grained garnet in sample FMC-2c is aligned with the fabric and contains abundant inclusions of rutile.In FMC-2c, the occurrence of fine-grained garnet within a fabric, which texturally post-dates Mesoproterozoic garnet and the observation of similar inclusion assemblages in fine-grained garnet in FMC-1b, strongly suggests finegrained garnet in FMC-2c formed during Cambrian-aged subduction.Furthermore, the matrix assemblages in FMC-1b and FMC-2c are relatively hydrous, containing varying proportions of muscovite and biotite (Figure 3).Therefore, it is conceivable that the matrix assemblages developed at some point during the Cambrian subduction cycle, as preservation of such assemblages through a subduction event is unlikely.Table 1 presents the interpreted Mesoproterozoic and Cambrian mineral assemblages in the metapelitic samples.

| Raman spectroscopy
Raman spectra of aluminosilicate inclusions in Mesoproterozoic-aged coarse-grained garnet were collected using a Horiba LabRam HR Evolution spectrometer (Jobin Yvon), equipped with an Olympus microscope, a Marzhauser XYZ controller, a Horiba Synapse cooled CCD detector, and a 100 mW, frequency-doubled, 532 nm Nd:YAG laser.The Raman spectrometer, housed at the University of Adelaide, was calibrated daily using a silica wafer.Spectra in the range of 400-1400 cm À1 were collected using a grating of 1800 g/mm.Three 5 s accumulations on a single spot were averaged to produce a spectrum for each inclusion.Data were acquired and processed using the Horiba LabSpec 6 Spectroscopy software.
Samples were 180 μm thick sections.A 100Â optical objective was used to locate aluminosilicate inclusions within garnet, with both reflected-and plane-polarized light used to determine whether inclusions were completely armoured or sat along fractures.In sample FMC-1b, six aluminosilicate inclusions armoured within coarse-grained garnet were observed across six thick sections.A representative Raman spectrum of an armoured aluminosilicate inclusion is provided in Figure S1 and imagery of the inclusion is shown in Figure 3d.Given the inclusion is very small ($30 μm), a minor proportion of garnet contamination is present in the Raman spectrum.Despite this, the highest per cent database match for the inclusion is kyanite (80%) from the RRUFF database (ID R050450; Figure S1).Aluminosilicate inclusions were not found in coarse-grained garnet from three thick sections of sample FMC-2c.
Zircons from FMC-17 were first analysed for U-Pb compositions using LA-ICP-MS at the University of Adelaide, before analysis for U-Pb-Hf compositions using split-stream/multicollector methods (LA (SS)-ICP-MS) at Monash University, Melbourne.U-Pb-Hf isotopic results for FMC-17 zircon and reference materials are given in Tables S1-S3.Additional details for all laser-ablation and ICP-MS instruments used in this study are provided in Table S4.
Zircons from both FMC-1b and FMC-2c were analysed for U-Pb-REE compositions using LA-ICP-MS at the University of Adelaide.U-Pb and REE results for FMC-1b zircon, FMC-2c zircon, and reference materials are given in Table S4.

| U-Pb-Hf (LA (SS)-ICP-MS)
Zircons were separated from quartzite sample FMC-17 using traditional techniques of crushing, milling, washing, and magnetic procedures.Zircon grains were handpicked and cast onto epoxy disks which were polished to approximately half-grain thickness.Zircon grains were imaged using back-scatter electron and cathodoluminescence (CL) methods at the University of Adelaide using an FEI Quanta MLA-600 scanning electron microscope.Representative CL imagery is given in Figure 4.

| U-Pb-REE (LA-ICP-MS)
Zircons were separated from samples FMC-1b and FMC-2c using the same techniques as outlined for sample FMC-17.Zircons were imaged using back-scatter electron and CL methods with an FEI Quanta MLA-600 scanning electron microscope (FMC-1b) and a Hitachi SU3800 scanning electron microscope (FMC-2c), both located at the University of Adelaide.Representative CL imagery is given in Figure 4.
Zircon U-Pb isotopes and REE compositions were acquired using a RESOlution LR 193 nm Excimer laserablation system and an Agilent 7900x ICP-MS at the University of Adelaide.Data were acquired over two sessions, with spot sizes of 19 μm (FMC-1b) and 29 μm (FMC-2c) used.For both samples, each analysis was undertaken with repetition rate of 5 Hz, a fluence of $2 J/cm 2 , and a total analysis time of 60 s, encompassing 30 s of background measurement and 30 s acquisition time.GJ-1 was utilized as the primary reference material for U-Pb isotopic corrections, yielding a 206 Pb/ 238 U weighted-average date of 601.8 ± 1.4 Ma (MSWD = 0.98, n = 56/59) for Session 1 and a 206 Pb/ 238 U weighted-average date of 601.8 ± 1.5 Ma (MSWD = 0.10, n = 65) for Session 2. Plešovice and 91500 zircon were used as secondary standards to test the accuracy of GJ-1.Plešovice gave a Zircon REE concentrations were calibrated to synthetic glass standard, NIST-610, which was analysed using 51, 43, and 19 μm spot sizes.Data reduction was undertaken using Iolite software, with the U_Pb_Geo-chronology4 and Trace_Elements_IS data reduction schemes for U-Pb isotopes and REE, respectively.Zr was used as the internal reference element for REE concentration calibration, with a stoichiometric concentration of 43.14 wt% used.

| Garnet REE (LA-ICP-MS)
LA-ICP-MS analysis for REE concentrations was undertaken on c. 1240 and c. 1285 Ma porphyroblastic garnet in samples FMC-1b and FMC-2c, respectively, using a RESOlution LR 193 nm Excimer laser-ablation system and an Agilent 7900x ICP-MS at the University of Adelaide.REE concentrations were obtained from garnet mounted into epoxy disks.A spot size of 43 μm, a repetition rate of 5 Hz, a fluence of $3.5 J/cm 2 , and a 70 s acquisition, with 30 s background measurement, were the operating parameters.Synthetic glass standard NIST-612 was used as the primary reference material (Jochum et al., 2011), ablated using a spot size of 51 μm.Glass standard GSD-1G was used as the secondary reference (Jochum et al., 2010), ablated using spot sizes of 51 and 43 μm.REE data were reduced using Iolite software (Trace_Elements_IS), and averaged electron microprobe concentrations of Al in wt% were used for REE corrections.REE results for FMC-1b garnet, FMC-2c garnet, and reference materials are given in Table S6.

