The Beishan orogen, a significant component of the southern Altaids, presents an opportunity for investigating the intracontinental deformation and exhumation history of the Altaids during the Mesozoic era. Although previous studies indicated that the Beishan orogen has experienced multiple reactivation since the late Mesozoic, the precise extent of these events remains poorly constrained. Here, we provide a comprehensive synthesis of field observations and apatite fission track (AFT) thermochronological dating throughout the Beishan orogen. Detailed field observations confirmed four major E-W trending thrusts in our study area. Based on the youngest truncated strata associated with the thrusts and previous dating results from neighboring regions, we propose that these thrust sheets likely developed in the late Middle Jurassic. AFT dating results from seven pre-Mesozoic granitoid samples and associated with thermal history modeling demonstrate that the Beishan orogen experienced a rapid basement cooling during the mid-Cretaceous (~115–80 Ma). Moreover, a compilation of previously published and newly gained AFT data reveals a comparable mid-Cretaceous cooling event in other parts of Central Asia, such as Qilian Shan, Eastern Tianshan, and Altai-Sayan. This observation suggests that the mid-Cretaceous cooling event is more likely to be regional rather than localized. This mid-Cretaceous cooling pulse is interpreted as a tectonic exhumation controlled by boundary faults and related to the rotation of the Junggar and Tarim basins. These processes are linked to distant plate-margin events along the Eurasian continent.

The Altaids, also known as the Central Asian orogenic belt, is sandwiched between the Tarim and North China cratons to the south and the Siberian and Eastern European cratons to the north and is one of the largest Phanerozoic accretionary orogens in the world [1-4]. It is widely accepted that the Altaids were initially established during the Neoproterozoic-Paleozoic [1, 5, 6], and then parts of Altaids were reactivated in response to accretionary and collisional events along the Eurasian margin during the Meso-Cenozoic [7-9]. Located in the southernmost part of the Altaids, the Beishan orogen connects to the Solonker suture to the east and the South Tianshan suture to the west (Figure 1). Initiating in the late Paleozoic, the Beishan orogen amalgamated from a series of accretions of island arcs during the Devonian to Triassic [10-12]. Following the Paleozoic-early Mesozoic orogenic event, the Beishan orogen was deformed and reactivated during the Jurassic in response to the collision of Cimmerian continental fragments with the southern margin of Eurasia [13-15]. As a result, the Beishan orogen has been in a critical tectonic position for addressing the thermotectonic evolution of the southern Altaids since the closure of the Paleo-Asian ocean [10, 16].

Over the past two decades, a series of investigations have been conducted in the Beishan orogen, from Paleozoic-early Mesozoic accretionary and collisional processes [10, 12, 16, 17] to Cenozoic reactivation [18-20]. Nevertheless, some debates remain regarding the late Mesozoic tectonic evolution of the Beishan orogen. For instance, Zheng et al. [21] discovered several great Jurassic thrust sheets in the Beishan-Southern Gobi with minimum displacements of ~120–180 km. Massive Mesoproterozoic limestones were carried by the thrusts over layers spanning the Neoproterozoic to Lower-Middle Jurassic [21]. These authors preliminarily suggested that the thrust sheets in the Beishan orogen might originate in the Middle Jurassic [21]. Faults significantly control intracontinental crustal evolution by absorbing deformation and stress; however, the activities of these thrusts are poorly investigated.

On the other hand, based on low-temperature thermochronological data, some researchers suggested that the Beishan orogen has suffered three main episodes of exhumation since the Mesozoic, including (1) Late Triassic-Early Jurassic (~220–180 Ma [13, 14]), (2) Early Cretaceous (ca.~130–100 Ma [14]), and (3) Late Cretaceous-Early Paleogene (~85–60 Ma [14]). By comparison, recent thermochronological ages and inverse thermal modeling unveiled a distinct mid-Cretaceous (~115–80 Ma) phase of rapid basement cooling in the west parts of Beishan orogen and in the neighboring Eastern Tianshan [22-28]. Nevertheless, the possible causes for this apparent discrepancy regarding the Cretaceous cooling histories surrounding the Beishan orogen have not been comprehensively understood. Furthermore, scholars hold different views on the distribution and extent of the mid-Cretaceous cooling event. Some scholars suggest that this mid-Cretaceous cooling event is relatively localized near major faults [23, 24], while others lay stress on its widespread occurrence and more frequent recognition in the Beishan orogen and its adjacent regions [28].

