Introduction
The Tibetan Plateau is the most structurally active area in the world (Lai et al., Reference Lai, Zhao, Huang, Wang, Ju, Li, Wang, Xu and Zhao2022), where many large rivers develop (Zhan et al., Reference Zhan, Chen, Zhang and Han2018). Influenced by collision of the Indian and Eurasian plates (Zhang et al., Reference Zhang, Liu and Bing2000; Li et al., Reference Li, Zhou, Zhao and Zhang2015), the Tibetan Plateau and its surrounding areas have been uplifting rapidly since the middle to late Pleistocene (Li et al., Reference Li, Chen, Zhou, Song, Zhang, Gu and Cao2020), which was accompanied by several glacial periods (Shi et al., Reference Shi, Liu, Li and Yao1999, Reference Shi, Yu, Liu, Li and Yan2001). Consequently, deposits are extensively distributed on both sides of the river valleys due to the combined effects of tectonic uplift, river incision, and climate change. The formation and evolution of deposits have been widely recognized as crucial aspects in understanding geological history (Zhan et al., Reference Zhan, Chen, Zhang and Han2018; Li et al., Reference Li, Chen, Zhou, Song, Zhang, Gu and Cao2020). Research indicates that the Quaternary allochthonous deposits in the southeastern margin of the Qinghai–Tibet Plateau are primarily attributed to surface processes such as debris flows, landslides, or rock collapses (Korup and Montomery, Reference Korup and Montomery2008; Wu et al., Reference Wu, Zhao, Liu, Granger, Wang, Cohen and Wu2016). Under natural conditions, deposits are often difficult to preserve in a well-maintained state due to the influence of dynamic surface processes (Li et al., Reference Li, Chen, Zhou, Song, Zhang, Gu and Cao2020). However, certain factors such as terrain, scale, and material composition play significant roles in the long-term preservation of some deposits (Fan et al., Reference Fan, Dufresne, Siva Subramanian, Strom, Hermanns, Tacconi Stefanelli and Hewitt2020).
The interaction between geological structure and climate is recognized as the primary factor influencing surface processes (Li et al., Reference Li, Chen, Li, Han, Zhai, Li, Zhang and Gu2022a). Consequently, the focal points and challenges of paleo-deposit research revolve around comprehending paleoenvironmental characteristics, elucidating genetic mechanisms, and analyzing the patterns of formation and evolution (Trauth and Strecker, Reference Trauth and Strecker1999; Korup and Clague, Reference Korup and Clague2009; Wang et al., Reference Wang, Chen, Dai, Long and Xu2014; Bao et al., Reference Bao, Zhai, Chen, Xu, Sun, Zhan, Zhang and Zhou2020; Li et al., Reference Li, Chen, Yan, Zhou, Wang, Li and Zhang2022b). Field investigation and chronological research serve as the foundation for addressing these issues. Field investigation allows for the effective collection of geological characteristics of deposits, enabling preliminary assessments of deposit types and causes (Costa and Schuster, Reference Costa and Schuster1988; Strom, Reference Strom2004; Dunning, Reference Dunning2006; Chen et al., Reference Chen, Dai, Lv and Cui2013). Chronological dating has made significant advancements through the development of luminescence and cosmogenic nuclide techniques (Hermanns et al., Reference Hermanns, Niedermann, Ivy-Ochs and Kubik2004; Chen et al., Reference Chen, Zhou, Cui, Li, Wu and Ma2018). These advancements have greatly enhanced our ability to analyze the structural and climatic background of paleo deposits. Additionally, progress in science and computer technology has facilitated research in various fields such as geophysics, geochemistry, and numerical simulation. These disciplines have further contributed to the understanding of paleoenvironments related to deposits, causal mechanisms, and geomorphological evolution (Kong et al., Reference Kong, Na, Fink, Zhao and Xiao2009; Yan et al., Reference Yan, Chen, Zhou, Li, Zhang, Gu and Zhang2022; Y.S. Zhang et al., Reference Zhang, Chen, Zhou, Bao, Yan, Zhang, Li, Gu and Wang2022).
The Shigu section of the Jinsha River (the First Bend of the Yangtze River–FBYR), located in Hengduan Mountains belt (Fig. 1a), has gained considerable attention due to its unique geomorphological feature known as the “First Bend” (Kong et al., Reference Kong, Zheng and Caffee2012; Li et al., Reference Li, Chen, Yan, Zhou, Wang, Li and Zhang2022b). In this study, we focus on the Hongwen allochthonous deposit (HAD) within the FBYR as a crucial window to understand the regional history of the geological environment. We delve into the formation and evolution mechanisms of the HAD through terrain interpretation, dating techniques, and paleoenvironmental analysis. Furthermore, we discuss the intertwined influences of tectonic activity and climate on the deposit. By integrating these insights with the developmental features of fluvial sediments in the FBYR, a sequence of river landform evolution is proposed.
