Amphibolite facies supracrustal rocks interleaved with granite mylonites constitute a shallowly dipping carapace overlying granulite facies anatectic basement gneisses in the Giridih-Dumka-Deoghar-Chakai area that spans ~11,000 km2 in the Chottanagpur Gneiss Complex (CGC). Steep N-trending tectonic fabrics in the gneisses include recumbent folds adjacent to the overlying carapace. The basement and carapace are dissected by steep-dipping sinistral shear zones with shallow/moderately plunging stretching lineations. The shear zones trend NNE in the north (north-down kinematics) and ESE in the south (south-down kinematics). Chemical ages in metamorphic monazites in the lithodemic units are overwhelmingly Grenvillian in age (1.0–0.9 Ga), with rafts of older domains in the basement gneisses (1.7–1.45 Ga), granitoids (1.4–1.3 Ga), and the supracrustal rock (1.2–1.1 Ga). P-T pseudosection analysis indicates the supracrustal rocks within the carapace experienced postthrusting midcrustal heating (640–690°C); the Grenvillian-age P-T path is distinct from the existing Early Mesoproterozoic P-T path reconstructed for the basement gneisses. Quartz opening angle thermometry indicates that high temperature (~600°C) persisted during deformation in the southern shear zone. Kinematic vorticity values in carapace-hosted granitoid mylonites and in steep-dipping shear zones suggest transpressional deformation involved a considerable pure shear component. Crystallographic vorticity axis analysis also indicates heterogeneous deformation, with some samples recording a triclinic strain. The basement-carapace composite was extruded along an inclined channel bound by the steep left-lateral transpressional shear zones. Differential viscous extrusion during crustal shortening coupled with the collapse of the thickened crust caused midcrustal flow along flat-lying detachments in the carapace.

The formation of Metamorphic Core Complexes (MCCs) is conventionally related to extensional tectonic processes ([1] and references therein), although proposed models vary in the details [29]. In addition to regions dominated by extension, such as the Basin and Range province, MCCs have been reported from accretionary margins [10] and in convergent regimes [1, 11, 12]. Despite the convergent nature of these tectonic settings, MCC development is associated with extension-related processes such as slab rollback, intrusion driven extension [13], or orogenic collapse under fixed boundary conditions or slow plate convergence [14]. Searle and Lamont [15] demonstrate that MCCs can form in entirely compression regimes unaffected by any extensional tectonism.

Synconvergent to postconvergent gravitational collapse of a previously thickened orogenic crust [1623] may lead to the development of extensional shallow-dipping midcrustal detachment zones [19, 2430] that may exhume deep-seated lower crustal rocks within a MCC. Long et al. [31] demonstrate that regions of upper crustal thickening directly control the spatial location of synorogenic extension. Additionally, vertical partitioning of strain in the crust during convergence can localize extension in the upper crust and contraction in the ductile lower crust (e.g., [1, 7, 3237]).

The presence of regional scale strike-slip dominated shear zones is noted in a number of MCCs [10, 3845]. Although these shear zones are known to aid exhumation of the deep crustal rocks [6, 4654], limited data exist on the field relations of these shear zones and their role in the formation of MCCs [55].

This work documents possibly the first MCC recognized in peninsular India (cf. Figure 2 of [1]), based on detailed structural mapping in the eastern part of the Chottanagpur Gneiss Complex (CGC) (Figure 1). The proposed Early Neoproterozoic MCC lies in the foreland region of a contemporaneous convergent boundary between the Meso-/Neoproterozoic CGC and the Meso-/Neoarchean Singhbhum Craton in the south (Figure 1(a)) and is closely associated with steep-dipping transpressional shear zones. Kinematic analyses coupled with crystallographic data, monazite dating, and P-T pseudosection analyses illustrate the evolution of the proposed MCC.

The Precambrian crystalline rocks of the CGC constitute a vast area (~80,000 sq. km). The rocks may be classified into three lithodemic units: (i) regionally extensive basement anatectic gneisses that include granulite facies felsic (charnockite-enderbite) orthogneisses, garnet-biotite-sillimanite-K-feldspar-bearing metapelite, calc silicate granulite, mafic granulite, massif anorthosites, and silica undersaturated syenites; (ii) blastoporphyritic granitoids structurally varying from massive to foliated to mylonitic rocks predominant in the CGC; and (iii) lower/middle-amphibolite facies supracrustal rocks that include muscovite-biotite schist (with rare garnet and sillimanite), micaceous and ferruginous quartzites, amphibolites (hornblende-plagioclase±epidote), and minor amounts of metamarls and metadolomite ([56, 57] and references therein). The high-grade gneisses and granulites occur as outcrop scale enclaves within the blastoporphyritic granitoids that comprise the basement of the CGC [5861]. The supracrustal rocks are prominently exposed in E-trending belts that coincide with topographic highs along and neighboring regional scale curvilinear ductile shear zones that trend E/ENE in central/northern CGC and E/ESE in southern CGC [62] (Figure 1(a)). The three lithodemic units are unconformably overlain by rift-related Gondwana basins that host Permo-Carboniferous to Early Cretaceous glacial-glaciogenic sediments that transition to fluvial sedimentary successions [63]. The Gondwana basin sediments are intruded by mafic and lamprophyre dykes and overlain by basalt flows [64, 65]. This study investigates a quadrangular area in the eastern part of the CGC covering ~11,000 km2 near the towns of Deoghar, Dumka, Giridih, and Chakai (Figure 1(b)).

U-Pb zircon geochronology [6670] and pooled monazite chemical ages [59] suggest the anatectic basement gneisses of the CGC are the oldest (1.60–1.45 Ga) lithodemic unit in the study area. This age range corresponds to high-grade metamorphism-anatexis [61, 66, 69, 71] and emplacement of felsic intrusives (Mukherjee et al. [67, 68]. Considerable variations exist in the U-Pb zircon ages obtained by Mukherjee et al. [6769] in the felsic lithologies in the Dumka-Deoghar area. Near-concordant and discordant U-Pb systematics in zircons from a charnockite gneiss/felsic orthogneiss sample (AS-100) yield an upper intercept age of 1450 Ma (emplacement age) and a lower intercept 943 Ma age from the intermittent and weak metamorphic overgrowth of zircon rims [67]. These authors suggest the weak Early Neoproterozoic overgrowths relate to a granulite-facies clockwise P-T path culminating with near-isothermal decompression in the retrograde sector. This contradicts the earlier findings [60, 61, 66, 72] that the granulite facies metamorphism in the basement gneisses is at least older than 1.4 Ga. It will be shown later that the field relationship and mesoscale structures adopted by Mukherjee et al. [67] in their interpretation are contentious.

In areas neighboring Dumka, Mukherjee et al. [68] report two amphibole-biotite-gneiss/granite samples (AS-114A and AS-34/1) with a mean U-Pb zircon age of 1465±17 Ma; this age is interpreted to correspond with the emplacement of the rocks. Based on the Lu-Hf isotopic compositions in zircon and major and trace element geochemistry of similar rocks, Mukherjee et al. [68] suggest that the protolith (ferroan A-type granitoid) for the rocks was derived from a Paleoproterozoic crustal source. Mukherjee et al. [69] determine U-Pb zircon ages of two biotite-amphibole-garnet bearing felsic augen gneisses from Deoghar, e.g., 1709±17 Ma for AS-37 and 1626±17 for AS-83A; the authors suggest that these ages represent the time of ultrahigh T granulite facies metamorphism.

There are no existing age data available in the amphibolite facies supracrustal rocks near Dumka. However, in the southern and central CGC, monazite chemical ages in the muscovite-biotite schists show two nonoverlapping age groups: 1300–1200 Ma ages are found in strongly embayed and fragmented cores mantled by younger (1050–880 Ma) metamorphic rims [59, 61, 62, 73, 74]. Metamorphic P-T path reconstructions in the supracrustal rocks in central CGC are not available.