| Electron probe microanalysis
Major element oxide compositions of solid-solution minerals were measured using a CAMECA SXFive electron microprobe at the University of Adelaide.Representative major element oxide compositions are provided in Table S7.Single-spot analyses were obtained using an accelerating voltage of 15 kV, a beam current of 20 nA, and a beam diameter of 5 μm.Calibration procedures were undertaken on a mixture of synthetic and natural standards from Astimex Pty Ltd, C.M. Taylor, P&H and Associates, and the National Museum of Natural History.The software Probe for electron probe microanalysis (EPMA) by Probe Software Inc. was used for data reduction.Qualitative major element X-ray compositional maps of garnet were obtained using both energy and wavelength dispersive spectrometers (EDS and WDS), an accelerating voltage of 20 kV, and a beam current of 200 nA.Dwell times and step sizes for mapping ranged between 30 and 40 ms and 2 and 4 μm, respectively.
The electron microprobe (CAMECA SXFive) was also used to acquire Ti (ppm) concentrations in quartz for the purpose of Ti-in-quartz thermometry.Results are provided in Table S8.Quartz inclusions exposed at the polished surface of Mesoproterozoic-aged coarse-grained garnet were targeted, with samples prepared as 30 μm thin sections, 180 μm thick sections, and epoxy discs.Inclusions located away from fractures (i.e., completely armoured within garnet) were identified using both reflected-and plane-polarized-light optical microscopy.
Both PET and LPET crystals on five tuneable wavelength dispersive spectrometers were used to count Ti Kα X-rays, with the intensities from all five spectrometers aggregated for optimal sensitivity.Ti Kα was calibrated using a synthetic Ti standard and corrections for continuum artefacts were undertaken using the quantitative blank correction method of Donovan et al. (2011).Blank corrections were performed using Spectrosil Glass 3 (Ti = 3.94 ± 0.33 ppm) with a specified SiO 2 concentration of 100% for matrix corrections.The operating conditions were a 20 kV accelerating voltage, a beam current of 200 nA, and a 'defocussed' beam with a diameter of 2 μm.Traditional two-pt background was performed using the alternating On/Off peak method, with a peak counting time of 60 s and background of 30 s. Eight alternating On/Off peak measurements were made for a total count time per spot of $1500 s.Individual measurements have a 2 ppm detection limit.To avoid secondary fluorescence effects, quartz inclusions in the vicinity of rutile and ilmenite were avoided entirely and multiple measurements were made on individual quartz grains to test consistency.In addition, secondary fluorescence boundary profiles were modelled using the Penepma module in Probe for EPMA.The low Ti content of the surrounding garnet (generally <100 ppm; Table S6) meant that the contribution of Ti signal to the quartz measurement from secondary fluorescence of enclosing garnet was on the order of sub ppm and not detectable for most larger quartz grains.Measurements yielding anomalous Ti (ppm) concentrations on individual spectrometers were interpreted as within-grain outliers and omitted (Table S8).Data were reduced using the Probe for EPMA software.Ti-in-quartz temperatures were calculated using the pressure-dependent calibration of Osborne et al. (2022), which is similar to the thermometer of Thomas et al. (2010) but is based on a wider range of experimental conditions.A TiO 2 activity less than 1 (0.75) was used for the temperature calculations given no Ti-bearing minerals were observed as inclusions in the coarse-grained garnet porphyroblasts.An aTiO 2 value lower than 0.75 was not considered plausible given the possibility that Tibearing phases were present in the Mesoproterozoic assemblages, as precited by mineral equilibria forward modelling (Section 5.4).Similar aTiO 2 values have been adopted by Glorie et al. (2023) and Ehrlich et al. (2012) for rutile-absent assemblages.

| Mineral equilibria forward modelling
Mineral equilibria modelling was undertaken using THERMOCALC (v.340) and the internally consistent data set ds62 (Holland & Powell, 2011).The activity-composition (a-x) models of White et al. (2014) were implemented for calculations in the MnNCKFMASHTO chemical system.The a-x models of Holland and Powell (2003) and Green et al. (2016) were implemented for plagioclase and clinopyroxene (jadeite), respectively.Mineral abbreviations in the phase diagrams are consistent with those in THERMOCALC.Mineral equilibria as a function of pressure and temperature were calculated using bulk-rock compositions determined using X-ray fluorescence spectrometry, undertaken by Bureau Veritas Minerals, Adelaide, Australia (Table 2).The bulk-rock compositions were corrected to account for the presence of minor tourmaline in FMC-1b and minor apatite in both samples-the compositions of which cannot be modelled using THERMOCALC.The corrections required the modal proportions of these minerals to be known.The modal proportions of all minerals were calculated using Mineral Liberation Analysis, with an FEI Quanta MLA-600 SEM (Table 3).Mineral modal proportions were converted from vol% to atom-normalized molar volume (mol%) to compare them to calculated proportions in THERMOCALC.The bulk-rock compositions were recalculated to atom-normalized molar volume for compatibility with THERMOCALC (Table 2).The oxidation states of the samples were determined from EPMA T A B L E 2 Calculated bulk-rock compositions given in both wt % and Mol%, the latter relevant to mineral equilibria calculations using THERMOCALC.

FMC-1b
FMC analyses of ferric iron-bearing minerals (Table S7).The Fe 2 O 3 /FeO ratios are 0.04 and 0.03 for FMC-1b and FMC-2c, respectively.Reduced compositions result from the fact that garnet is the primary host of Fe 3+ , with a small proportion from biotite in minor abundance.The H 2 O content in each sample was determined using P-M H2O models which were calculated between anhydrous (H 2 O = 1.5 wt%) and hydrous (H 2 O = 2.5 wt%) compositions (Section 5.4.1 and Figure 9).Based on the P-M H2O models, H 2 O concentrations of 1.6 and 2.2 wt% were determined for samples FMC-1b and FMC-2c, respectively (Section 5.4.1, Figure 9, and Table 2).

| Zircon U-Pb-Hf
Zircons in sample FMC-17 are elongated ($100-200 μm c-axis lengths), subhedral, and typically semiprismatic (Figure 4a-c).Terminations are mostly subrounded to rounded.Grains typically show oscillatory or broadly zoned cores, overgrown by narrow (<10 μm) rims with a mottled response in CL.Many grains have margins which appear altered or partially metamict.The morphologies of the zircon cores are consistent with a detrital origin (Corfu et al., 2003;Hoskin & Schaltegger, 2003).A total of 153 analyses were acquired from sample FMC-17 using LA-ICP-MS (Table S1), with 94 concordant analyses (i.e., ellipses intersecting concordia; Figure 5a).A total of 110 analyses were acquired also from previously unablated zircon grains from FMC-17 using LA (SS)-ICP-MS (Table S2), with 61 concordant analyses (Figure 5a).Probability density distribution plots for zircon analyses acquired using both methods are near-identical, with both showing a significant proportion of analyses with dates between c. 1910 and 1550 Ma and a primary peak comprising analyses with dates between c. 1500 and 1415 Ma.Zircons from these two age groups are semiprismatic and show oscillatory and broad zoning patterns (Figure 4a,b).Additionally, a small number of concordant zircons give dates between c. 2555 and 2505 Ma.These zircons exhibit ellipsoidal and rounded morphologies and are generally broadly zoned (Figure 4c).In the probability density distribution plots in Figure 5a, the analyses with concordant dates between c. 1500 and 1415 Ma (contributing to the youngest peaks) return a high MSWD of 2.20, indicating the analyses have excess scatter beyond their analytical uncertainties and therefore do not conform to a single population at the 95% confidence level.Assuming lead loss and omitting the youngest six analyses such that the probability of fit is 0.05, a 207 Pb/ 206 Pb weighted-average date of 1470 ± 5 Ma (MSWD = 1.40, n = 33/39) is returned.This is interpreted as the maximum depositional age.
Figure 5a also shows ε Hf data from detrital zircons in sample FMC-17 (Franklin Metamorphic Complex; Table S3).Only data from concordant U-Pb analyses are shown.Detrital zircons from the Franklin Metamorphic Complex show a range of ε Hf values between À9.13 and +10.36 (Table S3).Most zircons with dates between c. 1500 and 1415 Ma have positive ε Hf values between +4.32 and +9.03, indicative of magmatic zircon growth from melts with a significant mantle contribution (green cluster).Older zircons with dates between c. 1750 and 1550 Ma similarly show positive ε Hf values, although the spread is greater, ranging from +1.42 to +10.36 (yellow cluster).Zircons with c. 1900-1700 Ma dates have predominantly negative ε Hf values and few positive values between À2.36 and +1.38, consistent with crustal melts (grey cluster).
The detrital zircon age spectra, maximum depositional age, and ε Hf data from the Franklin Metamorphic Complex are similar to the lower middle Rocky Cape Group in northwest Tasmania and the Clark Group in southern Tasmania (Figure 5b; Black et al., 2004;Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015;Mulder et al., 2018), supporting an interpretation that these sedimentary sequences may be part of the same basin system.Mulder, Halpin, and Daczko (2015) pointed to the similarities in detrital zircon dates and ε Hf values between the lower middle Rocky Cape Group, the upper Belt-Purcell Basin, and the basement rocks from southwest Laurentia and the central Transantarctic Mountains in East Antarctica.This information has been re-emphasized in Figure S2 with the inclusion of the ε Hf data from the Franklin Metamorphic Complex.5.2 | Zircon U-Pb-REE