(a) Digital elevation model map showing major rivers of Southeast Asia. (b) Tectonic framework of southeastern Tibetan Plateau.
Geological Setting
Regional tectonics and stratigraphy
The southeastern margin of the Tibetan Plateau has exhibited intense tectonic activity since the Cenozoic, with the Three-River belt and the Ailaoshan–Red River shear zone serving as representative regions in this context (Burchfiel and Chen, Reference Burchfiel and Chen2012). These fault zones, along with other large faults, establish the fundamental structural framework of the study area (Fig. 1b). The Ailaoshan–Red River shear zone commenced ductile left-slip movement at circa 27–17 Ma (Searle et al., Reference Searle, Yeh, Lin and Chung2010; Wei et al., Reference Wei, Wang, Wu and Meng2016). However, in the Quaternary, three important NE-trending left-lateral faults (Lijiang, Heqing, and Jianchuan faults; Allen et al., Reference Allen, Gillespie, Yuan, Sieh, Buchun and Chengnan1984) were derived from the shear zone under the influence of brittle right-lateral faulting (Allen et al., Reference Allen, Gillespie, Yuan, Sieh, Buchun and Chengnan1984) (Fig. 2a). These faults exhibit left-lateral shearing, which can be attributed to either right-lateral movement along the Ailaoshan–Red River shear zone or subduction of the Indian plate beneath the Eurasian plate (Tapponnier et al., Reference Tapponnier, Peltzer, Armijo, Coward and Ries1986; Wang and Burchfiel, Reference Wang and Burchfiel2000). In addition, the internal dynamic geological processes generated by large faults such as Jinshajiang and Xianshuihe also have a significant effect on the regional geological evolution. Alkali magmatism occurred in western Yunnan during the late Eocene to Oligocene, which is effective evidence of active crust and lithospheric extension (Chung et al., Reference Chung, Lee, Lo, Wang, Chen, Yem, Hoa and Genyao1997).
(a) Simplified geological map of southeastern Tibetan Plateau showing main faults, distributions of strata and intrusions, and the course of the Jinsha River (revised from Wei et al., Reference Wei, Wang, Wu and Meng2016); location of Figure 2b shown with red box. (b) Geological map of the First Bend of the Yangtze River (FBYR).
The study area exhibits a well-developed stratigraphy. Dividing the area along the Shigu and Yangbi valleys, the western side is dominated by Cenozoic and Paleozoic strata, whereas the eastern side consists of Mesozoic and Paleozoic strata (Fig. 2a). The exposed strata in the First Bend primarily are pre-Triassic marine sedimentary strata (Fig. 2b). The predominant lithologies include limestone, slate, and schist. Additionally, the valley is characterized by widespread distribution of fluvial sediments, primarily of Cenozoic age.
Plateau uplift and climate background
Plateau uplift, which is the most important driving force of geomorphic evolution in southeastern Tibet, has been hotly debated over the past few years (Li et al., Reference Li, Zhou, Zhao and Zhang2015; Zhan et al., Reference Zhan, Chen, Zhang and Han2018). A credible view is that uplift of the Tibetan Plateau is caused by the continuous thickening of the lithosphere due to the subduction of the Indian plate under the Eurasian plate (Li et al., Reference Li, Zhou, Zhao and Zhang2015; Y.S. Zhang et al., Reference Zhang, Chen, Zhou, Bao, Yan, Zhang, Li, Gu and Wang2022). Although the geological evolution of the plateau continues, the recognition that it uplifted rapidly in the Quaternary is widely accepted (Li et al., Reference Li, Zhou, Zhao and Zhang2015). The average uplift rate of the Tibetan Plateau after the Early Pleistocene was 1.60–5.35 mm/yr, and the uplift rate of some areas exceeded 10 mm/yr in the Holocene (Fig. 3a) (Xiao and Wang, Reference Xiao and Wang1998). According to precision-leveling data of the surrounding area from 1970 to 2012 (Hao et al., Reference Hao, Wang, Shen, Cui, Ji, Li and Qin2014; Zhan et al., Reference Zhan, Chen, Zhang and Han2018), the annual average uplift rate of the First Bend section and the Yangbi valley is determined to be less than 3 mm/yr (Fig. 3b).