The abovementioned unclear issues call for additional investigations of the late Mesozoic tectonic-thermal evolution of the Beishan orogen. Considering that previous studies primarily focused on the western part of the Beishan orogen (Figure 2) [14, 15, 26, 29], the available data may not be able to reveal the complex thermotectonic evolution of the entire Beishan region. In this contribution, we conduct detailed field observations across the middle section of the Beishan orogen, as well as apatite fission track (AFT) analysis and associated thermal history modeling on seven pre-Mesozoic granitoid bedrock samples collected from the thrust-bounder areas. By integrating new with published data from the adjacent tectonic units, we aim to enhance our understanding of the influence of thrusts on exhumation processes in the Beishan orogen and reconstruct the thermal history and geodynamic evolution of the Beishan orogen since the late Mesozoic.

As a main constituent of the southern Altaids, the Beishan orogen is a huge and long-lived accretionary system that was active from around the late Precambrian to Mesozoic [10, 15, 29]. It is located on the Eastern Tianshan to the west, the Qilian Shan to the south, the Alxa block to the east, and the Gobi-Mongolia to the north (Figure 1(b)) [10, 30]. It has lower relief and elevations (~1500–2583 m) than neighboring orogenic belts (e.g., Qilianshan, Tianshan, and Altai), fewer and smaller late Cenozoic alluvial basins, and comparatively moderate levels of historical seismicity [19]. Tectonically, the Beishan orogen is divided into five tectonic units (Queershan, Heiyingshan, Mazongshan, Shuangyingshan, and Shibanshan arcs) from north to south by ophiolitic belts and larger-scale faults (Figure 2(a)) [10, 12]. Magmatism in the Beishan orogen is characterized by two major pulses. The first pulse occurred during the Silurian-Early Devonian period and was mainly associated with subduction processes [10]. The second pulse took place in the Late Permian-Triassic period, likely linked to a postorogenic setting [31]. Archaean to Mesozoic stratigraphic units can be found in Beishan orogen, including our study area (Figure 2). These stratigraphic units can be divided into two categories depending on the types of rocks and metamorphic grades [13, 32]. The Proterozoic and Paleozoic strata, which make up the majority of the lower portion, are composed of penetrative fabrics and low-grade marine sedimentary and submarine volcanic rocks [13, 32]. The Upper Triassic molasse, Lower-Middle Jurassic coal-bearing unit, Upper Jurassic elastic unit, and Cretaceous red beds comprise the upper section [21, 33]. In contrast to the lower units, the rocks of the top section did not undergo metamorphism and primarily exhibited brittle deformation behavior without penetrative fabrics (Figures 3 and 4) [21, 34].

The Paleozoic Beishan orogen recorded the opening and subsequent closure of several oceans of the Paleo-Asian domain as multiple terranes collided with the northern margin of the Tarim-North China cratons [10, 11, 15]. It is hypothesized that the eastern parts of the Tarim Craton underwent a collision with the Chinese Eastern Tianshan during the Late Carboniferous to Permian [10]. This collision resulted in forming a new active continental margin, which interacted with the Beishan archipelago, giving rise to a complex subduction-accretionary orogen [10]. The closure of the Paleo-Asian ocean and the subsequent collision between the Beishan and Eastern Tianshan resulted in the formation of the Xingxingxia Fault [10, 30] and triggered significant crustal thickening, accompanied by magmatism and transpressional strike-slip reactivation [10, 14, 30].

Mesozoic intracontinental deformation in the Beishan orogen is expressed as thrusting and strike-slip faulting [21, 30, 34, 35]. For example, the Eastern Tianshan in the west and the Beishan orogen are divided by the northeast-striking Xingxingxia Fault [30]. The time of left-lateral motion of the Xingxingxia Fault was constrained at~235–240 Ma by 40Ar/39Ar dating of muscovite and biotite from mylonite, and a displacement of ~30–35 km was estimated [30]. In addition, stratigraphic and sedimentological studies identified Jurassic north-south contractional deformation in the Beishan orogen with a minimum displacement of ~120–180 km [21, 35]. This Jurassic regional contraction was also confirmed in the Alxa Block, Gobi Altai, Hexi Corridor, and Longshoushan [21, 34-37].