The Early Neoproterozoic metamorphism is broadly coeval with the emplacement (1.1–0.9 Ga) of the expansive blastoporphyritic granitoids throughout the CGC ([66], and references therein; [58, 61, 62, 71, 7375]). Goswami and Bhattacharyya [75] used petrographic and geochemical characters (major and trace elements, mineral chemistry, and 87Sr/86Sr ratios) to classify such granitoids in the 1071±64 Ma Raghunathpur batholith to be postcollisional shoshonites formed due to the delamination of the subducting slab.

The Giridih-Dumka-Deoghar-Chakai area exposes the three lithodemic units of the CGC, i.e., the granulite facies gneisses, blastoporphyritic granitoids, and amphibolite facies supracrustal rocks (Figure 1(b)). The blastoporphyritic granitoids in the area are either pink or grey in color and commonly contain K-feldspar porphyroclasts and submeter to kilometer scale enclaves of the gneisses. The supracrustal rocks are restricted to the northern and southern parts of the area (Figure 1(b)). The Gondwana Supergroup sediments unconformably overlie the three lithodemic units.

As the anatectic gneisses are the oldest lithodemic unit within the CGC, they record all of the deformation events affecting the CGC (Table 1). Four structural domains are identified, based on observations in 1002 field stations (Figure 2). Domain I outcrops in the east-central part of the area, lacks supracrustal rocks, and is dominated by basement gneisses and granitoids that exhibit steeply dipping (>50°), sinuous, N-trending tectonic fabrics. The three other domains include blastoporphyritic granitoids and supracrustal rocks in addition to the basement anatectic gneisses. Domain II rocks in the west-central and southeastern parts (Figure 2(a)) are characterized by shallow-dipping (dip<35°) fabrics; these occur both in discontinuous lenses in and neighboring the northern and the southern shear zones and as irregular patches structurally overlying Domain I. Areas where these fabrics have steepened due to folding neighboring the steep-dipping shear zones are excluded from Domain II. Domain III and Domain IV are characterized by deformational structures within and near the steep shear zones that trend ENE in the north and ESE in the south. Lower hemisphere stereographic projections for the deformation fabrics in the lithodemic units within the four domains are summarized in Figure 3. The two shear zones in Domains III and IV appear to converge in the west, but deep weathering and alluvial cover obscured their projected intersection north and west of Giridih (Figure 2).

The conventional three-part classification of continental MCCs described by Platt et al. [6] consists of (1) the metamorphic core consisting of high-grade ductilely deformed metamorphic rocks, (2) the overlying detachment zone comprising highly sheared shallow-dipping rocks that range in thickness from a few meters to several kilometers, and (3) the unmetamorphosed/lower metamorphic grade rocks of the hanging-wall overlying the detachment zone. The identification of these features in the Precambrian crystalline rocks of the CGC is used in this study to propose a MCC structure for the area. Domains I and II represent the metamorphic core and detachment zone of the MCC, respectively. The striking difference in the metamorphic grade and structure of the granulite facies basement gneisses and the overlying amphibolite facies supracrustal rocks suggests that the supracrustal unit formed part of the hanging wall of the MCC and preserves deformation associated with the detachment zone formation. Unfortunately, due to extensive weathering and erosion, the only surviving remnants of the supracrustal unit are exposed in Domains III and IV. Although not included in conventional MCCs, steep-dipping shear zones are increasingly recognized as integral parts of MCC evolution [10, 38, 4042, 45].

3.1. Domain I (the Core of the Complex)

In the anatectic basement gneisses, the earliest recognizable planar fabrics are leucosome layers that are concordant with the pyroxene±hornblende±biotite segregations (SN1; D1 deformation; Table 1). Garnet is ubiquitous in both the leucosomes and the matrix. Coarse grains of orthopyroxene, clinopyroxene, and hornblende are common in the leucosomes in the felsic orthogneisses. The SN1 layers are typically isoclinally folded and are intrafolial to the pervasive gneissic layering (SN2: D2 deformation; Table 1) in the basement gneisses, resembling metatexite layers. The folds on SN1 leucosomes are characterized by thickened hinges and strongly attenuated and commonly boudinaged limbs. The SN2 metatexite layering in the anatectic gneisses is steeply dipping (dip>65°), sinuous yet broadly N-trending in the east-central part of the area, and swings to NW through WNW to W in the NW (Figures 2(a) and 3(a)). The SN1-SN2 layers in the gneisses are disrupted and/or truncated by coarser-grained syn-D2 diatexite pods typically 15–20 cm in diameter (also garnet±pyroxene bearing) at the necks of pre-SN2 leucosome boudins (Figure 4(a)). These diatexite melt pods comprise a small fraction (<10 area %) of the rock in any outcrop.

The basement gneisses were intruded by granitoids in two stages. The structurally older granitoids share the SN2 fabric in the gneisses but do not contain the intrafolial folded leucosome layers (SN1) and are thus post-D1 but pre-D2. The fabric within these granitoids is defined by schlieren of ferromagnesian phases (biotite, hornblende, and pyroxene), quartz ribbons, and drawn-out aggregates of dynamically recrystallized feldspars. A suite of granitoids that lack mesoscale fabrics cross-cut the SN2 fabric in the gneisses and are thus inferred to be post-D2 intrusives.

The outcrop and regional scale sinuosity in the SN2 layers in the gneisses and the variations in LN2 fold axes (estimated from the SN1-SN2 intersection lineations) are related to folding induced by locally penetrative discontinuous subvertical E-/ENE-/WNW-trending shears (Figures 2(a) and 4(b)). These shear zones are possibly genetically related to the steep-dipping shear zones in Domains III and IV (D4, Table 1, discussed later). The steeply plunging β-axis obtained from the SN2 pole girdle in Domain I (Figure 3(a)) suggests that the shear-related fold axes are steeply inclined to recline in the regional scale. Sheath folds are locally observed (Figure 4(c)) in these E-trending shear zones. These shears produce a locally penetrative fabric in the post-D2 granitoids wherever affected, but melting was not synchronous with shearing. The field features taken together demonstrate that pre-D2 melt productivity (vol% melt) during high-grade metamorphism in the basement gneisses waned considerably during D2, and leucosome formation in the basement gneisses did not occur during the development of the E-trending shears.

The structural context of the samples with U-Pb age determinations [6769] is lacking (see later), and this complicates attempts to infer the age of formation of N-striking steeply inclined D1-D2 fabrics in the Dumka-Deoghar area. Since the lithologies discussed by Mukherjee and coworkers possess at least one penetrative gneissic fabric which is N-striking (cf. [67]), it appears that the D1-D2 composite fabric formation and associated high-grade metamorphism and felsic magmatism in Domain I may have occurred between 1.46 and 1.6 Ga.

3.2. Domain II (the Detachment Zone)

All three lithodemic units in Domain II exhibit a shallow-dipping foliation. D1-D2 fabrics in the screens of basement gneisses interleaved with granitoid mylonites in this domain exhibit recumbent to gently inclined folds (D3; Table 1); however, axial planar fabrics (SN3) corresponding to these folds (Figure 3(b)) are absent or only locally developed (Figures 5(a) and 5(b)). These recumbent folds affect the anatectic gneisses only at structurally higher levels in the crust; at deeper levels, the unmodified D1-D2 fabrics are steep dipping (cf. [62]). The D3 deformation is 0.9–1.1 Ga (see section on monazite dating), limited to a carapace over the basement. The basement remains largely unaffected by this deformation, as expected for the detachment zone formation in an MCC. In close-spaced outcrops, the axial planes (SN3) of the recumbent/gently inclined folds in the gneisses are broadly coplanar with the shallow-dipping mylonite foliation (SG1) in the granitoids (Figures 3(b) and 3(c)). The LN3 fold axes of the gently inclined folds in the gneisses straddle the mean SN3 axial plane (Figure 3(b)).