(a) Uplift rates of the Qinghai Tibet Plateau, revised from Zhan et al. (Reference Zhan, Chen, Zhang and Han2018). (b) Vertical uplift velocity in northwest Yunnan with respect to the Sichuan basin (referenced from Hao et al., Reference Hao, Wang, Shen, Cui, Ji, Li and Qin2014); inset shows photographic image of the FBYR.
The uplift history of the Tibetan Plateau has a profound effect on the development and evolution of Quaternary glaciers. The coupling of mountain uplift and global glacial climate is the fundamental reason for the development of glaciers in the Tibetan Plateau and its surrounding mountain areas (Fig. 4a). The Kunhuang Movement that occurred during 1.1–0.6 Ma was characterized by a large-scale regional uplift in the early stage. In the later stage, there was primarily sudden uplift or subsidence of fault blocks, from which the Tibetan Plateau rose to more than 3000 m and developed the largest glaciation—the Kunlun glaciation (MIS 16) (Cui et al., Reference Cui, Wu, Liu, Ge, Pang and Xu1998). About 0.15 Ma ago, under the influence of the Gonghe Movement (Li et al., Reference Li, Fang, Ma, Zhu, Pan and Chen1996), many mountains on the southeastern margin of the Tibetan Plateau were uplifted above the glacial equilibrium line, and the penultimate (MIS 6) and last (MIS 4–2) glaciations occurred successively. The glacial and interglacial cycles represent the alternation of cold and warm (EPICA community members, 2004). Under the action of glacial freezing–thawing and scouring, the Yulong Snow Mountain area, situated in the lower reaches of the study area, witnessed the development of extensive moraines. The activity range of these moraines extended up to an elevation of nearly 2800 m (Fig. 4b).
(a) Schematic diagram of the coupling between uplift of the Qinghai-Tibet Plateau and the Quaternary glaciation, revised from Zhou et al. (Reference Zhou, Wang, Wang and Xu2006). (b) Characteristics of glacier development in Yulong Snow Mountain region.
Quaternary allochthonous deposit
The topography of the study area is undulating (Fig. 5a) due to regional uplift, river incision, and glacial activity. Quaternary allochthonous deposits are prevalent in the area (Fig. 5b). Considering the widespread existence of small-scale deposits and the limited accuracy of remote-sensing interpretations, only large-scale mass movements are listed in this study. Referring to the classification of movement types (Hungr et al., Reference Hungr, Leroueil and Picarelli2014), allochthonous deposits are simplified as landslide (6), debris flow (13), and bedrock deformation body (6). All deposits were confirmed by field survey and remote-sensing interpretation. 3D imagery was analyzed to determine the location and dimensions of the deposits (Fig. 5b). The HAD, located at the junction of the Jinsha and the Yangbi valley, with an average length of 4 km and average width of 1–1.5 km, is a multi-source, multi-stage, and complex deposit (Fig. 5c). The HAD is an excellent window to understand the paleoenvironment and geomorphic evolution, due to its long history, complex geological features, and special geographical location.
(a) Topographic characteristics of the FBYR. (b) Distribution of allochthonous deposits in the FBYR; yellow rectangle indicates location of Figure 5c. (c) Hongwen allochthonous deposit (HAD).
Methodology
Field observations
The geological characteristics of the HAD and the traceability of river sediments were the primary foci of the field investigation, providing essential foundations for the analysis of geomorphological evolution. Based on morphological and sedimentological characteristics, the HAD is divided into three distinct parts (Fig. 6a): Part I (3.52 km2), Part II (0.53 km2), and Part III (0.97 km2) (Fig. 6b). The valleys located on the eastern side of these parts serve as the source areas (Fig. 6c). In order to obtain more specific geological characteristics of the allochthonous deposit and the river sediment, a detailed survey of representative outcrops was conducted in each part. Additionally, we traced the development of river terraces upstream of the deposit.
(a) Characteristics of zoning in the HAD. (b) Geometric features of the different parts in the HAD. (c) Topographic features of the source areas of the different parts.
