Post-Orogenic Tectonic Evolution of the Jiangnan-Xuefeng Orogenic Belt: Insights from Multiple Geochronometric Dating of the Mufushan Massif, South China

The Jiangnan-Xuefeng orogenic belt (JNXFOB) is a Triassic intra-continental orogen in central South China. It straddles the Neoproterozoic suture between the Cathaysia and Yangtze blocks, which constitute the South China Block (SCB). The JNXFOB is a key region to understand the tectonic evolution of the SCB. The SCB is a region of extensive Late Mesozoic-Early Cenozoic magmatism, with voluminous granitoids and associated world-class mineral deposits (Wu et al., 2020; Qian et al., 2019; Jiang et al., 2018; Hu et al., 2017; Li J W et al., 2014, 2010; Li X H et al., 2010). Earlier studies suggested various models for this magmatism: (1) Lithosphere thinning in the aftermath of the Triassic collision between Yangtze Block and North China Craton (Ratschbacher et al., 2003); (2) Pacific-Plate subduction (Jiang et al., 2009; Shen et al., 2009; Zhou et al., 2006) including subduction angle changes (He et al., 2010; Zhou and Li, 2000) and flat-slab underthrusting (Shen et al., 2012a; Li and Li, 2007); (3) lithosphere extension and magma ascent were controlled by major faults (Wan and Zhao, 2012). Most studies agree on the important role of lithosphere extension linked to the subduction of Pacific Plate beneath Asia for the magma generation in the SCB (e.g., Shen et al., 2012a, b ; Chen et al., 2009; Zhou et al., 2006). However, there is no consistent understanding of the magmatic activities and their genesis in the intra-continent orogen.

Here, in order to further understand magmatic activities in the intra-continental orogen, the Mufushan massif was investigated. New muscovite and biotite ⁴⁰Ar/³⁹Ar, apatite fission-track (AFT), and apatite (U-Th-Sm)/He (AHe) ages were reported, combining them with zircon U-Pb data for reconstructing their evolution from emplacement to exhumation, and understanding the genesis of the intra-plate magmatism and the architecture of the Jiangnan-Xuefeng intra-continental orogenic belt.

The Cathaysia Block and Yangtze Craton constitute the SCB, which is surrounded by the Tibetan Plateau (TP), the North China Craton (NCC) and the Pacific Plate. The JNXFOB is an Early Mesozoic intra-continental orogen (Chu et al., 2012a, b ; Shen et al., 2012c; Wang et al., 2005). Based on detailed structural geology, petrology, and isotopic-chronology, Chu et al. (2012b) developed a model of SE-directed continental underthrusting of the SCB. First, large thrust nappes thickened the SCB crust to form the JNXFOB orogen after ca. 245-225 Ma, and then voluminous post-orogenic granitoids stitched the thrust belt at 225-215 Ma. Thickening resumed in the Jurassic with folds and thrusts (Ding et al., 2017; Ge et al., 2016; Faure et al., 2014). In the Cretaceous, a series of ~NE-trending rift basins (Fig. 1) developed, e.g., the Dongting Lake, Jianghan, Boyang, and Yuanma basins, and the JNXFOB developed into a Basin-and-Range type extensional province (Wu et al., 2020; Yang et al., 2019; Li J H et al., 2012; Shen et al., 2012a, b ; Zhang et al., 2010). Seismic profiles across these basins (Fig. 2a) highlight low angle, large detachment faults cutting through the sub-basin basement and the pre-extensional thrusts and suggest that the basins are half-grabens and grabens controlled by normal faults (Fig. 2b). Some normal faults have come to be reversed since the Neogene (Fig. 2c).

Figure 1.  A brief geological map of the Jiangnan-Xuefeng orogenic belt (modifed from Shi et al., 2013). MFS. Mufushan granitoids; HR. Huarong granitoids; WX. Wangxiang granitoids; JHB. Jianghan Basin; YB. Yuanma Basin; DTLB. Dongting Lake Basin; BYB. Boyang Basin; Fault: F1. Jiangnan fault; F2. Jingshan fault; F3. Tan-lu fault; F4. Xiangguang fault; F5. Jiangshao fault. A, B, C. the location of seismic profile (see Fig. 2).

