Abstract
Damming-induced river dis-connectivity represents one of the most pervasive anthropogenic perturbations to global river systems. We report detrital apatite fission-track (AFT) thermochronology from modern sediments of the upper Yellow River and the Wei River, and compare these data with bedrock AFT signatures from the adjacent northeastern Tibetan Plateau as a potential primary sediment source. The results reveal pronounced spatial variability in age‑component proportions along the uppermost Yellow River but only minor downstream change along the Wei River. These patterns can be reasonably explained by extensive cascade damming and reservoir trapping in the uppermost Yellow River versus minimal impoundment in the Wei River. Impoundments have reduced sediment connectivity, disrupted sediment loads, and likely altered hydrological dynamics and associated ecosystems. The provenance trends of the upper Yellow River from the Guide region to the West Ordos Block segment documented here challenge interpretations that attribute sediment supply throughout the upper Yellow River solely to large-scale exhumation of the Tibetan Plateau; instead, they indicate substantial contributions from localized sources, including adjacent mountain ranges and Cenozoic basins. Collectively, our interpretations demonstrate that anthropogenic infrastructure has produced partially blocked sediment routing along discrete river reaches and, at the scale of local river systems, can dominate over natural tributary-driven sediment mixing. These findings underscore the urgent need to evaluate the socio‑economic and eco‑environmental consequences of disrupted sediment connectivity and to integrate connectivity preservation into sustainable river management and mitigation strategies.
Similar content being viewed by others
Introduction
The evolution of transcontinental rivers is governed by variation of topography, activation of tectonic structures1,2,3, climate variability4, and anthropogenic forcing5,6,7. Accelerating global urbanization has substantially altered the hydrological response in river basins around the world8,9,10 via urban land modification11, intensified agriculture12, and extensive dam construction13,14. Prior work demonstrates that the widespread construction of dams and reservoirs fragments river networks15, reduces longitudinal sediment and biotic connectivity16, alters the fluxes of sediment5 and nutrient17, and degrades riverine ecosystems18,19. A more detailed quantification of human impacts on the sediment supply and water discharge5,6,7 is therefore required, given expanding human populations and the proliferation of flow‑regulating infrastructure15. Global assessments indicate that sediment trapping behind large dams is already producing substantial storage loss: cumulative reservoir storage is estimated to have declined by 16% of initial capacity by 2022 and is projected to reach 26% by 2050 across 47,403 large dams in 150 countries20.
The river connectivity can be conceptualized as water-mediated transfer of sediment among storage reservoirs within a catchment sediment cascade over timeframes21, encompassing longitudinal (upstream-downstream and tributary-mainstem relationships)22, lateral (channel-floodplain and slope-channel systems)22, vertical (continuous groundwater-channel interactions)23, and temporal dimensions (seasonal-long term framework)21. Longitudinal connectivity is particularly critical for maintaining ecosystem processes, including ecological succession, genetic interaction, organism dispersal, material and energy flows, and riparian-zone resources24,25. However, physical barriers commonly impair longitudinal links by modifying channel base level and local bed profiles22. For instance, reservoir backfilling and pool aggradation upstream of bedrock controls trap sediment and create localized discontinuities in downstream sediment transport22. Meanwhile, reservoirs and other in‑stream barriers can intercept a large fraction of sediment fluxes, in some cases retaining up to 90% of incoming load22,26,27. As loss of connectivity alters habitat structure and degrades biodiversity28, preserving longitudinal sediment transfer has become a priority in contemporary water‑management and river restoration policy29.
The Yellow River, Chinese second-longest river, extends 5,464 km, sustains 12% of the national population and irrigates ~15% of arable land (Fig. 1a–c), and delivers an average suspended sediment discharge of ~1.08 × 109 tons/yr30,31. The Yellow River has undergone pronounced changes in sediment fluxes due to anthropogenic modification, principally dam and reservoir construction, as documented by hydrological, sedimentological and geochemical datasets27,31,32. A key intervention is the Liujiaxia Reservoir in the upper Yellow River (Fig. 1b), constructed in 1968; its cascade of impoundments has markedly altered downstream hydrology and sediment yield, with post‑2000 suspended sediment discharges substantially lower than those observed during 1950 – 199927,33. Consequently, sediment transport pathways, sediment budgets and erosion patterns along the Yellow River have been reorganized, with substantial sequestration behind dams27,34. By contrast, the Wei River lacks major impoundments along its course (Fig. 1a, b, d)35. Although China hosts numerous dams of varying size during the past several decades27, national assessments have largely emphasized the variation impacts in sediment discharge from large reservoirs32 or individual basins27, and the extent and spatial pattern of longitudinal connectivity loss at the national scale remain insufficiently constrained36. Identifying where connectivity has been fragmented and how provenance signals vary spatially therefore remains a critical research priority.
a Topographic map showing the contour of the Yellow River and Wei River. b Topographic map showing the locations of the reservoirs, dams and sampling sites. c Field picture showing the Yellow River in Zhongwei. d Field picture showing the Wei River in Tianshui. U: upper Yellow River. M: middle Yellow River. L: lower Yellow River.
In this study, we evaluate sediment transfer along a segment of the upper Yellow River by tracing variations in detrital apatite populations, an accessory mineral ubiquitous in magmatic, metamorphic, and sedimentary rocks37. Apatite populations were distinguished using fission-track thermochronology, which yields single-grain cooling ages that record closure temperature of approximately 110 °C38. Prior provenance work on the Yellow River has employed geochemical and geochronological proxies (e.g., zircon U-Pb dating39,40,41, isotopic geochemistry27,42, and heavy mineral analysis43). Zircon U-Pb typically resolves broad source populations but may lack sensitivity to fine‑scale spatial variation along the channel30,41. Detrital apatite fission-track analysis, however, offers the unique capability of capturing a wide spectrum of cooling ages derived from diverse bedrock sources44,45. This variability reflects the distinct geological histories of source rocks, allowing for a more precise comparison between the detrital compositions of modern river sand and the adjacent bedrock source domains46,47.