| FMC-1b
Zircons in sample FMC-1b have variable morphologies and internal features.Grains are typically subhedral and near-equant to elongated (less than $100 μm c-axis lengths; Figure 4d-f).Terminations are subrounded to rounded.Some grains have homogenous interiors, whereas others show core-rim relationships.Broad zoning is common, and faint oscillatory zoning is also observed.
A total of 318 analyses were acquired from FMC-1b zircon (Table S5), with 35 analyses omitted as they contained common lead ( 204 Pb ppm) concentrations greater than the detection limit.Zircon analyses with Al concentrations >500 ppm were also omitted as they were interpreted to reflect contamination from micro-inclusions.On a Wetherill concordia diagram (Figure 6a), the majority of analyses is discordant, with the most discordant analyses appearing to track along an array between c. 2400 and 1600 Ma.
Concordant zircon analyses fall into three populations that can be defined on the basis of age, composition, and zoning patterns (Figure 6a).The youngest of these (population 1) has a spread of dates between c. 545 and 500 Ma (n = 30; Figure 6c), with these analyses returning a 206 Pb/ 238 U weighted-average of 513 ± 3.2 Ma (MSWD = 3.90, n = 27/30; Figure 6c).The three oldest analyses are statistical outliers and therefore excluded from the weighted-average calculation, and the relatively high MSWD indicates an element of dispersion in the data.Zircon grains from population 1 have near-equant and elongated morphologies, with many grains preserving rounded margins consistent with a metamorphic origin (Figure 4d; Corfu et al., 2003;Hoskin & Schaltegger, 2003).The zircons also have either homogenous interiors or broad zonation.Most analyses are located at the near-rim and rim (Figure 4d).Chondritenormalized HREE trends for c. 513 Ma zircons (population 1) are near-flat (Figure 6c).Y (ppm) concentrations are between 125 and 465 ppm, Th/U values are between 0.004 and 0.015, and Yb/Gd values are between 1.38 and 26.67 (Figure 6f).Eu/Eu* values range from 0.083 to 0.620 (Table S5).
These analyses define a spread along concordia from c. 1480 to 1235 Ma, which could be a consequence of nonrecent lead loss (Figure 6d).As a result, a statistically coherent date could not be calculated for this zircon population.Analyses from population 2 typically come from weakly zoned cores or unzoned rims (Figure 4e).These morphologies and compositional zonation patterns differ from younger population 1 zircons.The low Th/U ratios, together with the CL imagery, are consistent with a metamorphic paragenesis for the population 2 zircons.
Zircons from the third population have significantly elevated LREE concentrations and prominent positive Ce anomalies, which are more pronounced than those preserved in younger zircons from populations 1 and 2 (Figure 6e).Chondrite-normalized HREE concentrations are elevated and positively sloped (Figure 6e), and Th/U values (0.108-0.751) and Y concentrations (375-1530 ppm) are greater than those in younger zircons (Figure 6f).Yb/Gd values are between 8.09 and 70.97 and much lower than those in population 2 zircons (Figure 6f), and Eu/Eu* values (0.017-0.243) are suggestive of strong negative Eu anomalies (Figure 6e and Table S5).These analyses define a spread along concordia from c. 2345 to 1435 Ma (n = 15; Figure 6e).Individual zircon grains from population 3 have morphologies and internal zoning features unlike zircons from populations 1 and 2, preserving largely near-equant to equant habits and irregular zoning, with occurrences of oscillatory and sector zoning (Figure 4f).Core-rim relationships are evident and grains are subrounded to rounded, indicative of surficial transport processes and typical of detrital zircon (Corfu et al., 2003;Hoskin & Schaltegger, 2003).

| FMC-2c
Zircons in sample FMC-2c are typically elongated, with subrounded to rounded terminations (Figure 4g).C-axis lengths do not exceed $100 μm.Grains commonly contain bright-CL, oscillatory zoned cores and comparatively dark-CL rims that are either homogenous or oscillatory zoned.Bright-CL outer rims are common but are too thin to analyse.
A total of 17 analyses were acquired from FMC-2c zircon cores (Table S5).On a Wetherill concordia diagram, analyses are largely discordant, with the few concordant analyses generally clustering at c. 1700 Ma (Figure 6b).Ellipses bordered by dashed lines contain common lead ( 204 Pb ppm) concentrations greater than the detection limit (Table S5).Chondrite-normalized REE concentrations are similar to those from c. 2345 to 1435 Ma population 3 zircons in sample FMC-1b, which are interpreted to be detrital (Figure 6e).Zircons from FMC-2c also have  S5).(c) 206 Pb/ 238 U weighted-average date for concordant analyses from population 1 (c.545-500 Ma, green ellipses).Unfilled ellipses with black borders were rejected from the calculation (513 ± 3.2 Ma, MWSD = 3.50, n = 27/30).The concordia showing concordant zircon analyses from population 1 is accompanied by an REE plot, normalized to chondrite values from Sun and McDonough (1989).REE trends indicate zircon growth at high pressures.(d) Concordant analyses from zircon population 2 (yellow ellipses).Analyses show a significant spread along concordia attributed to nonrecent lead loss.The accompanying chondrite-normalized REE plot for population 2 zircon indicates growth at comparatively low pressures with an absence of garnet in the system.(e) Concordant zircon analyses from population 3 (grey ellipses).The chondrite-normalized REE plot indicates population 3 zircon has magmatic signatures.(f) Plots of age against Th/U, Y (ppm), and Yb/Gd for all zircon populations.Eu/Eu* values are quoted throughout the text.The calculation for Eu/Eu* is provided in Table S5.

| Mesoproterozoic-aged garnet compositions
Representative major element oxide compositions and end-member proportions for solid-solution minerals in both FMC-1b and FMC-2c are provided in Table S7.Relative abundances of Fe, Mg, Ca, and Mn in Mesoproterozoic-aged garnet porphyroblasts from samples FMC-1b and FMC-2c are shown in qualitative X-ray maps (Figure 7a,b).In both samples, coarse-grained garnet is zoned in the major elements, although an undetermined degree of diffusional relaxation is apparent given the diffuse character of the compositional gradients.This is perhaps unsurprising in polymetamorphic samples and means using the garnet major element compositions to extract P-T information is not appropriate (Section 5.4).The possible exception is the garnet Ca composition in sample FMC-2c, with the sharp zoning in Figure 7b indicating diffusion-facilitated modification of Ca may have only been minor.
Figure 8 shows coarse-grained garnet REE rim-to-rim profiles and chondrite-normalized plots for both samples.Individual garnet zones (core, mantle, and rim) have been interpreted based on variations in HREE and MREE concentrations.Coarse-grained garnet has HREEenriched cores relative to the rims and HREE annuli at the mantle region (Figure 8a,b).Eu anomalies are negative and constant in magnitude (Eu/Eu*) across coarsegrained garnet (Figure 8a,b).

