Introduction

Thermochronological data may provide useful markers to detect normal and reverse fault activity1,2,3,4. However, detecting strike-slip faults by thermochronological data is challenging. In most cases, the timing of strike-slip fault activity was constrained through identifying the vertical offset triggered by strike-slip movements, if recorded, in thermochronological data5,6,7. However, this approach has limitations because most thermochronological markers are insensitive to lateral displacements of fault blocks, while the vertical offset caused by strike-slip movements is complex and often of minor importance. In such scenarios, the thermochronological clocks on both sides of the fault do not record the strike-slip movements but rather the earlier thermal events. Nevertheless, these records could be regarded as virtual strata, which ought to be dissected by the lateral displacement caused by strike-slip movements, providing opportunities to constrain strike-slip faults. Progressive deformation during plate convergence may produce thermochronological age trends that are suitably oriented and can be used to reveal subsequent strike-slip faulting events, allowing us to test the above assumption in real geology.

The Eastern Sichuan-Xuefeng fold-and-thrust belt (ESXFTB) is a part of the Yangtze Block in South China, spanning from the Xuefeng Uplift in the east to the Sichuan Basin in the west8 (Fig. 1a, b). It records the subduction of the Paleo-Pacific Plate underneath Eurasia during the Mesozoic9,10 and provides an excellent natural laboratory for studying fold-and-thrust deformation in intracontinental orogenic belts11. Previous studies highlighted a northwestward progressive deformation affecting this fold-and-thrust belt during the Mesozoic12,13,14 (Fig. 1c), which is responsible for the long-recognized trends of decreasing thermochronological ages from the SE to the NW8,15,16. These trends can be exploited as a marker to detect potential strike-slip movements in the subsequent stages of orogeny, in a region where tectonics is governed by varying Paleo-Pacific-Eurasia convergence directions17 through time.

Fig. 1: Geologic maps of the study region.
figure 1

a Simplified tectonic map of South China showing the location of the study area. The black arrows indicate Paleo-Pacific and India Plates motion. b Geologic map of the Eastern Sichuan-Xuefeng fold-and-thrust belt. Cenozoic faults (in red) after Wang et al.18. The blue dashed line L and its branch Lb represent the strike-slip fault revealed in this study. The asterisk and cross represent L and Lb, respectively. Line LO represents the initial front of regional Mesozoic deformation derived from the restorations of balanced cross sections16. c Model of progressive deformation propagation (red arrow, time in Ma) within the study area according to analogue modelling results14. XU, Xuefeng Uplift; WHHD, Western Hunan-Hubei Domain; ESD, Eastern Sichuan Domain. The red solid lines indicate the Cili-Baojing (CBF) and Qiyueshan (QF) Faults. The reddish dotted lines and associated time represent the approximate location and timing of the regional Mesozoic deformation. Image generated using Qgis 3.36.2 (https://qgis.org/en/site/) and Inkscape 1.3 (https://inkscape.org).

Full size image

Despite strike-slip faults were locally described in the field within the ESXFTB18,19, their role in this complex region is not fully assessed yet. To shed light on this issue, we present a dataset of 17 original ZHe ages. We exploit the regional thermochronological age pattern to reveal, and trace in map view, a large and previously undetected left-lateral strike-slip fault (Fig. 1b) and constrain its timing of deformation. The implications of the Late Cretaceous Paleo-Pacific Plate subduction pattern on intracontinental deformation are finally discussed.

Results

Geological settings

The ESXFTB is located in the central part of the Yangtze Block (Fig. 1a). Situated to the west of the Xuefeng uplift, it is marked by NW-directed thrusts, nappes, and NE-NNE-striking arcuate structures. It stretches from the Cili-Baojing Fault in the SE to the Huayingshan Fault in the NW, divided by the Qiyueshan Fault into the Western Hunan-Hubei Domain and the Eastern Sichuan Domain (Fig. 1b).

During the Mesozoic, the ESXFTB underwent northwestward-propagating deformation8,12, which began in the Late Triassic11,20 and proceeded northwestward from the Early Jurassic to Late Cretaceous13,15 due to the continued subduction of the Paleo-Pacific Plate beneath Eurasia. The Mesozoic progressive deformation is confirmed by thermal history reconstructions based on apatite fission-track (AFT) data15,16 and the decreasing number of Mesozoic unconformities northwestward21. Although dominated by thrust structures, the ESXFTB also underwent strike-slip movements since the Late Cretaceous (Fig. 1b), which were also documented in the adjacent Yuanma Basin within the Xuefeng Uplift22,23. However, the timing of those strike-slip faults remains poorly constrained due to the lack of Mesozoic stratigraphic record19,24 (Fig. 1b). The northwestward-decreasing trends of apatite and zircon fission-track ages16,20 (ZFT) generated during regional progressive deformation in the Mesozoic may provide suitable thermochronological markers to reveal younger strike-slip faults. However, AFT ages are mainly Late Cretaceous or younger, whereas ZFT ages are mainly Late Triassic. They do not cover the main deformation stages revealed by AFT thermal history modeling15, and no major age break in the AFT and ZFT map-view distribution patterns was highlighted by previous studies. The ZHe system, having a partial retention zone in between the partial annealing zones of the AFT and ZFT systems25, may provide a useful tool to unravel the puzzle of Mesozoic deformation within the central Yangtze Block and reveal subsequent strike-slip faulting during the Late Cretaceous and beyond.

