Constraining the Timing of Evolution of Shear Zones in Two Collisional Orogens: Fusing Structural Geology and Geochronology

: In recent decades, constraining the timing of shear activity has been one of the main topics of research about the tectono-metamorphic evolution of orogenic belts. We present a review of a combined structural and geochronological approach to two major ductile regional shear zones, in two collisional orogens: the ﬁrst one affecting the Variscan basement in northern Sardinia (Italy) and the External Crystalline Massifs of the Alps (East Variscan Shear Zone; EVSZ), and the second one de-forming the medium- to high-grade rocks of the metamorphic core of the Himalaya (High Himalayan Discontinuity). High-resolution, texturally and chemically controlled monazite geochronology ap-plied in separated shear zones of the Variscan belt allowed recognizing a similar timing of activity ranging between c. 340–330 and 300 Ma. This approach led to a better understanding of the evolution of the EVSZ, supporting a model where several branches were active according to a growth by linkage model. Following a similar approach, in situ U-Th-Pb analysis of monazite constrained the timing of top-to-the-S/SW shearing of a regional-scale High Himalayan Discontinuity in the Himalayan belt to between c. 28 Ma and 17 Ma. Earlier exhumation of the hanging wall was triggered by shear zone activity, whereas at the same time, the footwall was still experiencing burial with increasing P-T conditions. The timing of shearing of this shear zone ﬁts with an in-sequence shearing tectonic model for the exhumation of the Himalayan mid-crust.


Introduction
Dating deformation events in tectonites and therefore unraveling the tectonic and structural evolution of the deformation of the lithosphere in different geodynamic settings are major challenges in structural geology [1][2][3]. Crosscutting relationships among structures such as folds, foliations and shear zones, magmatic bodies, unconformities and the stratigraphic age of the deformed sequence, as well as the age of sedimentary rocks deposited during or after deformation and magmatism, allow relative dating of major deformation events in the deformed lithosphere [4,5]. However, deformed rocks often show several episodes of superposed deformation and metamorphism such that the use of basic geological criteria often does not allow for an adequate temporal resolution. To overcome this problem, radiometric dating of mineral (re)crystallization within different fabrics has been widely used in folded and sheared rocks to better unravel the deformation history (e.g., [6,7]). In addition, in situ dating techniques enable retention of the structural positions of the minerals with respect to the different deformation structures to be dated. This approach allows full use of textural information regarding the relationships between rock-forming minerals and fabrics.   [20]; Corsini and Rolland, 2009 [21]; Guillot and Ménot, 2009 [22]); (B) inferred   [20]; Corsini and Rolland, 2009 [21]; Guillot and Ménot, 2009 [22]); (B) inferred lateral relationships between the branches of EVSZ during the late Carboniferous (modified from Simonetti et al., 2020 [23]

Monazite Behavior during Metamorphism: A Synthesis
For a successful application of monazite geochronology to solve geological problems, an evaluation of the monazite-in reaction, as well as monazite growth/resorption compared with the tectono-metamorphic evolution of the study sample, is required [25][26][27]. It is fundamental to investigate the chemical changes of both rock-forming minerals and monazite (re)crystallization with respect to the P-T-D evolution of the studied rock ( Figure 3).
Monazite may display different internal foliations analogous to porphyroblasts [27] or asymmetric growth tails during shearing [28]. A close link between the monazite age and chemical domain has often been reported, and rare-earth element (REE) + Y concentrations, coupled with U and Th zoning and variations in Eu anomalies, have been used as proxies to link monazite ages to different phases of metamorphism [2,26,29,30]. The behavior of monazite during metamorphism and partial melting has been extensively investigated both on the basis of empirical and experimental studies ( [12] and references therein). Consequently, despite being far from complete, we have a reasonable understanding of monazite behavior and its chemical signature. These studies have shown that the distribution of Th, Y and heavy rare-earth elements (HREE) is strongly controlled by the coupled growth/dissolution of coexisting phases such as garnet, xenotime, apatite and/or allanite [12,25,31]. Common reactions forming monazite in the metamorphic assemblage during a prograde history of metapelites are related to the breakdown of allanite at~450-550 • C, especially in relatively Ca-rich pelites (e.g., [32]). Garnet is regarded as the main silicate phase controlling the Y + HREE budget during monazite growth (e.g., [25,32]). Consequently, the initial growth of monazite ( Figure 3) is expected to show higher Th and HREE + Y contents due to the breakdown (up to the complete consumption) of allanite and/or xenotime [33]. During continuing prograde metamorphism (Figure 3), together with garnet growth, coexisting monazite should show lower Y + HREE concentrations [32,33].   If, during prograde growth, garnet-consuming reactions occur (e.g., the appearance of staurolite), monazite domains should show an increase in Y + HREE. During partial melting, the high solubility of phosphate in the peraluminous melt results in monazite dissolution (Figure 3), whereas during crystallization of the melt by cooling and concomitant garnet breakdown, precipitation of high-Y + HREE monazite is expected (e.g., [26,30,39,40]). Variation in Eu anomalies coupled with the Ca/Sr ratio, in a suprasolidus condition, can further help to link monazite ages to melt-present and/or feldspar . Simplified (and idealized) sketch showing the possible P-T-D evolution of a metamorphic tectonite vs. monazite behavior/zoning. Note that several complexities (e.g., fluids, different bulk compositions) can modify this picture. T range for quartz dynamic recrystallization mechanisms from Law (2014) [34] and Stipp et al. (2002) [35] (BLG-bulging recrystallization; SGR-subgrain rotation recrystallization; GBM-grain boundary migration recrystallization). Mineral-in curves: Als + Kfs + melt-in and Mnz-in = Kohn (2016) [10]; Grt-in and Ky-in = Tinkham et al. (2001) [36] (MnNCKFMASH system); Aln + Chl-in = Spear (2010) [37]; Xtm-out = Spear and Pyle (2010) [32]. Mineral abbreviations as in Whitney and Evans (2010) [38].