Figure 2.  (a) Seismic profile in the Jianghan Basin; (b) seismic profile in the Jiangnan orogenic belt; (c) seismic profile across Jianghan Basin, Huarong granite and Dongting Lake Basin.

The Mufushan massif constitutes the largest Mesozoic intrusive complex, located in the center of the JNXFOB. They are surrounded by granitoid massifs, e.g., the Wangxiang and Lianyunshan massifs (Fig. 1). These granitoids mostly formed in Late Yanshanian Period between 180-140 Ma through zircon U-Pb and biotite ⁴⁰Ar/³⁹Ar dating (Wang et al., 2014; Shen et al., 2012b; Li P C, 2006; Peng et al., 2004). The Mufushan granitoids intruded at ~154-146 Ma, coevally with Late Jurassic to Early Cretaceous tectonic shift, changing from a regional transpression to extension in eastern China (Shi et al., 2013; Shen et al., 2012a; Li J W et al., 2010). Based on field geological surveys and zircon U-Pb geochronology (Wang et al., 2014), the Mufushan granitoids comprise gabbro-diorite (~154 Ma), two-mica monzogranite (~151.5 Ma), biotite monozogranite (~148.3), and granodiorite (~145.8 Ma). Biotite K-Ar dating indicated their ages varying from ~143 to 108 Ma (Li P C, 2006; IGSH, 1974); their AFT ages are ~70 to 30 Ma (Shi et al., 2013). These data summarized in Table 1 and suggest a complex and prolonged exhumation history.

The Mufushan granitoids intruded the locally preserved Mesoproterozoic Lengjiaxi Formation. This formation consists mainly of low-grade phyllite and micaschist. The earliest and most mafic intrusion is a gabbro-diorite (hornblende and plagioclase (An_42-51)) and is exposed over only one square kilometer; it has typical mantle-derived geochemical characteristics (Wang et al., 2014). The most voluminous, second-phase intrusions are two-mica monzogranites and two-mica granites with a medium- to fine-grained, massive texture (K-feldspar, quartz, plagioclase (An_17-41), biotite and minor muscovite). Epidote-bearing granodiorites intruded last (10%-18% K-feldspar, 42%-54% plagioclase (An_34-42), 21%-27% quartz, 6%-12% biotite, and minor amphibole; Zou et al., 2011).

The locations of samples are shown in Fig. 3. Samples were collected from two-mica monzogranites. Their positions and elevations, determined by GPS, are listed with the geo-thermochronologic data (Tables 2, 3, 4, S1). Two samples (WT8-1, WT9-1) were dated by biotite and muscovite ⁴⁰Ar/³⁹Ar analysis. Three samples (WT7-1, WT8-1 and WT9-1) were selected for AFT and AHe geo-thermochronology analyses.

Figure 3.  Sample locations and a brief geological map of the Mufushan massif (modified from Wang et al., 2014). Pt. Proterozoic.

Biotite and muscovite ⁴⁰Ar/³⁹Ar analyses were conducted in the Argon-Argon Laboratory Freiberg, Germany. The minerals were hand-picked from the 200-250 μm sieve fraction, ultrasonically cleaned using deionized water. And then these minerals were dried and packed in irradiation holders for neutron irradiation. The neutron irradiation details and the detailed procedures of ⁴⁰Ar/³⁹Ar experiments see Sperner et al. (2014). Age calculation and graphical representation used Isoplot 3 (Ludwig, 2003). The age standard used Fish Canyon tuff sanidine (28.3±0.5 Ma); the errors of ages are 1σ. AFT analyses were fulfilled in the Particle Track Laboratory Freiberg, Germany. Two apatite separates were individually used for age dating and confined track-length measurements. The detailed methods of age dating see Ge et al. (2018) and Yang et al.(2017, 2013). The detailed procedures of confined track-length measurement follow in Jonckheere et al. (2007). The individual grain apatite AHe ages were analyzed at the (U-Th-Sm)/He laboratory in University of Florida, USA. The detailed methods refer Min et al. (2017) and Shen et al. (2012b).