ZHe ages confirm primary Mesozoic deformation

Weighted mean ZHe ages are shown in Fig. 2a and mainly fall between the Late Triassic and the Late Cretaceous. Single-grain ZHe ages that are older than the corresponding stratigraphic ages were excluded to avoid biasing the dataset, and the derived age trends, with unreset or overcorrected ages (see Methods for details on data filtering). Kernel density estimates (KDE) of single-grain ZHe ages are reported in Fig. 2b (see Supplementary Tables 1 and 2 for ZHe data). Literature AFT and ZFT data are summarized in Fig. 2c (Supplementary Tables 3 and 4). AFT, ZFT, and ZHe ages are plotted in Fig. 2a, c versus their distance from the initial front of regional Mesozoic deformation, derived from the restorations of balanced cross sections16 (line LO in Fig. 1b), which passes through the towns of Huaihua and Taoyuan and is parallel to the trend of the Xuefeng uplift. From the Xuefeng Uplift to the Eastern Sichuan Domain, ZHe data indicates a general northwestward trend of decreasing ages (Fig. 2a), which confirms the two trends delineated by AFT and ZFT ages (Fig. 2c). All AFT ages are younger than the corresponding stratigraphic ages (Fig. 2c and Supplementary Table 3), indicating that the samples have been reset due to burial after deposition. The same applies to ZFT ages (Fig. 2c and Supplementary Table 4). The observed distribution for ZHe ages (Fig. 2a) systematically falls between the ranges observed for AFT and ZFT ages, as expected for thermochronological ages that record exhumation25. A few single-grain ZHe ages from the Jurassic strata overlap within error with the stratigraphic age, which might suggest the presence of magmatic-age components. However, these ZHe ages are younger than the youngest ZFT age component observed in the same strata20,26, which rules out any major impact of volcanic activity on the observed ZHe age trends. This is consistent with the observation that the study area lacks Mesozoic magmatic activity, which was confined, in South China, to the east of the Xuefeng Uplift during the Mesozoic27.

Fig. 2: ZHe data and published AFT, ZFT data.
figure 2

a Reset ZHe weighted mean ages versus sample distance from the regional Mesozoic deformation initial front (LO in Fig. 1b). Blue asterisk and cross indicate the main age breaks marking the strike-slip fault detected in this study (L and Lb in Fig. 1b). The red arrows show the observed age trends. Error bars indicate ±1 s.e. b Kernel density estimate of all accepted single grain ZHe ages. c AFT and ZFT ages versus distance from LO (Fig. 1b). Blue asterisks mark the location of strike-slip fault L. Image generated using IsoplotR49 and Inkscape 1.3 (https://inkscape.org).

Full size image

Left-lateral strike-slip fault detected by age breaks

Evident breaks are found in the ZHe age trend from younger to older ZHe ages, which are indicated by blue markers in Fig. 2a. Similar age breaks are also found in the AFT and ZFT age trends (asterisk in Fig. 2c). The locations of these breaks in map view define a branched line indicated in blue in Fig. 1b (L and its branch Lb). Ages suddenly become older when crossing the line, then restart the northwestward decreasing age trend on the western side of the line.

The reason for the observed age break is explained in the conceptual model of Fig. 3. In our interpretation, line L, together with its branch Lb represents an undetected left-lateral strike-slip fault, referred to as the Yidu-Hefeng branched strike-slip fault hereafter. The Yidu-Hefeng Fault juxtaposes rocks with old ZHe ages on the NW block to rocks with young ZHe ages on the SE block. This is because the original trend of decreasing ZHe ages generated by Mesozoic deformation was cut obliquely by this left-lateral strike-slip fault, which divided the region into two blocks with partly overlapping ZHe ages. After left-lateral strike-slip, the youngest ages in the SE block are found in contact with the oldest ages in the NW block (Fig. 3).