Samples for this study were collected from modern sand bars along an ~850 km reach of the Yellow River encompassing Liujiaxia dam–reservoir system and a cascade of smaller anthropogenic structures, primarily hydropower installations upstream of Lanzhou (Fig. 1a, b). For comparison, six modern sand samples were obtained from a ~ 420 km reach of the Wei River (Fig. 1b), a tributary that remains largely unimpacted by anthropogenic infrastructure.
To specifically evaluate the impact of human-made structures on functional sediment connectivity, we target a relatively short segment of the Yellow River. This contrasts with numerous previous studies that have addressed sediment dynamics at a broader spatial scale. Such studies have demonstrated that sediments in the upper Yellow River are predominantly sourced from the northeastern margin of the Tibetan Plateau41,42, and that, at the scale of the entire Yellow River basin, major sediment storage occurs in the middle reaches (Fig. 1a). In particular, a significant proportion of sediment from this region is ultimately incorporated into the Loess Plateau (Fig. 1b)41.
Despite the overall good sediment connectivity at the large scale along the upper Yellow River, our results indicate that, at a more local scale, the presence of anthropogenic sediment barriers can significantly alter downstream sediment composition. Specifically, impoundments and related retention preferentially reduce mainstem contributions, yielding downstream sediments that are comparatively enriched in material derived from proximal geological units delivered via tributaries, indicative of a disrupted sediment routing along the mainstem.
Results
The bedrock signature
The whole Yellow River is traditionally divided into three reaches characterized by distinct geomorphological features (Fig. 1a)48. The studied segment is a portion of the upper reach that covers the transition between the northeastern margin of the Tibetan Plateau and the Loess Plateau (Fig. 1b), and it is characterized to cross alluvial plains and tight gorges41,43,49. The large variability in geomorphological features also reflects the large variability of bedrock lithologies, including the Archean metamorphic rocks, Proterozoic – Paleozoic magmatic and metamorphic rocks, and Paleozoic and Mesozoic carbonate and clastic rocks30.
The bedrock differs for cooling ages (Figs. 2a and 3a, b), which reflect crustal exhumation histories associated with deformation events that have involved the northeastern margin of the Tibetan Plateau since the Mesozoic. Exhumation-induced cooling ages are diagnostic for distinguishing the different mountain range and provide an effective provenance signature for sediments. The spatial configuration of watershed divides for tributary drainage basins that join the sampled mainstem is a critical control on delineating potential sediment source areas within the study catchments. Detrital AFT ages in river sediments record the cooling-age spectra of tectonic units exposed in the source regions50 and imply statistical mixtures of bedrock and detrital AFT ages derived from distinct source domains (Fig. 4). Bedrock ages are not spatially homogeneously distributed, and the limited number of available data prevents a statistical comparison between the relative sizes of bedrock age populations and those of the modern detrital samples. Therefore, we used the occurrence of specific populations as diagnostic indicators of the provenance.
a AFT ages dataset from bedrocks in the northeastern Tibetan Plateau (details in Supplementary Data 3) are plotted with indication of watersheds and potential source regions. b Non-metric multi-dimensional scaling analysis (MDS) plot of detrital AFT ages of sediments from Yellow River and Wei River. Stress value: 4.17. Data of Yellow River sample GD25 is from Wang et al.56. AR: Arcuate range area (The Arcuate range area is bounded by four WNW- or NNW-trending mountain ranges in Liang et al.89), EWQN: eastern West Qinling mountains. c Non-metric multi-dimensional scaling analysis (MDS) plot of detrital AFT ages of sediments from Yellow River and Wei River, and bedrock AFT ages of rocks (details in Supplementary Data 3) outcropped in the inferred source regions. Stress value: 7.36.
Geological maps showing the sedimentary successions distribution and the magmatic out crop around the Guide Basin, Xunhua Basin, and Xining Basin (a) and around the AR (Arcuate Range) region and EWQN (eastern West Qinling mountains) region (b). The bedrock AFT compilation dataset and the related literatures are in Supplementary Data 3. Modified from the Geological Map of China (1:2,500,000) compiled by the China Geological Survey institute in Hebei. LWR.: Longwu River. ZLR.: Zuli River. SDR.: Sandu River. QSR.: Qingshui River.
The AFT age populations and the histogram were plotted using Density Plotter (bin width and bandwidth were set as 10). a Sample HH8. b Sample HH7. c Sample HH6. d Bedrocks from Guide Basin. e Bedrocks from Xunhua Basin. f Bedrocks from Xining Basin. g Detrital grains from the sedimentary sections in Guide Basin. The sedimentary sections are indicated in the Guide Basin in Fig. 3a. The detrital AFT data of Guide Basin are compiled from Wang et al.56. h Detrital grains from the sedimentary sections in Xunhua Basin. The sedimentary sections are indicated in the Xunhua Basin in Fig. 3a. The detrital AFT data of Xunhua Basin are compiled from Guo et al.90. i Detrital grains from the sedimentary sections in Xining Basin. The sedimentary sections are indicated in the Xining Basin in Fig. 3a. The detrital AFT data of Xining Basin are compiled from Wang et al.56. j Sample HH5. k Sample HH4. l Sample HH3. m Sample HH2. n Sample HH1. o Detrital grains from the sedimentary sections in AR region. The sedimentary sections are indicated in the AR region in Fig. 3b. The detrital AFT data of AR region are compiled from Jiao et al.54. p Bedrocks from AR region. q Sample WH1. r Sample WH2. s Sample WH3. t Sample WH4. u Sample WH5. v Sample WH6. w Detrital grains from the sedimentary section in the EWQN region. One sedimentary section is indicated in the EWQN region in Fig. 3b. The detrital AFT data of EWQN region are compiled from Jiao et al. (unpublished data). x Bedrocks from EWQN region. N shown in (d), (e), (f), (p), (x) represents the number of bedrock samples (each bedrock sample yielding an AFT central age) that were decomposed using Density Plotter. The bedrock AFT compilation dataset and the related literatures are in Supplementary Data 3. N shown in (g), (h), (i), (o), (w) represents the number of grains that were decomposed using Density Plotter.