The mica ⁴⁰Ar/³⁹Ar dating can reveal thermal history in the range of 300-400 ℃ (McDougall and Harrison, 1999). The ⁴⁰Ar/³⁹Ar ages and associated errors are listed in Table 2. The details of stepwise heating are summarized in Table S1. Figure 4 shows the age spectra. Weighted mean ages (WMA) of samples WT8-1 and WT9-1 were calculated by using the shaded steps. Their muscovite minerals yielded ages of 129.68±0.15 Ma (MSWD=0.26) and 128.16±0.31 Ma (MSWD=0.50) (Fig. 4), implying cooling from 800-900 ℃ of U-Pb zircon closure temperature (Cherniak and Watson, 2000) to 350-400 ℃ of muscovite ⁴⁰Ar/³⁹Ar system (Bai et al., 2018; Kelly, 2002) after magmatism. Similarly, the plateau ages of two biotite samples are 127.03±0.20 Ma (Sample WT8-1, MSWD=0.40) and 128.09± 0.32 Ma (Sample WT9-1, MSWD=0.56), which means it experienced through the biotite ⁴⁰Ar/³⁹Ar closure temperature interval of 300-350 ℃ (McDougall and Harrison, 1999) in Early Cretaceous.

Figure 4.  Samples' ⁴⁰Ar/³⁹Ar weighted mean ages (WMA) and K/Ca ration. TFA. Total fusion age; IA. isochron age; IIA. inverse isochron age. Errors are 1σ.

The detailed results of single grain AHe ages are available in Table 3. The determined AHe ages are widely scattered in the range of 46.7-123.5 Ma. Most of these AHe ages are older than their AFT ages, in contrary to expectations from the closure temperatures of these two systems. Only two aliquots from WT8-1 yielded AHe ages (mean=47.5 Ma) which are younger than their corresponding AFT ages. Such an inverted relationship between AHe and AFT ages has been reported in numerous studies, particularly for samples experienced prolonged cooling histories (Gleadow et al., 2002). The inverted age relationship can be explained by radiation damage (Shen et al., 2012b; Fitzgerald et al., 2006), U-Th rich inclusions (Farley, 2002), fluid inclusions or microtextures (Zeitler et al., 2017), crystal size (Reiners et al., 2004; Reiners and Farley, 2001), nearby U-Th rich phases (Murray et al., 2014; Spiegel et al., 2009), U-Th zonation (Farley et al., 1996) and fragmentation (Beucher et al., 2013; Brown et al., 2013). Our preliminary investigations on thin sections showed: (i) some zircons identified near apatite crystals; (ii) apatites included in biotite; (iii) Fe-oxide decorating apatite grain boundaries; (iv) apatite with anomalously high interference colors (Fig. 5). These observations suggest possibility of alpha implantation from the nearby U-Th rich phases into apatites (Reiners et al., 2014; Fitzgerald et al., 2006). In this case, the apparent AHe ages become older depending on the extent of alpha recoil implantation into apatites. The AHe ages do not display any clear relationship with eU, size and grain shape, hampering more clear demonstration of the old, scattered AHe ages. Shi et al. (2013) also reported widely scattered AHe ages (33.4 to 154.2 Ma); these AHe ages also are older than the AFT ages from granitoids in the Mufushan massif located north of our study area. Because it is unclear which caused the apparent old AHe ages, we excluded AHe ages older than AFT ages for our thermal history modeling.

Figure 5.  The images of thin section of apatite under microscope (Ap. apatite; Bi. biotite; Zrn. zircon).

The detailed results of AFT analyses are available in Table 4. The central ages are 53.5±1.6 Ma (WT7-1), 58.3±2.4 Ma (WT8-1), and 55.7±1.9 Ma (WT9-1). The track lengths show negatively skewed unimodal distributions. These samples' mean confined track lengths are 12.08±0.09 μm (SD: 1.33, WT7-1), 12.02±0.11 μm (SD: 1.41, WT8-1), and 11.83±0.10 μm (SD: 1.48, WT9-1), respectively. The mean values of the kinetic parameter D_par, that is c-axis parallel etch pit diameters (Donelick et al., 1999), are 1.57-1.71 μm, which are similar to 1.5-1.9 μm of Durango apatite standards (McDowell et al., 2005). It implied that these samples have a similar resistance to annealing as Durango apatite standards. The Track Age Spectrum Calculator (TASC) (Belton and Raab, 2010) was used to define the time when the AFT system is able to preserve tracks and thus record age information. Fig. 6 shows the track age spectra of our samples; the maximum ages of the onset of cooling are at 67±2 Ma (WT7-1), 79±3 Ma (WT8-1), and 79±3 Ma (WT9-1). Shi et al. (2013) obtained 8 AFT ages spanning ~27 to 71 Ma (Table 1; Fig. 3) from the Proterozoic micaschists and granitoids of the Mufushan massif north of our study area; their ages are not inconsistent with our results.