Due to its low mountain relief, the relative absence of seismicity, and the minor tilt of subhorizontal Cenozoic strata, the Beishan orogen has previously been classified as a stable Cenozoic crustal fragment [13, 18, 34]. Along the southern boundary of the Beishan orogen, recent investigations showed Cenozoic reactivation or active deformation [18-20]. For example, Late Pliocene to Early Pleistocene left-slip faults [19] and Late Quaternary left-slip faults with reverse- or normal-slip components [18-20] were both reported from the Beishan orogen based on structural and remote sensing analysis, paleoseismological trench studies, and Quaternary dating of alluvial sediments. However, there is no thermochronological signal of exhumation in reaction to the India-Asia collision, indicating possible tectonic quiescence over the Cenozoic [13, 14].

3.1. Field Observations

Four major E-W trending thrusts occur in our study area. From north to south, they are named the Pochengshan, Mazongshan-Niujuanzi, Jiangjuntai, and Laojunmiao thrusts, respectively (Figures 2 and 3). Between these thrusts, there are several klippes of various sizes. These klippes are mostly made of Mesoproterozoic dolomitic limestone and a few mylonite series (Figures 3 and 4).

Among them, the northernmost one is the Pochengshan klippe (Figures 3(a) and 4(a)-(b),3,4), which mainly consists of massive dolomitic limestone dipping steeply to the north. Neoproterozoic tillite and Cambrian sandstone are exposed in the north of the primary klippe, forming a half-window with an opening to the west. The Mazongshan (Figures 3(b) and 4(c)–4(d)) and Niujuanz thrusts (Figures 3(c) and 4(e)–4(f)) are located southeast of the Pochengshan klippe. The former replaces Permian sandy mudstone on top of Jurassic coal-bearing sandstone (Figure 3(b)). The fault surface dips to the north at angles of 52°–72° (Figure 3(b)). In the Niujuanzi thrust, Jurassic strata on the lower plate gradually steepen northward toward the fault surface, generating a “drag fold” within a zone around 20 m thick (Figures 3(c) and 4(e)–4(f)). Such a flexure indicates that the hanging wall moved southward in relation to the footwall. The Laojunmiao klippe lies southeast of the Niujuanzi thrusts, exhibiting Mesoproterozoic metamorphosed granite and granitic mylonite thrust over the Permian granite and sandy mudstone (Figures 3(d) and 4(g)–4(h)). The upper plate granitic mylonite takes the shape of an antiform adjacent to the thrust and is compatible with southward thrusting (Figures 3(d) and 4(g)). Additionally, several specific structures, such as microfold (Figure 4(b)), asymmetric folds (Figures 4(e) and 4(f)), and slicken lines or fiber lineations (Figure 4(h)), can be observed in these thrusts. These structures consistently demonstrate that the higher plate of the thrusts generally moved northward in relation to the bottom plate.

We also discovered that a Lower-Middle Jurassic coal-bearing unit in the Mazongshan and Niujuanzi sections is the youngest stratigraphic unit truncated by the thrusts (Figures 3(b)–3(c)). In conjunction with other studies [21], it is suggested that Cretaceous red beds and Upper Jurassic elastic strata unconformably cover the thrusts [21], and this angular unconformity marked the cessation of thrusting events. Additionally, Zheng et al. [21] reported a metamorphic core complex in the nearby South Gobi and later obtained the Rb-Sr isochron date of ~153 Ma and the 40Ar-39Ar plateau age of ~155 Ma. The formation age of thrusts can be determined in part by the metamorphic core complex, which is the product of an extensional event after the thrust event [21]. Based on the above-detailed observations, we suggest that the thrust sheets in the Beishan orogen may have developed in the late Middle Jurassic.

3.2. Sampling Sites

In this study, we supplemented samples from the thrust-bounder areas across the middle parts of the Beishan orogen. Seven pre-Mesozoic granitoid samples were collected along a ~300 km long, near N-S transect that was perpendicular to the thrusts’ (main direction of) strike across the study region. To assess the impact of these thrusting structures on the deformation and exhumation of Beishan orogen since the late Mesozoic, major thrusts and klippes were targeted during sampling. Six samples were collected from the Paleozoic granites within the thrusts, while sample 17BL-1 was taken from the Mesoproterozoic mylonite. Detailed sample descriptions are shown in Table 1, and their locations are exhibited in Figure 2 as well.