| Modelling approach and rationalization
To investigate the Mesoproterozoic metamorphic history of the polymetamorphic metapelitic samples, bulk-rock compositions determined using X-ray fluorescence spectrometry (Table 2) were favoured over compositions determined using other methods (i.e., EPMA bulk-rock compositions).This approach ensures the entire range of mineralogical compositional variability is captured and means the P-T history of the protoliths to the metapelites can be explored.The approach is also favoured given the samples show minimal evidence for retrogression and no evidence of melt loss (Figure 3), suggesting the bulk-rock compositions have not been modified.A significant uncertainty with respect to the bulk-rock compositions is the amount of H 2 O in the samples.It is conceivable the protoliths to the metapelites were metamorphosed under water-saturated (or near water-saturated) conditions during prograde burial in the Mesoproterozoic.This means F I G U R E 7 Qualitative electron microprobe X-ray compositional maps for coarse-grained garnet in (a) FMC-1b and (b) FMC-2c.Compositional maps are provided for Fe, Mg, Ca, and Mn.Warmer colours correspond to higher elemental abundances (counts).
that the amount of H 2 O residing in hydrous minerals preserved in the samples ($1.7 wt% H 2 O for FMC-1b and 2.2 wt% H 2 O for FMC-2c) is not a reliable indicator of the H 2 O environment in the Mesoproterozoic but instead relates to conditions during Cambrian subduction.The P-M H2O models (Section 4.5 and Figure 9) were calculated to test the above hypothesis of Mesoproterozoic water-saturated conditions.The P-M H2O models were calculated at a fixed temperature of 620 C given this temperature approximately satisfies the average Ti-in-quartz temperatures recorded by both FMC-1b and FMC-2c at arbitrary pressures of 7-8 kbar (Ti-in-quartz temperatures calculated from quartz inclusions armoured in Mesoproterozoic-aged garnet; Sections 4.5 and 5.4.2).For FMC-1b, the P-M H2O model does not stabilize kyanite at H 2 O concentrations above 1.7-1.9wt% between pressures of 6 and 12 kbar (Figure 9a).Given kyanite is included within Mesoproterozoic-aged garnet (Figure 3), a lower H 2 O content of 1.6 wt% was selected, which is consistent with the stabilization of other inferred Mesoproterozoic minerals, garnet, quartz, and plagioclase (Figure 9a and Table 1).In contrast, because kyanite inclusions were not identified in Mesoproterozoic-aged garnet from sample FMC-2c, a higher H 2 O content of 2.2 wt% which sits outside of kyanite stability was chosen for this sample (Figure 9b) and is in-line with the hypothesis of Mesoproterozoic water-saturated conditions.In the P-M H2O models for both samples, the respective modelled modal proportions of Mesoproterozoic-aged coarse-grained garnet intersect the selected H 2 O compositions (Figure 9).

| Thermobarometric constraints on Mesoproterozoic-aged metamorphism
As a consequence of Cambrian-aged subduction and mineralogical recrystallization during high-pressure metamorphism, the primary record of Mesoproterozoicaged metamorphism in samples FMC-1b and FMC-2c is relict coarse-grained garnet (Brown et al., 2022;this study).Therefore, c. 1285-1240 Ma coarse-grained garnet and its inclusion assemblages represent the only repository from which P-T information can be extracted to explore the metamorphic character of Mesoproterozoicaged tectonism.P-T pseudosections for samples FMC-1b and FMC-2c are presented in Figure 10.Simplified versions of the pseudosections are given in Figure 11.
For sample FMC-1b, the modal proportion of coarsegrained garnet is 8 mol%, which occurs in the model between pressures of $8 and 8.5 kbar at temperatures above 500 C and below the solidus (Figure 11a and Table 3).Garnet modal proportions are sensitive to pressure (Figure 11a).Given the sample is polymetamorphic and not all relict garnet may be preserved, a conservative interpretation is that the garnet modal proportion can be used to provide a minimum pressure estimate.At pressures of 8-8.5 kbar, Ti-in-quartz temperatures recovered from armoured quartz inclusions in coarse-grained garnet span $590-680 C (n = 23/26; Table S8).Below the solidus, the garnet 8 mol% contour intersects the range of temperatures recorded by the Ti-in-quartz thermometer at $8.5 kbar, which represents the minimum peak metamorphic pressure (Figure 11a).Kyanite is also stabilized within this region of P-T space, which is consistent with the observation of armoured inclusions in coarse-grained garnet (Figure 3d).Thus, the P-T pseudosection predicts the protolith to the high-pressure assemblage currently preserved in FMC-1b likely contained garnet + kyanite + biotite + muscovite + plagioclase + rutile + quartz ± ilmenite (Figure 11a).
For sample FMC-2c, the modal proportion of coarsegrained garnet is 3 mol%, with the modelled 3 mol% contour spanning pressures of $7-7.5 kbar at temperatures above 500 C and below the solidus (Figure 11b and Table 3).Temperatures calculated from Ti (ppm) concentrations in quartz inclusions armoured in coarse-grained garnet span a large range ($515-610 C at 7.5 kbar; n = 61/66; Figure 11b and Table S8).Like the P-T pseudosection for FMC-1b, garnet modes are sensitive to pressure (Figure 11b).Where the modelled proportion of coarse-grained garnet intersects the approximate maximum range of these temperatures and the associated 2σ uncertainty envelope, the pressure does not exceed 7.5 kbar, which is $1 kbar lower than the minimum pressure recorded by sample FMC-1b (Figure 11a,b).It is also evident that the average X grs composition recorded in the mantle region of FMC-2c coarse-grained garnet ($0.04) is modelled to occur at slightly higher pressures, near-consistent with those pressures recorded by FMC-1b.The significance of these results, and therefore the possible Mesoproterozoic P-T conditions of the protolith to FMC-2c, is explored further in Section 6.
F I G U R E 9 P-M H2O sections for (a) FMC-1b and (b) FMC-2c, calculated between anhydrous and hydrous compositions.The chemical system used for all calculations is MnNCKFMASHTO.V signifies the variance of each mineral assemblage.
This interpretation agrees with zircon petrochronological data from mafic eclogite enclosed by the Franklin Metamorphic Complex metapelitic schists which points to c. 500 Ma eclogitic zircon (Black et al., 1997;Brown, Hand, & Morrissey, 2021).The high-pressure zircon dates from the Franklin Metamorphic Complex are consistent with existing high-pressure dates from metapelitic rocks in western Tasmania (Berry et al., 2007;Brown, Hand, & Morrissey, 2021;Chmielowski & Berry, 2012;Fergusson et al., 2013;Turner et al., 1998).This high-pressure record is attributed to the subduction of the eastern continental margin of Tasmania beneath an intraoceanic arc during the Tyennan Orogeny (Figure 1a; Berry & Crawford, 1988).Although the character of Cambrianaged high-pressure metamorphism in western Tasmania has been relatively well-constrained (Brown, Hand, & Morrissey, 2021;Chmielowski & Berry, 2012;Mulder, Berry, & Scott, 2015;Palmeri et al., 2009), peering through this metamorphic event to unravel the thermobarometric character of Mesoproterozoic-aged metamorphism is more complex.