If, during prograde growth, garnet-consuming reactions occur (e.g., the appearance of staurolite), monazite domains should show an increase in Y + HREE. During partial melting, the high solubility of phosphate in the peraluminous melt results in monazite dissolution (Figure 3), whereas during crystallization of the melt by cooling and concomitant garnet breakdown, precipitation of high-Y + HREE monazite is expected (e.g., [26,30,39,40]).
Variation in Eu anomalies coupled with the Ca/Sr ratio, in a suprasolidus condition, can further help to link monazite ages to melt-present and/or feldspar growth/dissolution stages [41][42][43]. During further cooling, subsolidus HREE + Y-rich monazite (and xenotime) can be formed at the expense of the ongoing garnet (and fluid)-consuming reactions. From this summary, it is evident that the correlation of microstructural positions alone is no longer sufficient to interpret the tectonic significance of monazite ages [10]. Since the relative growth of rock-forming minerals (e.g., garnet, feldspars) or the crystallization of the melt can be "predicted" within the equilibrium thermodynamic framework (i.e., pseudosections), it is possible, in principle, to link monazite chemical fingerprints to a particular stage of the P-T-D path (e.g., [41][42][43]). Moreover, recent advances in experimental and theoretical work have also tried to include monazite in pseudosection modeling that, despite far from being perfect, is really promising (e.g., [31,32,44,45]).

Geological Setting of the Southern Variscan Belt
Between c. 380 Ma and 280 Ma [6,[46][47][48][49], the continent-continent collision involving Laurentia-Baltica and Gondwana ( Figure 1) led to the final assemblage of Pangea and to the development of the Variscan belt. It is represented by a 1000 km-wide and 8000 km-long mountain system that extends from the Caucasus to North America [49]. The main debate concerning the Variscan collision in Europe is about the presence or absence of microplates between Laurentia-Baltica and Gondwana [49,50].
The portion of the belt exposed in Central and Western Europe is known as the Variscides. The architecture of the belt in this sector has been defined by many authors [49,[51][52][53][54]: the Saxothuringian Zone (southwestern Iberian Massif, northern Armorican Massif, Vosges and Bohemian Massif) and the Moldanubian Zone (southern Iberian Massif, central and southern Vosges, Massif Central, southern Armorican Massif and southern Bohemian Massif) ( Figure 1). The latter is considered as the metamorphic root of the belt [55].
The south-eastern sector of the Variscides underwent reworking and fragmentation during the Alpine Orogeny. In the present day, the correlation between the now dispersed sectors of the belt is still debated and is mostly based on lithological and stratigraphic affinities and paleomagnetic data ( [74][75][76][77] and references therein).

The East Variscan Shear Zone
The EVSZ should be seen as a network of interconnected transpressive shear zones that display a similar tectono-metamorphic evolution [62]. In the present day, evidence of the EVSZ is represented by the shear zones occurring in the dispersed fragments of the Variscan belt in the Mediterranean area (Table 1), such as in the Corsica-Sardinia Block, in the Maures Massif in Southern France and in the Alpine External Crystalline Massifs ( Figure 1).