Sampling and dating
Chronology of the HAD and river sediments is of great significance for analysis of the genetic mechanism of the HAD and the evolution of fluvial geomorphology. Thermoluminescence (TL) is a process in which substances that have previously absorbed radiant energy release photons under excitation by heat. In sediments, the annual radiation dose received by crystalline material that can store energy in a stable radiation field is a constant. Therefore, the burial time (T) of deposits can be calculated according to the number of electrons accumulated (N) by mineral grains since burial time as a result of such radiation and the rate (B) at which the radiation has been received, namely T = N/B. Samples used for dating were collected from typical locations in different parts of the deposit (Fig. 7a). All deposits were sampled using stainless steel tubes on clean surfaces and sealed with aluminum foil to avoid exposure to light and moisture loss (Xu et al., Reference Xu, Jiang, Yu, Yang and Chen2015; W. Zhang et al., Reference Zhang, Wang, Chen, Soltanian, Dai and WoldeGabriel2022). The samples were numbered RSG1–RSG5. Among them, RSG1 and RSG2 were collected from the local shear zone of Part III and Part II, respectively, representing the age of the debris-flow and the rock landslide (Fig. 7b). Three samples were collected from Part I due to its large scale (Fig. 7b). RSG3 came from the local ductile shear zone formed by the difference in velocity at the top of the deposit during movement of the moraine landslide, reflecting the time of the landslide (Part I). RSG4 came from the upper broken moraine rock block, reflecting the formation time of the moraine. RSG5 was collected from the riverbank of the front edge of Part I, which is a comparative sample of modern fluvial sediment.
(a) Distribution of dated samples in parts I, II, and III. (b) Locations and ages of dated samples.
All samples were used for fine particle component preparation under laboratory red light conditions (wavelength 640 ± 10 nm). Firstly, about 20 g were used to determine water content and the content of U, Th, and K. A central sample was then taken for screening (180 mesh). Organic matter and carbonates were removed using 40% hydrogen peroxide (H2O2) and 30% hydrochloric acid (HCl), respectively, after which 30% fluorosilicic acid (H2[SiF6]) was added. After five days, the sample was washed with distilled water until it reached a neutral pH. The suspension was placed in a beaker and allowed to stand to obtain 4–11 μm sized particles. The separated fine particle components were fully shaken and injected into the prepared funnel with stainless steel sheets. Finally, the experimental sample was obtained by drying fine-grained components precipitated in a funnel at a low temperature.
The test instruments used in this study were Daybreak 2000 Luminometer, 801E irradiator, and inductively coupled plasma mass spectrometry (ICP-MS). The sample equivalent dose and environmental dose were measured to eliminate errors. The equivalent dose was determined using a simple multi-aliquot regeneration method. After measuring the luminescent signals for both natural and regenerative doses, the samples were irradiated with an experimental dose, and the changes in sensitivity were monitored by measuring the luminescent signals corresponding to the experimental dose. The equivalent dose was then determined by calculating the ratio of the luminescence signal from the natural/regenerative dose to that of the subsequent experimental dose. The environmental dose was obtained by measuring the levels of radionuclides and water content, and it was further adjusted using the Fleming method (Fleming, Reference Fleming1971).
Results and Analysis
Geological characteristics of the deposit
Part I
Part I, the main part of the HAD, is a calcareous cement-rich deposit formed by a moraine landslide that has undergone extended physical and chemical alteration. There are mainly three judgment bases for moraine landslide: (1) the study area is significantly affected by Quaternary glaciation (Fig. 4b); (2) the cirque-shaped source area is characterized by typical early-glacial geomorphology and is a natural place for loose material storage (Fig. 8a); and (3) the sedimentological characteristics of the deposit are consistent with a moraine landslide. Several huge and complete rock masses (carbonate rocks) are embedded in the deposit (Fig. 8b), and some rock blocks are also scattered on the surface of the deposit. In addition, there are numerous unrounded, unsorted, and poorly graded rock blocks inside the deposit, which conforms to characteristics of moraine landslides (Fig. 8c). Part I has experienced a long evolution since its formation and is currently predominantly composed of calcareous cement. Affected by weathering and rainfall, these calcareous cements generally show significant signs of erosion (Fig. 8d).
Field investigation of the HAD. (a) Source area of Part I; (b) huge rock mass embedded in Part I; (c) typical outcrop of Part I; (d) calcareous cementation and corrosion; (e) loose landslide deposits from Part II; (f) typical outcrop of Part II; (g) debris-flow deposits; (h) alluvial deposits; (i) material flow in debris-flow deposits; (j) stratification characteristics of debris flow deposits.
Part II
Part II is a landslide deposit composed of loose and fully weathered carbonate rocks (Fig. 8e, f), whose lithology is consistent with that of the source area on its SE side (Fig. 6a). Because the strata of the source area are nearly upright with a trend that is nearly parallel to the Jinsha River, a high and steep slope structure has formed. Over time, geological forces acting on the rock mass have caused bending deformation and the formation of tensile cracks. Eventually, this deformation progressed to a toppling failure mode, resulting in the occurrence of the landslide. In terms of vertical sequence, Part I is situated above Part II, indicating that Part I was formed subsequent to the formation of Part II.