4.1. AFT Dating Method

Sample preparation for AFT dating was performed at the State Key Laboratory for Mineral Deposits Research, Nanjing University, China. Apatite grains were embedded in epoxy resin on glass slides, polished to expose internal surfaces, and etched in 5M HNO3 at 20°C for 20 seconds to reveal the spontaneous fission tracks. Apatite grains with polished surfaces parallel to prismatic crystal faces and homogeneous track distributions were selected and then imaged by the Autoscan fission-track analysis system [38] based on a Zeiss Axio Imager M2m microscope. The Track Works software [38] was used to capture stacks of high-resolution digital images of selected grains with a 100× dry objective under both transmitted and reflected light using a highly sensitive and fast iDS camera. Fission tracks were manually counted using the FastTracks software [39]. True confined track lengths [40] were measured on this software. Track etch pit diameters parallel to the crystallographic c-axis (Dpar) and angles with respect to the crystallographic c-axis direction. The 238U concentration of each analyzed grain was measured using the LA-ICP-MS (Laser Ablation Inductively Coupled Plasma Mass Spectrometry) method by an Agilent 8900 ICP-QQQ coupled to an ESL (Elemental Scientific Lasers) New Wave NWR 193UC (TwoVol2) laser ablation system at the Beijing Quick-Thermo Science and Technology Co., Ltd. Ablation was carried out over 25 seconds with a 30 µm diameter beam size, ~3 J/cm2 energy and 5 Hz repetition rate on the selected grains and reference materials (primary NIST SRM 612 glass and secondary apatite Durango) after approximately 10 seconds baseline signal collection. Uranium concentrations were calculated using calcium (assumed stoichiometric in apatite) as an internal elemental standard in the Iolite data reduction package within the Wavemetrics Igor Pro data analysis software [41]. Single-grain AFT ages were calculated using an absolute calibration based on primary constants (aggregate ξ = 2.001 × 10-3). RadialPlotter [42] was used to calculate a central age for each sample. It also visualized the dispersion of individual grain ages and corresponding Dpar values, which reflect different annealing kinetics [42].

Inverse thermal history modeling was performed using the HeFTy software [43] with the annealing model [44] based on the apparent AFT ages, Dpar values, and confined track length data. HeFTy presents candidate time-temperature paths using a constrained Monte Carlo scheme, enabling the user to specify t-T box constraints for each statistically acceptable t-T path. Old AFT apparent ages were used as high-temperature constraints for samples without Zircon (U-Th)/He (ZHe) data. Therefore, the initial modeling constraints in this study are set at the oldest single AFT ages at a temperature ranging from 120°C to 200°C. All models were set to continue creating and evaluating paths until 500 paths had passed the “good” criterion. The weighted mean of all good paths was considered the most representative thermal history.

4.2. AFT Age and Thermal History Modeling Results

Seven samples were analyzed using the AFT method, and the results are presented in Table 2. Radial plots and confined fission track length histograms are shown in Figure 5. The AFT central ages obtained in this study are dominantly Cretaceous, ranging between 139 and 93 Ma, except for one sample (17BW-5) displaying Jurassic central age (~200 Ma). Five samples pass the χ2 test (P2] > 5 %) and can thus be treated as statistically single population samples, except for samples 17BL-1 and 17BW-8. Samples 17BL-1 and 17BW-8 yield central ages of 92.7 ± 5.0 and 107.7 ± 8.2 Ma, respectively, correlating well with the other central ages for this study. The mean confined track lengths (MTLs) of all samples show unimodal distributions and yield mean lengths between ~13.4 and ~13.7 μm with standard deviations of ~1.0 to ~1.1 μm, indicating certain degrees of thermal annealing. In addition, samples that display younger AFT central ages are closer to major structures (Figure 2).

Thermal history modeling was conducted on four samples (samples 17BL-1, 17BW-1, 17BW-6, and 17BW-12) from which at least one hundred horizontal confined tracks were measured (Figure 6). Due to poor sample quality (e.g., small grain size, limited crystals yielded, or low track densities), other samples yielded insufficient numbers of confined tracks to build convincing track length distributions for inverse modeling. The spatial distribution (Figure 7(a)) and the weighted mean paths (Figure 7(b)) of all the modeled samples are compiled here and used to represent the most probable cooling history in the discussion. In more detail, samples 17BL-1 and 17BW-6 display similar cooling paths, and they record rapid cooling through the apatite partial annealing zone (APAZ) [45] during ~110–95 Ma, and the accelerated cooling lasted until the mid-Cretaceous at ~90 Ma (Figures 6 and 7(b)). By comparison, the cooling pulse of samples 17BW-1 and 17BW-12 initiated slightly earlier at the Early Cretaceous (~130 Ma), and they rapidly cooled through the APAZ during ~120–110 Ma, and this rapid cooling lasted until the mid-Cretaceous at ~100 Ma (Figures 6 and 7(b)). It is worth noting that a final rapid cooling event during the late Cenozoic is observed in the modeled samples. However, this cooling occurred outside the APAZ window, which may be associated with a well-known modeling artifact [46]. Hence, this final cooling step is excluded from further discussion.