Asinara Shear Zone (PASZ; Figures 1, 4a, 5A and 6a,b; Table 1), located between the High-Grade Metamorphic Complex (HGMC) and the Low-to Medium-Grade Metamorphic Complex (L-MGMC), has been well documented by several authors [6,67,[78][79][80][81]. The transpressive deformation linked to the PASZ activity was recently investigated in detail by Carosi et al. (2020) [67] and Graziani et al. (2020) [82] in the Baronie region.  Transpression affected the boundary between the HGMC and L-MGMC after collision-related deformation and metamorphism (D1 phase) [67,79]. An earlier HP event occurred in the HGMC ( [83,84] with references), in metabasites of the PASZ [85] and in metapelites a dozen kilometers south of the PASZ [83]. The PASZ developed under decreasing temperature and pressure, from amphibolite facies conditions down to greenschist facies conditions [82][83][84] in the time span of ~25 Ma [66,67]. The amphibolite facies is confirmed by the presence of syn-kinematic sillimanite + biotite or biotite + white mica parallel to the mylonitic foliation [66,67]. This is also in agreement with the occurrence of GBM as the main dynamic recrystallization mechanism of quartz, indicative of temperatures higher Grade Metamorphic Complex (HGMC) and the Low-to Medium-Grade Metamorphic Complex (L-MGMC), has been well documented by several authors [6,67,[78][79][80][81]. The transpressive deformation linked to the PASZ activity was recently investigated in detail by Carosi et al. (2020) [67] and Graziani et al. (2020) [82] in the Baronie region.  Transpression affected the boundary between the HGMC and L-MGMC after collision-related deformation and metamorphism (D1 phase) [67,79]. An earlier HP event occurred in the HGMC ( [83,84] with references), in metabasites of the PASZ [85] and in metapelites a dozen kilometers south of the PASZ [83]. The PASZ developed under decreasing temperature and pressure, from amphibolite facies conditions down to greenschist facies conditions [82][83][84] in the time span of ~25 Ma [66,67]. The amphibolite facies is confirmed by the presence of syn-kinematic sillimanite + biotite or biotite + white mica parallel to the mylonitic foliation [66,67]. This is also in agreement with the occurrence of GBM as the main dynamic recrystallization mechanism of quartz, indicative of temperatures higher Transpression affected the boundary between the HGMC and L-MGMC after collisionrelated deformation and metamorphism (D1 phase) [67,79]. An earlier HP event occurred in the HGMC ( [83,84] with references), in metabasites of the PASZ [85] and in metapelites a dozen kilometers south of the PASZ [83]. The PASZ developed under decreasing temperature and pressure, from amphibolite facies conditions down to greenschist facies conditions [82][83][84] in the time span of~25 Ma [66,67]. The amphibolite facies is confirmed by the presence of syn-kinematic sillimanite + biotite or biotite + white mica parallel to the mylonitic foliation [66,67]. This is also in agreement with the occurrence of GBM as the main dynamic recrystallization mechanism of quartz, indicative of temperatures higher than c. 500 • C. The presence of chlorite testifies to a decrease in the metamorphic conditions, supported by incipient SGR overprinting GBM in quartz [82].
Geosciences 2022, 12, x FOR PEER REVIEW 8 of 28 than c. 500 °C. The presence of chlorite testifies to a decrease in the metamorphic conditions, supported by incipient SGR overprinting GBM in quartz [82]. The PASZ displays a strong deformation gradient and strain partitioning, with the strain intensity increasing towards the N and a dextral top-to-the-W/NW sense of shear [66,67]. The PASZ developed under conditions of decreasing temperature, starting in amphibolite and evolving in greenschist facies conditions (D2 deformation phase [67]; D2 early and D2 late deformation phases [82]), as suggested by the changes in the syn-kinematic mineral assemblage and the shift in quartz recrystallization mechanisms [67,82]. The kinematic vorticity data highlight a transpressional deformation regime where the simple shear component of the flow increases along the deformation gradient (i.e., toward the N). The age of the PASZ has been constrained by monazite U-Th-Pb petrochronology [67]. Several monazite generations in different microstructural positions (along the mylonitic foliation or included in porphyroclasts) and with different chemical features and zoning were recognized. The first group is composed of grains with low-to medium-Y cores grown during the prograde D1 stage. The second group is composed of newly formed grains and rims around older monazite grains (that are not included in porphyroclasts) with a high Y content grown during the D2 phase. The last group includes low-Y rims formed during D3 around monazite grains that are not shielded by other phases. Analyses of monazite domains grown during the D2 transpression revealed that shearing along the PASZ started at ~325 and lasted up to ~300 Ma (Table 1). The PASZ displays a strong deformation gradient and strain partitioning, with the strain intensity increasing towards the N and a dextral top-to-the-W/NW sense of shear [66,67]. The PASZ developed under conditions of decreasing temperature, starting in amphibolite and evolving in greenschist facies conditions (D2 deformation phase [67]; D2 early and D2 late deformation phases [82]), as suggested by the changes in the syn-kinematic mineral assemblage and the shift in quartz recrystallization mechanisms [67,82]. The kinematic vorticity data highlight a transpressional deformation regime where the simple shear component of the flow increases along the deformation gradient (i.e., toward the N). The age of the PASZ has been constrained by monazite U-Th-Pb petrochronology [67]. Several monazite generations in different microstructural positions (along the mylonitic foliation or included in porphyroclasts) and with different chemical features and zoning were recognized. The first group is composed of grains with low-to medium-Y cores grown during the prograde D1 stage. The second group is composed of newly formed grains and rims around older monazite grains (that are not included in porphyroclasts) with a high Y content grown during the D2 phase. The last group includes low-Y rims formed during D3 around monazite grains that are not shielded by other phases. Analyses of monazite domains grown during the D2 transpression revealed that shearing along the PASZ started at~325 and lasted up to~300 Ma (Table 1).