Part III
Part III consists of debris-flow (Fig. 8g) or alluvial deposits (Fig. 8h), which exhibit characteristics of material flow (Fig. 8i) and stratification (Fig. 8j). The presence of sorted and rounded gravel within the deposit serves as evidence that it has been transported by water flow. Situated at the starting point of the Yangbi valley, Part III showcases a prominent feature of the valley—its Quaternary sediment fill. Regarding vertical arrangement, Part II is positioned above Part III, indicating that the geomorphic pattern of the Yangbi valley and Part III was established prior to the formation of Part II.
Developmental characteristics of river sediments
The investigation of terraces from Shigu to HAD reveals the presence of three terraces (T1, T2, and T3) in Shigu (Fig. 9a). Terraces T2 and T3 where the Shigu is located are 10–30 m higher than the river surface (Fig. 9b) and mainly composed of breccia sand or gravel sand (Fig. 9c). Terrace T1 is less than 10 m higher than the river surface (Fig. 9b) and mainly composed of medium- to coarse-grained sand. Only one continuous terrace (T1) is observed in the lower reaches of Shigu. Furthermore, river sediments were observed on the surface of the HAD. However, due to extensive artificial reconstruction, only certain portions of river sediments are exposed in the middle and upper sections of the deposit. This indicates that the ancient Jinsha River may have flowed through the middle and upper regions of the HAD in the past.
(a) Aerial view of the three terraces at Shigu. (b) Profile of the river terraces developed at Shigu (revised from Tong, Reference Tong2019). (c) Typical outcrop of T3. (d) Longitudinal profile of the First Bend reach showing river terraces (revised from Tong, Reference Tong2019); arrows for SSW and NNE indicate the approximate flow direction of the river around the First Bend reach.
In a study conducted by Tong (Reference Tong2019) on river terraces, it was revealed that river sediments are well developed in the FBYR (Fig. 9d). The Qizong–Shigu section generally exhibits the presence of three terraces. However, the continuity of terrace development downstream of Shigu is relatively poor, which aligns with our existing understanding. By combining the river terrace profiles depicted by Tong (Reference Tong2019) for the Qizong–Shigu section with our investigation results on terrace distribution between Shigu and HAD, a comprehensive terrace profile of the FBYR has been established (Fig. 9d).
Chronology characteristics of the deposit and river sediments
Chronological measurements indicate that all samples, except for RSG1, exhibited normal luminescence signals and had corresponding equivalent dose values (Table 1). The dating results allowed us to establish that the moraine formed before 85 ka. The Part II and Part I deposits formed before 46.4 ± 2.8 ka and 33.5 ± 1.4 ka, respectively. Unfortunately, the age of Part III could not be determined directly due to insufficient luminescence signals. However, field investigation confirmed that the formation of Part III occurred earlier than that of Part II, indicating an earlier timeframe for the deposition of Part III.
TL dating results.
The age estimation of river terraces is based on previous studies conducted by Tong (Reference Tong2019) (Table 2). Both Shigu and Qizong exhibit the presence of three terraces, while all corresponding terraces demonstrate comparable ages, with the exception of secondary terraces. Taking the Shigu area near the study site as an example, the ages of T1, T2, and T3 are estimated to be 0.42 ka, 7.10 ka, and 10.53–14.18 ka, respectively. The vertical distance between two adjacent terraces ranges from 11–13 m.
Age of river terraces at Qizong and Shigu (Tong, Reference Tong2019).