Bedrock age population P1 refers to the youngest cooling ages (<30 Ma) identified in the Arcuate Range region (AR) of the West Ordos Block51,52,53,54 and in mountains north of the Xunhua Basin55 (Fig. 3a, b), which underwent a rapid and localized exhumation along the tectonic features active during the late Cenozoic.
Early Cenozoic cooling ages (30 – 49 Ma) found in bedrocks are grouped into population P2, which are found in mountains north to the Xunhua Basin55,56, the Central Qilian mountains56,57,58,59, and the eastern West Qinling mountains (EWQN)60,61, which underwent rapid exhumation associated with tectonic deformations during the early Cenozoic (Fig. 3a, b).
Bedrock population P3 corresponds to cooling ages from the late Cretaceous to early Cenozoic (50 – 79 Ma), when the mountains north of the Xunhua Basin55,56,57, the Central Qilian mountains56,57,58,59, and the eastern West Qinling mountains experienced rapid exhumation60,61, whereas the Arcuate Range region52,62 and the western West Qinling mountains63 underwent steady-state regional exhumation (Fig. 3a, b).
Bedrock age population P4 that encompasses ages between 80 and 139 Ma, is found in the mountains north to the Xunhua Basin55,56,57, Central Qilian mountains57, West Qinling mountains59,63 and Arcuate Range region51,52,64, which underwent a rapid or regional exhumation during the Cretaceous (Fig. 3a, b).
Cooling ages between 140 and 179 Ma from bedrocks, referred to as population P5, are in response to the rapid or steady-state exhumation of the Central Qilian mountains56,57 and the Arcuate Range region51,52 during the middle Jurassic – late Cretaceous (Fig. 3a, b).
Based on published data, population P2 appears to be absent in the Arcuate Range (AR) region (Fig. 4o, p), whereas it is present in exposed rocks from Qinling Mountains (EWQN region) (Fig. 4x).
Fission-track age populations as a signal for connectivity
Eight fluvial modern sand samples from the Yellow River and six samples from the Wei River were processed to extract apatite grains, which were subsequently dated using the fission-track method (Supplementary Data 1). Each sample yielded a multimodal distribution of single-grain ages, which was decomposed into statistically significant age populations using Density plotter65 and Binomfit66. To evaluate sediments (dis)connectivity along the studied segment of the upper Yellow River, we examined how these age populations vary longitudinally downstream along the river. We then compared the identified populations with a compilation of previously published AFT ages from bedrock exposed in the river’s source regions. We distinguished AFT ages derived from pre-Cenozoic units and from magmatic rocks (here collectively referred to as bedrock samples) from those derived from sedimentary rock representing Paleogene and Neogene basin fill. The Wei River provides a useful comparison, as it is largely undisturbed by dams or hydropower stations and drains lithologies similar to those of the Yellow River.
To assess similarity among the analyzed samples, we employed non-parametric multidimensional scaling (MDS), in which samples with similar age distributions plot in close proximity to one another67. The MDS results revealed substantial variability among the Yellow River samples, which are widely dispersed in MDS space, in contrast to the more tightly clustered and uniform age distributions observed in the Wei River samples (Fig. 2a, b).
The three uppermost Yellow River samples (HH8, HH7, and HH6) occupy distinctly separate positions in MDS space, with HH6 and HH8 located on opposite sides and HH7 positioned intermediately, indicating significant differences among them (Fig. 2a, b). In terms of age populations, HH8 is characterized by a dominant age population at ~33 Ma (P2, accounting for 34% of grains), which is absent in the two downstream samples (Fig. 4a, Supplementary Table 1). These downstream samples, in turn, display a younger age population (P1 < 30 Ma) not present in HH8 (Fig. 4a, b, c). Notably, two age populations identified in HH8 (P3 and P4, accounting for 62% and 4% of grains, respectively) are also present in HH7 but in different age proportions (P3 and P4 account for 39% and 44% of grains, respectively) (Fig. 4a, b). HH6, however, exhibits distinct age populations except for the age population about 110 Ma (Fig. 4a, b, c).
Despite these differences, each of the three samples shows a good correspondence with AFT bedrock ages from the surrounding source rock (Fig. 4d, e, f). The match between modern sand and crystalline bedrock AFT ages is particularly notable, despite the limited number of available bedrock datasets. In contrast, the comparison of modern sands with AFT ages from the sedimentary successions of Xunhua Basin (stratigraphic age: ca. 7 – 30 Ma) is less consistent, owing to the absence of the youngest age populations. These younger populations are related to post-depositional Miocene – Pliocene exhumation and are therefore not recorded in the older basin fill (Fig. 4h).
In the lowermost studied reach of the Yellow River, from sample HH5 to HH1, the samples cluster closely in MDS space, indicating broadly similar age distributions (Fig. 2a, b). Samples from HH5 to HH2 share the presence of population P2 (36 – 44 Ma, accounting for 11% – 43% of grains), a population P3 (62 – 70 Ma), often similar in abundance to P2, and population P4 (29% – 53% proportion), which frequently represents the most abundant component and is consistently present across these samples (Fig. 4j, k, l, m). In contrast, population P2 is absent in upstream samples HH6 and HH7 (Fig. 4b, c).
Wei River samples differ from those of the Yellow River in that they display more homogeneous age distributions, dominated by population P2 (36 – 39 Ma, accounting for 20% – 40% of grains) and population P3 (52 – 76 Ma, accounting for 40% – 69% of grains) (Fig. 4q, r, s, t, u, v). These populations are similar to those observed in the downstream Yellow River samples (below HH5) (Fig. 4j, k, l, m, n), although some downstream Yellow River samples (HH2, HH3, HH4) also contain a substantial contribution from population P4 (Supplementary Table 1). More specifically, Wei River samples show very similar AFT age distribution to upstream Yellow River sample HH8 and downstream sample HH5 (Fig. 2b), suggesting a potential common provenance.