In the Variscan portion of the MTM (Southern France; Table 1), the Cavalaire Fault (CF; [86]), a ductile shear zone, separates the High-Grade Complex to the east from the Medium-Grade Complex to the west ( Figure 5B). The basement underwent HP conditions  [86] found that, in the transition zone of the Maures Massif, where the CF is localized, a retrograde metamorphism occurred between~600 • and 400 • C. This is testified by miscrostructural observations such as the presence of GBM in quartz and syn-tectonic mineral assemblages along the mylonitic foliation such as Bt + Sill. Subsequently, deformation evolved under decreasing temperature conditions [62,86,87] testified by the partial and local overprint of GBM by SGR and by the growth of white mica. The CF is characterized by the presence of a mylonitic belt, similar to the PASZ, at the boundary between the External Zone and the Internal Zone (Figures 1, 5B and 6c,d). Recent kinematic vorticity analysis performed on sheared rocks of the CF revealed general shear conditions with a major component of pure shear acting together with simple shear under a dextral transpressional deformation regime [62]. The shear zone evolved under decreasing temperature conditions starting from amphibolite facies. To constrain the time of activity of the CF, several monazite generations in different microstructural positions (along the mylonitic foliation or included in porphyroclasts) and with different chemical features and zoning were analyzed. Monazite grains commonly show complex Y zoning [62]. In particular, three main compositional domains have been recognized: (1) high-to medium-Y cores; (2) [23] with references). In the alpine basement, HP rocks also occur. They are mainly mafic eclogites embedded within migmatites. HP metamorphism occurred at~341 Ma in the Argentera Massif (U-Pb on zircon [90,91]) and around~345 Ma (U-Pb on zircon [92]) down to~335-330 Ma (U-Pb on zircon and rutile [93]) in the Aiguilles Rouge Massif.
The FMSZ is a dextral shear zone (Figure 7a,b) and separates two migmatitic complexes: the Gesso Valle Stura Vesubiè to the NE and the Tiné Complex to the SW. The FMSZ shows a subvertical attitude and extends for about 20 km with a variable thickness between 100 m and more than 1000 m. Deformation along the FMSZ started shortly after peak metamorphism in the migmatite [94]. The FMSZ evolution is characterized by a type II growth [95] under different thermal stages recognized by quartz c-axis thermometers, microstructures and syn-tectonic mineral assemblages. The shear zone started at a temperature above~600 • C and evolved under decreasing temperature down to~400 • C.   [96] recognized that the FMSZ shows a well-developed deformation gradient, with the strain intensity increasing from the wall rocks towards the center of the shear zone along with a gradual transition from protomylonite up to ultramylonite. Quartz opening-angle thermometers and syn-tectonic mineral assemblages suggest a decreasing temperature during deformation from HT amphibolite facies down to greenschist facies [94]. Kinematic vorticity analysis revealed a transpressional deformation with a prevalent pure shear component and a variable component of simple shear along the deformation gradient [63,94] (Table 1)   In order to unravel the timing of deformation, monazite grains, in different microstructural positions (included in porphyroclasts or along the mylonitic foliation), from the sheared rocks of the FMSZ were charachterized [63,94]. The Y content and zoning revealed that grains along the mylonitic foliation usually have a low-Y core, developed during the prograde stage of metamorphism, surrounded by a high-Y rim formed during syn-shear retrometamorphism. Some grains have an asymmetrical shape with the high-Y domains developed in the extensional sites around the grain. Differently, newly formed syn-tectonic grains present a homogeneous high-Y composition. Monazite included within porphyroclasts was shielded, and no high-Y rims are present. The ages obtained from syn-tectonic high-Y domains and grains revealed that shearing started at~340-330 Ma in the protomylonites at the margins of the FMSZ. From mylonites, ages of~320 Ma were obtained, suggesting a progressive concentration of the deformation in the core of the shear zone [63,94].
In the Aiguilles Rouge Massif, a detailed study of the EBSZ [23,97] revealed that it is a subvertical, high-strain mylonitic zone, developed in metasediment and orthogneiss, accommodating a pure-shear-dominated transpressional deformation ( Figures 5D and 7c,d).
The deformation temperatures of the EBSZ were recently determined using a quartz c-axis opening-angle thermometer (temperature between~630 • C and~550 • C [23]) supported by microstructures and syn-tectonic mineral assemblages along the mylonitic foliation. The data suggest that deformation occurred under amphibolite facies conditions. Subsequently, deformation was active under lower temperature conditions even if it is difficult to clearly separate the low-T activity of the Variscan Shear Zone and the Alpine overprint/reactivation [23]. The age of deformation of the EBSZ was obtained by U-Th-Pb monazite petrochronology. According to the microstructural position, dated monazites are divided in two groups: the first group is represented by grains included in porphyroclasts (garnet or biotite), and the second group includes grains located along the mylonitic foliation [23]. The two groups differ in terms of chemical characteristics: the Y content is low in monazite included in porphyroclasts, while it is higher in grains located along the main foliation. From high-Y syn-tectonic monazites along the mylonitic foliation, ages between 320 Ma and 310 Ma were obtained, suggesting that shearing started at~320 Ma [23] ( Table 1).