5.1. Cretaceous Reactivation of the Beishan Orogen

The present-day geometry of the Beishan orogen is generally considered to have resulted from multistage Meso-Cenozoic deformation and reactivation of older Paleozoic structures [13, 14, 18-21, 28, 34]. However, one controversial issue remains regarding whether a mid-Cretaceous reactivation occurred in the Beishan orogen. Here, we plot our data versus a compilation (Figure 7) of available AFT data for the Beishan orogen (including the Xingxingxia faults) using results from Tian et al. [13], Gillespie et al. [14], Wang et al. [25], and Luo et al. [28]. The compiled thermal history modeling results demonstrate a consistent pattern across the Beishan orogen, with a substantial mid-Cretaceous (~115–80 Ma) rapid cooling event (Figure 7).

A boomerang plot [45, 47] displaying the relationships between AFT age and mean track length for each sample is presented in Figure 8. More specifically, samples yielding long (>13.5 μm) mean track lengths indicate rapid basement cooling through the APAZ, while samples with lower track lengths have an extensive residence in the APAZ. Here, we plot our data versus a compilation of available AFT data (see Figure 8) in the Beishan orogen, which aims to interpret the significance of the newly gained AFT data in a more regional context. The plotting results show a systematic trend for all areas, indicating a significant rapid cooling event at ~115–80 Ma (Figure 8). The extensive Cretaceous sediments observed in the adjacent basin of the eastern Beishan orogen [33] support the correlation between the~115 and 80 Ma cooling signal and a substantial exhumation phase. Brittle reactivation of inherited structures or the overall uplift controlled by boundary faults may play a role in this phase of exhumation [14].

Previous studies suggested that a rapid cooling was initially recorded in the Late Triassic and Early Jurassic, indicating a major pulse of orogenesis in the Beishan orogen induced by the closure of the Meso-Tethyan ocean to the south [13, 14]. This event was followed by slow cooling (with shorter mean track lengths) during the Middle to Late Jurassic, fitting well with this period with moderate tectonic activity interpreted to a late Middle Jurassic thrust event in the Beishan orogen. Additionally, samples from this contribution and previously published studies in the Beishan orogen reveal a secondary trend of decreasing mean track length values with decreasing age since ~90 Ma (Figure 8). Samples with AFT ages younger than ~80 Ma are again associated with lower mean track values and broader fission track distributions, suggesting that they are likely partially reset by a more youthful Cenozoic event (Figure 8). However, the available data cannot constrain the time of this younger event.

Although this mid-Cretaceous rapid cooling event (~115–80Ma) is significant in the Beishan orogen, there is a debate about whether this cooling event is controlled by compressional or extensional deformation. Recent studies indicated that an Early Cretaceous tectonic transition from contraction to extension has been well recognized in a vast region of the southern Altaids, from the Beishan orogen in the west to the Yanshan-Yinshan belt in the east [21, 48-55]. Evidence for this tectonic transition is the widespread occurrence of Early Cretaceous extensional basin systems on fold-thrust belts and previously thickened crust [48, 56, 57]. For example, two Early Cretaceous normal faults are reported in the Hongliudaquan area [55], postdating out-of-sequence thrusting until ~140–133 Ma in the Beishan fold-thrust belt [55]. In addition, the Yagan metamorphic core complex in southern Mongolia has biotite 40Ar/ 39Ar ages of ~129–126 Ma for mylonitic rocks associated with detachment faulting [58]. Furthermore, Early Cretaceous rifting-related basalts with the age ranges of ~126–99 Ma are determined in the Alxa block, southern Mongolia, Hexi Corridor, and northern Qilian Shan [56, 59, 60]. However, Liu et al. [55] suggested that the extensional deformation in the Beishan orogen, which probably started at ~133–129 Ma, had ceased before the end of the Early Cretaceous, and its influence is impossible to persist until ~80 Ma. Also, a series of late Cretaceous thrust nappe structures or strike-slip deformations have been reported in the eastern margin of Alxa [61-63]. Therefore, the mid-Cretaceous tectonic setting of the Beishan orogen is still debated due to the lack of records of structural deformation (in field observations and seismic profiles) and sedimentary (e.g., growth strata). More detailed studies are required to be performed in the future to answer this question.