Discussion
(Dis)connectivity along the Yellow River
The large variation in AFT age population along the studied portion of the Yellow River may be related to the presence of 21 hydropower stations and dams and three reservoirs. The degree of the sediment disconnectivity appears to vary along the river, with reaches where the sediment transfer is partially blocked and others where the downstream sediment transport appears more efficient. Perturbations of the river sediment chain are also evident in published zircon U-Pb datasets, where differences in age population proportion are observed across the Liujiaxia reservoir and dam41 (Fig. 1b). However, zircon U-Pb dating alone may not fully capture all variations in age populations along the upper Yellow River, as zircon U-Pb ages from river sediments and bedrock cluster mainly around two populations, at ca. 250 Ma and ca. 450 Ma30,41.
Severe impaired river connectivity is observed in the upstream Yellow River, as inferred from variation in AFT age populations in samples HH8, HH7 and HH6. Between samples HH8 and HH7, the abrupt decrease in P1 + P2 and the increase in P4 (Fig. 5b) suggest that sediment supply along the mainstem was altered and that the contribution from tributaries became dominant, as P4 represents a diagnostic population derived from local rocks (Fig. 4b, e). Several studies have proposed that the age composition changes significantly downstream of the tributary confluence when the mainstem sediment load is limited by anthropogenic structures68,69,70. Between samples HH8 and HH7, only small tributaries, such as Longwu River (Fig. 5a), join the Yellow River, contributing a total drainage area of 4,990 Km2. This is in contrasts with the drainage area of the Yellow River upstream of HH8 (56,169 Km2) (Supplementary Fig. 1). When a tributary catchment and the mainstem share similar relief, lithology (and therefore similar apatite fertility71), and climate conditions, the sandy sediment load is expected, to first order, to scale with drainage area68,70,72,73. The large difference in drainage area between Yellow River upstream of HH8 and the tributaries upstream of HH7 suggests that the sandy sediment contribution from these tributaries should be limited. Moreover, no evidence for young local erosion is observed in rocks drained by these tributaries, such as Longwu River (LWR, Fig. 3a), as indicated by the dominance of relatively old P4 population (80 – 139 Ma), which could otherwise supply young AFT ages50. Together, these observations suggest that primary cause of the change in age population proportion is blockage of sediment transport along the mainstem due to the presence of hydropower stations. A similar interpretation has been proposed for the Missouri system68. The natural mixing of sediments cannot be rule out, but its contribution is likely secondary. Moreover, our sampling was carried out several kilometers downstream of tributaries confluences to minimize potential bias from incomplete sediments mixing between tributaries and mainstem source.
a Topographic map showing the river connectivity extent among the river samples. b Distribution of the AFT age populations of Yellow River samples and Wei River samples along the mainstem. Yellow R.: Yellow River. Wei R.: Wei River. LWR.: Longwu River. DXR.: Daxia River. TR.: Tao River. HSR.: Huangshui River. YCR.: Yuanchuan River. ZLR.: Zuli River. QSR.: Qingshui River. DR.: Dahe River. XHR.: Xianhe River. SDR.: Sandu River. HR.: Hulu River. NR.: Niutou River. TGR.: Tongguan River. The lengths of the river along the sampling sections are shown in Supplementary Data 4.
From sample HH7 to HH6, the AFT age population changes again, marked by a decrease in the P1 + P2 and P3 populations and the appearance of a new P5 population. Variability in detrital AFT age populations is interpreted primarily as a signal of anthropogenic disruption to sediment connectivity, with subsidiary modification by natural tributary mixing. When both the river trunk and its tributaries share similar lithologies (with no significant variation in mineral concentrations71), reliefs, and other controlling factors, sediment load is expected to scale with drainage area72,73,74. The Huangshui River joins the mainstem between samples HH7 and HH6. The difference in drainage area between the Huangshui tributary (32,954 km²) and the river trunk upstream of HH7 (63,378 km²) suggests that sediment input from the Huangshui River should be limited. However, the Huangshui River (HSR, Fig. 5a, b) may have transported sediments characterized by the P5 AFT age population (140 – 179 Ma) to sample HH6, as these ages are well represented in the bedrock of the Huangshui catchment, whereas erosion associated with P5 is not evident upstream of samples HH7 and HH8 (Figs. 4a–c and 3a). These observations suggest that the primary cause of the observed change in age population proportions is sediment blockage along the mainstem, induced by the presence of numerous hydropower stations and the Liujiaxia Reservoir (approximately 130 Km2), which enhances the relative sediment contribution from Huanshuang tributary.
A further marked change in age population proportions occurs between sample HH6 and sample HH5 (Fig. 5b), a reach where the Yellow River flows through deeply incised gorges and where numerous hydropower stations have been constructed. Gorges are generally considered favorable geomorphic settings for efficient sediment transfer because of limited sediment storage and high stream power70. The pronounced sediment disconnectivity observed between sample HH6 and sample HH5 therefore suggests that sediment blockage by hydropower station is the dominant controlling process. Possible contribution from natural sediment mixing with tributaries along this reach are considered secondary due to the small drainage areas of these tributaries (Supplementary Fig. 1). Relative to the HH6, sample HH5 shows an increase in P1 + P2 populations and the disappearance of P5 population (Fig. 5b).
Downstream to HH5 sample, the Yellow River flows through gorges, but flow is no longer interrupted by human structures. This condition is reflected in the limited changes in age population proportions between sample HH5 and HH4 (Fig. 5b). Sample HH4 displays AFT age populations consistent with those of HH5 (Fig. 4j, k), implying the lack of tributaries contribution and the intact sediment connectivity along this river segment (Fig. 5a, b).
The sediments chain in the lowermost studied reach shows overall continuity, as indicated by the close proximity of samples HH1 to HH4 in MDS space (Fig. 2b). However, lightly impaired river connectivity is inferred among samples HH4, HH3, and HH2. Sample HH3 lacks the P3 population present in HH4, whereas samples HH2 and HH1 contain four age populations, including additional P3 and P5 components (Fig. 4k–n). Minor differences in heavy mineral proportions are likely related to dam construction and tributary confluences (sites d, e, g, h, i; Supplementary Fig. 2)43.