Domain II is dominated by shallow-dipping granitoid mylonites best described as S and S>L tectonites (Figures 5(c) and 5(d)). The shallow-dipping mylonitic foliation (SG1; D3 deformation, Table 1) in the basement granitoids generally dips northwards, with S-dipping planes more common in the south (Figures 2(a) and 3(c)). The regionally warped foliation is monophase in the majority of the granitoids; the pre-D2 granitoids, however, show rare recumbent folds on the former layering in the intrafolial domains of the penetrative SG1 foliation. Stretching lineations (LG1) are uncommon in the shallow-dipping granitoid mylonites, possibly due to intense dynamic recrystallization accommodating size reduction in the feldspar clasts.

At a regional scale, the orientations of LG1 straddling the mean SG1 plane (Figure 3(c)) are similar to the LN3 fold axes in the anatectic gneisses (Figure 3(b)). Mesoscale S-C fabrics are rare within these rocks, and asymmetry in feldspar clasts is observed only in few outcrops (Figure 5(e)). As a result of the lack of stretching lineations and mesoscale shear sense indicators in the shallow-dipping granitoids (D3), the kinematics of shearing associated with the formation of the detachment zone could not be ascertained. In a few outcrops, top-to-the-north kinematics in sections parallel to lineation and perpendicular to foliation (X-Z sections of strain ellipsoid) is inferred from S-C fabrics and asymmetry of feldspar clasts in the northern parts of the domain. In S>L tectonites, in the southern parts of Domain II, top-to-the-south kinematics is inferred (Figure 5(e)).

The supracrustal rocks in Domain II exhibit a shallowly dipping penetrative schistosity (SS2; Figure 5(f)) broadly coplanar with the shallow-dipping SN3 and SG1 fabrics in the anatectic gneisses and granitoids, respectively (Figure 3(d)). This penetrative SS2 schistosity in the mica schists (quartz, biotite≥muscovite, ilmenite±garnet±sillimanite (altered to pyrophyllite)±plagioclase) within the supracrustal unit is a crenulation cleavage (D3 deformation, Table 1); the pre-SS2 fabrics, largely obliterated, occur as oblique strands and relic hinges of microfolds on muscovite-biotite aggregates in the SS2 intrafolial domains. In amphibolites associated with the mica schists, both the former and the SS2 fabrics are defined by aggregates of hornblende±epidote in a polygonized mosaic of plagioclase-quartz. In Domain II, the SS2 foliation is commonly crenulated possibly due to the D4 deformation, especially approaching the flanking Domains III and IV (Figure 6(a)). Garnet porphyroblasts, uncommon in the mica schists, overgrow the SS2 schistosity (Figures 6(a) and 6(b)). We tentatively infer the garnets to be post-SS2, although garnet-hosted inclusion trails are lacking, and therefore, Si-Se relations critical for establishing garnet growth in relation to the external fabric could not be unequivocally ascertained. The garnet porphyroblasts hosted in the shear zones are pretectonic with respect to SS3, D4 deformation (Figure 6(c)).

3.3. Domains III and IV (the Steeply Dipping Shear Zones)

Domain III encompasses the ENE-trending shear zone (Figures 2 and 7(a)–7(c)) in the north and the structures neighboring it, while Domain IV constitutes an ESE-trending shear zone (Figures 2 and 7(d)–7(f)) in the south. The formation of these basement-penetrative shear zones represents the final deformation event (D4), which is also Early Neoproterozoic in age (see later). Proximal to Domains III and IV, the shallowly dipping D3 fabrics in the detachment zone exhibit open to close and upright to steeply inclined E-/ESE-trending D4 folds with gently plunging axes. The long limbs of the asymmetric folds are north-dipping in Domain III (Figure 7(c)) and south-dipping in Domain IV. By contrast, D4 folds on steeply dipping SN2 fabric in the basement gneisses below the detachment zone—unaffected by D3 recumbent folding—exhibit steeply dipping E-/ESE-trending axial planes (SN4) with steeply plunging fold axis LN4 (Figures 3(f) and 4(d)).

S and S>L tectonite granitoids within both shear zones are steeply dipping mylonites (SG2 foliation), with stretching lineations defined by drawn out K-feldspar clasts and stretched quartz ribbons oriented subhorizontally in the north (LG2; Figures 3(e) and 7(b)) and moderately plunging in the south (LG2; Figure 3(i)). Porphyroclast-matrix relations indicate that both the northern (Figure 7(a)) and the southern shear zones (Figures 7(d) and 7(e)) record a sinistral shear sense on horizontal surfaces (approximating the X-Z section of the strain ellipsoid). In vertical sections oriented perpendicular to the mylonite foliation and stretching lineation (i.e., the Y-Z section of the strain ellipsoid), asymmetry in feldspar clasts is less prominent; however, north-down kinematics are inferred for the Domain III northern shear zone (Figure 7(c)) and south-down motion in the Domain IV southern shear zone (Figure 7(f)). In high-strain domains within the ENE-trending northern shear zone, a set of SE-trending C” shear bands with dextral kinematics is persistently developed.

As the two flanking shear zones are approached, subhorizontal to moderately plunging asymmetric folds (D4 deformation, Table 1) are developed on the shallow-dipping gneisses (SN3) and supracrustal rocks (SS2) in the detachment zone. These features become progressively more tightly folded (Figure 3(g)) within the shear zones, where the SS2 schistosity is ultimately transposed to SS3 schistosity (Figures 3(h) and 3(j)). Locally, however, the shallowly dipping supracrustal rocks, albeit crenulated, occur as lens-shaped rafts of low DS3 strain within the shear zones (Figure 2). Synthetic CS3 shear bands (Figure 7(c)) exhibit sinistral and north-down kinematics in Domain III and sinistral and south-down sense of movement in Domain IV.

The northern and southern shear zones appear to converge in the west of Giridih (Figure 1(b)), but no coherent analysis could be made about the temporal relationship between the two sinistral shear zones. However, because of the overlapping monazite ages (see later), these two shear zones are interpreted to be synchronous and formed during the same D4 deformation event.

Crystallographic fabrics from plastically strained rocks reflect the kinematics of deformation. To better understand the structural evolution of the CGC, we collected crystallographic data from six samples, two each from Domains II, III, and IV. Polished thin sections of samples oriented perpendicular to foliation and parallel to lineation were analyzed using electron back-scatter diffraction (EBSD) at Washington and Lee University. The data were collected with an Oxford EBSD with Aztec software on a Zeiss EVO-MA 15 SEM. To avoid charging, analyses were conducted in variable pressure mode with a chamber pressure of 30 Pa. Machine conditions included an accelerating voltage of 25 kV, a probe current of 20 nA, and a working distance between 24 and 28 mm. Data were collected on a rectangular grid with a spacing between 12 and 30 μm (chosen based on sample grain size); observations from a series of 500μm×375μm scan regions were montaged into a single map within Aztec for analysis. Scan areas ranged from 175 to 300 mm2. The crystallographic data were processed using v. 5.1.1 of the MTEX toolbox [76].

Crystallographic fabrics are generally viewed in a reference frame defined by the foliation and lineation. In many cases (e.g., simple shear), the vorticity axis of a deformation is oriented perpendicular to lineation within the foliation plane, so this reference frame is best for interpreting the deformation kinematics. However, in some situations (e.g., transpression), the vorticity axis may be aligned parallel or oblique to lineation [77]. Since the D3-D4 deformations involve transpression (see later), we apply crystallographic vorticity axis (CVA) analysis to determine an appropriate reference frame to view the crystallographic data [78]. In this technique, the dispersion of crystallographic orientations within each grain is used to determine the vorticity axis for that crystal. The average of many individual observations is interpreted to represent the vorticity axis for the bulk deformation.

