Illite K‐Ar Dating of the Leibo Fault Zone, Southeastern Margin of the Tibetan Plateau: Implications for the Quasi‐Synchronous Far‐Field Tectonic Response to the India‐Asia Collision

Whether tectonic strain from the early stage India‐Asia collision has synchronously affected the far‐field margin of the Tibetan Plateau is crucial for understanding plateau deformation and growth processes. However, direct evidence for early far‐field deformation remains scarce. Utilizing illite K‐Ar dating of three fault gouge samples, we established the faulting history of the Leibo fault zone (LFZ) at the southeastern margin of the Tibetan Plateau (SEMTP). Consistent authigenic illite ages of 52 ± 2, 54 ± 12 and 55 ± 6 Ma suggest the reactivated thrust faulting of the LFZ in the Early Cenozoic. Positioned ∼700 km east of the collisional boundary and at the intersection of three blocks with distinct lithospheric rheology in strength/viscosity, this event suggests a quasi‐synchronous far‐field tectonic response in the SEMTP to the India‐Asia collision.


Introduction
Whether tectonic strain from continental collision can be transmitted to a far-field terrane boundary rapidly during the initial collision or progressively long after the collision, is a key question in continental tectonics and remains hotly debated.A classic example is the modern Tibetan Plateau, the world's broadest and highest plateau (∼5 km in mean elevation; Figure 1), which developed during the India-Asia collision, starting at 65-50 Ma (Ding et al., 2022;Najman et al., 2010;Rowley, 1996).It comprises several terranes (e.g., the Lhasa, Qiangtang and Songpan-Ganzi terranes) accreted during successive subduction of the Paleo-, Bangong-Nujiang, and Neo-Tethyan Oceans along the southward younging suture zones (Figure 1b; Li, Yin, et al., 2019;Liu et al., 2016;Song et al., 2015).Understanding the process of tectonic strain transmission across different terranes of the Tibetan Plateau is essential for exploring its uplift history and associated geodynamic mechanisms (Ding et al., 2022;Shi et al., 2014).
Two existing end-member models, continuum (England & Houseman, 1986) and step-wise growth (Tapponnier et al., 2001), are widely accepted as contrasting end members for the formation and evolution of the Tibetan Plateau.Both models propose that high strain rates and crustal thickening due to the India-Asia collision were initially concentrated at the collisional plate boundary and then gradually propagated outward (Duvall et al., 2011).By contrast, some previous studies suggest that tectonic strain is likely to be transmitted to the farfield, present-day plateau margin, simultaneously with the continental collision.For example, the northern and northeastern boundaries of the Tibetan Plateau were interpreted to have been established by 60-45 Ma (e.g., Clark et al., 2010;Duvall et al., 2011;He et al., 2022;Yin et al., 2002), indicating far-field strain transmission synchronous with the initial India-Asia collision.These interpretations are consistent with numerous rapid cooling events around the Tibetan Plateau margin, synchronous with the initial collision (Figure 1b; Figure S1b and Table S1 in Supporting Information S1).Moreover, some numerical modeling results reveal the synchronous, localized shearing along the far-field lithospheric strength discontinuity during continental collision (Dayem, Houseman, & Molnar, 2009;Dayem, Molnar, et al., 2009), which partly support the synchronous far-field tectonic deformation.However, direct evidence of such a far-field tectonic response (e.g., direct dating of faulting at the initial collision stage) remains lacking along most of the plateau margin, and particularly along the southeastern margin of the Tibetan Plateau (SEMTP).
Here, we address the above question by determining and interpreting the timing of faulting along the Leibo fault zone (LFZ) at the SEMTP, via K-Ar isotopic dating of authigenic illite in fault gouge samples.We then evaluate the quasi-synchronous far-field tectonic response and its areal extension around the margin of the Tibetan Plateau based on our K-Ar ages and comprehensive analysis of regional geochronological data.