Wei river samples show consistent age distributions along the analyzed reach, dominated by populations P2 and P3, which together account for 67% – 92% of grains (Figs. 4q–v; and 5b). Sediment transfer along the mainstem therefore appears efficient and largely unaltered, suggesting intact river connectivity. The analyzed reach of the Wei River extends for ~420 km and crosses lithologies with highly variable cooling history, reflected in complex bedrock AFT ages pattern75,76. Under such conditions, tributaries would be expected to deliver sediments with distinct detrital AFT signatures, potentially producing large downstream variation in age populations, similar to those observed in the Yellow River. In contrast, the AFT populations remain consistent along the Wei River, indicating limited tributaries influence.
Sediment provenance pattern
The broad spectrum of fission-track ages reflects the heterogeneity of cooling ages among potential sources and their distinct geological histories. Consequently, detrital AFT age populations in river sediments can capture both sediment transfer efficiency and sediment source location, making provenance comparison between modern river sands and potential sources feasible.
Sediments in the uppermost sample (HH8) were likely derived primarily from the upstream mainstem draining the northeastern Tibetan Plateau27,42 and the Guide Basin. Few bedrock AFT ages are available from the uppermost Yellow River, making its contribution difficult to constrain. However, the AFT ages signal of HH8 is consistent with AFT ages from bedrock and sedimentary sequences of the Guide Basin, which are dominated by populations P2, P3, and P4 (Fig. 4a, d, g, Fig. 3a and Fig. 2c). It is therefore likely that the main provenance of sediments in HH8 is from local sources within the Guide Basin, particularly given that the river reach upstream of HH8 is characterized by hydropower stations that may have interrupted sandy sediment transport along the mainstem draining the uppermost Yellow River.
A provenance contribution from the Xunhua Basin and the mountains north to the Xunhua Basin is detected in sample HH7, as indicated by the occurrence of population P1 and the dominance of population P4 (Figs. 3a and 5a, b). Population P4 is very abundant in the sedimentary sequences of the Xunhua basin, which are widely exposed along the river reach between HH7 and HH8. This suggests an increased contribution from very local sources, rather than from large tributaries. Such a local provenance implied enhanced erosion of river bedrock or small tributaries, likely induced by channel incision downstream of an artificial barrier, a geomorphic response commonly observed downstream the dams69. The presence of population P1, which is well represented in bedrock proximal to the sampling location (Fig. 3), further supports increased local erosion.
Sample HH6 contains a substantial contribution of sediments sourced from the Xining region (Figs. 2c and 3a), adjacent to the Central Qilian mountains, as evidenced by the presence of population P5 (Figs. 4c and Fig. 5a, b). Bedrock AFT ages contributing to population P5 in the Xining Basin closely match those observed in sample HH6 (Fig. 4f). This interpretation is further supported by modern sediments from the Huangshui River (HSR), which drains the Xining region (Fig. 5a) and display a detrital AFT age population centered at ~145 Ma77. Zircon U-Pb age distribution and K-feldspar Pb isotope data indicate similar sediment contributions between the Huangshui River and the Yellow River at Lanzhou (Fig. 2a) consistent with the proximity of sample HH6 to the Lanzhou sample reported by Lin et al.42.
The detrital signal changes abruptly between samples HH6 and HH5, marked by the disappearance of the oldest population and a relative increase in the population P2 and P3. The provenance of these two populations remains poorly constrained owing to the lack of bedrock and sedimentary AFT data from areas proximal to this river reach. Notably, the age distribution of sample HH5 closely resembles that of Wei Rivers samples WH3 and WH4, as indicated by their proximity in MDS space (Fig. 2b). This similarity can be explained by increased sediment input from the region located in the middle between Yellow River and Wei River (Tianshui Basin, Fig. 3b), which is drained northward by the Zuli River and southward by the Sandu River (ZLR and SDR in Fig. 3b). These two tributaries traverse or border the Tianshui Basin, and connect hydrologically to the Yellow River near sample HH5 and to the Wei River near sample WH3, respectively. They likely acted as transport pathways for sediments derived from the Tianshui Basin to both river system, resulting in the similar provenance signals observed in samples HH5 and WH3 (Fig. 2b).
Support for sediment transport via the Zuli River to the Yellow River trunk downstream of HH5 is provided by clay mineral analyses, which show that both the Zuli River tributary (site b, Supplementary Fig. 2) and the downstream Yellow River (site l) are dominated by illite, with subordinate kaolinite, chlorite, and smectite78. Petrographic analyses further indicate that sediments from the Zuli River are enriched in quartz and feldspar, consistent with downstream Yellow River samples78. However, the presence of muscovite and dolomite in downstream trunk samples (sites f and l), but not in Zuli River sediments (site b), suggests that the Zuli River contributes only a partial fraction of the downstream sediment load (Fig. 3b; Supplementary Fig. 2). Although the representativeness of apatite as a provenance tracer for sand-sized sediments can be questioned, the consistency between AFT data, petrographic observations, and clay mineral assemblage strengthens our provenance interpretation.
Downstream of sample HH5, sediment transfer appears more efficient, as indicated by the absence of abrupt changes in age population proportions (Figs. 2, and 5). From sample HH5 to HH1, sediments sourced from the Arcuate Range region (AR in Fig. 2) are consistently present and appear not to have been significantly sequestrated by the reservoir and hydropower stations along the mainstem. Together with contribution from the Arcuate Range, the persistent presence of population P2 indicates a largely intact sediment transfer system. Notably, bedrock in the Arcuate Range region lacks population P2 (Fig. 4o, p), whereas the Tianshui Basin exhibits a dominant P2 population (Fig. 3b), characterized by AFT ages between 20 and 60 Ma, as documented by detrital AFT data in Wang et al.60. The proportion of P2 in samples from HH5 to HH2 decreases from ~35% to ~11%, suggesting a primary contribution from the Zuli River and limited dilution along the mainstem.