We utilized an MTEX algorithm developed by Michels et al. [78] to perform a CVA analysis for each sample. For four of our six samples, the estimated vorticity axis is positioned nearly perpendicular to the lineation and within the foliation plane (i.e., as would be expected for a monoclinic deformation). However, for two of the samples: GM 2 (Domain II) and GM 57 (Domain IV), the calculated vorticity axis is within the foliation plane but oblique to the lineation. This is a characteristic of triclinic deformation [79, 80]. Misalignment of the quartz CPO patterns with the lineation/foliation reference frame has been observed previously (e.g., [8183]). In these studies, the crystallographic data are rotated around the pole to foliation until a conventional quartz CPO pattern is observed. Rather than apply an arbitrary rotation, we achieve a similar effect by instead rotating the data to place the CVA in the center of the pole figure.

4.1. Domain II

The CVA analysis for GM 2 reveals a vorticity axis within the foliation plane but oblique to the lineation, indicative of a triclinic deformation. Rotation of these data to place the vorticity axis in the center of the pole figure produces a c-axis fabric (Figure 8(a)), albeit one that is difficult to interpret. Most prominent is a strong girdle stretching from the bottom to top of the pole figure; additionally, two poorly defined limbs are oriented within the foliation plane (along the left-to-right axis in the pole figure), giving an appearance reminiscent of a cross-girdle. If this is a cross-girdle, it has an unusual orientation, since cross-girdle limbs generally split around the pole to foliation [84]. Alternatively, the fabric may represent a poorly defined single girdle, with the subordinate limbs observed in the foliation plane corresponding to noise.

In contrast to GM 2, the c-axis pattern for GM 24 is poorly defined. We attribute the lack of a strong fabric, at least in part, to an insufficiently large sample. Although data were collected from an area>175 mm2, the large grain size (500-2200 μm) (Figure 8(b)) limited the number of grains in the analysis and prevented us from defining a clear c-axis fabric. Nevertheless, sample GM 24 does show a preferred orientation in the positive and negative rhombs (r and z). For instance, the z orientation lacks orientations parallel to the vorticity axis (center of pole figure) as well as along the upper left and lower right margins of the pole figure. Both experimental [85] and observational [8688] studies have shown that rhomb-dominated fabrics may develop even in the absence of strong c-axis fabrics. This pattern reflects Dauphiné twinning (a 60° rotation around the quartz crystallographic c-axis) triggered by a response to stress. Because the elastic modulus of quartz varies with crystallographic direction, crystals loaded against the relatively inelastic z face will twin to bring the more compliant r-face towards the compression axis. In noncoaxial shear zones, the instantaneous shortening direction is perpendicular to the vorticity axis and inclined with respect to the foliation plane. The preferred orientations of the rhombs (Figure 8(b)) are therefore consistent with dextral shear (top-to-the-ESE), in which shortening on an axis from the upper left to lower right of the pole figure triggered Dauphiné twinning that created the observed void in z-faces in that direction.

4.2. Domains III and IV

Four samples, two from each of the steeply dipping, high-strain shear zones that bound the complex, were analyzed. The samples from Domain III (the northern shear zone) show patterns reminiscent of sample GM 24. Both samples lack an obvious c-axis fabric, a fact we again attribute at least in part to the relatively large grain size of these samples. GM 75 (Figure 8(c)) displays a preferred orientation within the rhombs in a nearly identical fashion to GM 24; accordingly, we infer that this sample may have been affected by dextral shear. Although GM 76 also displays a similar pattern in the rhombs (Figure 8(d)), in this case, the z-rhomb maxima are oriented perpendicular to the foliation and parallel to the lineation. Tentatively, we suggest that this pattern indicates a greater contribution of pure shear deformation.

Like sample GM 2, CVA analysis for GM 57 demonstrates a vorticity axis oblique to the sample lineation. Centering the data on the vorticity axis reveals a typical type I cross-girdle (Figure 8(e); [89]) suggesting a significant pure shear component to the deformation. This type of fabric is amenable to quartz opening angle thermometry, an empirical approach based on the observation that quartz opening angles increase with deformation temperature [90, 91]. We quantified the opening angle for GM 57 using an MTEX routine that quantifies fabric intensities around the perimeter of a pole figure [92]. The measured opening angle of 80° indicates a temperature of ~600°C using the calibration of Faleiros et al. [93].

Sample GM 47 shows an exceedingly weak crystallographic fabric (Figure 8(f)), with c-axis maxima no more than 1.8 times uniform distribution. Unlike for samples GM 24, 75, and 76, a large area was scanned relative to the grain size (median of ~100 μm from the EBSD dataset), and we regard the random fabric as real rather than a small-sample artifact. The lack of a quartz CPO suggests an important role for a mechanism other than dislocation creep. The small grain size suggests an important role for grain boundary sliding [94]. Numerous studies have shown that grain size reduction during recrystallization can cause a transition to diffusion creep and associated grain boundary sliding that can weaken and randomize an existing CPO (e.g., [9597]). Experimental work suggests that very high shear strains, on the order of γ>17, are required to disaggregate mineralogic layering and generate well-mixed ultramylonites that lack a CPO [98]. The anticorrelation between the positive and negative rhombs is again interpreted to reflect stress-induced Dauphiné twinning.

Kinematic vorticity estimates provide a measure of the contribution of pure shear (Wk=0) and simple shear (Wk=1) during deformation [99]. Fossen and Tikoff [100] consider equal contributions of pure shear and simple shear at Wk=0.81, whereas Law et al. [101] consider the value to be Wk=0.71 for shear zones with monoclinic symmetry. Vorticity estimation from rotation of rigid porphyroclasts assumes steady-state deformation and monoclinic symmetry with the VNS (vorticity normal section) parallel to the XZ section in the rock [102]. Although CVA analysis indicates that some samples record triclinic deformation, many samples exhibit monoclinic symmetry. We therefore applied the porphyroclast aspect ratio (PAR; [103]) method to measure the mean vorticity number (Wm; [102]) for three granitoid mylonites in Domain III and one sample from Domain II to better estimate the kinematics of formation of the steep-dipping shear zones (D4) and the detachment zone (D3).

The PAR method estimates the mean kinematic vorticity number (Wm) by plotting the aspect ratio (R=longaxis/shortaxis) of rotated rigid clasts against φ, the angle between the long axis of the clast and the flow plane (taken as the foliation plane in our study) [102]. The rock matrix is assumed to behave like a Newtonian linear-viscous fluid with freely rotating rigid porphyroclasts perfectly bonded with the matrix [102]. The plot is divided into two fields by the critical aspect ratio (Rc) which separates the clasts that rotated continuously, from the clasts that obtained stable sink orientations. The Rc value is then used in the equation Wm=Rc21/Rc2+1 [104] to calculate the mean vorticity number associated with the deformation. Although vorticity estimates using rigid porphyroclasts are widely used, the results may have large uncertainties [105, 106].

A mylonitic granitoid sample (Figure 5(e)) from the detachment zone (Domain II) containing a well-developed N-trending stretching lineation with the size of the long axis of the clasts varying from a couple of mm to more than 2 cm in diameter was chosen. Kinematic analyses should be performed on the vorticity normal section (VNS), identified as the plane of maximum asymmetry perpendicular to the foliation plane. Several sections at varying angles from the strike were cut to determine the VNS (cf. [107, 108]) in the Domain II sample but no perceptible difference in asymmetry was observed. However, Rc estimations were made in three sections cut normal to the foliation (Figures 9(a)–9(c)): (a) parallel to the lineation (XZ section), (b) perpendicular to the lineation (YZ section), and (c) 45° from the lineation. The clast measurements made on photographs of polished slabs yielded overlapping ranges of Rc values that translate to Wm values 0.59–0.74. Taken simply, these values indicate that the deformation was marginally dominated by pure shear.