Discussion
Paleoseismic events and paleoclimate characteristics
The chronological results obtained from the deposits necessitate analysis within the framework of paleostructural and paleoenvironmental evolution. Affected by the influence of tectonic movement and regional tectonic stress adjustment, an extremely rapid uplift phase began after the late Pleistocene (Fig. 3a) (Zhan et al., Reference Zhan, Chen, Zhang and Han2018; Li et al., Reference Li, Chen, Zhou, Song, Zhang, Gu and Cao2020), which resulted in strong tectonic activities and several regional differential uplift events in the study area (Ming and Shi, Reference Ming and Shi2006). Many faults have been reactivated by tectonic activities. The Longpan–Qiaohou fault exerts the most significant influence on the HAD. This fault is located along the posterior margin of the HAD, spanning approximately 120 km in length, and exhibiting an overall strike of 015–020°. Since the Late Quaternary, this fault has been characterized by strike-slip movement in mountainous regions and dip-slip movement in the basin areas. The maximum horizontal slip rate ranges from 3.10–6.45 mm/yr (Tang et al., Reference Tang, Hu, Tian, Wang, Yang and Xiong2014). The vertical slip rate has been estimated to be approximately 0.2 mm/yr to approximately 0.45 mm/yr since the late Pleistocene. According to the dating analysis of a paleoseismic trench in the Jianchuan basin, three seismic events have been identified along the Longpan–Qiaohou fault since the late Pleistocene (Tang et al., Reference Tang, Hu, Tian, Wang, Yang and Xiong2014). Notably, the timing of one paleo-earthquake event (33,860 ± 280 yr BP) (Tang et al., Reference Tang, Hu, Tian, Wang, Yang and Xiong2014; Li et al., Reference Li, Chen, Yan, Zhou, Wang, Li and Zhang2022b) corresponds to the formation time (33.5 ka) for primary body (Part I) of the HAD.
The formation period of the HAD, ranging from 46.4–33.5 ka, represents a very interesting time node. Paleoclimate studies indicate that 40–30 ka falls within the interglacial epoch of the last glaciation in China (Shi et al., Reference Shi, Liu, Li and Yao1999), and it is the warmest period within the last glaciation (Shi et al., Reference Shi, Yu, Liu, Li and Yan2001). During this period, the most remarkable climatic characteristics were increasing temperatures, glacier retreat, and plentiful rainfall (Shi et al., Reference Shi, Liu, Li and Yao1999, Reference Shi, Yu, Liu, Li and Yan2001). This conclusion is drawn from various sources of evidence, such as the analysis of the Guliya ice core and examination of pollen records (Berger, Reference Berger1978; Thompson et al., Reference Thompson, Yao and Davis1997; Shi et al., Reference Shi, Yu, Liu, Li and Yan2001; Xiao et al., Reference Xiao, Zhao, Wang and Zhou2013). These studies indicate that the temperature of the Tibetan Plateau was approximately 2–4℃ higher than present-day temperatures, which has been attributed primarily to enhanced solar radiation during the summer season as a result of the precession cycle. High lake surface records from numerous large lakes in the Tibetan Plateau provide further evidence of abundant precipitation during that period, with rainfall levels estimated to be 1.4–2.0 times higher than the present (Shi et al., Reference Shi, Yu, Liu, Li and Yan2001). Prolonged rainfall and subsequent erosion and seepage have the potential to weaken the stability of rocks and soil in the source area. These climatic conditions create favorable circumstances for geological processes such as landslides or debris flows to occur.
Incision rate of the Jinsha River
The formation and evolution of the HAD cannot be discussed without considering the influence of river undercutting, the most direct manifestation of which is incision rate. River sediments exposed on the surface of the HAD can provide some insight into the incision rate of the Jinsha. Fluvial cobble–gravel was observed near the RSG3 sampling point during field work (Fig. 10a), indicating that the ancient Jinsha flowed through this area after the deposit formed. The river sediments exhibit good cementation characteristics but poor stratification. The distribution of cobble–gravel is discontinuous and is only visible in locally excavated outcrops. A reasonable explanation is that instability of the ancient Jinsha River during formation of the HAD, along with subsequent human activities, has made it exceptionally challenging to trace the deposition of river sediments. The only two outcrops available are roughly aligned with the elevation of RSG3. Therefore, the age of RSG3 (formation time of the HAD) can be assumed to be the same as that of the Jinsha sediments at the same elevation. RSG5 represents the age of modern river sediments. Combined with the elevation difference between RSG3 (1867 m) and RSG5 (1825 m), the incision rate of the Jinsha from 33.5–13 ka in the First Bend reach can be calculated as 2.05 mm/yr (Fig. 10b). To support this inference, it is necessary to gather chronological data and conduct incision-feature analysis derived from the study area's river terrace evolution. Table 2 shows that the terraces at Qizong and Shigu primarily represent three levels (T1, T2, and T3) (Fig. 9d). The average incision rate of the river has greater significance over a longer time span. Therefore, taking the T1 and T3 terraces as a reference (Table 2), the average river incision rates of Shigu and Qizong are 2.566 mm/yr and 2.277 mm/yr, respectively. This result is similar to the incision rate inferred from the exposed river sediments on the surface of the HAD.
(a) River sediments exposed on the south side of the HAD. (b) Topographic characteristics of the Yangbi valley; inset table shows the incision rate of the Jinsha River, determined by examining the age and elevation of the river sediments exposed on the surface of the HAD (referenced from Li et al., Reference Li, Chen, Yan, Zhou, Wang, Li and Zhang2022b).