Therefore, the timing of shearing along the PASZ (Figure 8a), the CF (Figure 8b), the FMSZ ( Figure 8c) and the EBSZ (Figure 8d) is well constrained thanks to detailed data obtained from in situ texturally controlled U-(Th)-Pb geochronology on monazite [23,62,63,66,67,94]. The microstructural position, texture, zoning and chemical analyses highlight a complex history of monazite growth during the tectono-metamorphic history. Several generations of grains and chemical domains have been recognized. In general, monazite cores and grains included within porphyroclasts preserve older ages, between c. 340 and 333 Ma (Figure 8), that can be linked to the early collisional stage (between c. 380 Ma and 330 Ma; [98]) during the nappe stacking and prograde metamorphism. On the contrary, the growth of monazite rims around older cores or the growth of new homogeneous grains along the mylonitic foliation is linked to the decompression and retrogression stages that occurred during the transpressional phase of deformation. From those chemical domains and grains, ages of shearing between c. 340-330 Ma and 320 Ma have been obtained (Figure 8). Even if a detailed and complete study is not available at the moment, evidence of transpressional deformation has also been documented in the Belledonne Massif by Grandjean et al. (1996) [99], recognizing a transpressional deformation regime dated to 324 ± 12 Ma (K-Ar on hornblende), in agreement with the chronological data of the branches of the EVSZ. The chronological data obtained from in situ high-resolution monazite geochronology allow better constraining the progressive evolution over time of different branches of shear zones within a crustal-scale system-the EVSZ-in agreement with a growth by linkage model [94].

Geological Setting of the Himalaya
The Himalayan belt (Figure 2) developed following the collision between the and Eurasian Plates at ~59 Ma [100]. The tectonic units derived from the northern Plate were stacked, building up the highest mountains in the world (Figure 2). metamorphosed rocks occur in the Greater Himalayan Sequence (GHS), a 3-30 km sequence of medium-to high-grade metasedimentary and meta-igneous rocks. Th part of the GHS is composed of garnet-to kyanite-bearing paragneiss and micaschi subordinate calc-schist, quartzite, impure marble and migmatite. The middle por the GHS is composed of calcsilicate gneiss and marble with minor pelitic and psa

Geological Setting of the Himalaya
The Himalayan belt (Figure 2) developed following the collision between the Indian and Eurasian Plates at~59 Ma [100]. The tectonic units derived from the northern Indian Plate were stacked, building up the highest mountains in the world (Figure 2). Deeply metamorphosed rocks occur in the Greater Himalayan Sequence (GHS), a 3-30 km-thick sequence of medium-to high-grade metasedimentary and meta-igneous rocks. The lower part of the GHS is composed of garnet-to kyanite-bearing paragneiss and micaschist with subordinate calc-schist, quartzite, impure marble and migmatite. The middle portion of the GHS is composed of calcsilicate gneiss and marble with minor pelitic and psammitic rocks. The uppermost part of the GHS consists of orthogneiss and aluminosilicate-bearing migmatitic rocks. The GHS is intruded by Oligo-Miocene leucogranites [24]. The Lesser Himalayan Sequence (LHS) and Tethyan Himalayan Sequence (THS) lay, respectively, structurally below and above the GHS. These sequences were formerly deposited on the Indian passive margin [101].
The GHS is bounded by a northward-dipping wide shear zone, the Main Central Thrust (MCT), separating it from the LHS [102,103], and by the South Tibetan Detachment System (STDS) separating it from the upper THS [104][105][106][107] (Figure 2). High-strain zones occur within the GHS and have been classically interpreted as out-of-sequence thrusts, affecting the GHS after a southward motion of the MCT ( [108] and references therein).
The discovery of the contemporaneous activity of the normal-sense STDS at the top of the GHS and of the reverse-sense MCT at its base [106,109] had major consequences for formulating tectonic models for the exhumation of the mid-crust and led to the proposal of the following models: (i) channel flow [110], (ii) wedge extrusion [109,111], (iii) channel flow followed by wedge extrusion ( [112,113] with references), (iv) wedge insertion [114] and (v) critical taper wedge [115]. All these models, except for (v), require the contemporaneous activity of the MCT and STDS to exhume the GHS between~23 and 16 Ma ( [112,116] with references).

High Himalayan Discontinuity in Western and Central Himalaya
The GHS records deformation associated with prograde and retrograde metamorphic events ( [117][118][119] with references). Deformation is heterogeneously partitioned, and portions affected by folds and related foliations are separated by portions affected by discrete shear zones. The prominent fabric in the GHS is a transpositive foliation formed during a second deformation phase (D2; [120][121][122]). The S2 foliation strikes NW-SE and dips 30-60 • toward the NE. It is marked by the preferred orientation of medium-to high-grade metamorphic minerals and recrystallized quartz ribbons. Kyanite, staurolite, muscovite and biotite are occasionally bent or kinked along shear bands. Sillimanite is frequently stretched within shear bands. Grain and aggregate lineations [4] (L2) trend NE-SW and plunge NE moderately to steeply (20 • , 60 • ). The S1 foliation is preserved as a relict in S2 microlithons or as an internal foliation in porphyroblasts.