5.2. Comparison with Adjacent Regions

Figure 9 displays the spatial distribution of published and new AFT data on basement rocks in Central Asia. To the west of the Beishan orogen, there is evidence pointing to Cretaceous cooling events in the Chinese Tianshan, as indicated by Luo et al. [28] and Yin et al. [64]. Sedimentological records further support the existence of an ancient mountain range that separated the Tarim and Junggar basins throughout the Cretaceous [65, 66]. Rapid cooling during the Cretaceous was demonstrated in the Talas-Fergana [67, 68], the Issyk-Kul and Song-Kul [69, 70], and in the Balkhash [71] of the Kyrgyz Tianshan. Due to the significant distance between these ranges and the study region, it is challenging to establish a precise correlation between the varying thermal histories of the Kyrgyz Tianshan and Beishan orogen, as highlighted by Gillespie et al. [14]. It is noteworthy that Kyrgyz Tianshan and Chinese Western Tianshan mainly underwent the Early Cretaceous (>~120 Ma) or/and Late Cretaceous to Early Paleogene (~75–60 Ma) rapid exhumation. A mid-Cretaceous exhumation, however, was mainly reported from the Eastern Tianshan (e.g., in the Moqinwula mountain [23]; in the Harlik mountain [22, 24, 26]; in the Yamansu arc [26]; in the Bogda mountain [27]).

In the Altai region, situated north of the Beishan orogen, the Cretaceous basement cooling event has been extensively documented based on AFT data (Figure 9). Early Cretaceous (~140–115 Ma) rapid exhumation can be observed in the Mongolian and Gobi Altai [72-74]. Similar findings have been obtained from the East Sayan-Lake Baikal, demonstrating the Early Cretaceous (~140–120 Ma) rapid exhumation [75-77]. In contrast to the thermal histories postulated for the Mongolian and Gobi Altai, AFT data have revealed a considerable mid-Cretaceous fast cooling in the Chinese Altai [26, 78-80] and the Siberian Altai [8, 81-86]. Samples taken along the Fuyun fault and the Irtysh shear zone in the Chinese Altai recorded a rapid mid-Cretaceous exhumation [80].

To the south of the Beishan orogen, the Altyn Tagh-Qilian Shan provides evidence of Cretaceous basement cooling phases based on AFT data. For example, the Early Cretaceous (~120–100 Ma) cooling phase was recorded by AFT data obtained from the Altyn Tagh fault [87, 88]. In comparison, AFT data from the Qaidam basin to the south and the North Qilian Shan foreland to the north indicate cooling events that occurred during the Late Cretaceous (~90–60 Ma) [37, 89]. Inverse modeling results based on AFT thermochronology support a mid-Cretaceous cooling event that took place in the inner Qilian Shan [90, 91]. The North Tianshan [64] and Alxa block [36] also exhibit scattered evidence of a mid-Cretaceous cooling event. Conclusively, this study adds solid evidence of a mid-Cretaceous cooling pulse in the southern Altaids, which also occurred in the surrounding blocks, albeit with variations in age and magnitude (Figure 9). Hence, the mid-Cretaceous cooling can be regarded as a regional event in the southern Altaids.

5.3. Dynamic Sources of Mid-Cretaceous Cooling

5.3.1. Response to Far-Field Effects

Determining the primary geodynamic triggering factor responsible for the mid-Cretaceous cooling in the Beishan orogen and surrounding areas poses a challenging issue. The aforementioned tectonic episodes, which are typically regarded as major plate convergence sites in Central and East Asia during the late Mesozoic (Figure 10), are most likely what caused the late Mesozoic intracontinental deformation and reactivation in the Altaids [92-95]. Previous studies have put forward several distant tectonic processes that could potentially account for the Cretaceous exhumation events in the Central Asian highlands: (1) the collision between the Lhasa block and Eurasia [69, 78, 87], (2) the closure of the Mongol-Okhotsk ocean [75-77, 84]; (3) the accretion of the Kohistan-Dras arc to Eurasia [14, 22, 26, 65].