The average (2.3 mm/yr) of the incision rates mentioned above is adopted as the reference value for the incision rate in the study area. Considering the formation time of the main body of the HAD, it can be estimated that the river has experienced a downcutting depth of approximately 77 m since 33.5 ka. This value is similar to the thickness of the deposit, suggesting that the formation of the large-scale HAD may not have caused long-term blockage of the river. Furthermore, it indicates that the downcutting process of the river has been continuous rather than impeded by the deposition.
Genetic mechanism and geomorphological evolution of the HAD
The occurrence of mass movement events is typically influenced by various factors, especially the combined effects of heavy precipitation and seismic shaking (Dortch et al., Reference Dortch, Owen, Haneberg, Caffee, Dietsch and Kamp2009). Prolonged increases in precipitation can raise pore water pressure and reduce the strength of rock and soil, causing previously stable slopes to become unstable due to unloading, erosion, and tectonic activity. This process persists over the long term until the slope is pushed beyond its stability threshold (Owen et al., Reference Owen, Kamp, Khattak, Harp, Keefer and Bauer2008). Simultaneously, intense precipitation also can lower the seismic acceleration required to trigger landslides. The paleoenvironmental analysis confirms that both tectonic activity and climate change indeed have significant effects on the formation of the HAD. Earthquakes, rainfall, snowmelt, and glacial processes may have all played important roles in the formation of the HAD. Therefore, it can be concluded that the HAD is a multi-source and multi-stage product of the synergistic effect of tectonic activity and climate change.
Based on field investigation, terrain interpretation, dating, and paleoenvironment analysis, an evolutionary sequence for the HAD has been proposed (Fig. 11), which indicates that 46.4 ka ago, a phase of the last glaciation was reaching its conclusion. With the tectonic activity and climate changes, the source areas of Part III and Part II underwent destruction and accumulated to the south of Hongwen. At 33.5 ka, a great interglacial stage of the last glaciation (the MIS 3 interstadial) began. This period held significant importance in the formation of gravity-driven deposits along the southeastern margin of the Qinghai–Tibet Plateau, as glaciers receded, temperatures increased, and rainfall significantly intensified.
Schematic diagram of the formation and evolution of the HAD. (a) During the transition to the last glacial period, glacial advance gradually began and moraines were produced; (b) Part III was formed (since no effective dating data are available, the formation time in the figure is an estimate); (c) Part II was formed; (d) Part I was formed; (e) evolution of river landform after HAD formation. The numbers 1, 2, and 3 represent Part III, Part II, and Part I, respectively.
Part I was formed by downslope movement of moraines, possibly triggered by some kind of geological force (earthquake, rainfall, rapid snowmelt, rapid glacial retreat, or a combination thereof), that had undergone erosion by rain and snowmelt over an extended period of time. Of course, the influence of river incision and downcutting cannot be ignored. Finally, a huge deposit (Part I) formed in the Jinsha River course. This deposit did not result in long-term river blockage, but it does not exclude the possibility of temporary channel obstruction. There are three observations supporting this judgment. (1) Lack of lacustrine deposits upstream of the deposit indicates that long-term river blockage did not occur. (2) The deposit front extends directly to the opposite bank, but no residual deposits are observed at the same elevation on the opposite bank. (3) The downcutting depth calculated based on incision rates since the deposit formation is similar to the thickness of the deposit itself. Although Part I did not cause a complete river blockage, its large scale still led to a temporary rise in river water level within a short period of time. With water level upstream rising gradually and eventually overtopping the front of the deposit, the leading edge was eroded. After formation of the new river course, the leading part of the deposit was rapidly carried downstream by high-energy water currents. Lateral erosion by these currents primarily affected the left bank, which was related to the topography of landslide deposit. The left bank was considerably eroded and scoured, resulting in almost no residual deposits, whereas the main body of the landslide remained relatively intact on the right bank. After that, with continuous downcutting of the river and supergene dynamic action, the current geomorphic features were formed.