Background
The LFZ is situated at the intersection of the Sichuan Basin, South China, and Chuan-Dian blocks, southeastern Tibetan Plateau (Figures 1 and 2a), a tectonically and seismically active region (Chen et al., 2022;Shi et al., 2024;Su et al., 2023;Wang et al., 2023;Zhang et al., 2022).The fault displays highly pronounced NE-striking subparallel lineaments, causing the displacement of multiple streams by ∼320 m on the northernmost branch (Han et al., 2009), within a zone of ∼40 km long and ∼10 km wide, near Leibo County (Figure 2b).This fault has been inferred to extend northeastward blindly into the southern Sichuan Basin, with a total length exceeding 500 km (Wang et al., 2014).This fault traverses the Late Permian Emeishan basalts (P 3 e) that constitute a part of the Emeishan large igneous province in southwestern China, with a crystalline age of ∼260 Ma (Zhou et al., 2002).These rocks' petrochemistry classifies them as compact alkaline basalts with high titanium (Ti) and very low potassium (K) contents (Liu et al., 2020).The thickness of the basalt strata in the Leibo area reaches up to approximately 700 m (Liu et al., 2020), and they are underlain by the Lower Permian limestone.
The LFZ shows a complex tectonic evolution, with an earlier reverse faulting and a later dextral-slip faulting history (Han et al., 2009).The earlier reverse faulting likely formed in the Early Mesozoic due to the subduction of the (Paleo-) Pacific Plate along with the Lianfeng and Zhaotong-Ludian faults (Figure 2a) and widespread NNE-trending fold-and-thrust belts in the South China block (Li, Suo, et al., 2019).The onset of dextral slip along the LFZ is unknown, but was inferred to be ∼4-2 Ma driven by shear along the Xianshuihe-Xiaojiang fault (Su et al., 2023;Wang et al., 1998).In general, the faulting history of the LFZ since the Early Mesozoic remains largely uncertain.

Field Sampling and Methods
We targeted two sites situated along linear fault traces with clearly visible fault planes in outcrops, in the northern part of the LFZ (Figure 2b; Figure S2 in Supporting Information S1).At Site 1, which exposes an outcrop hosting a thin layer of brownish fault gouge within the fault zone, we obtained sample LB-gouge-2 (Figures 2c-2e).At Site 2, positioned along a stream offset valley, we collected two samples, namely LB-gouge-1A and -1B.These two chromatic varieties (yellowish and bluish) of fault gouge samples were taken from a single fault core (Figure 2f; Figures S2b and S3 in Supporting Information S1).Notably, these two samples are located ∼770 m north of the first sample (Figure 2b).These two sampled locations represent two different branches of the LFZ, and detailed information on sample locations is provided in Table S2 in Supporting Information S1.
We utilized authigenic illite K-Ar dating as a key method to constrain the timing of shallow crustal faulting (van der Pluijm et al., 2001).Two distinct end-member polytypes of illite are commonly found in nature: detrital mica (2M 1 ) and authigenic illite (1M d ) phases (Pevear, 1999;van der Pluijm et al., 2001).The former is mainly derived from surrounding wall rocks, while the latter is a newly grown product developing simultaneously with fault activity (Haines & van der Pluijm, 2012).These two illite polytypes are usually found to be mixed together in fault gouges.The proportion of 1M d relative to the 2M 1 polytype increases with decreasing grain size, and the corresponding K-Ar age of the fault gouge decreases, accordingly (Haines & van der Pluijm, 2008;Pevear, 1999;Scheiber et al., 2019).The ages representing the pure authigenic/detrital materials can be obtained by calculating the percentages of isolated illite of varying sizes and establishing a linear relationship between the percentages of detrital illite and corresponding K-Ar ages (i.e., illite age analysis; Pevear, 1999;van der Pluijm et al., 2001).The extrapolated age of the 0% 2M 1 intercept is generally interpreted as the timing of fault activity (Duvall et al., 2011;van der Pluijm et al., 2001;Zwingmann et al., 2010), while the extrapolated age of the 100% 2M 1 intercept may reveal the timing when wall rock-derived micas were exhumed through the isotherm at 280°C (Haines & van der Pluijm, 2023) or the age of an individual tectonic event (Haines & van der Pluijm, 2023;Torgersen et al., 2022;Zwingmann et al., 2010).
We investigated the microstructures of fault gouge samples using optical and scanning electron microscopes (SEM).The samples were separated into four clay-sized fractions (i.e., 2-1, 1-0.5, 0.5-0.25, and <0.25 μm), and the ratios between detrital and newly formed illite concentrations were measured for each aliquot using X-ray diffraction (XRD).Finally, we determined corresponding ages by K-Ar dating methods (see Text S1 in Supporting Information S1 for detail description on methods).