| Character of Mesoproterozoic-aged metamorphism in central-west Tasmania
While a middle Mesoproterozoic (c.1285-1240 Ma) metamorphic episode was recently identified from Lu-Hf age data from coarse-grained garnet (Brown et al., 2022), compositional data from zircon suggest the Franklin Metamorphic Complex may record an earlier, distinct metamorphic episode in the early Mesoproterozoic (c.1480-1235 Ma; Figure 6d,f).The thermobarometric character of both interpreted Mesoproterozoic-aged metamorphic events is explored further using information from mineral compositions and mineral equilibria forward modelling.

| Early Mesoproterozoic (c. 1480-1235 Ma) metamorphism
Zircons in FMC-1b have a range of dates from c. 1480 to 1235 Ma (population 2; Figure 6d), with the bulk of the data older than c. 1300 Ma.The large age range is potentially indicative of nonrecent lead loss attributable to c. 1285-1240 Ma garnet-grade metamorphism.If this interpretation is correct, this could mean the timing of zircon growth is closer to c. 1480 Ma; however, the dispersion in the zircon data precludes determination of a precise date.While a metamorphic paragenesis for these zircons is interpreted based primarily on their low Th/U ratios (0.002-0.064; Figure 6f; Rubatto & Gebauer, 2000), there is some uncertainty with this interpretation given the oldest zircons from this population overlap with the calculated maximum depositional age for quartzite in the Franklin Metamorphic Complex (c.1470 Ma; Section 5.1 and Figure 5a).However, the overlap in dates is not substantial and the vast majority of interpreted metamorphic zircons from population 2 gives dates younger than c. 1480 Ma (Figure 6d).Additionally, population 2 zircons have distinct internal zoning features and different compositional patterns compared with older, interpreted detrital zircons in both FMC-1b and FMC-2c (e.g., c. 2345-1435 Ma, population 3; Figures 4 and 6e), such as lower Th/U ratios, smaller positive Ce anomalies, and much larger Yb/Gd ratios (Figure 6d-f).Further to this, the compositions of population 2 zircons are relatively consistent compared with the interpreted detrital zircons which show a greater degree of variability, presumably as a result of being derived from multiple sources.
The elevated LREE and HREE concentrations and pronounced negative Eu/Eu* signatures (0.068-0.264) compared with Cambrian-aged zircon (Figure 6c,d,f and Table S5) suggest c. 1480-1235 Ma zircon formed at lower pressures relative to eclogite facies zircon, in a plagioclase-present and garnet-limited system (Rubatto, 2017).This supports the idea that metamorphism at c. 1480-1235 Ma was distinct from garnet-grade metamorphism at c. 1285-1240 Ma (Brown et al., 2022).To test this hypothesis, REE partition coefficients were calculated between the average composition of c. 1480-1235 Ma zircon and the average compositions of c. 1240 Ma garnet rim, mantle, and core (Figure 12 and Table S9).Independent zircon-garnet D HREE values are not available for amphibolite facies rocks, although extrapolation of natural and experimentally determined high temperature D HREE estimates to subgranulite facies temperatures (e.g., Rubatto & Hermann, 2007;Taylor et al., 2017) suggests that the zircon-garnet-core D HREE values in this study may approach a possible equilibrium pair (Figure 12).However, it is clear that zircon-garnet D HREE values for the garnet rim and mantle domains are F I G U R E 1 0 Calculated P-T pseudosections for the bulk-rock compositions of (a) sample FMC-1b and (b) sample FMC-2c.The dashed yellow line indicates garnet stable (up-pressure).The chemical system used for all calculations is MnNCKFMASHTO.Bulk-rock compositions are provided in Table 2. V signifies the variance of each mineral assemblage.
F I G U R E 1 1 Simplified P-T pseudosections for the bulk-rock compositions of (a) sample FMC-1b and (b) sample FMC-2c.The thick lightyellow contours signify the calculated modal proportions of coarse-grained garnet (FMC-1b = $8 mol%, FMC-2c = $3 mol%).The thin dashed yellow lines show how garnet proportions vary in the pseudosections.Average X grs compositions in the mantle regions of Mesoproterozoic-aged coarse-grained garnet are also annotated on the pseudosections.The range of Tiin-quartz temperatures and associated 2σ uncertainties for both samples is represented by the yellow shaded regions.
not consistent with those from the rock record or determined experimentally (7-10 kbar, 800-900 C; Harley & Kelly, 2007;Hermann & Rubatto, 2003;Hokada & Harley, 2004;Taylor et al., 2015).Therefore, there is a high probability that zircon formed at low pressures with minimal garnet influence in the system, especially considering the zircon REE patterns outline above (i.e., positively sloping HREE; Figure 6d).The D REE array plot in Figure 12 supports the interpretation that the early Mesoproterozoic metamorphic event had negligible garnet influence given the disagreement with 7 kbar experimental data from Taylor et al. (2015).
In the P-T pseudosection in Figure 10a, there are several modelled fields at comparatively low pressures which contain plagioclase and do not stabilize garnet.The majority of c. 1480-1235 Ma zircon analyses has Ti (ppm) concentrations below the analytical detection limit (Table S5), which precludes the use of Ti-in-zircon thermometry.Without temperature information from zircon, it is not possible to constrain the temperatures of early Mesoproterozoic mineral assemblage development.However, the maximum pressure associated with metamorphism related to the growth of c. 1480-1235 Ma zircon is analogous to the approximate maximum pressure in P-T space where garnet is not stable (Figure 11a).Therefore, mineral equilibria modelling indicates this metamorphic event did not exceed pressures of $5-5.5 kbar (Figure 11a).Pressures less than $5-5.5 kbar suggest metamorphism probably proceeded along high thermal gradients.Additionally, the low Th/U values in c. 1480-1235 Ma zircon suggest the coexistence of a high-Th accessory phase given the Th/U budget of metamorphic zircon is partly a function of accessory mineral partitioning (e.g., Rubatto, 2017).Monazite is present in both sample FMC-1b and other metapelitic lithologies in the Franklin Metamorphic Complex and returns similar dates between c. 1420-1280 Ma (Figure 13; Brown et al., 2022).Monazite is interpreted to have a metamorphic origin and similar to zircon; monazite growth occurred in a system that stabilized plagioclase but not garnet (Brown et al., 2022).The age and compositional similarities between zircon and monazite from Franklin Metamorphic Complex metapelitic rocks suggest they record ages related to the same interpreted high thermal gradient metamorphic event.Furthermore, HREE-enriched zircons from mafic eclogite in the Franklin Complex record similar ages (c.1495-1410 Ma; Figure 13), further supporting the occurrence of early Mesoproterozoic high thermal gradient metamorphism.These ages are interpreted as either metamorphic or magmatic (Brown, Hand, & Morrissey, 2021).The combination of mineral equilibria modelling and compositional data from relict coarse-grained garnet and its inclusion assemblages (i.e., Ti-in-quartz thermometry) constrains both the pressures and temperatures of younger middle Mesoproterozoic metamorphism (c.1285-1240 Ma), despite Cambrian-aged mineralogical overprinting.First, considering sample FMC-1b, the constrained P-T conditions of $8.5 kbar and $590-680 C intersect a kyanite-quartz-plagioclase-bearing modelled field (Figure 11a).While kyanite and quartz are both observed within coarse-grained garnet (Figure 3d), plagioclase inclusions have not been detected.However, Eu/Eu* values in c. 1240 Ma coarse-grained garnet (0.15-0.26) are indicative of moderate to strong Eu anomalies and show no substantial variation from rim to rim (Figure 8a).This implies garnet formed in an environment in which plagioclase remained stable (Warren et al., 2019).Additionally, coarse-grained garnet preserves REE growth zoning consistent with development as temperature increased (Figure 8a).Decreasing HREE concentrations from core to rim is likely a result of Rayleigh fractionation whereby progressive garnet growth depleted the local environment in HREEs (Otamendi et al., 2002).The prominent HREE annuli in the mantle region may be a consequence of resorption and HREE release during exhumation and subsequent garnet growth (Jedlicka et al., 2015).Another possibility is that the breakdown of accessory apatite and, to a lesser extent, monazite produced the HREE annuli during garnet growth (Gieré et al., 2011;Moore et al., 2013;Spandler et al., 2003).This is supported by the presence of apatite and Mesoproterozoic-aged monazite in this sample (Brown et al., 2022).Apatite is also found as inclusions in coarse-grained garnet.
Mesoproterozoic coarse-grained garnet in FMC-2c shows similar HREE trends to coarse-grained garnet in FMC-1b (Figure 8).However, constraining the middle Mesoproterozoic P-T conditions of the protolith to sample FMC-2c using the abundance of coarse-grained garnet and Ti-in-quartz thermometry is less straightforward than for sample FMC-1b given the larger range of calculated Ti-in-quartz temperature and the lack of diagnostic mineral inclusions.Compared with sample FMC-1b, which records middle Mesoproterozoic minimum pressures of $8.5 kbar and temperatures of $590-680 C (Figure 11a), sample FMC-2c appears to record Mesoproterozoic conditions characterized by lower temperatures and pressures (Figure 11b).Specifically, the P-T pseudosection predicts that the 3 mol% contour representing the proportion of coarse-grained garnet in FMC-2c intersects the approximate maximum range of Ti-in-quartz temperatures and the associated 2σ uncertainties at pressures of $7-7.5 kbar (Figure 11b).The lower Ti-in-quartz temperatures relative to those recorded in FMC-1b may relate to the capture of quartz inclusions during the earlier stages of garnet growth.This is supported by the observation that targeted quartz inclusions for Ti-in-quartz thermometry are located at the interpreted core-mantle region of coarse-grained garnet (Figure 3i).Eu/Eu* values across the length of coarse-grained garnet are indicative of moderately to strongly negative Eu anomalies (Figure 8b), which suggests coarse-grained garnet formed in equilibrium with plagioclase (e.g., Warren et al., 2019).Therefore, a burial path can be envisaged that tracks through plagioclase-stable P-T fields towards a point characterized by the intersection of the 3 mol% garnet contour and the highest recorded temperatures from Ti-in-quartz thermometry (Figure 11b).At this point, the minimum pressure is $1.5 kbar lower than that of FMC-1b ($8.5 kbar).
Although not a substantial pressure discrepancy, the simplest interpretation is that unlike sample FMC-1b, the proportion of coarse-grained garnet in sample FMC-2c underestimates the minimum pressure conditions reached by the protoliths to the metapelites.This could be a result of the uncertainties associated with determining garnet modal proportions.It is notable that the average X grs composition recorded in the mantle region of FMC-2c coarse-grained garnet ($0.04) is modelled to occur at pressures better approximating those recorded by FMC-1b (Figure 11), suggesting the Ca composition has not been greatly modified and potentially represents a superior indicator of peak metamorphic pressure in FMC-2c.This interpretation is supported by the sharp Ca zoning in garnet (Figure 7b).