Vorticity estimations for the shear zone forming D4 deformation were obtained from three steep-dipping granitoid mylonites with a subhorizontal stretching lineation and well-defined clasts from the northern shear zone (Figure 7(b)). After making several sections perpendicular to foliation, the maximum asymmetry displaying sinistral shear sense was observed in sections close to the horizontal surface (XZ section). This confirms that the northern shear zone approximates a monoclinic symmetry, consistent with the crystallographic analysis discussed above. The Wm estimates (Figures 9(d)–9(f)) in the three samples are comparable, ranging between 0.60 and 0.74, with an average value of Wm=0.67, indicating a marginally larger component of a pure shear than a simple shear associated with the deformation.

The metamorphic P-T evolutionary history of supracrustal rocks has been reconstructed along the southern accretion zone the CGC shares with the Singhbhum Craton [73, 74]. However the P-T path of the supracrustal rocks in the central part of the CGC distal from the accretion is unknown. One of the reasons for the lack of detailed P-T work stems from the fact that key minerals in these rocks have been altered due to intense weathering (garnet especially is commonly replaced by martite, and sillimanite is altered and replaced by pyrophyllite). Here, we provide a new P-T path evolutionary history retrieved from a muscovite-biotite mica schist (SM-22B; Figure 10) consisting of garnet, sillimanite, and minor amounts of chlorite, plagioclase, and ilmenite. The sample is located in Domain II, but close to the boundary with Domain III (Figure 1(b)).

The mica schist exhibits spaced muscovite-biotite-defined SS3 disjunctive cleavage axial planar to D4 crenulations on muscovite-biotite-sillimanite-defined SS2 schistosity. Chlorite replacing biotite randomly overgrows the SS3 crenulations. Garnet porphyroblasts truncate the SS2 mica-sillimanite aggregates in the SS3 interfolial domains but are pretectonic with respect to SS3 (Figure 6(c)). No post-SS3 sillimanite was observed. Two garnet porphyroblasts were examined in detail (Figures 10(a) and 10(b)). The margins of garnet grains are straight and not corroded (Figures 10(a) and 10(b)). In Figure 10(a) (bottom left), the garnet appears to mimic the shape-preferred aggregates of muscovite (±biotite). There is no evidence to suggest that the garnets were resorbed via retrograde reaction involving biotite/chlorite (Figures 10(a) and 10(b)). The profiles for the two garnet porphyroblasts (Figures 10(c) and 10(d)) are chosen so that the garnet rims are juxtaposed with muscovite and not biotite. This ensures that the postgrowth chemical modifications in garnet composition due to Fe-Mg diffusive exchanges are minimal across garnet-biotite interfaces. In other words, the profiles indicate garnet growth. The core compositions of the garnets are Py8Alm86Grs3Sps3 and Py8Alm85Grs4Sps3, and the corresponding rim compositions are Py7Alm86Grs5Sps1and Py7Alm85Grs6Sps2, respectively. The garnets are, for the most part, chemically homogenous (Figures 10(c) and 10(d)) but exhibit an increase in XCa (=Ca/(Mg+Fe+Ca)) and a decrease in XMn (=Mn/(Mg+Fe+Ca+Mn)) towards the rim. XFe (=Fe/Fe+Mg) and XFe (=Fe/(Fe+Mg+Ca+Mn)) show a gentle rimward increase (Figures 10(e) and 10(f)).

The bulk composition of the mica schist (Figures 10(g) and 10(h)) is approximated to the system MnO–Na2O-CaO–K2O–FeO–MgO–Al2O3–SiO2–TiO2–H2O (MnNCKFMASHT). Fe2O3 contents in ilmenite are negligible. The MnNCKFMASHT P-T pseudosection for the sample was constructed using Perple_X_6.8.3 software [109] and the updated version of the internally consistent thermodynamic dataset of Holland and Powell [110].

The following phases and the phase components (in square brackets) were chosen from those listed in the software: garnet (Gt) [alm, py, gr, and sps], chloritoid (Ctd) [fctd, mctd, and mnctd], staurolite (St) [fst, mst, and mnst], Crd [crd, fcrd, hcrd, and mncrd], biotite (Bio) [ann, phl, east, and mnbi], chlorite (Chl) [ames, clin, daph, and mnchl], and muscovite (Mica) [mu, pa, cel, and fcel]. For both K-feldspar and plagioclase, the ternary feldspar (abh, an, and san) model was adopted. Clinozoisite (cz), zoisite (zo), andalusite (and), sillimanite (sill), kyanite (ky), ilmenite, rutile, and quartz (q) are taken to be pure phases. The solution model of White et al. [111] is used for garnet, biotite, chlorite, chloritoid, and staurolite. The solution model CHA1 [112, 113] is used for muscovite. The feldspar solution model is after Fuhrman and Lindsley [114]. In the computations, quartz is taken to be in excess and the fluid was assumed to be pure H2O.

For interpreting the P-T pseudosection (Figures 10(g) and 10(h)), it was assumed that no part of the garnet dissolved during growth. The core to rim variations in measured compositions of garnet combined with the topology of XFe, XCa, and XMn isopleths suggests that the post-SS2 and pre-SS3 garnets grew in quartz-muscovite-biotite-garnet-sillimanite-ilmenite phase field due to prograde heating (640–690°C) at midcrustal pressures of 5.2–5.5 kbar (Figure 10(h)). The P-T path is drawn in the direction of increasing XCa (=Ca/(Ca+Fe+Mg) and XFe (=Fe/Fe+Mg), decreasing XMn (Mn/Mn+Ca+Fe+Mg), and higher volume% of garnet (Figure 10(h)). The reconstructed P-T path for the supracrustal rocks is similar to the one recorded by Maji et al. [60] but is strikingly different from the Early Mesoproterozoic granulite facies clockwise P-T path documented by Karmakar et al. [71], which is characterized by postpeak decompression and decompression-cooling for the Dumka anatectic basement gneisses.

High-quality U-Pb (zircon) data exist for the anatectic basement gneisses and granitoids (Domain I) in and around Dumka [6670]. Monazite chemical dating by electron probe at the National Facility, Indian Institute of Technology, Kharagpur, was used to constrain the timing of the formation of the detachment zone and the steep-dipping shear zones. Th-U-Pb (total) ages in monazites were obtained following Montel et al. [115] from 9 samples selected from the three lithodemic units that make up Domains II–IV. Monazites in four mica schists of the supracrustal unit and three anatectic quartzofeldspathic gneisses including a garnet-sillimanite-K-feldspar metapelite (in the ensemble of high-grade basement rocks) and two blastoporphyritic granitoids were analyzed. X-ray element maps and back-scattered electron images were used to identify chemically distinct domains in representative monazites. A total of 213 spot ages in nine samples with error%=1002σerrorinMa/ageinMa<6 were statistically resolved using Isoplot 3.0 [116]. The spot ages were calibrated against the standard Moacyr monazite dated to be 497±10 Ma (EMP age, [117], p. 228); 487 ± Ma (TIMS age, [118]), and 509.3±0.5 Ma (TIMS age [117], p. 228) (details in [119]). The details of analytical procedure are the same as in protocol I of Prabhakar [119]. The analytical data on monazites, spot ages, and 2σ errors are presented in Supplementary Material 1. Y and Th maps and textural contexts of age data are presented in Supplementary Materials 2, 3 respectively. Summary probability-density plots and selected X-ray element maps of monazites in the three lithodemic units are shown in Figure 11.