Fig. 3: Conceptual and tectonic models.
figure 3

The conceptual model on the left illustrates how left-lateral strike-slip faulting produces age breaks in pre-existing thermochronological age trends. The tectonic model and plate tectonic interpretation show the orogenic belt caused by NW subduction of the Paleo-Pacific Plate since the Early Jurassic (geological cross-section after Dong et al51.), basin formation during slab rollback in the Cretaceous, and the activation of the Yidu-Hefeng strike-slip fault after the change of subduction direction in the Late Cretaceous. XU, Xuefeng Uplift; YB, Yuanma Basin; CBF, Cili-Baojing Fault; YHF, Yidu-Hefeng branched strike-slip Fault (L and its branch Lb); WHHD, Western Hunan-Hubei Domain; ESD, Eastern Sichuan Domain. Image generated using Inkscape 1.3 (https://inkscape.org).

Full size image

The conceptual model of Fig. 3 also explains other major features revealed by our analysis. They are the trend of progressively older ages from the SW to the NE observed on the northern block of the L fault (Fig. 1b), and the occurrence of two age peaks, in the Early Jurassic ( ~180 Ma) and in the Early Cretaceous ( ~ 125 Ma), observed in the KDE of Fig. 2b. A trend of progressively older ages is indeed predicted by our conceptual model moving parallel to the fault (from the left to the right in Fig. 3). And the fact that the KDE is polymodal, despite the starting age distribution being more gradual, is a direct effect of the repetition in outcrops of rocks characterized by a similar thermochronological imprint on the opposite sides of the fault (Fig. 3). These repeated ages, being more represented in the distribution, determine peaks where a more gradual distribution would have been expected.

Obviously, only trends generated before faulting are offset. Therefore, the Yidu-Hefeng Fault was necessarily active after the youngest displaced age was set. This is the 89 Ma AFT age shown in Fig. 2c, which is supportive of fault activity starting not earlier than the Late Cretaceous. Previous regional geological studies have revealed several NE to ENE small-scale and discontinuous left-lateral faults in the study region. Some authors have interpreted these faults within the framework of a NE-SW zone of distributed right-lateral shear accommodating India-Asia convergence during the Cenozoic18. However, the location of these Cenozoic strike-slip faults (in red in Fig. 1b) does not fit with the location of the faults revealed by thermochronological age breaks, despite the orientation being virtually the same.

Discussion

The activation of the Yidu-Hefeng strike-slip fault detected in this work is interpreted in the light of the Late Cretaceous evolution of the Paleo-Pacific subduction zone. As shown in the tectonic model in Fig. 3, the progressive deformation within the ESXFTB was likely triggered by the subduction of the Paleo-Pacific Plate since the Early Jurassic28 and the formation of a regional large-scale fold-and-thrust belt in the Jurassic – Early Cretaceous (in red in Fig. 3) as recorded by ZHe ages. This event is extensively documented across South China and includes magmatic intrusions east of the Xuefeng Uplift during the Early Jurassic9,27,29, the widespread development of NE-directed thrust faults12, and large-scale copper, lead-zinc, and tungsten mineralizations30.

A major tectonic change took place in the Cretaceous due to the rollback of the Paleo-Pacific Plate (in grey in Fig. 3), which resulted in extensive magmatic activity across East Asia10,31, especially in the Cathasia Block, accompanied by formation of ore deposits32,33,34 and sedimentary basins21,35 as evidenced in the Xuefeng Uplift23. By utilizing the age decreasing trends in regional thermochronological datasets, a previously undetected left-lateral strike-slip fault is revealed at this stage. Its onset time is after ~89 Ma, as constrained by the AFT data (Fig. 2c). We interpret the left-lateral strike-slip in the light of a change of convergence direction between Eurasia and the Paleo-Pacific Plate, from NW-SE to around N-S (Fig. 3). The change in subduction direction of the Paleo-Pacific Plate31 triggered left-lateral strike-slip motion along the entire Tanlu Fault36 (~97–82 Ma, Fig. 1a), and also induced widespread transpression in the South China Block17. Notably, the ESXFTB hosts several strike-slip faults that were active since the Late Cretaceous18,19 (Fig. 1b). The western Xuefeng Uplift experienced major transpression and left-lateral slip during the Late Cretaceous23,37, when exhumation revealed by AFT analysis led to the cessation of hydrocarbon generation38 and to the influx of meteoric water with further shale gas dispersion in the study area39.

The activity of the Yidu-Hefeng strike-slip fault necessarily precedes the main stages of Indo-Asian collision ( ~ 50Ma40,41), because India-Eurasia convergence is expected to produce right-lateral strike-slip tectonics in the Yangtze Block, as documented for example within the Western Hunan-Hubei Domain and in the Sichuan Basin18,42, not left-lateral strike-slip tectonics as observed in the ESXFTB. Geologic maps of the eastern margin of the Sichuan Basin show a general S-shaped folding pattern, which originates from Early Cretaceous NE-trending linear folds that have been subsequently deformed within a framework of distributed NE-SW right-lateral shear. The minimum age of right-lateral shear is constrained to the Cenozoic, coeval with Indo-Asian collision18,42. The major post 89 Ma NE-SW left-lateral strike-slip documented in this study was thus concluded before the onset of distributed NE-SW right-lateral shearing described by Wang et al.18.