| The case for metamorphosed Laurentian crust in southern Australia
The similar detrital zircon provenance of the lower middle Rocky Cape Group in northwest Tasmania and the upper Belt-Purcell Basin in western North America is used to propose a connection between much of presentday western Tasmania and western North America during deposition of the c. 1450-1370 Ma upper Belt-Purcell Basin (e.g., Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015).Deposition occurred in an extensional system associated with rifting between proto-Australia, East Antarctica, and western Laurentia within supercontinent Nuna (Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015;Pisarevsky et al., 2014).The new provenance data from the Franklin Metamorphic Complex  in this study suggest the protoliths to the rocks in central-west Tasmania correlate with those in the lower middle Rocky Cape Group in northwest Tasmania and the Fraser and Surprise Bay formations on King Island (Figure 5; Black et al., 2004;Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015) and the upper Belt-Purcell Basin of western Laurentia (Figures 5 and S2; Halpin et al., 2014;Jones et al., 2011;Mulder, Halpin, & Daczko, 2015;Stewart et al., 2010).Furthermore, the interpreted correlation between the upper Rocky Cape Group, the Clark Group in southern Tasmania, and the Unkar Group in Arizona further supports the notion that much of western Tasmania has a Laurentian heritage.The latter interpretation is based on both detrital zircon dates and ε Hf data (Figures 5 and S2;Mulder et al., 2018), which has been used to infer proximity between western Tasmania and southwest Laurentia within Rodinia (Mulder et al., 2018).Unlike the upper Rocky Cape Group strata from northwest Tasmania and correlatives in the Clark Group in southern Tasmania, the centralwest Tasmanian strata do not contain late Mesoproterozoic (<1300 Ma) detrital zircon.The absence of late Mesoproterozoic detrital zircon in central-west Tasmania suggests the protoliths to the strata preserved in presentday central-west Tasmania were deposited prior to the assembly of Rodinia.This suggests that central-west Tasmania represents a deeper and older part of the basin that has been structurally juxtaposed between shallower and younger parts of the basin (i.e., the upper Rocky Cape group in northwest Tasmania and the Clark Group in southern Tasmania; Figure 1c).
While existing provenance data point to a Laurentian source for the Mesoproterozoic strata in western Tasmania, the record of garnet-grade metamorphism suggests the central-west Tasmanian rock system is in fact a fragment of western Laurentia (Brown et al., 2022).Therefore, the new constraints on the timing and style of Mesoproterozoic-aged metamorphism recorded in Tasmania allow a comparison with known metamorphic events in the Belt-Purcell Basin of western North America.
Metamorphic Complex metapelitic schists (Figures 13 and 14b;Brown et al., 2022; this study), inferred to reflect high thermal gradient metamorphism at pressures less than $5-5.5 kbar (Figure 11a), could potentially be similar to the record of high thermal gradient metamorphism preserved in the Belt-Purcell Basin rocks.Supporting this interpretation is the age agreement between the tholeiitic Moyie sills in southern British Columbia (1468 ± 2 Ma; Anderson & Davis, 1995) and the potentially magmatic zircons in mafic eclogite from the Franklin Metamorphic Complex (1457 ± 16 Ma; Brown, Hand, & Morrissey, 2021).Interestingly, this record of high thermal gradient metamorphism and mafic magmatism is not documented in northwest or southern Tasmania.
Metamorphism associated with the cryptic East Kootenay Orogeny (c.1400-1255 Ma) Subsequent metamorphism, deformation, and magmatism recorded in the Belt-Purcell Basin from c. 1400 to 1300 Ma is broadly referred to as the East Kootenay Orogeny (Doughty & Chamberlain, 1996;McFarlane & Pattison, 2000).The effects of the East Kootenay Orogeny cover central-north Idaho, southern British Columbia, and western Montana (Nesheim et al., 2012), although the character of metamorphism and deformation, and therefore the tectonic setting and palaeogeographical significance of the East Kootenay Orogeny, remains unclear.This ambiguity reflects the strong Cordilleran overprint in the region (Zirakparvar et al., 2010) and the scarcity of exposed rocks in western North America that record evidence of the East Kootenay Orogeny (McFarlane, 2015).Despite the cryptic nature of the East Kootenay Orogeny, it has been proposed to reflect renewed expression of basin activity involving local bimodal magmatism (Figure 14c and Table S10; Aleinikoff, Lack, et al., 2012;Lewis et al., 2007;McFarlane, 2015), high-grade metamorphism and partial melting (Doughty & Chamberlain, 1996, 2008), garnet-grade amphibolite facies metamorphism (McFarlane, 2015;Nesheim et al., 2012;Zirakparvar et al., 2010), and more regional hydrothermal fluid flow (Aleinikoff, Hayes, et al., 2012;Aleinikoff, Lack, et al., 2012;Saintilan et al., 2017).Some workers interpret the East Kootenay Orogeny to have been characterized predominantly by basin subsidence and elevated geotherm extensional processes that were concurrent with ongoing subsidence and deposition of the c. 1450-1370 Ma upper Belt-Purcell Basin within an extensional setting (Doughty & Chamberlain, 1996;Zirakparvar et al., 2010).A zircon U-Pb date of 1379 ± 1 Ma from a mafic intrusive rock and a metamorphic zircon date of 1370 ± 2 Ma from a migmatitic metapelite in central-east Idaho is interpreted to reflect mantle-derived magmatism, partial melting of the lowest units of the basin, and high-grade metamorphism (Figure 14c; Doughty & Chamberlain, 1996, 2008;Evans et al., 2000;Stewart et al., 2010).Possible correlative magmatism in Tasmania is preserved on King Island where the basaltic protolith to amphibolite in the Surprise Bay Formation yields a magmatic zircon U-Pb age of c. 1370 Ma (Everard, 2022).Furthermore, c. 1370-1270 Ma dates from monazite, xenotime, and cobaltite in low-grade rocks from the Belt-Purcell Basin are similar to c. 1345-1265 Ma authigenic monazite dates from the correlative lower middle Rocky Cape Group in northwest Tasmania-all of which are indicative of regional hydrothermal fluid flow (Figure 14a,c; Aleinikoff, Hayes, et al., 2012;Aleinikoff, Lack, et al., 2012;Halpin et al., 2014).Deformation in the correlative Rocky Cape Group was likely syn-sedimentary, occurring during riftfill sedimentation as evidenced by extensional structures (Everard et al., 2007;Halpin et al., 2014).
The extensional model for the East Kootenay Orogeny is contentious, in part due to the cryptic nature of its metamorphic record and the lack of mesoscale and macroscale structures that can be confidently tied to the event.Doughty and Chamberlain (1996) constrained conventional thermobarometric pressures and temperatures of $4.5-6.5 kbar and $680-690 C for c. 1370 Ma highgrade metamorphism and partial melting of metapelitic migmatites intercalated with mafic intrusions in centraleast Idaho (Figure 14c).Consistent with the extensional model, this episode of magmatism and metamorphism is interpreted to reflect rifting, basin deepening, and basal magma injection (Doughty & Chamberlain, 1996).In addition to the evidence for basin subsidence and highgrade metamorphism in central-east Idaho, metapelitic garnet-staurolite-aluminosilicate-bearing schists in other parts of the Belt-Purcell Basin record low-to mediumpressure and medium-temperature amphibolite facies metamorphism during the East Kootenay Orogeny (McFarlane, 2015;McFarlane & Pattison, 2000;Nesheim et al., 2012;Zirakparvar et al., 2010).In northern Idaho, c. 1380 Ma garnet-grade metamorphism is interpreted to reflect extension and basin deepening (Figure 14c; Zirakparvar et al., 2010).In contrast, c. 1350-1255 Ma garnet in nearby staurolite-aluminosilicate-bearing rocks from northern Idaho and isoclinal folds in the region (Lang & Rice, 1985;Reid et al., 1981) is interpreted to document a period of contractional deformation and crustal thickening (Nesheim et al., 2012).Garnet-staurolite-kyanite-bearing rocks in this area record peak metamorphic conditions of $5-9 kbar and $570-645 C (Johnson, 2015).Interestingly, deformation characterized by tight folding is also observed in the Surprise Bay formation on King Island, with c. 1290 Ma garnet-andalusite-bearing schists recording lower P-T conditions of $3 kbar and $500 C (Figure 14a; Berry et al., 2005).Evidently, the record of c. 1285-1240 Ma garnet-grade metamorphism with P-T conditions of $8.5 kbar and $590-680 C in central-west Tasmania (Figures 11a and 14b; Brown et al., 2022; this study) is similar in terms of age and metamorphic style to the record in northern Idaho (Johnson, 2015;Nesheim et al., 2012).The significance of garnet-grade metamorphism recorded in central-west Tasmania is discussed further in the next section of the discussion.
There is also a record of Mesoproterozoic shortening and crustal imbrication in southern British Columbia.Coupled with a 1365 ± 10 Ma monazite U-Pb date and constrained P-T conditions of $3.5-4.5 kbar and $610 C from a garnet-sillimanite schist, this record is ascribed to transpressive-transtentional tectonics and crustal thickening (Figure 14c;McFarlane, 2015;McFarlane & Pattison, 2000).Amphibolite facies metamorphism and interpreted compressional deformation in British Columbia at this time is coeval with the emplacement of voluminous granodiorite of apparent active margin geochemical character (McFarlane, 2015;Soloviev, 2011).However, structural evidence for extensional tectonics and shearing in southern British Columbia, which overlaps in timing with pegmatite emplacement and the development of retrograde mineral assemblages between c. 1335 and 1320 Ma, ultimately points to contrasting deformational regimes existing in southern British Columbia and northern Idaho around c. 1330 Ma (McFarlane, 2015;Nesheim et al., 2012).
Late Mesoproterozoic (c.1200-1050 Ma) metamorphism Younger, late Mesoproterozoic metamorphism and deformation is also recorded in the Belt-Purcell Basin.Low-pressure hydrothermal events are interpreted from xenotime (U-Pb) and arsenopyrite (Re-Os) dates from northern Idaho and western Montana spanning c. 1220-1060 Ma (Figure 14c; Aleinikoff et al., 2015;Arkadakskiy et al., 2009).These dates are similar to an authigenic monazite date of 1085 ± 9 Ma from the Rocky Cape Group (Figure 14a; Halpin et al., 2014).The record of late Mesoproterozoic metamorphism in the Belt-Purcell Basin is primarily from c. 1200 to 1065 Ma garnet in metapelitic rocks from northern Idaho (Figure 14c; Nesheim et al., 2012;Zirakparvar et al., 2010), and there are also late Mesoproterozoic zircon dates from xenolithic gneisses located in northern Yukon, Canada (Milidragovic et al., 2011).The late Mesoproterozoic metamorphic record has been interpreted to reflect tectonism related to the assembly of Rodinia (Zirakparvar et al., 2010).Additionally, titanite dates between c. 1090 and 1030 Ma from the Moyie sills in southern British Columbia have an inferred metamorphic origin and demonstrate that the effects of late Mesoproterozoic tectonism in western North America extend north into Canada (Anderson & Davis, 1995).