The HHD has been studied in detail in Nepal and NW India (Figures 2, 9 and 10) following a structural, petrological and petrochronological approach ( Table 2).
Two main shear zones have been mapped in western Nepal at the same structural level in the GHS (Figures 2, 9 and 11): the Toijem [120,123] and the Mangri [116,120,124] Shear Zones. The mylonite thickness varies from a few dozen meters up to 4 km. In central Nepal, a shear zone has been recognized at the same structural level in the Kali Gandaki valley, named the Chomrong "Thrust" [125,126] (Figures 2 and 10a; Table 2). These three shear zones are localized in the lower portion of the GHS, close to the boundary between migmatite and kyanite-bearing gneiss, with hanging wall rocks showing a higher degree of melting than the footwall rocks. These structures are interpreted to be strands of the same high-temperature shear zone and are referred to as the HHD, showing a regional-scale occurrence [116,127].
In Garhwal Himalaya (NW India), the Badrinath Shear Zone mainly affects sillimanitegarnet-bearing gneiss of the GHS (Figures 2 and 10a; Table 2), showing former evidence of partial melting. The shear zone shows a hectometric thickness [128]. Shear sense indicators in the HHD, such as C-S fabric, shear bands, asymmetric tails around porphyroclasts, mica fish, asymmetric boudins and rotated garnets, indicate a top-to-the-S/SW sense of shear (Figure 11a-f). Shear planes dip to the N/NE, and the related object lineation plunges to the N/NE; these are inferred to be contractional shear zones. Quartz in mylonite shows evidence of high-T grain boundary migration (GBMII) recrystallization with often chessboard extinction [34]. Feldspar shows evidence of ductile deformation by undulose extinction, deformation twins and myrmekite. Deformation microstructures in the mylonite suggest a high-temperature deformation regime (T > 650 • C; [4]), in agreement with the syn-kinematic growth of sillimanite + biotite.     The Badrinath Shear Zone rocks followed a clockwise path achieving a maximum pressure of 1.0 GPa and then re-equilibrated at peak conditions around 700-720 • C and 0.65-0.8 GPa (Figure 12). In situ U-Th-Pb dating on zoned monazites revealed that the Badrinath Shear Zone followed a prograde metamorphic path from 34 to 24 Ma, followed by decompression at c. 23 Ma in the sillimanite stability field, which continued at least until 19 Ma [128]. The Badrinath Shear Zone was responsible for a difference in pressure conditions (c. 0.2-0.3 GPa) and led to the onset of exhumation (c. 3 Ma earlier), between the upper GHS and the lower GHS [128].
The activity of the Mangri Shear Zone (western Nepal; Figure 9, Table 2) divided the GHS into two portions: an upper and a lower one [124]. It was associated with the decompression in the sillimanite stability field of the upper GHS rock package, showing relicts of relatively high-pressure metamorphism within the kyanite stability field. P-T conditions indicate that the final pattern due to the Mangri Shear Zone is that the footwall rocks record a higher pressure (1.0-0.9 GPa) than the hanging wall (0.7 GPa), at similar temperatures (675 • -700 • C) ( Figure 12). In situ LA-ICP-MS U-(Th)-Pb dating of zoned monazite in metapelites constrained the shearing related to retrograde high-Y monazite growth to between c. 25 and 18 Ma. Samples from the lower GHS experienced ductile shearing related to retrograde zoning in monazite (high-Y rims) later, only at <17-13 Ma. These ages and the associated P-T-t paths reveal that peak metamorphic conditions were reached c. 5-7 myr later in the footwall of the shear zone with respect to the hanging wall [121,124], pointing to a diachronism in the metamorphism triggered by the occurring Mangri Shear Zone. What is important here is to stress that, due to the shear zone activity, when the footwall rocks experienced prograde metamorphic conditions, associated with burial, the hanging wall rocks experienced decompression.   In western Nepal, the activity of the Tojiem Shear Zone (Figures 2 and 9; Table 2) started at c. 26 Ma, constrained by in situ SHRIMP (sensitive high-resolution ion microprobe) monazite dating, coupled with LA-ICP-MS (laser ablation inductively coupled plasma mass spectrometry) trace element analyses on monazite in mylonites of the shear zone [120]. Monazite inside garnet exhibits older ages between c. 43 and 30 Ma that are related to prograde metamorphism during and following the collisional stage. The end of the shearing has been constrained by the age of an undeformed leucogranite dyke to c. 17 Ma, crosscutting the foliation. Geotermobarometry suggests that the Tojiem Shear Zone developed after the metamorphic peak, during the retrograde path of the sheared rocks: pressures of c. 0.9 GPa are recored in the footwall and 0.7 GPa in the hanging wall, with similar temperatures of 675-700 • C. Therefore, such a situation is comparable to the Mangri Shear Zone.