Thermochronology age breaks across strike-slip faults are common in nature, as evidenced by examples in the Alps43, Tibet44, and Tianshan45. Although these age breaks could be attributed to differential exhumation across faults44,45, our case study suggests that they can also result from the horizontal displacement of fault blocks if the thermochronological signals induced by earlier regional thermal events are not reset. We suggest taking this effect into careful consideration when reconstructing thermal histories and calculating exhumation rates of strike-slip fault blocks, because age offsets caused by horizontal displacement may have a major impact on thermal modeling. This is particularly important in the light of the capability of the latest version of modeling software such as QTQt, which offers thermochronologists the capability to perform regional planar thermal history inversion models46. This advancement raises the bar for understanding the impact of planar deformation on the spatial distribution patterns of thermochronological ages.

In conclusion, we successfully use thermochronological markers and a simple conceptual model to reveal an undetected strike-slip fault and constrain its onset time. Our ZHe thermochronology data record the northwestward propagating deformation within the central Yangtze Block, which is caused by the NW subduction of the Paleo-Pacific Plate during the Early Jurassic to the Late Cretaceous. Together with previous AFT and ZFT datasets, regional thermochronological datasets collectively reveal a major left-lateral strike-slip fault and limit its onset time after ~89 Ma. Strike-slip motion requires a change in the regional compression direction during the Late Cretaceous, which is likely controlled by the change of subduction direction of the Paleo-Pacific Plate beneath Eurasia. Our results provide evidence for the Paleo-Pacific Plate subduction evolution during the Late Cretaceous and reflect the impact of plate subduction processes on intracontinental tectonic deformation, confirming the potential of thermochronology in studying the relationships between surface processes in complex geodynamic environments.

Methods

Zircon (U-Th)/He

We collected 17 sandstone samples for ZHe analysis across the ESXFTB (Supplementary Table 1), from strata ranging in age between the Upper Paleozoic and the Upper Cretaceous. Samples were crushed and processed with standard magnetic and density separation methods25 to retrieve zircon grains for analysis. U and Th isotopic ratios were measured in the Thermochronology Lab at the University of Florida. Zircon crystals free of impurities, inclusions, and fractures were selected under a stereomicroscope, and their dimensions were measured to calculate alpha-ejection corrections47. Grains were encapsulated in Nb tubes, degassed with a diode laser, and heated at least twice (each heating process lasted 20 minutes at 9 Amp) for complete helium extraction. The extracted gas, mixed with >99.99% pure 3He spike, underwent purification with an NP-10 getter. 4He/3He ratios were measured with a Pfeiffer-Blazers Prisma quadrupole mass spectrometer after a two-minute equilibration. Degassed zircon packets were then spiked, and bomb-dissolved using strong acids of HNO3, HF, and HCl, and their U and Th isotopic ratios were measured by Thermo-Finnigan Element2 ICP-MS. Fish Canyon Tuff zircon served as age standards. Single grain ZHe ages are listed in Supplementary Table 2.

Data reduction and distance calibration

We performed data filtering and abandoned single-grain ZHe ages older than the stratigraphic deposition age to avoid bias in the analyzed age trends related to overcorrected or unreset ages. Although some single-grain ZHe ages from the Jurassic strata coincided with the stratigraphic deposition age after filtering, all these ages remained younger than the provenance-related ZFT age components from the same strata, with the minimum age being 216 Ma20. One grain in sample S4 yielded an age of 58.1 ± 1.4 Ma which is much younger than other grains in this sample and previous AFT ages in this area (within Western Hunan-Hubei Domain, Fig. 1b and Supplementary Table 2). Considering the partial annealing zone of AFT is usually lower than the partial retention zone of ZHe, this grain is regarded as an outlier and thus ignored. In particular, the eU of this grain is higher than that of other grains in the same sample, and there is no evidence of major reheating around ~58 Ma in the study area that could cause the zircon (U-Th)/He partial retention zone to fall below the AFT partial annealing zone48. This further supports our decision to consider this age as an outlier. The weighted mean ages of reset ZHe grains for all samples were calculated using IsoPlotR software49. Published AFT and ZFT data are collected and listed in Supplementary Tables 3 and 4. Only outcrop sample data were collected. Unreset ages are ignored following the procedure as in ZHe data processing to avoid effect of distant source thermal history.

All samples location were imported into QGIS. Their distance from the regional Mesozoic deformation initial front line, i.e. the LO in Fig. 1b, were calculated with Nearest neighbor algorithm by NNjoin plugin (version 3.1.3) in QGIS software (version 3.28.8).