Results
We identified numerous sharp, sub-parallel dextral stream offsets in the northern LFZ, ranging from 330 to 360 m (Figure S2 in Supporting Information S1).These offsets, consistently of a similar magnitude, support the hypothesis that the LFZ's dextral slip likely commenced during the Late Cenozoic, for example, ∼4-2 Ma (Su et al., 2023;Wang et al., 1998).In the outcrop at Site 1, a narrow fault zone (<0.5 m wide) was observed, cutting through the Late Permian basalts (Figures 2c-2e).The cleavage structures intersect the fault plane at an acute angle, with their pointing direction indicating a relative downward movement of the footwall (Figures 2c and 2d).In addition, the plunge side of striations on the fault gouge surface is almost parallel to the fault strike, with the fining direction suggesting a relative rightward sliding of the footwall (Figures 2c and 2d).These observations suggest an initial phase of reverse faulting, followed by a subsequent episode of younger dextral-slip faulting that partially exploits the pre-existing fault plane.Thus, field observations demonstrate an overall transpressional regime within the LFZ.
Microstructures observed under an optical microscope show that the fault gouge in the three samples is mainly composed of clay minerals and subhedral to euhedral quartz, hornblende and plagioclase grains (Figures 3a-3c).The SEM secondary electron (SE) images with an energy dispersive X-ray spectroscopy (EDS) spectrum show small euhedral particles with fibrous-and slab-like shape, and high K-content, indicating the presence of authigenic illite (Figures 3d-3f).
The XRD analyses reveal that kaolinite is the predominant clay mineral phase in the separated fractions.Additionally, illite-smectite and illite could be identified as an additional phase within the separated clay fractions.The X-ray diffractograms indicate that only quartz and minor kaolinite are mixed with clay minerals in the coarse fractions, and there is no contamination of potassium (K)-bearing minerals observed in any of the separated fractions (Figures S4-S6 in Supporting Information S1).The percentages of detrital illite (2M 1 ) and the K-Ar ages for the four-size fractions of illite separated from each fault gouge sample are listed in Table S3 in Supporting Information S1.Our results reveal a clear relationship of ages with varying grain sizes, with the smallest fraction (<0.25 μm) yielding the youngest ages.A least squares regression (York et al., 2004) of the 2M 1 percentage versus illite K-Ar age constrains ages of the authigenic components (0% 2M 1 ), at 55 ± 6, 52 ± 2, and 54 ± 12 Ma, for samples LBgouge-1A, -1B and 2, respectively (Figure 4).The detrital illite ages (100% 2M 1 ) are 125 ± 24, 86 ± 6 and 181 ± 36 Ma, respectively (Figure 4; see Text S1 in Supporting Information S1 for more details).Importantly, authigenic illite ages of the three samples are relatively consistent, and the detrital illite ages are much younger than the age of the wall rock (∼260 Ma; Zhou et al., 2002).