A far-field effect from the collision between the Lhasa and Qiangtang blocks at ~150–120 Ma [96-99] is a possible cause for the Early Cretaceous tectonic reactivation in the southern Altaids. However, the mid-Cretaceous rapid cooling event (~115–80 Ma) recognized in this study does not coincide with the Early Cretaceous rapid cooling (~140–110) discovered in the other regions, as indicated in Figure 9. Collision of the Lhasa block thus could not be a reasonable tectonic driver for the mid-Cretaceous cooling. Some scholars suggested that the slab rollback and break-off after subduction of the Bangong-Nujiang oceanic lithosphere under Eurasia at ~120–100 Ma [100, 101] may have mainly driven mid-Cretaceous exhumation around the Beishan orogen [14, 22]. However, recent research revealed that these models related to the slab rollback and break-off are not only highly speculative [102] but also inapplicable to interpret to mid-Cretaceous exhumation in the Qiangtang block [103]. In addition, the slab rollback and break-off of the Bangong-Nujiang ocean took place in a short period [98, 99], and its far-filed effects are unlikely to have sustained until ~80 Ma. As an alternative, this ~115–80 Ma cooling event is also contemporaneous with low-angle or flat-slab subduction of the Neo-Tethyan ocean lithosphere at ~120–80 Ma [104-108]. It is a more reasonable driver of stress propagation to the continental interior and significant reactivation in the South Altaids during the mid-Cretaceous.

Some scholars suggested that the triggering mechanisms for this mid-Cretaceous cooling event in the South Altaids may be attributed to the collision between the Kohistan-Ladakh arc and Karakorum terrane during ~100–80 Ma [109, 110]. This may be a plausible explanation for reactivation in the western Tianshan [8, 65] and even in the Eastern Tianshan and Beishan orogen [14, 22, 26, 27], but this event is unlikely to have caused dramatic tectonic effects as far afield as Siberia Altai [22]. Furthermore, it is worth noting that the collision of the Kohistan-Ladakh arc took place slightly later than the mid-Cretaceous rapid cooling event (~115–80 Ma) identified in this study, suggesting that it may not have been a significant dynamic driver for this cooling event.

Other scholars proposed that the closure of the Mongol-Okhotsk ocean (Figure 10) could be a potential tectonic driver for the Cretaceous cooling in the Altaids [78, 85, 86]. However, significant deformation associated with the closure of the Mongol-Okhotsk ocean had ceased at least in the Early Cretaceous, making it unlikely to be the cause of the cooling that occurred in the Altaids during the mid-Cretaceous period unless denudation took a very long time to respond [111, 112]. Also, whether the closure of the Mongol-Okhotsk ocean caused an orogenic episode and contractional deformation is still up for debate [77]. Alternatively, the mid-Cretaceous rapid cooling phase and exhumation facilitated by reactivated faults in the South Altaids can be attributed to the extensional collapse of the Mongol-Okhotsk orogeny [14, 79, 80, 84].

In the eastern margin of Alxa, a series of Late Cretaceous thrust nappe structures or shortening or strike-slip deformations have been reported [61-63], which are explained as changes in the subduction direction of the Paleo-Pacific oceanic plate along the eastern margin of the Eurasian continent, oblique collision of blocks, or low-angle subduction of Paleo-Pacific oceanic plate (Figure 10). It is not yet known whether this period of deformation affected the Beishan area [14]. However, on the geological and geomorphological maps, there were obvious superimposed folds in the Jurassic strata of the Beishan area [55]. The interpretation of these superimposed folds is different, but it is entirely possible that they were late superimposed folds in the mid-Cretaceous period, which may be the same as in the Alxa area [61-63]. We further postulate that westward subduction of the Paleo-Pacific plate [113-115] may have caused or contributed to the synchronous deformation during the mid-Cretaceous in the Beishan orogen [55].

The proximity of the Altai to the Mongol-Okhotsk suture has led to the frequent attribution of the mid-Cretaceous cooling and exhumation events in the Altai region to the Mongol-Okhotsk orogeny rather than the Cimmerian orogeny [79, 80, 84]. However, the impact of the Mongol-Okhotsk orogeny on the evolution of Tianshan and Beishan has also been a topic of discussion, with the debate still ongoing [9, 77]. Differentiating between the effects of the Cimmerian and Mongol-Okhotsk orogenies proves challenging as they occurred during a similar time frame [9]. Therefore, it is plausible that the flat-slab subduction of the Neo-Tethyan ocean lithosphere, the collapse of the Mongol-Okhotsk orogeny, and the westward subduction of the Paleo-Pacific plate influenced the mid-Cretaceous (~115–80 Ma) evolution of the South Altaids.