shearing related to retrograde zoning in monazite (high-Y rims) later, only at <17-13 Ma. These ages and the associated P-T-t paths reveal that peak metamorphic conditions were reached c. 5-7 myr later in the footwall of the shear zone with respect to the hanging wall [121,124], pointing to a diachronism in the metamorphism triggered by the occurring Mangri Shear Zone. What is important here is to stress that, due to the shear zone activity, when the footwall rocks experienced prograde metamorphic conditions, associated with burial, the hanging wall rocks experienced decompression.  In the Kali Gandaki valley (central Nepal), in situ LA-ICP-MS monazite dating in kyanite-bearing migmatitic paragneiss allowed constraining the activity of the Chomrong Thrust (Figures 2 and 10b) to c.   (Table 2). Zoned monazites in Y and Th revealed that sheared rocks underwent prograde metamorphism, reaching partial melting within the kyanite stability field (710-720 • C/1.0-1.1 GPa) at c. 43 to 28 Ma. Melting occurred, during prograde metamorphism close to the maximum pressure, between c. 36 and 28 Ma [129]. When the Chomrong Thrust was active (c. 25 and 18 Ma), these rocks changed their P-T trajectory and underwent decompression associated with pervasive hightemperature shearing, reaching P-T conditions of 650-670 • C and 0.7-0.8 GPa [125,130] ( Figure 12).

Discussion and Conclusions
The study of shear zones in recent years has mainly focused on the geometry, kinematics and deformation mechanisms of relatively small-scale shear zones [5,67]. The study of km-size or regional-scale shear zones is often hampered by incomplete geological records due to the lack of continuous outcrops, the difficulty in correlating different fragments or the same shear zone [23,94,116,127] due to the large distance separating them or to the occurrence of subsequent tectonic events. Once the geometry and the kinematics of a shear zone have been detected, the timing of shearing remains the main topic to be addressed due to the availability of different chronometers, and assessing the correct shearing age is often challenging. In this framework, tracing first-order lithospheric structures in ancient and modern collisional orogens is rather difficult due to the complex tectonic and metamorphic history of the orogens. An integrated multidisciplinary technique, including field mapping, structural analysis at different scales and geochronology or, better, petrochronology, represents a very useful tool to better unravel the tectonic evolution of regional-scale shear zones.
In the case of the EVSZ, a dextral, transpressive shear zone in the Variscan belt, we were able to correlate different strands of the same shear zone system from the Corsica-Sardinia block, through the Maures Massif, up to the External Crystalline Massifs in the Western Alps. This also provides new reliable constraints that should be considered for paleogeographic reconstructions. Furthermore, data documented that the EVSZ is not a simple late Variscan (or even Permian) strike-slip shear zone developed sincronously along all the different strands, but rather a crustal-scale system of transpressive shear zones with a complex and progressive evolution over time. Some of the branches were activated earlier than others (e.g., the FMSZ in the Argentera Massif), and therefore they record a long deformation history, while the younger branches (e.g., the PASZ in northern Sardinia) record a relatively shorter history. In this view, Simonetti et al. (2020) [62] identified two main generations of branches: the first one started at c. 340-330 Ma, and the second one started at c. 320 Ma. It is worth noting that the older shear zones of the system are fully localized in the migmatitic crust. This suggests that between c. 330 and 320 Ma, the Variscan lower crust was weak and easier to deform because of the ongoing anatexis. A detailed investigation of other shear zones, potentially linked to the EVSZ, could lead to the recognition of other generations of shear zones and more complex architectures of a possible anastomosing pattern of ductile shear zones. An anastomosing pattern can in fact accommodate the compatibility problem caused by the pure shear component in the development of transpressional shear zones.
The integration of vorticity data with detailed monazite ages along transects of the FMSZ and PVSZ shows that deformation evolved from a pure-shear-dominated transpression towards a simple shear-dominated transpression, from 330/340 Ma to 305/300 Ma [62,67] (Table 1). This can be related to a change from nearly orthogonal collision during D1 to the progressive initiation and increasing curvature of the two regionalscale arcs in the Variscan belt (Ibero-Armorican Arc and the eastern arc delimited to the east by the EVSZ) [79]. By increasing the curvature of the arc over time, the slip along the flanks of the arc increased, and this is reflected by the increasing component of simple shear in the younger mylonites of the EVSZ in the segments of the FMSZ and PVSZ and by the dominant pure shear component detected in the older mylonites [62,67].
The case of the HHD in the metamorphic core of the Himalaya is even more puzzling because the regional extent of this shear zone was completely neglected in the literature until a few years ago, even if some isolated strands were identified in a few parts of the orogen. Moreover, by establishing the duration of the shearing at c. 10-11 Ma, it was possible to highlight the effect of this protracted activity on the P-T-t paths of the footwall and hanging wall of the HHD. They reached the maximum P-T condition diachronically, showing different timings of exhumation caused by the kinematics of the HHD. Last but not least, the correct dating of the HHD allows unequivocally separating it from the lower MCT (often mixed in the past literature) and contributing to better discussing the models of assembly and of exhumation of the metamorphic core.