In Domains II to IV, Late Grenvillian chemical ages (1020–900 Ma) obtained in both mantles and outermost rims in zoned monazite grains, as well as in chemically homogenous monazites, are the most dominant age population in the samples (Figure 11). These monazites are typically parallel to and/or postdating (overgrowing?) metamorphic fabrics in the detachment zone (D3) foliations and steep-dipping shear zone foliation (D4 deformation) in granitoids and mica schists (Figure 11). These Grenvillian chemical ages also overlap with the U-Pb (zircon) lower intercept dates obtained in the basement gneisses from the core of the proposed MCC [61, 67, 68, 71]. Smaller populations of relict older ages (Figure 11), e.g., 1.6–1.4 Ga in the basement gneisses, 1.4–1.3 Ga in blastoporphyritic granitoids, and rare 1.3–1.2 Ga in mica schists are obtained in embayed rafts in the interiors of monazite grains. The Early Mesoproterozoic ages are interpreted to correspond with the early D1-D2 granulite facies metamorphism in the basement gneisses. The Mid Mesoproterozoic (1.4–1.3 Ga) cores possibly relate to the age of post-D2, pre-D3 emplacement of the granitoids intrusive into the basement gneisses. The 1.3–1.2 Ga ages in the mica schists are difficult to interpret; these could be detrital ages or instead mark an episode of midcrustal contact heating induced by granitoid emplacement.

The older ages are common in monazite grains lodged in dynamically recrystallized quartz-feldspar aggregates; monazites within mica aggregates record Grenvillian ages throughout all or most of the monazite grain. P-T pseudosection analyses in the mica schist and deformation microtextures indicate that these midcrustal shallowly dipping fabrics (Domain II) formed at T between 640 and 690°C (Figure 10(h)). The T values are lower (100–150°C) than the blocking temperature for intracrystalline U-Pb diffusion in monazite (>800°C; [120122]). Clearly, therefore, in polychronous monazites, the Grenvillian age mantles and rims around older cores and entire grains formed due to the fluid-induced dissolution-precipitation process [123125]. We suggest that monazites lodged in the “dry” quartz-feldspar aggregates better preserve older ages than those hosted in mica aggregates because of a lack of infiltrating fluids. In summary, the detachment zone and the steep shear zones were broadly coeval (within the 2σ errors of spot ages) with MCC formation in the Early Neoproterozoic age. The small subsidiary age peak at ~1060 Ma (Figure 11(c)) is difficult to interpret; however, Bhattacharya et al. [74] correlate this age with the early phase of crustal thickening during accretion of the CGC with the Singhbhum Craton.

The Early Neoproterozoic accretion between the CGC with the Singhbhum Craton in the south occurred between 1.1 and 0.9 Ga [61, 73, 74]. The monazite chemical ages (this study) suggest that the D3 and D4 deformations that produced the CGC-MCC were contemporaneous with the convergent strain regime during the accretion. Documentation of domains of shallowly dipping foliations and recumbent/gently inclined folds truncated by steep-dipping ENE- to ESE-trending shear zones (Figures 1 and 2) has recently been made in large tracts of the CGC [62, 74]; these features are strikingly similar to the detachment zone of the proposed MCC in the Deoghar-Dumka-Chakai-Giridih area. Additionally, there are many areas in the CGC where the shallow-dipping fabrics are absent, exposing older basement rocks with steep, N-trending Mesoproterozoic fabrics. We propose that the CGC as a whole represents a coalescent agglomerate of regional-scale MCCs that developed within a convergent tectonic setting (Figure 12). After summarizing the major characteristics of each segment of the CGC-MCC, we propose a model for the formation of the CGC-MCC that integrates detachment zone formation (D3) and exhumation with transpressional shear zone development (D4).

8.1. The Core of the CGC-MCC

The granulite facies D1-D2 fabric-forming events in the basement gneisses (Domain I) involved pre-/syn-D1 voluminous melt generation manifested by metatexite layers in gneisses; the limited volumes of locally developed diatexites suggest that melt productivity waned considerably during the D2 deformation that produced the pervasive N-trending, steeply dipping SN2 gneissic layers in the anatectic gneisses. The Early Mesoproterozoic ages in the cores of monazites (1.51.4Ga) in the anatectic gneisses (Figure 11(c)), consistent with U-Pb (zircon) upper intercept ages and Pb-Pb zircon ages (1.6-1.45 Ga), are inferred to be the age of emplacement of felsic orthogneisses [6769] and granulite facies metamorphism in the gneisses in different parts of the CGC [61, 66, 71]. The 1.45Ga cores in monazites in granitoids that share the D2 fabric with the gneisses are similar to the U-Pb (zircon) ages documented by Mukherjee et al. [68] in felsic orthogneisses from the Dumka area. Possibly, this age marks the culmination of Early Mesoproterozoic felsic emplacement and granulite facies metamorphism in the basement gneisses.

The CGC-MCC formation occurred in the Early Neoproterozoic, and therefore, the Early Mesoproterozoic gneisses and granitoids testify to earlier deformation, anatectic and metamorphic events unrelated to MCC formation; these rocks occur as enclaves within the Early Neoproterozoic granitoids exhumed beneath the shallowly dipping detachment zone (Figure 12). Mukherjee et al. [67] interpret the N-trending oldest D1-D2 composite fabric in the gneisses to correspond to the youngest deformation, despite the fact that two subsequent Early Neoproterozoic deformation events affected the CGC, as documented in this study, manifested by the shallowly dipping D3 fabric and the E-striking basement piercing D4 shear zones. For this reason, we consider the interpretation of Mukherjee et al. [67] to be simplistic.

8.2. The Hanging Wall of the CGC-MCC

Extensive erosion in the CGC has removed much of the upper part of the hanging wall of the CGC-MCC. The surviving rocks comprise the amphibolite facies supracrustal unit that was deformed ductilely during the formation of the detachment zone, producing the penetrative shallowly dipping SS2 foliation in the rocks. The muscovite-biotite schist in the amphibolite facies supracrustal unit structurally overlying the basement gneisses is devoid of anatexis, and the rocks do not exhibit an intrusive relationship with the Mid Mesoproterozoic to Early Neoproterozoic granitoids [61, 62, 73] intrusive into the Early Mesoproterozoic basement gneisses. The shallow foliation in the supracrustal unit is a crenulation cleavage (SS2) and is coplanar with the shallowly dipping D3 fabrics in the gneisses and granitoids (Figures 3(b)–3(d)). This suggests that the earlier history in the supracrustal unit was not shared with the granitoids that lack the earlier fabrics or the basement gneisses in which the earlier fabrics formed at granulite facies conditions. In the muscovite-biotite schists, the lack of 1.6–1.5 Ga monazite ages corresponding to the D1-D2 anatectic conditions experienced by the Early Mesoproterozoic basement gneisses suggests that the Grenvillian age supracrustal rocks in the flat-lying/shallow-dipping carapace (Figure 11(a)) constitute an allochthonous unit transported over the basement gneisses and granitoids. Also, the dP/dT slope of the post-D3 prograde P-T path reconstructed for the supracrustal unit is gentler than those recorded in zones of accretion of cold crustal blocks. It is necessary therefore to invoke heat sources to explain the metamorphic evolution in the nonanatectic amphibolite facies supracrustal unit. We suggest that the tectonic emplacement of the supracrustal unit over the basement closely followed intrusion of the Grenvillian age granitoids. Release of heat from the cooling granitoids intrusive into the basement gneisses provided the necessary heat for the prograde heating.