Figure 6.  The cooling onset ages determined using TASC (TASC, the track age spectrum calculator, from Belton and Raab, 2010, was used to define the time when the AFT system is able to preserve tracks and thus record age information).

The thermal histories (time-temperature (T-t) paths) of samples WT7-1, WT8-1 and WT9-1 were modeled using HeFTy 1.9.3 software (Ketcham et al., 2009, 2005). In the modeling, we used the fanning curvilinear annealing model (Ketcham et al., 1999) with the measured D_par values and c-axis projected lengths (Donelick et al., 2005, 1999). The initial mean track length determined from measurements of induced confined tracks in Durango apatite length standards (Ketcham, 2015) is L₀=16.0±0.8 μm (Shen et al., 2012b). Using Monte Carlo method generate one hundred good-fit T-t paths. In order to assess the fit between the modeled and measured results, we use the Kolmogorov-Smirnov (K-S) test (GOF). The thresholds of 0.05 and 0.5 are used for acceptable and good fits (Ketcham, 2005). Except for constraints on the annealing temperature of AFT and the surface temperature at present, we also constrain the age of AHe. The inverse modeling results (Fig. 7) are similar for all three samples with: (1) rapid cooling from ~80 to ~55 Ma, with samples entering into the AFT partial annealing zone (PAZ) and the AHe partial retention zone (PRZ); (2) very slow cooling from ~55 to ~18 Ma; and (3) accelerated cooling since ~18 Ma to the present surface temperature conditions (Fig. 7). This evolution process is coincidence with the Huarong granite (Shen et al., 2012b) although their ages of emplacement are different, which clearly illustrate their late exhumation history controlled by tectonic activity is analogical.

Figure 7.  The thermal histories (time-temperature paths) modeling results by using HeFTy 1.9.3 software (Ketcham et al., 2009, 2005). The red paths represent the statistical good paths, while green paths show the statistically acceptable paths.

Multiple geochronometers with different closure temperatures can be used to reconstruct thermal evolutions of studied area (Heberer et al., 2014; Shen et al., 2012b; Dong et al., 2011). These new data provide thermal history from plutonic emplacement (zircon U-Pb) to upper crustal exhumation (AFT and AHe) via intermediate cooling (mica ⁴⁰Ar/³⁹Ar) as shown in Fig. 8. The mica ⁴⁰Ar/³⁹Ar ages can reveal the timing when the rock cooled through 300-400 ℃ (McDougall and Harrison, 1999), connecting high-temperature conditions from zircon U-Pb (800-900 ℃) with low-temperature cooling path from AFT and AHe (45-120 ℃). To construct the low-T cooling path, the HeFTy inverse modeling results (Fig. 7) can be used to supplement and improve.

Figure 8.  The geochronological framework of the Mufushan granitoids from emplacement to exhumation (data are given in Table 1). The closure-T intervals for the applied methods (zircon U-Pb; mica ⁴⁰Ar/³⁹Ar; ZHe; AFT; AHe) are indicated in fading grey. The constrained thermal history are confined by dash line and shaded. The data of zircon U-Pb is from Wang et al. (2014); the ages of ZHe is from Shi et al. (2013).

The revealed time-temperature thermal histories of the Mufushan massif can be divided into four stages: (1) Stage I, constrained from the zircon U-Pb and mica ⁴⁰Ar/³⁹Ar data, shows almost linear cooling between ~148 Ma (~800-900 ℃) and 127 Ma (300-400 ℃) with having a high ~24 ℃/Ma cooling rate; (2) Stage II, recorded from the mica ⁴⁰Ar/³⁹Ar and the AFT systems between ca. 127 (300-400 ℃) and ca. 55 Ma (AFT partial annealing zone: 60-120 ℃) with ca. 3.5 ℃/Ma cooling rate; (3) Stage III, the slow cooing stage from 75 to 60 ℃, occurred at ca. 55-18 Ma, with having a cooling rate of ca. 0.4℃/Ma; and (4) Stage IV, since ca. 18 Ma the final accelerated cooling with a rate of ca. 2.2 ℃/Ma, from ~60 ℃ to the surface temperature.