Discussion and Conclusions
The extrapolated ages of authigenic components (55-52 Ma) from three fault gouge samples indicate a particular faulting event of the LFZ.This event is interpreted as representing a major episode of reverse faulting that evidently predates the dextral-slip faulting inferred to have initiated at ∼4-2 Ma (Su et al., 2023;Wang et al., 1998).The inferred reverse faulting event at 55-52 Ma is contemporaneous with the initial India-Asia collision.This observation likely reflects the quasi-synchronous far-field tectonic response of the SEMTP to the India-Asia collision, for several reasons.First, Duvall et al. (2011) have interpreted 40 Ar/ 39 Ar ages (∼50 Ma) of fault gouge-derived illite to reflect the initiation of the West Qinling fault (WQF) that is coincident with the timing of the India-Asia collision, together with thermochronologic data (Clark et al., 2010).Noteworthily, the distance from the present plate boundary of the Eastern Himalayan Syntaxis (EHS) to the WQF (∼1,000 km) is greater than to the LFZ (∼750 km; Figure 1b).Second and more importantly, the LFZ is located at the intersection zone among the Sichuan Basin, South China and Chuan-Dian blocks, with obvious rheological (strength/ viscosity) differences (Bao et al., 2015;Molnar & Dayem, 2010).Numerical experiments show that shear strain and fault reactivation can concentrate in weaker zones adjacent to such boundaries within a continuously deforming zone (Dayem, Houseman, & Molnar, 2009;Huangfu et al., 2023).Moreover, a strong region analogous to the Tarim Basin or other relatively strong cratonic lithosphere would facilitate quick strain transmission from the collision boundary to the northern margin of the Tibetan Plateau (Dayem, Houseman, & Molnar, 2009;Dayem, Molnar, et al., 2009).Therefore, a process of strain concentration may also occur in the Leibo area (the boundary of the Sichuan Basin, South China and Chuan-Dian blocks), quasi-synchronously with the continental collision.This process could contribute to the reactivation of the LFZ.
Our critical inference of the far-field tectonic response to the India-Asia collision along the SEMTP is supported by numerous contemporaneous geological events (∼65-50 Ma) reported around the Tibetan Plateau (Figure 1b).These events include: (a) onset of the WQF (Clark et al., 2010;Duvall et al., 2011) and the Altyn Tagh fault (Yin et al., 2002); (b) initiation of metamorphism near the Sagaing fault recording a high pressure-low temperature oceanic subduction-continental collision tectonic event (Min et al., 2022;Morishita et al., 2023); and (c) rapid cooling events in the SEMTP (Liu-Zeng et al., 2018), Qilian Shan (He et al., 2022), Tian Shan (De Grave et al., 2011;Jolivet et al., 2010) and Pamir (Cao et al., 2013; see Figure S1b and Table S1 in Supporting Information S1 for more details).Given that the late Cretaceous-early Cenozoic climate of the East Asia was subtropical arid/semiarid (Wu et al., 2022), these rapid cooling events are more likely to be attributed to tectonic uplift/exhumation rather than rapid erosion caused by rainfall.Therefore, we propose that these contemporaneous ages signify the tectonic activity of associated faults.Although these ages are widespread across a large region, they are primarily concentrated along the boundaries of the Tibetan Plateau and the rigid cratons (the Tarim, North China and South China blocks), exhibiting obvious rheological (strength/viscosity) differences (Bao et al., 2015).They also occur within ancient orogenic belts, which can be easily reactivated by tectonic stress, such as the Tian Shan.Based on our new data of authigenic illite ages and regional contemporaneous geological events, we suggest that such a far-field tectonic response, caused by the India-Asia collision, propagated quasisynchronously, at least to portions of the margins of the Tibetan Plateau, if not to its entire periphery.
Notably, the K-Ar ages of authigenic illite in three fault gouge samples are restricted to the Early Cenozoic, lacking the record of younger faulting.One plausible explanation is that, due to successive thrust faulting and erosion of the uplifted hanging wall, rocks were exhumed through the thermal window (∼110°C; Duvall et al., 2011) conducive to the formation of illite.Alternatively, the relatively weaker nature (at an average slip rate of ∼0.09-0.18mm/yr since ∼4-2 Ma) of the younger dextral-slip faulting of the LFZ may have resulted in limited fluid-rock reaction, leading to the absence of younger authigenic illite (Haines & van der Pluijm, 2012).
Explanation for the detrital illite end member (2M 1 ) ages is complex and needs to be examined comprehensively together with the deposition/crystallization age of wall rocks and subsequent cooling history for a clearer geological meaning (Haines & van der Pluijm, 2023).Some previous studies (e.g., Haines & van der Pluijm, 2023;Torgersen et al., 2022;Zwingmann et al., 2010) interpreted the detrital illite (2M 1 ) components as high-temperature illite grown within the fault gouge during associated faulting.Here, we adopt their interpretation and suggest that each detrital illite age (∼181, ∼125, and ∼86 Ma) likely reveal an individual tectonic event, for two reasons: (a) the three groups of detrital illite ages are significantly younger than the crystalline age of the wall rock (∼260 Ma; Zhou et al., 2002), and (b) the wall rock consists of compact basalts, which means negligible precursor clay minerals existed prior to faulting, and thus, all clay minerals must be authigenic.If meaningful, these events might represent thrust fault reactivation linked to the far-field effects of terrane collage during the Mesozoic (Li, Yin, et al., 2019;Li, Suo, et al., 2019).However, a more solid interpretation on these Mesozoic detrital illite ages requires further investigation, which is beyond the scope of this study.
The temporal correlation between the continental collision and intracontinental faulting does not always mean causation.Some presumably collision-induced geological structures may have different and/or complex tectonic origins.A previous study (Fan et al., 2019) indicates that the present-day northeastern margin of the Tibetan Plateau underwent NW-SE extension in the Early Cenozoic.This is more possibly a result of far-field effect of subduction of the western Pacific Plate than the India-Asia collision.Geodynamic experiments reveal that the synchronous activity and interaction of the Indian indentation and Pacific and Sunda slab rollback provide a better explanation for modern Asian deformation (Schellart et al., 2019).The complex tectonic evolution backgrounds and geodynamic mechanisms make it difficult to attribute independent geological events to a single factor.However, two noteworthy factors are considered: (a) the LFZ's proximity to the continental collision boundary (∼750 km) compared to the Pacific Plate's subduction boundary (∼1,400 km; Figure 1a), and (b) potential contemporaneous geological events at the Tibetan Plateau's periphery (Figure S1b and Table S1 in Supporting Information S1).These factors suggest that the interpreted reactivated thrusting of the LFZ during 55-52 Ma, driven by NW-SE-oriented transpression, may be predominantly affected by the quasi-synchronous far-field tectonic response to the India-Asia collision.
In summary, our illite K-Ar dating of three gouge samples from the LFZ fills a critical knowledge gap regarding the direct dating of fault reactivation in the SEMTP.The resulting consistent authigenic illite ages of 55-52 Ma align with the timing of the initial India-Asia collision (65-50 Ma).Together with contemporaneous geological events reflected by tectonic deformation and thermochronologic data, our results suggest the quasisynchronous far-field tectonic response to the India-Asia collision, at least along portions of the margins of the Tibetan Plateau.