5.3.2. Impacts of Block Rotations and Climate Change

According to a few pieces of literature [23, 64], the mid-Cretaceous exhumation in the Tianshan is not widely acknowledged. These authors propose that this exhumation event was associated with localized relief development rather than widespread mountain building. A theory has been put forth suggesting that the formation of Mesozoic local relief in Tianshan and its adjacent areas resulted from strike-slip deformation induced by the differential rotation of blocks, such as the Junggar [116-118], Tarim [116, 119], and Yili basins [120]. The rotation of these basins was accommodated by strike-slip faults in the Altai (e.g., the Irtysh shear zone [79, 80]), Tianshan (e.g., the North Tianshan Fault [23, 64]), and Southeastern margin of Tarim (e.g., the Xingxingxia [14, 28] and the Altyn Tagh faults [87, 90]), leading to the Cretaceous reactivation of these faults. Moreover, the vertical rotation axes of the Junggar and Tarim basins were located in the western part of these basins [116, 118], suggesting a more significant response on the eastern side compared with the western side. The spatial distribution of the mid-Cretaceous rapid cooling primarily occurred along the eastern boundary of the Junggar and Tarim basins, which might be explained by a hypothesis related to the rotation of the Junggar and Tarim basins.

Some scholars suggested that crustal deformation alone does not drive exhumation, and topographic and denudational evolution of orogens are strongly coupled with climate [121-123]. A decoupling between tectonics and denudation may be one reason why identifying a tectonic driver for the mid-Cretaceous cooling in Central Asia is challenging [79]. Sedimentary records in the Junggar basin suggest a climatic transition from arid to semiarid conditions during the Jurassic-Early Cretaceous to seasonally humid conditions during the mid-Cretaceous [65, 124, 125]. This climatic variability may have resulted in mid-Cretaceous increased cooling or erosion rates in the Eastern Tianshan and Altai [65, 79, 93, 124, 125]. They also show that related denudation was produced by the orographic effects on precipitation from moisture derived from the Eurasian continental interior. In the Beishan orogen, the lower part of the Cretaceous strata lacks coal but features calcareous paleosols, indicative of arid conditions [33]. The upper part of the Cretaceous strata unit is distinguished by extensive coal deposits, indicating humid conditions [33]. Consequently, the sedimentary records from the Beishan orogen provide evidence of a climatic shift from arid to humid conditions around the mid-Cretaceous period. However, further studies are required to determine whether this transition from a semiarid to a humid climate around the mid-Cretaceous influenced the denudation processes in the Beishan orogen.

Although the current physiography of Central Asia is due to the recent tectonics, the Mesozoic episode has probably played a significant role in the present geometric pattern. However, these interpretations and conclusions are still preliminary, and additional thermochronological constraints are necessary to enhance the kinematic scenario proposed in this study.

Detailed field observations confirmed four major E-W trending and thrusting northward faults in our study area and further proved that these thrusts may have initially developed in the late Middle Jurassic. The new AFT data from this study and a compilation of previous AFT modeling datasets identify a regional mid-Cretaceous (~115–80 Ma) rapid cooling event throughout the Beishan orogen and its adjacent mountains. The tectonic driver response for this cooling pulse is debated. Still, it may be linked to the rotation of the Junggar and Tarim blocks, as well as the multiple collisions and convergence along the Eurasian margins during the Cretaceous (e.g., the flat-slab subduction of the Neo-Tethyan ocean, the collapse of the Mongol-Okhotsk orogeny, and the westward subduction of the Paleo-Pacific plate).

The authors confirm that the data supporting the findings of this study are available within the article.

The authors declare that there is no conflict of interest regarding the publication of this article.

We thank two anonymous reviewers for detailed and constructive comments and associate editor Xiaoming Shen for editorial handling. This study was financially supported by the National Natural Science Foundation of China [Grant No. 92162211]. The support provided by the China Scholarship Council [CSC, No. 202206190029] is appreciated for financing the research stay of F.J. Wang in Germany.

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