Minor variations in AFT ages populations among downstream samples (HH5, HH4, HH3, HH2, HH1) may reflect differential sediment input from tributaries draining the Arcuate Range region (Figs. 2c, 3b and 4j–p), although these samples share a common dominant provenance from the same region as discussed in the Results section. Population P4 is consistently the most abundant component in samples HH5 through HH1, mirroring the dominance of P4 in Arcuate Range source rock (Fig. 4j–p). Samples HH5, HH4, HH2, HH1 share population P3 with the Arcuate Range region, whereas the population P1, characteristic of the Arcuate Range region, occurs only in samples HH3 and HH1. Consistent heavy mineral assemblage along the Yellow River trunk (site a, c, d, e, l) further supports derivation from the Arcuate Range43. In addition, the heavy mineral, petrographic and clay mineral composition of Qingshui River tributary (QSR, site f, g, h) closely resemble those of the downstream trunk sample (site l), reinforcing a shared provenance signal (Supplementary Fig. 2)43,78.
Sediments provenance in the Wei River appears relatively uniform along the analyzed reach (Fig. 5a, b). Dominant populations P2 and P3 account for approximately 67% to 92% of total grain in samples WH1 through WH6 and closely match AFT age distributions observed in both bedrock and detrital apatite from the eastern West Qinling Mountains region (EWQN) (Fig. 2c), where most of ages range from 30 Ma to 79 Ma (Figs. 3b and 4x). This interpretation is supported by published detrital AFT dataset (e.g., Wang et al.60, Fig. 4q–x).
Several geological uncertainties and challenges must be considered when interpreting provenance variations along the river network: (i) Sediment source fingerprints delivered by tributaries can be biased by differences in tributary drainage area and by variations in apatite fertility of source rocks68,71,73,74. This effect is illustrated by the comparison between samples HH6 and HH7. Sample HH6 is enriched in population P5 largely because upstream dams intercepted sediment carrying populations P1 + P2 and P3, while P5 is predominantly sourced from the Huangshui River, which drains the larger Xining Basin (~ 32,954 Km2) compared to the Xunhua Basin (~4,990 Km2) near HH7. Even assuming similar lithologies and erosion rate, tributaries draining larger basins are expected to deliver higher mineral fluxes to the mainstem72,73,74. (ii) Active fluvial plains may dilute signals related to human impact on sediment transfer by redistributing recycled sediments, leading to similar detrital signatures across different source regions. For example, despite the significant distance (~502 Km) and the presence of dams between samples HH8 and HH5 (Figs. 4a, j and 2b), similar age populations occur in both samples, even though they originate from different source regions (the Guide Basin and Tibetan Plateau upstream of HH8 versus the Longzhong-Tianshui region for HH5; Fig. 3a, b). The active fluvial plain of the West Qinling Mountains transported source material to Miocene depositional centers such as the Tianshui and Longzhong regions79, where recycled sediments share similar AFT age components with those of the West Qinling Mountains60,80 (Fig. 3). Consequently, sample HH5, derived from the Tianshui and Longzhong regions, exhibits an AFT age distribution similar to that of the West Qinling Mountains, which also supply sediments to the Guide Basin56. Our results suggest that the Guide Basin contributed sediments to sample HH8, resulting in similar AFT age composition between HH5 and HH8, even though dams and reservoirs along the intervening river reaches have modified sediment transfer and source contributions. (iii) The uncertainties in the timing of sediment deposition relative to dam construction could exist. Our samples were collected from fluvial bars with a concave-up surface along the river section, which typically represent recent deposition rather than sediments deposited decades ago. Nevertheless, temporary storage of sediments in floodplains and channel bars over timescales of years to decades cannot be excluded. As a result, some grains may predate dam construction and have been subsequently reworked, potentially impairing the signal of anthropogenic impacts. Future work can be further explored by comparing the age composition between modern river sediments and river terrace sediments.
Conclusions
By contrasting the upper Yellow River with the largely undisturbed Wei River, this study underscores the dominant role of anthropogenic barriers in reshaping sediment provenance signals, demonstrating that human-induced disruptions can outweigh natural sediment mixing processes. Our work is focused on the upper Yellow River that shows pronounced downstream variability in detrital AFT age‑component proportions, thus defining two reaches. The upstream reach above the Lanzhou Basin exhibits substantially greater AFT heterogeneity than the downstream reach below Lanzhou. This contrast coincides with a much higher density of hydraulic infrastructure upstream (16 hydropower stations and three reservoirs versus five dams downstream), indicating that a severe regulation has disrupted sediment connectivity and altered sediment‑routing signals.
Sample HH8 (upper reach) preserves age-component assemblages characteristic of the upstream mainstem (Guide Basin) and the Tibetan Plateau. Downstream samples HH7 and HH6 record incremental incorporation of age contributions from the Xunhua and Xining basins, respectively. Under end‑member natural mixing, downstream increases in discrete age populations should scale with tributary drainage area (modulated by mineral fertility or apatite abundance); because relief, lithology, and climate are broadly similar across the catchments examined, we assume comparable apatite fertility and therefore first‑order scaling of sandy sediment supply with drainage area.
Given that tributary catchments here are orders of magnitude smaller than the Yellow River mainstem, the abrupt increase in population P4 between HH8 and HH7 is unlikely to reflect natural mixing alone. We infer that sediment blockages on the mainstem–trap and retention behind dams or reservoirs–attenuated contributions from populations P1 – P3 and relatively amplified locally derived P4 input from small tributaries in Xunhua.
An analogous process explains the transition from HH7 to HH6: the mainstem provenance signal is progressively overwritten by population P5, attributable to enhanced input from the Xining Basin via the Huangshui tributary. This provenance shift is magnified by sediment trapping in the Liujiaxia Reservoir and associated hydropower cascade, which elevates the proportional contribution of tributary-sourced detritus. Between HH6 and HH5, a further pronounced reorganization of age components occurs across a reach regulated by three intervening dams and a large reservoir; tributary mixing here is secondary due to limited contributing area.