Figure 2 .
Figure 2. Structural setting of the Leibo fault zone (LFZ) and location of fault gouge samples.(a) Simplified map showing major faults of the southeastern Tibetan Plateau.(b) Geological map of the study area and location of the LFZ.Solid black circles show locations of fault gouge samples.See Figure S2 in Supporting Information S1 for detailed analysis of spatial distribution of the LFZ.(c) Exposure of LB-gouge-2 outcrop (at Site 1) with basaltic wall rocks.The inset equal-area lower hemisphere stereonet shows field-measured attitudes of the fault, cleavage and joint planes, as well as fault striae and basalt bedding.(d, e) Close-up views of structural features in (c).The cleavage structures and fault striations in (d) suggest an initial phase of reverse faulting followed by a later episode of younger dextral-slip faulting.(f) Field photo of samples LB-gouge-1A (yellow-colored) and LB-gouge-1B (blue-colored) at Site 2. See Figure S3 in Supporting Information S1 for more details.

Figure 3 .
Figure 3. Photomicrographs of representative microstructures of three fault gouge samples.Clay minerals forming around primary rock-forming minerals of basalts: quartz (a) and hornblende (b) clast under cross-polarized light, and plagioclase (c) under plane-polarized light.(d-f) Scanning electron microscope (SEM) secondary electron (SE) images with energy dispersive X-ray spectroscopy (EDS) analyses (inset) at red crosses.Fibrous-(d, f) and slab-like (e) illite particles are shown by orange arrows.

Figure 4 .
Figure 4. Illite age analysis plot for the three fault gouge samples.Percentage of detrital (2M 1 ) illite for each size fraction is plotted against measured K-Ar age (expressed as e λt -1, black circles).Horizontal error bars indicate uncertainties on %2M 1 .Vertical errors on e λt -1 are smaller than circle sizes.Straight lines and shaded areas show linear regressions and 2σ confidence intervals of the data, respectively.The function e λt -1 (λ ∼ decay constant of 40 K, t ∼ apparent age), is plotted rather than age because it is the decay constant of 40 K that is linearly proportional to the percentage of detrital mica (2M 1 ).Extrapolated ages for 0% and 100% 2M 1 indicate the age of newly formed illite in the gouge (timing of the last faulting at temperatures of 60-180°C; Haines & van der Pluijm, 2012) and ages of illite formed during earlier faulting events, respectively.