| Significance of high thermal gradient and garnet-grade metamorphism in centralwest Tasmania
Four general conclusions can be drawn from the information synthesized in the previous section: 1. Considering provenance data from much of presentday western Tasmania and both western and southern North America, a strong case can be made for a Laurentian source for many of the Mesoproterozoic successions in western Tasmania (Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015;Mulder et al., 2018; this study).2. The record of metamorphism and deformation in western North America during deposition of (i) the lower Belt-Purcell Basin (c.1470-1450 Ma), (ii) the upper Belt-Purcell Basin and during the East Kootenay Orogeny (c.1450-1300 Ma), and (iii) during late Mesoproterozoic tectonism (c.1200-1050 Ma) has striking similarities with the Mesoproterozoic-aged metamorphic record in western Tasmania.3.These similarities suggest high thermal gradient, garnet-grade, and hydrothermal metamorphic events documented in southern Australia are the same events as documented in western North America, indicating that metamorphosed Mesoproterozoic-aged crust in southern Australia is likely relict western Laurentian crust, metamorphosed before and during the cryptic East Kootenay Orogeny.4.Although there is evidence supporting both interpretations for the tectonic character of the middle Mesoproterozoic East Kootenay Orogeny (i.e., that it was characterized by either rifting and extension or crustal thickening and convergence), extension is likely the more probable model, as discussed below.
In more detail, the deposition of the Belt-Purcell Basin-and by extension, deposition of Mesoproterozoic western Tasmanian strata-occurred during the separation of Laurentia from proto-Australia and the Mawson Continent during the breakup of supercontinent Nuna (Halpin et al., 2014;Mulder, Halpin, & Daczko, 2015;Pisarevsky et al., 2014).Therefore, the record of pre-East Kootenay c. 1400 Ma high thermal gradient metamorphism preserved in central-west Tasmania and western North America (Aleinikoff, Hayes, et al., 2012;Brown et al., 2022;Doughty & Chamberlain, 2006, 2008;McFarlane, 2015;Zirakparvar et al., 2010; this study) was conceivably a consequence of rifting associated with Nuna breakup.Despite contrasting rifting and crustal thickening tectonic models put forward for the East Kootenay Orogeny, the evidence for the rifting model is more abundant and compelling, with extension-related metamorphism and deformation occurring concurrently with the final stages of deposition of the upper Belt-Purcell Basin (Doughty & Chamberlain, 1996;Zirakparvar et al., 2010) and evidence for post-sedimentation extension and fluid flow events preserved in southern British Columbia and east-central Idaho, respectively (Aleinikoff, Lack, et al., 2012;McFarlane, 2015).Additionally, c. 1350-1240 Ma amphibolite facies metamorphism involving garnet development at crustal depths of $30 km ($8.5 kbar) need not be attributed to crustal thickening, given that medium-pressure metamorphic terranes can develop within deep ($20-30 km), intracratonic rift basins (Tucker et al., 2015).Indeed, the presently preserved thickness of the Belt-Purcell Supergroup (≥20 km; Lydon, 2007) and Rocky Cape Group (>10 km; Halpin et al., 2014) indicates that medium-pressure amphibolite facies conditions could be achieved in the lower parts of these basins without significant structural thickening.Therefore, it is likely that this record, preserved in both central-west Tasmania and northern Idaho (Nesheim et al., 2012;this study), is an expression of extensional or transtentional tectonics (McFarlane, 2015).This implies that the margin of western Laurentia, and possibly southern Australia, was undergoing extension following the apparent breakup of supercontinent Nuna, with some estimates between c. 1500 and 1300 Ma (e.g., Evans & Mitchell, 2011;Pisarevsky et al., 2014;Zhang et al., 2012;Zhao et al., 2004), and during the apparent assembly of supercontinent Rodina, with some estimates between c. 1300 and 900 Ma (e.g., Goodge et al., 2008;Kee et al., 2019;Li et al., 2008).Supporting this hypothesis are recent palaeomagnetic data suggesting that Laurentia and Australia were stable as part of the Nuna core (e.g., Pisarevsky et al., 2014) until c. 1300 Ma and disassembled between c. 1300 and 1200 Ma (Kirscher et al., 2020).