8.3. The Detachment Zone

The deformation microstructures in granitoid mylonites in the detachment zone juxtaposed against the supracrustal unit are noteworthy. Domain II granitoid mylonites comprise core-mantle-structured feldspar porphyroclasts in finer grained matrices consisting of aggregates of moderately inequant grains of polygonized quartz and feldspar and myrmekites. The grains in the matrix are internally strained (undulatory extinction, subgrains, deformation twins, and bent twin lamellae) and share mildly curved grain/phase boundaries, although triple junctions are common. The porphyroclasts exhibit undulatory extinction and rare subgrains in magmatic cores of feldspars. The microstructural relations, in combination with the crystallographic data, suggest deformation by dislocation creep with some recrystallization. However, deformation temperature in excess of 650°C is precluded due to the absence of high-T deformation microstructures (such as chessboard twinning in quartz [126], randomly dispersed interphase boundaries [127], suprasolidus deformation microstructures (such as rationally developed faces of minerals growing into neighboring quartz and K-feldspar; [128]), imbrications of euhedral feldspar aggregates [129], and optically homogenous quartz grain piercing feldspars [130, 131]. We speculate therefore that emplacement of the supracrustal unit and formation of the detachment zone postdated intrusion of Early Neoproterozoic granitoids, but the cooling of the granitoids outlasted emplacement.

Results of P-T pseudosection analysis imply that the supracrustal unit experienced crustal thickening up to 16-18 km (5.25.5kbar lithostatic pressure; Figure 10(h)) syn-to-post-D3 deformation/metamorphism in the detachment zone. Crustal thickening can result in collapse manifest as vertical thinning, horizontal extension, and the formation of midcrustal flat-lying detachment zone even in convergent settings [6, 21, 132135]. In the CGC, vorticity estimates for the shallow-dipping granitoids (Figures 9(a)–9(c)) point to considerable influence of a pure shear component during D3 fabric formation in the detachment zone (cf., [136]). Also, rare shear sense indicators in the carapace consistently show a core-up sense of movement (Figure 5(e)). We suggest therefore that the line of demarcation between the detachment zone that experienced D3 deformation and the underlying Domain I basement gneisses (Figure 2) unaffected by D3 is a decollement that formed due to extension [136138]. Similar Grenvillian age structures are reported from the central parts of the CGC [62], indicating that most of the crustal block making up the CGC evolved together and was involved in the formation of a regional-scale MCC.

Results of our crystallographic analyses and vorticity estimations from rigid porphyroclasts indicate that deformation within the detachment zone was heterogeneous. The rare stretching lineations in the detachment zone do not show any decipherable map pattern, neither linear nor radial (Figures 2(b) and 3(e)), and the axes of D3 recumbent folds in the basement gneisses are similarly scattered (Figure 3(d)). This variable distribution of stretching lineations and recumbent fold axes is unusual in continental MCCs, in which stretching lineations normally display linear map patterns related to the flow direction of material during extension [1]. In the central CGC, the stretching lineations in the shallow-dipping detachment zone consistently plunge towards NW and SE [62], suggesting definite strain patterns associated with the detachment zone in these regions. The variable orientations of the shallow lineations in the Giridih-Dumka-Deoghar-Chakai quadrant are therefore difficult to explain and are possibly best justified by considering local variations in strain, although radial distribution of stretching lineation could be related to a diapiric rise of plutons [139]. Also, deeper down from the detachment, the bulk strain and relative contribution of simple shear is likely to decrease, and this could influence the dispersal of the stretching axis if different depths exposed by erosion are sampled.

The map pattern of the study area (Figures 1(b) and 2) with lithostructural ‘islands’ of Domain II rocks within Domain I and vice versa is similar to map patterns of regional-scale MCCs [140, 141] in which the detachment zone has been folded on a kilometer- scale by steeply inclined to upright open folds commonly termed ‘corrugations’ [142]. The fold axes of these corrugations are subparallel with the stretching direction in the detachment zones of the MCC [140, 142, 143]. However, although the shallowly dipping SS2 fabric in the supracrustal rocks and the SN3 fabric in the gneisses within the area display outcrop-scale upright folds, no large-scale corrugations were observed or could be inferred at the present level of exposure of the detachment zone in Domain II. We interpret the locally developed upright folds in the supracrustal rocks and basement gneisses as effects of the formation of the steeply dipping transpressional shear zones (D4).

8.4. The Steep Shear Zones

The orientations of folds in transtensional and transpressional shear zones have been investigated using numerical strain modeling [144148]. In vertical/steep-dipping transtensional shear zones, folds on flat/shallow-dipping planes initiate at angles >45° with the shear zone wall, and the hinge rotation determined by the instantaneous stretching axes and the divergence vector are small for pure shear-dominated systems even at infinitely high strain and moderate for wrench-dominated system [146]. By contrast, in vertical/steep-dipping transpressional shear zones, fold hinges rotate increasingly towards the shear zone wall [100, 149151] with higher bulk strain, although significant simple shear strain is necessary to rotate fold hinges parallel to the maximum finite stretch ([147], and references therein). In the ENE-/NE-trending and ESE-trending shear zones in Domains III and IV of the Giridih-Dumka-Deoghar-Chakai area (Figure 2(a)), the subhorizontal/gently plunging and noncylindrical hinge lines of E-/ESE-trending D4 upright folds on flat-lying/shallow-dipping schistosity in the supracrustal unit (Figures 3(h) and 3(j)) share low-angle obliquity with the stretching lineations of the granitoid mylonites (Figures 3(e) and 3(i)) and the walls of the two hosting shear zones. The shear zones therefore are transpressive in nature. The small variation in the plunges of the stretching lineations in the Domain III and IV granitoids (Figures 3(e) and 3(i)) indicates regional-scale heterogeneity in deformation strain within the shear zones (cf. [152]). Both shear zones exhibit sinistral kinematics. In the northern shear zone, the VNS is subhorizontal, implying that the shear zone follows a monoclinic symmetry that is consistent with the results of our crystallographic vorticity analyses; however, stretching lineation orientations (Figure 3(i)) and crystallographic data (Figure 8(e)) indicate that the deformation in the southern shear zone was at least locally triclinic. Although both shear zones dip steeply to the north, opposite sense of shear is exhibited in the section of the two shear zones; N-down for the northern shear zone and S-down for the southern shear zone (Figure 12). Since Grenvillian monazite ages in mica schists hosted within both shear zones overlap, we suggest the two shear zones were contemporaneous. The shear zones possibly constitute a sinistral freeway [153], although deep weathering and soil cover near the projected junction (NW of Giridih) prevent field confirmation.

Subhorizontal stretching lineations in transpressive shear zones with significant finite strain indicate a high simple shear component of deformation [77, 144, 154]. Vorticity analysis of the northern shear zone granitoids on the other hand produced Wm values intermediate to pure shear and simple shear (Figures 9(d)–9(f)). Li and Jiang [106] have demonstrated that the vorticity estimates from rigid porphyroclasts could be simple shear dominated even when the Wm values range between 0.50 and 0.85. The EBSD results (Figures 8(c)–8(f)) support the vorticity estimations and suggest a significant pure shear component for the D4 deformation in both Domains III and IV.

Therefore, if the Wm values are considered to be relatively indicative of the vorticity for the northern shear zone, the average of the range estimated is Wm=0.67. This value is tentatively used in the kinematic modeling of the shear zone following the triclinic transpression with oblique extrusion model of Fernandez and Diaz-Azpiroz [155] using the software GTOE (version 2) (Carlos Fernandez, personal communication). The following parameters are used in modeling: φ is the angle between the simple shear direction and the strike of the shear zone, υ is the angle between dip of the shear zone and the extrusion direction due to the coaxial component, Wk is Truesdell’s [99] kinematic vorticity number, and K is the log shape parameter of a coaxial component. The model yields the principal quadratic extensions (λ1λ2λ3) and their orientations with respect to the external reference frame fixed to the shear zone under investigation. The stretching lineation is approximated to the X axis (λ1) of the finite strain ellipsoid and the foliation is approximated to the XY plane (perpendicular to the Z axis, λ3).