Late Jurassic-Early Cretaceous Extensive magmatism in the SCB shows a systematic change of geochemical signatures with the oceanic subduction affinity near the plate boundary and the intra-plate affinity in the hinterland of the South China (e.g., Wang et al., 2014; Duggan et al., 2005; Li X H, 2000). This large scale pattern is also confirmed in the Mufushan granitoids, which is located behind the orogenic belt, displaying a typical intra-plate affinity (Fig. 9). Pirajno et al. (2009) suggested that most of the anorogenic intra-plate magmatism is likely to be related to asthenospheric upwelling, caused by the lower crust delamination and subcontinental lithosphere. The magmatism of Mufushan massif is also likely associated with far field effect of the subduction.

Figure 9.  The comprehensive genesis model of magmatism in the South China (revised from Li X H et al., 2010; Pirajno et al., 2009; Sun et al., 2008).

With the subduction of the Paleo-Pacific Plate, the oceanic lithosphere gradually dips into the continental lithosphere carrying abundant water. As the subduction angle increases, the subducting slab experiences dehydration more close to the plate boundary (Kuritani et al., 2011). As the distance from the plate boundary increases, the overlying asthenosphere experiences upwelling and decompression melting, transferring mantle signatures to the shallow crustal levels. Based on these schemes, the igneous rocks of the SE China can be divided into three types depending on the distance from the plate boundary: (1) the island arc magmatism in the coast of the SE China (close to the plate boundary) (Fig. 9) corresponds to the offshore magmatism in the Cathaysia Block (Sun et al., 2008); (2) the subduction-related igneous activities in the back-arc environment occurred inland (intermediate distance from the plate boundary), caused by the asthenospheric upwelling and decompression melting (Li X H et al., 2010; Jiang et al., 2009); (3) the intra-plate magmatism in studied area (away from the plate boundary).

The Mufushan granitoids, the first-phase, mantle-derived gabbro diorites and subsequent crust-derived granites (Wang et al., 2014), belonging to intermediate intrusive rocks may suggest that the magma formation occurred in the lower crust or upper mantle. Similar intra-plate magmatism also occurred in the Central Asian orogenic belt (Yarmolyuk et al., 2014) with almost identical thermal evolutions. The Mufushan area is probably least affected by asthenospheric upwelling because (1) it is distant from the back-arc basin which is believed to have formed by asthenospheric upwelling, and (2) Mesozoic magmatism occurred in the western of JNXFOB, which belongs to a stable continental crust (Wan and Zhao, 2012). Although a flat-slab subduction model can be used to explain the Mesozoic intra-

continental orogen and magmatic activities in the SCB (Li and Li, 2007), we also suggest an possibility that the subduction zones as the important part of plate tectonics is inclined to making the oceanic lithosphere into deep mantle (Walter et al., 2011) rather than subducting under the continental crust as a flat slab. The exceptionally thick crust in JNXFOB was produced by intra-continent orogeny during early Mesozoic, and most of magmatic activities occurred in the margin of the orogenic belt. Large-scale thrust faults developed during the active orogenic period cut the continental crust providing space for emplacement of the magma of the Mufushan granitoids. Seismic profiles across the JNXFOB (Fig. 2b) suggest that the large Jiangnan faults (Fig. 1) control the tectonic movement. This is also clear from the geographical distribution of these granitoids in JNXFOB occurring exclusively in the vicinity of major faults (Fig. 1; Wan and Zhao, 2012).