Conversely, samples HH5 and HH4 exhibit invariant AFT age populations consistent with unimpeded sediment transfer where no major barriers intervene. Further downstream, HH3 diverges from HH2–HH1, a pattern attributable to combined input from the Arcuate Range region and upstream dam-induced interception of mainstem sediment. By contrast, the Wei River preserves largely intact sediment connectivity: its detrital AFT age populations remain consistent along the course where major impoundments are absent, providing a useful comparative baseline for unregulated sediment routing.
Overall, this study demonstrates that anthropogenic barriers significantly impair sediment connectivity along the Yellow River, producing downstream sediments increasingly dominated by locally derived tributary inputs. Detrital apatite fission‑track thermochronology emerges as a sensitive proxy for detecting these human‑driven disruptions in large fluvial systems, with important implications for quantifying anthropogenic modifications to sediment routing in tectonically active regions.
Methods
Sampling
Fourteen samples of modern river sand were collected from the fluvial bar ca. 10 cm below the surface along the trunk of the rivers (Fig. 1b, Supplementary Table 2). The collected grain size is variable, including the silt sand, fine sand, and medium to coarse sand. The samplings were collected from fluvial bars with a concave-up surface along the river section, which are widely recognized as representative of actively transported sediments (Supplementary Fig. 3). The sampling sites are at least 1.5 km away from the human constructions to avoid the uncertain effect of human activities such as the sand carry.
AFT dating
Fission-tracks are linear damage zones produced in apatite and in U-bearing minerals by the spontaneous nuclear fission decay of 238U. A fission-track age is determined by the U content and track density that are totally retained in mineral above the closure temperature assumed to be about 110 °C. For its low closure temperature, AFT method is routinely used to constraint tectonic processes and exhumation occurring in the uppermost kilometers of the earth crust81,82,83. For the good mechanical and chemical resistance and for the widespread occurrence in different rocks, detrital apatites are a good tools to link the detritus to the source rocks in provenance studies45,84,85. Eighty to one hundred apatite grains have been separated from a 2 – 3 kg of river sand and were dated by the external detector method at the University of Padua (Italy) using a Zeta calibration factor with the Fish Canyon and Durango age standards86,87. Ages were obtained after fission track counting (zeta=348.85 ± 14.67 for the CN5 dosimeter glass; analyst Xiaoqin Jiao) on an Olympus optical microscope (magnification ×1250). A chi (χ)-squared test was employed to evaluate the statistical homogeneity of the age distribution88. Ages with a chi-squared (χ2) probability of <5% were taken as indicative of a heterogenous population, which is present in all the analyzed samples (Supplementary Data 2).
Quantitative estimation of the age distributions
Kernel Density Estimate (KDE) is a standard statistical technique that involves summing a set of Gaussian distributions and solid theoretical foundations65. The Density plotter (KDE analysis)65 and BinomFit software were used to decompose the age distribution of our samples into best-fitting age populations66. The fitting of age populations in the BinomFit was calculated with the “automatic setting” (Supplementary Fig. 4, Supplementary Data 2), which fits well with the populations plotted with the Density plotter in this study. It is noteworthy that the modern sands of fluvial system provide a statistical sampling over a wide area, wherein peaks in AFT grain-age distributions of river sands represent the bedrock fingerprints in the source areas47,50. The population boundaries of AFT ages are defined through the age ranges of each population and the numbers of age populations plotted by the Density plotter and BinomFit. The non-matrix Multidimensional scaling (MDS) analysis (using the Kolmogorov-Smirnov statistical tool, Vermeesch65) is a robust method to evaluate similarities in the age distributions among samples through the dimensional reduction in exploratory data analysis. MDS method produces the plot with dots (samples) to visualize the similarities among these dots based on their distance67.
Data availability
Supplementary note, figures, and data that support the findings of this study are provided in the Supplementary Information and Supplementary Data 1–4 files. All datasets collected in this study are available at https://doi.org/10.5281/zenodo.19176938.
References
Maruyama, T. & Lin, A. Tectonic history of the Rokko active fault zone (southwest Japan) as inferred from cumulative offsets of stream channels and basement rocks. Tectonophysics 323, 197–216 (2000).
Schoenbohm, L. M., Whipple, K. X., Burchfiel, B. C. & Chen, L. Geomorphic constraints on surface uplift, exhumation, and plateau growth in the Red River region, Yunnan Province, China. Bull. Geol. Soc. Am. 116, 895–909 (2004).
Clark, M. K. et al. Surface uplift, tectonics, and erosion of eastern Tibet from large-scale drainage patterns. Tectonics 23, 1–21 (2004).
Van Den Berg, A.-P. H., Saparoea, V. & Postma, G. Control of Climate Change on the Yield of River Systems. in Recent Advances in Models of Siliciclastic Shallow-Marine Stratigraphy (eds. Hampson, G. J., Steel, R. J., Burgess, P. M. & Dalrymple, R. W.) 90 15–33 https://doi.org/10.2110/pec.08.90.0015 (2008).
Walling, D. E. Human impact on land-ocean sediment transfer by the world’s rivers. Geomorphology 79, 192–216 (2006).
Vercruysse, K. & Grabowski, R. C. Human impact on river planform within the context of multi-timescale river channel dynamics in a Himalayan river system. Geomorphology 381, 107659 (2021).
Downs, P. W. & Piégay, H. Catchment-scale cumulative impact of human activities on river channels in the late Anthropocene: implications, limitations, prospect. Geomorphology 338, 88–104 (2019).
Liu, X. et al. High-spatiotemporal-resolution mapping of global urban change from 1985 to 2015. Nat. Sustain. 3, 564–570 (2020).
Vliet, V. J., Eitelberg, D. A. & Verburg, P. H. A global analysis of land take in cropland areas and production displacement from urbanization. Glob. Environ. Chang. 43, 107–115 (2017).
Ceola, S., Laio, F. & Montanari, A. Satellite nighttime lights reveal increasing human exposure to floods worldwide. Geophys. Res. Lett. 41, 7184–7190 (2014).