The related analysis of the HAD can serve as a reference for studying the ancient deposits, recording both paleotectonic and paleoclimatic conditions. This contributes to a better understanding of the evolution of river systems in regions with complex geological conditions. As widely acknowledged, mass movement events, such as collapses, landslides, and debris flows, frequently occur in mountainous river valleys globally (Haque et al., Reference Haque, da Silva, Devoli, Pilz, Zhao, Khaloua and Wilopo2019; Fan et al., Reference Fan, Dufresne, Siva Subramanian, Strom, Hermanns, Tacconi Stefanelli and Hewitt2020). In certain regions, these events often concentrate and exhibit a cascading effect (Zhan et al., Reference Zhan, Chen, Zhang and Han2018; Li et al., Reference Li, Chen, Zhou, Song, Zhang, Gu and Cao2020). They are usually caused by complex geological environmental factors, such as tectonic activities, temperature fluctuations, rainfall patterns, glacial activity, and erosion and weathering processes (Shulmeister et al., Reference Shulmeister, Davies, Evans, Hyatt and Tovar2009; Korup et al., Reference Korup, Gorum and Hayakawa2012; Reznichenko et al., Reference Reznichenko, Davies, Shulmeister and Larsen2012; Kirschbaum et al., Reference Kirschbaum, Kapnick, Stanley and Pascale2020; Lin et al., Reference Lin, Steger, Pittore, Zhang, Wang, Jiang and Wang2022; Zhao et al., Reference Zhao, Su, Xu, Li, Xu and Wang2023). The residual deposits accompany the subsequent processes of geomorphic evolution. Therefore, when investigating complex river geomorphic evolution issues, in addition to the geometric and sedimentological characteristics of the river system itself, dating analysis and the analysis of the genesis mechanism of ancient deposits can provide insights into paleoenvironments. This contributes to reconstructing geological history and enhancing the accuracy of the analytical results. In fact, researchers have already noted that the formation of extensive deposits often heralds extreme climate or tectonic events (W. Zhang et al., Reference Zhang, Wang, Chen, Soltanian, Dai and WoldeGabriel2022). For example, in the Qinghai–Tibet Plateau and the surrounding mountain belts, the distribution frequency of ancient deposits over time often corresponds to periods of climate change and active tectonics (Li et al., Reference Li, Chen, Wang, Chen, Bao, Yan and Li2023). Therefore, the unique geological historical period during which the HAD formed, as presented in this study, is likely just a microcosm of widespread mass movement on the Qinghai–Tibet Plateau. Future research on a larger spatial scale will contribute to reshaping this history, and the HAD can offer insights for related studies.
Conclusions
The formation of deposits plays a pivotal role in the evolutionary dynamics of a valley's geomorphology. In this study, we focused on investigating the allochthonous deposits and the river sediments of the First Bend of the Yangtze River (FBYR). Through analyzing the evolutionary process of the Hongwen allochthonous deposit (HAD) and the incision rate of the Jinsha River, we gained valuable insights into the paleostructure and paleoclimate characteristics of the region.
The HAD is a complex allochthonous deposit with diverse geological characteristics, resulting from a multi-source and multi-stage landslide influenced by the combined effects of tectonic activity and climate.
The HAD is categorized into three distinct parts, distinguished by their morphological and sedimentological features. Part I (3.52 km2), the primary component of the HAD, is a calcareous cement-rich deposit resulting from a moraine landslide. Part II (0.53 km2) consists of loose and fully weathered carbonate rocks forming a landslide deposit. Lastly, Part III (0.97 km2) encompasses debris-flow or alluvial deposits characterized by material flow and stratification.
Following the late Pleistocene, a notable phase of rapid plateau uplift commenced, leading to intensified tectonic activity and several regional uplift events in the study area. The formation of the HAD took place between 46.4–33.5 ka, aligning with the significant interglacial stage of the last glacial period or the transitional period leading up to it. This timeframe was characterized by frequent tectonic activity, rising temperatures, increased rainfall, and heightened surface erosion.
The evolution of the HAD and the river sediments of the First Bend reach reveal that the average incision rate of the Jinsha over the past 33.5 thousand years was about 2.3 mm/yr.
The HAD represents the outcome of the combined influence of tectonic activity and climate change, serving as the primary mechanism driving surface processes in the First Bend region. The HAD has undergone a significant evolutionary process. Subsequently, with the continuous downcutting of the river and supergene dynamic action, the current geomorphic features were formed.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant No. 42307204, 41941017). The authors would like to thank the editor and anonymous reviewers for their comments and suggestions, which helped a lot in making this paper better.
Conflict of interest
The authors declare that they have no conflicts of interest related to this work.
Author contributions
Yuchao Li: conceptualization, methodology, investigation, writing-original draft; Jianping Chen: conceptualization and methodology; Qing Wang: resources, investigation, supervision, validation, review, and editing; Zhihai Li: investigation, data curation; Yansong Zhang: investigation, data curation; Jianhua Yan: language modification, investigation.