The tectonic evolutions of the Mufushan massif can be divided into four phases as illustrated in Fig. 10. In the initial stage (Stage 1 in Fig. 10), the intrusion of the Mufushan granitoid is followed by rapid cooling through heat exchange with the surrounding rock (Yu et al., 2013) during ca. 148 to 127 Ma as revealed from thermochronology (Phase I in Fig. 8). The magma ascended though the thrust faults and emplaced into the surrounding rocks. We connected the genesis of Mufushan granite with regional tectonic regime change from Late Jurassic transpression to Early Cretaceous extension (Li X H et al., 2010) caused by the movement of Pacific Plate. U-Pb Zircon dating of other granites in the adjacent regions (Shen et al., 2012b, c ; Li J W et al., 2010; Li P C, 2006) correspond to ~164-117 Ma, suggesting that the post-orogenic regional magmatism prevailed during Late Jurassic-Early Cretaceous in the entire SCB (Xie et al., 2011; Li X H et al., 2010; Jiang et al., 2009; Li X H, 2000). Because of the changing of tectonic regime, the intense rifting was initiated in ~120 Ma, followed by large scale extension until ca. 30 Ma (Shen et al., 2012b). Also, the intrusion ages become younger toward the subduction plate boundary, which can be explained by lithospheric extension associated with the subduction of the Paleo-Pacific Plate beneath the Eurasian Plate.

Figure 10.  The tectonic evolution model of Mufushan massif.

Many evidences support that the SCB was under an extensional setting since the enhanced subduction of the Paleo-Pacific Plate (Li X H et al., 2010). A series of NE trending rift basins (Fig. 1) may developed in adjacent regions, including the well-known basin-range structures in the XFSB (Shen et al., 2012b; Zhang et al., 2010).

Stage 2 corresponding to ca. 127 to 55 Ma (Fig. 10) can be explained by the edge shoulder surrounding the deposited depression caused rapid tectonic denudation and weathering on account of regional extension (Shi et al., 2013). Molasse deposits in peripheral basin in Early-Middle Cretaceous (Li et al., 2012; Zhang et al., 2010; Yuan et al., 2007) may reflect this denudation based on basin-mountain coupling (Shen, 2012c; Wang et al., 2005). In this stage, the root of XFSB experienced partial

delamination and subsidence to make lithosphere thinning. Numerous normal faults, which were inherited from prior structural boundaries, developed the rift basin (Figs. 2a, 10). The extension was enhanced during the Late Cretaceous when the subduction angle of the Pacific Plate increased (Wei et al., 2012; Zhou and Li, 2000). Zhang et al. (2010) suggested that the Yuanma Basin (Fig. 1) located in the western foothill of XFSB experienced regional extension transforming existing foreland basin to rift basin since the Cretaceous. Similarly, Bai et al. (2014) concluded that the Xiangyin depression of the Dongting Lake Basin (Fig. 1) situated in the left of Mufushan experienced a rapid rifting during the Late Cretaceous-Paleocene.

Stage 3 can be represented by a slow denudation in Paleogene (Fig. 8). Such a slow denudation is also well known similar in the eastern Tibetan Plateau areas including Sichuan Basin (Yang et al., 2017; Tian et al., 2015, 2013; Deng et al., 2013; Richardson et al., 2008), the eastern and western Qinling-Dabie Orogen (Heberer et al., 2014; Ratschbacher et al., 2006; Hu et al., 2006b), and the middle Yangtze Craton (Shen et al., 2018; Ge et al., 2016; Hu et al., 2006a). Since the Indian Plate collided with Eurasian continent ~55 million years ago, the rise and growth of the Tibet Plateau was initiated (Tapponnier, 2001). Also, the drifting direction of the Pacific Plate changed from NNW to NWW after ca. 50 Ma (Sun et al., 2008, 2007; Sharp and Clague, 2006). Tectonic evolutions of the SCB were significantly controlled by these two processes, such as the Tan-Lu fault evolution, the basin sedimentation and the peneplanation of Mesozoic granitoids. A set of stable continental deposits composed of conglomerate to mudstone formed in adjacent basin since Eocene (Huang and Hinnov, 2014; Li et al., 2012). These deposit formation is related to the widespread slow exhumation and peneplanation in the Mufushan granitoids toward the end of the extensional regime in the SCB, which is likely resulting in the development of the low-amplitude, long-wavelength topography of the eastern TP (Shen et al., 2018; Yang et al., 2017; Zhang et al., 2015; Deng et al., 2013) and the SCB (Shen et al., 2012b; Ratschbacher et al., 2006, 2003).