Considering a vertical dip, a strike of 70°N and an average value Wk=0.67 for the northern shear zone, a value of υ=80° (taking into account the gently plunging stretching lineations) was chosen. Three values of φ, φ=0, 10, and 20 were modeled for, to account for the any uncertainty in the VNS estimation and three values of K were chosen, K=1, 2, and 105 with 105 assumed to be tending to infinity. The results of the modeling are shown in Figures 9(g)–9(j). Figure 9(g) shows the log-Flinn diagram while Figures 9(h)–9(j) display the orientations of the X axis (λ1) and XY planes (λ3) of the resultant strain ellipsoids. The orientations of the X axis and the XY planes of all considered models show a fairly good fit to the natural data (Figure 3(e)), indicating that the shear zone may not be distinctly monoclinic but could incorporate some amount of triclinicity as well. The log-Flinn diagram (Figure 9(g)) shows that only small values of K (K1) are compatible with the natural data because as K values are increased, the plots move into the constrictional field of the diagram. This would lead to the formation of L-tectonites, which were not observed in the field.

Thus, both the northern and the southern shear zones are transpressional shear zones with a significant contribution of pure shear, in contrast to the strike-slip shear zones reported in other MCCs [10, 38, 40, 42, 45, 54]. Although structurally the shear zones (D4) developed subsequent to the detachment zone (D3) of the MCC, the contemporaneity in the monazite age data in the two segments suggests that the shear zones played an integral part in the formation of the MCC in the Early Neoproterozoic.

8.5. The Model

The Early Neoproterozoic (1.1–0.9 Ga) D3-D4 deformations in the CGC are contemporaneous with the convergent strain regime during the accretion of the CGC with the composite of the Singhbhum Craton and the North Singhbhum Mobile belt [61, 73, 74]. The large-scale emplacement of granitoids in the CGC during the Grenvillian age is well documented [61, 62, 73]. Field relations and microstructural evidence suggest the deformation in these Grenvillian age granitoids (intrusive within the Early Mesoproterozoic basement gneisses and the older granitoids) outlasted their intrusion, but emplacement of the supracrustal rocks over the basement occurred with the cooling granitoids still being hot (Figure 10(h)). In other words, the intrusion of the granitoids preceded the formation of the MCC but the time interval between intrusion, thrusting, and detachment formation was likely to have been small (cf. [156]). Voluminous granitoid intrusion accelerates thermal weakening of the crust and renders it buoyant [157], promoting orogenic collapse and strain localization along shallow-dipping normal faults [1, 158]. Large-scale magmatic emplacements are thus likely associated with exhumation of the ductile crust [13, 159], and the Grenvillian-age granitoid intrusion within the CGC was possibly the trigger for the formation of the MCC.

Traditional models for the formation of the shallow dipping carapace within MCCs such as the “rolling-hinge detachments” [2, 9, 160] or “channel detachment” [8] models are based upon the relationships between the hanging wall and detachment. Within these models, the detachment fault results from the shallowing of originally steep-dipping normal faults which originate within the brittle upper crust. This mechanism is doubtful for MCCs with spatially expansive detachments such as the CGC, since the displacement along the envisaged normal faults would need to run hundreds of kilometers with exceedingly high rates of motion [6].

In the CGC, the detachment zone was formed in the ductile middle crust under high geothermal gradients and involved the allochthonous supracrustal unit and the screen of basement rocks immediately underlying it. We propose that the vertical shortening resulting from the collapse of a thickened crust combined with the buoyant rise of the basement dominated by granitoids was accommodated by strain localization and horizontal extension at the contact between the supracrustal unit and the basement, producing the shallow-dipping foliations in the granitoids (SG1) and supracrustal rocks (SS2) and recumbent folds within the anatectic gneisses (SN3). Such localized extension during an overall convergent regime has been observed in other MCCs around the world ([1, 14, 161] and references therein). The shallow foliations produced within this zone would thus show variations such as in the amount of displacement, orientation of stretching lineations (Figure 3(c)) and recumbent fold axis (Figure 3(b)), and in shear sense indicators in different parts of the detachment, as observed within the CGC and in other MCCs as well [6, 162].

The crucial factor governing the mechanics of the CGC-MCC nevertheless is the presence of the regional-scale, subvertical sinistral transpressional Grenvillian age interconnected ENE- and ESE-trending shear zones (Domains III and IV, this study; HFSZ, [62]). The shear zones cut across all three lithodemic units of the CGC and reorient all earlier fabrics (Figure 12). Most numerical models for subvertical transpressional shear zones consider the shortening component of stress to be normal to the shear zone which is accommodated by vertical [150], lateral [144, 163], or oblique [155, 164166] extrusion parallel to shear zone walls. Within the investigated N-down, sinistral, subvertical northern shear zone, gently plunging stretching lineations nearly parallel to the VNS imply a shear zone symmetry close to monoclinic symmetry. The coeval ENE-trending shear zone in the south shows identical sinistral kinematics and moderately plunging stretching lineations except for a S-down (reverse) sense of movement on the steeply N-dipping shear plane (Figures 3(i) and 7(f)). Although the subhorizontal stretching lineations suggest a lateral component of extrusion, the opposite movements in the profile sections are reminiscent of escape tectonics [167170], wherein the entire block between the two shear zones is considered competent enough to be displaced vertically en masse, especially in a convergent regime.

We suggest the basement-piercing transpressional shear zones accommodated the bulk of the compression strain in the ductile middle and lower crust during the accretion of the CGC with the Singhbhum Craton in the Grenvillian age and developed contemporaneously or closely following the formation of the detachment. Vertical extrusion of the MCC segments between the shear zones was possibly at least partially compensated by the vertical shortening associated with orogenic collapse of the thickened crust. This would account for the absence of considerable changes in the metamorphic grade of the basement rocks or the supracrustal rocks due to exhumation. There was also a considerable component of lateral stretch associated with the subvertical shear zones as modeled in Figures 9(g) and 9(h). MCCs with greater horizontal movement within the core as compared to vertical motion have been previously modeled [2, 171] but without considering the contribution of shear zones.

Thus, the CGC provides a new example of a metamorphic core complex that was developed during synconvergent extension (Figure 12). This type of deformation is well documented within the modern Himalaya (e.g., [172]) but is increasingly recognized in the geologic record, whether in core complexes previously traditionally thought to form entirely postconvergence (e.g., Naxos; [41]) or in midcrustal domes interpreted to form only during collision (e.g., the southern Appalachians; [173]). With this context, recognition of synconvergent extension in the CGC suggests that it may be worthwhile to reconsider the structural evolution of high-grade Precambrian complexes around the world.

All the data used in this study are incorporated within the article and in the supplementary material provided along with the article. No other data from any other source was used in the research described in the article.

The article forms a part of the doctoral dissertation of NS. SM and SS carried out parts of the work during their tenure at IIT Kharagpur as part of their Masters programs.

The authors declare that they have no conflicts of interest.

Financial support for the work was provided by University Grants Commission (New Delhi) through a Research Fellowship to NS. AB acknowledges financial support for fieldwork and chemical analyses provided by the host institute through the CPDA funding scheme for the block years 2017–2020. The authors appreciate the comments of W.M. Schwerdtner and an anonymous reviewer; their suggestions greatly improved the style and presentation of the manuscript. The editorial handling of the manuscript by Damien Nance is greatly appreciated.

Exclusive Licensee GeoScienceWorld. Distributed under a Creative Commons Attribution License (CC BY 4.0).