An integrated technique, as described here, enables the detection of what is often referred to as "hidden" or "cryptic" tectono-metamorphic discontinuity with the mid-crust exposed in the GHS in western and central Nepal. It has been recognized in Mugu-Karnali, Lower Dolpo, the Kaligandaki valley and NW India as well as elsewhere along the orogen (e.g., [113,118,127,131]) ( Figure 2). A key point here is the sense of shear, which is top-tothe-S/SW. Linking the mineral growth along the mylonitic foliation and in shear bands, the HHD is inferred to have been active during decompression, after peak metamorphism and related deformation (at least in the HHD hanging wall) and later retrograde metamorphism, of the deformed rocks within the sillimanite stability field (Figure 11a,b). It is worth noting that sillimanite grew along C shear planes (Figure 11b), whereas kyanite occurs in microlithons, pointing to how the top-to-the-S/SW sense of shear is associated with decompression ( Figure 12).
Monazite petrochronology constrains the timing of shearing to c. 28-27 and 17 Ma (Figure 13a) and provides direct timing constraints on both metamorphism and deformation [ [127] with references]. The youngest monazite dates suggest that, for a duration of c. 11-10 Ma, the footwall moved downward while the hanging wall was simultaneously exhumed. The hanging wall underwent retrograde metamorphism and exhumation over a duration of c. 11-10 Ma before the footwall. The HHD was active for several m.y. and deeply affected the metamorphic history of the wall rocks. Movement on this shear zone at 28-27-17 Ma caused a discontinuity in the metamorphic paths of the hanging wall and footwall rocks which, until that point, shared similar P-T-t paths, which indicates a remarkably different timing of exhumation (Figure 13b,c). The earlier exhumation of the hanging wall rocks was triggered by the activation of the contractional top-to-the-S/SW HHD (Figure 13b). Exhumation by thrusting and/or contractional shear zones in the GHS was possible due to the contemporaneous activity of erosion, extensional tectonics or even the thinning of the GHS caused by non-coaxial deformation [132].  (c) The HHD, due to its top-to-the-SW kinematics, triggered the earlier exhumation of the GHSU (dots a1 and a2 in (c)), while the GHSL was still experiencing prograde burial (dots b1 and b2 in (c)). Exhumation by the thrust-sense shear zone was possible due to a combination of the contemporaneous/episodic activity of the STDS, focused erosion in the frontal part of the belt and tectonic thinning of the GHS caused by non-coaxial flow.
Taking into account all of the common features of the recognized discontinuities in western-central Nepal up to Sikkim and Bhutan to the east, Montomoli et al. (2015) [116], Wang et al. (2016) [131] and Carosi et al. (2018;2019) [127,130] correlated the discontinuities in the Himalaya along-strike length for~1000 km (Figure 2). Benetti et al. (2021) [128], by recognizing the Badrinath Shear Zone as a possible prosecution of the HHD to the NW, further stretched its occurrence up to~1500 km along the strike of the belt. Montomoli et al. (2015) [116] showed that deformation shifted in space and time from the HHD to the structurally lower MCT at c. 17 Ma. However, in order to unambiguously distinguish between the HHD and MCT, it is necessary to integrate structural, metamorphic and geochronological investigations of both hanging wall and footwall rocks and the intervening high-strain zone [113,116,120,121,124,125,127,131,133].
This framework agrees with the complex structural architecture of the GHS along the belt. The progressive exhumation of the GHS occurred due to activation of first-order ductile shear zones localized within the GHS, such as the HHD. Recent Ar/Ar datings of the upper MCT (Vaikrita Thrust: VT) and lower MCT (Munsiari Thrust: MT) in NW India [134,135] have allowed us to recognize two different timings of activity of these two shear zones at c.  [125] proposed that this shift could be linked to the hardening (due to cooling and melt crystallization) of hanging wall rocks, compared to the footwall rocks, the latter being warmer and weaker in the same span of time. What is important to note is that, when deformation shifts downward, the upper (older) shear zones become inactive, as testified by undeformed crosscutting granite dykes in the HHD or by the growth of static, post-tectonic, white mica in the MCT (VT). This tectonic framework highlights the occurrence and development of an asymmetric shear zone in an orogenic wedge propagating to the foreland in addition to the four types of evolution of symmetric shear zones proposed by Fossen and Cavalcante (2017) [5].
Studies of the HHD in the Himalaya and of the EVSZ in the southern Variscan belt are two examples of a detailed characterization of high-temperature shear zones localized in the mid-crust showing a regional extent, occurring in different tectonic settings (including ortogonal vs. oblique collision/indentation), and of Earth's history. Several main aspects remain to be further investigated, such as why deformation localizes along the HHD in the GHS of the Himalaya and along the EVSZ in the Variscan belt. Furthermore, the integration of structural geology (at a different scale), geochronology and/or petrochronology and petrology strongly contributes to better understanding their role in the evolution of two collisional belts and to better clarifying their importance for the tectonic mechanisms active during the deformation of the continental lithosphere.