Stage 4 corresponds to the final accelerated cooling since ca. 18 Ma (Fig. 8), which can be attribute to the uplift and exhumation of Mufushan granitoids as a result of the eastward growth of the TP. The collision between the Indian and Eurasian plates resulted the uplift of the TP which has significantly affected geological, tectonic, geodynamic and climatic evolution in Asia during Cenozoic (Tian et al., 2012; Liu et al., 2004; Tapponnie, 2001). After the tectonic inversion, the compressional regime became dominated developing reverse faults along the existing normal fault planes (Fig. 10). The effect of TP is large enough to propagate to the Mufushan area just as that it affect regions as far north as Northern Kyrgyz Tien Shan (Macaulay et al., 2014) and the Baikal rift (de Grave et al., 2008). The geodetic data suggest that the northeastern TP is growing faster eastward than the SCB is extruding to the east (Enkelmann et al., 2006); the Pacific Plate subduction is also a constraint to the escaped structure (Avouac and Tapponnier, 1993), so the compression stress field is widespread in the JNXFOB and SCB. Numerous lines of evidence from adjacent basins support this compressive event, such as, the unconformity between Paleogene and Neogene in adjacent Jianghan Basin (Shen et al., 2009); the development of reverse faults and tectonic inversion in Dongting Lake Basin (Fig. 2c); the uplift and erosion of these rift basins (Zhang et al., 2010; Yuan et al., 2007). There are some geological events occurring in Miocene. The southeastern Tibetan Plateau experienced rapid cooling (Yang et al., 2017) from as early as ca. 23 Ma. The northern part of Longmenshan (Arne et al., 1997) and the Eastern Kunlun Mountains (Yuan et al., 2006) underwent rapid uplifting since ca. 20 Ma. The Sichuan Basin experienced final exhumation since ca. 20 Ma (Deng et al., 2013). The NE Sichuan Basin (Shen et al., 2009) and Michangshan (Ge et al., 2018; Yang et al., 2013; Tian et al., 2012) experienced uplift and denudation since ca. 15 Ma, which is coeval with the Qinling-Dabie Orogen (Shen et al., 2018; Hu et al., 2006). These ages reflect the intra-continental progressive expansion of the Tibetan Plateau to some extent.

The newly obtained muscovite ⁴⁰Ar/³⁹Ar, biotite ⁴⁰Ar/³⁹Ar, AFT and AHe data, combined with previously zircon U-Pb and (U-Th-Sm)/He thermochronology ages, yield thermal evolutions of the Mufushan granitoids from their emplacement to near-surface exhumation. The intrusion of this granitoids occurred at ca. 154-146 Ma and following splitting natural cooling from emplacement to ca. 127 Ma, which indicate the Late Jurassic to Early Cretaceous regional tectonic switch from transpression to extension. The muscovite and biotite ⁴⁰Ar/³⁹Ar ages suggest the denudation of the edge shoulder like the Mufushan rocks until passing though the closure temperature of apatite at ca. 55 Ma (Stage 2 and 3). According to the new AFT and AHe data with thermal modeling results, slow cooling occurred during the Paleogene (ca. 55-18 Ma) suggesting insignificant denudation, peneplanation, and local burial during the late stage extension in the SCB. The final accelerated cooling since ca. 18 Ma is attributed to the uplift and denudation of the Mufushan massif as a result of the TP uplift and eastward growth which is corresponding to the collision of the Indian and Eurasian continents. The intra-plate, anorogenic magmatism was generated by partial melting of the lower crust, which was caused from asthenospheric upwelling and large-scale thrust faulting.

This study was supported by the NSFC projects (Nos. 41972152, 41672140, 41372140), the "Thirteenth Five-Year" Major National Science and Technology Programs (Nos. 2017ZX05032-002-004, 2016ZX05024-002-005), the Special Fund for Basic Scientific Research of Central Colleges, China University of Geosciences, Wuhan (No. CUGCJ 1820), the "111" Program (No. B14031) and the Natural Science Foundation of Hubei Province (No. 2016CFA055). The authors are grateful to Professor Wanming Yuan and Dr. Bastian Wauschkuhn for supporting fission-track analysis, Professor Lothar Ratschbacher and Dr. Raymond Jonckheere for polishing the English. The final publication is available at Springer via https://doi.org/10.1007/s12583-020-1346-2.

Electronic Supplementary Material: Supplementary material (Table S1) is available in the online version of this article at https://doi.org/10.1007/s12583-020-1346-2.