| CONCLUSIONS
This contribution demonstrates that even in extensively reworked terranes, P-T information can be obtained from relict garnet and its inclusion assemblages, despite its residence in younger hydrous mineral assemblages formed during subduction.This was achieved through targeted compositional analysis complimented by a detailed petrological investigation, mineral equilibria forward modelling, and petrochronology.The extracted P-T information from superimposed mineral assemblages allows an evaluation of crustal tectonometamorphic processes which are otherwise cryptic in nature.Specifically, the thermobarometric conditions of two Mesoproterozoic metamorphic events were identified in the western Tasmanian relict assemblages, with these P-T conditions consistent with a high thermal gradient rift-related extensional setting, potentially evolving to more of a transtentional setting with basin deepening.The Mesoproterozoic Tasmania-Laurentia connection proposed by previous detrital zircon studies is supported by the newly recognized Mesoproterozoic metamorphic record in western Tasmania, which is similar in terms of age and style to the metamorphic record in western North America.Linking these metamorphic records has implications for the breakup and assembly of supercontinents Nuna and Rodina, respectively, with the results from this study providing evidence for the breakup of Nuna later than c. 1300 Ma.

F
I G U R E 4 (a-f) Representative cathodoluminescence images of zircon from samples FMC-17 and FMC-1b.In (a)-(c), (e), and (f), annotated dates are 207 Pb/ 206 Pb and, in (d), are 206 Pb/ 238 U. (g) Cathodoluminescence images for all zircons recovered from sample FMC-2c.Annotated dates are 207 Pb/ 206 Pb.Refer to Figures 5 and 7 for zircon age results.

F
I G U R E 6 Zircon LA-ICP-MS results for samples FMC-1b and FMC-2c.(a) Wetherill concordia plots showing all analyses and emphasizing concordant analyses.In the concordia showing concordant analyses, 2σ uncertainty ellipses have been colour-coded based on age.(b) Wetherill concordia plot showing all analyses from sample FMC-2c.Ellipses with dashed borders have common Pb ( 204 Pb ppm) concentrations above the detection limit (Table

F
I G U R E 8 LA-ICP-MS rim-to-rim profiles and chondrite-normalized REE plots for (a) FMC-1b coarse-grained garnet (c.1240 Ma) and (b) FMC-2c coarse-grained garnet (c.1285 Ma).Normalization after Sun and McDonough (1989).Calculated Eu anomalies (Eu/Eu*) are shown for every analysis across the diameter of each garnet.F I G U R E 9 Legend on next page.

F
I G U R E 1 2 REE partition coefficients between the average composition of population 2 zircon (c.1480-1235 Ma) and individual analyses from the core, mantle, and rim of c. 1240 Ma coarse-grained garnet in sample FMC-1b.REE partition coefficients are compared with those from experiments and the rock record.(1) Hermann and Rubatto (2003), (2) Hokada and Harley (2004), (3) Harley and Kelly (2007), and (4) Taylor et al. (2015).Array plot with D Yb/Gd against D Yb for the D REE between the average composition of population 2 zircon and individual analyses from the core, mantle, and rim of c. 1240 Ma coarse-grained garnet.Experimentally determined (at 7 kbar) D Yb/Gd and D Yb from Taylor et al. (2017).

F
I G U R E 1 3 Summary of early to middle Mesoproterozoic ages from central-west Tasmania (Franklin Metamorphic Complex).

F
I G U R E 1 4 Summary of Mesoproterozoic metamorphic (blue), magmatic (pink), and hydrothermal (yellow) ages from (a) northwest Tasmania and King Island, (b) central-west Tasmania (Franklin Metamorphic Complex), and (c) the Belt-Purcell Supergroup in western North America.Geology of the Rocky Cape Group and King Island sourced from Mineral Resources Tasmania.Franklin Metamorphic Complex geology after Palmeri et al. (2009).Belt-Purcell Supergroup geology after Höy et al. (1995), Vuke et al. ( Observed and inferred Cambrian-and Mesoproterozoic-aged mineral assemblages for the metapelitic samples.
T A B L E 3 Calculated mineral modal proportions.
Note: mol% refers to atom-normalized molar volume, which is equivalent to the modal proportions calculated in THERMOCALC.