Rogger, M. et al. Land use change impacts on floods at the catchment scale: Challenges and opportunities for future research. Water Resour. Res. 53, 5209–5219 (2017).
Tao, S. et al. Rapid loss of lakes on the Mongolian Plateau. Proc. Natl. Acad. Sci. USA. 112, 2281–2286 (2015).
Timpe, K. & Kaplan, D. The changing hydrology of a dammed Amazon. Sci. Adv. 3, 1–13 (2017).
Cooley, S. W., Ryan, J. C. & Smith, L. C. Human alteration of global surface water storage variability. Nature 591, 78–81 (2021).
Spinti, R. A., Condon, L. E. & Zhang, J. The evolution of dam induced river fragmentation in the United States. Nat. Commun. 14, 3820 (2023).
Grill, G. et al. Mapping the world’s free-flowing rivers. Nature 569, 215–221 (2019).
Chen, Q. et al. Does a hydropower reservoir cascade really harm downstream nutrient regimes. Sci. Bull. 69, 661–670 (2024).
Winemiller, K. O. et al. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science 351, 128–129 (2016).
Latrubesse, E. M. et al. Damming the rivers of the Amazon basin. Nature 546, 363–369 (2017).
Perera, D., Williams, S. & Smakhtin, V. Present and Future Losses of Storage in Large Reservoirs Due to Sedimentation: A Country-Wise Global Assessment. Sustain 15, 219 (2023).
Ward, J. V. The four-dimensional nature of lotic ecosystems. J. North Am. Benthol. Soc. 8, 2–8 (1989).
Fryirs, K. Dis)Connectivity in catchment sediment cascades: A fresh look at the sediment delivery problem. Earth Surf. Process. Landforms 38, 30–46 (2013).
Brunke, M. & Gonser, T. The ecological significance of exchange processes between rivers and groundwater. Freshw. Biol. 37, 1–33 (1997).
Panagiotou, A., Zogaris, S., Dimitriou, E., Mentzafou, A. & Tsihrintzis, V. A. Anthropogenic barriers to longitudinal river connectivity in Greece: A review. Ecohydrol. Hydrobiol. 22, 295–309 (2022).
Belletti, B. et al. More than one million barriers fragment Europe’s rivers. Nature 588, 436–441 (2020).
Graf, W. L. Downstream hydrologic and geomorphic effects of large dams on American rivers. Geomorphology 79, 336–360 (2006).
Guo, Y. et al. Sediment Routing and Anthropogenic Impact in the Huanghe River Catchment, China: An Investigation Using Nd Isotopes of River Sediments. Water Resour. Res. 57, 1–14 (2021).
Schmitt, R. J. P., Bizzi, S., Castelletti, A., Opperman, J. J. & Kondolf, G. M. Planning dam portfolios for low sediment trapping shows limits for sustainable hydropower in the Mekong. Sci. Adv. 5, 1–12 (2019).
Carvajal-Quintero, J. D. et al. Damming Fragments Species’ Ranges and Heightens Extinction Risk. Conserv. Lett. 10, 708–716 (2017).
Zhang, Z. et al. Sedimentary provenance perspectives on the evolution of the major rivers draining the eastern Tibetan Plateau. Earth-Science Rev 232, 104151 (2022).
Wang, H. et al. Stepwise decreases of the Huanghe (Yellow River) sediment load (1950-2005): Impacts of climate change and human activities. Glob. Planet. Change 57, 331–354 (2007).
Wang, S. et al. Reduced sediment transport in the Yellow River due to anthropogenic changes. Nat. Geosci. 9, 38–41 (2016).
Wohlfart, C., Kuenzer, C., Chen, C. & Liu, G. Social–ecological challenges in the Yellow River basin (China): a review. Environ. Earth Sci. 75, 1–20 (2016).
Milliman, J., Yun-Shan, Q., Mei-E, R. & Saito, Y. Man’s influence on the erosion and transport of sediment by Asian Rivers: The Yellow River (Huanghe) Example. J. Phys. Chem. 99, 3411–3415 (1995).
Song, C. et al. A comprehensive geospatial database of nearly 100,000 reservoirs in China. Zenodo https://doi.org/10.1016/j.chemgeo.2018.04.010 (2022).
Fan, C. et al. Exacerbating dam-induced fragmentation in China’s river systems. Commun. Earth Environ. 6, 1–13 (2025).
O’Sullivan, G. O., Chew, D., Kenny, G., Henrichs, I. & Mulligan, D. The trace element composition of apatite and its application to detrital provenance studies. Earth-Science Rev 201, 103044 (2020).
Ketcham, R. A., Donelick, R. A. & Carlson, W. D. Variability of apatite fission-track annealing kinetics: II. Crystallographic orientation effects. Am. Mineral. 84, 1224–1234 (1999).
Yang, J. et al. Episodic crustal growth of North China as revealed by U – Pb age and Hf isotopes of detrital zircons from modern rivers. Geochim. Cosmochim. Acta 73, 2660–2673 (2009).
Kong, P., Jia, J. & Zheng, Y. Time constraints for the Yellow River traversing the Sanmen Gorge. Geochem. Geophys. Geosyst. 15, 395–407 (2014).
Nie, J. et al. Loess plateau storage of northeastern Tibetan plateau-derived Yellow River sediment. Nat. Commun. 6, 8511 (2015).
Lin, X. et al. Tracing the source areas of detrital zircon and K-feldspar in the Yellow River Basin. Earth-Science Rev 251, 104718 (2024).
Pan, B. et al. Heavy-mineral analysis and provenance of Yellow River sediments around the China Loess Plateau. J. Asian Earth Sci. 127, 1–11 (2016).
Hurford, A. J., Fitch, F. J. & Clarke, A. Resolution of the age structure of the detrital zircon populations of two Lower Cretaceous sandstones from the Weald of England by fission track dating. Geol. Mag. 121, 269–277 (1984).
Carter, A. Present status and future avenues of source region discrimination and characterization using fission track analysis. Sediment. Geol. 124, 31–45 (1999).
Clift, P. D. et al. Thermochronology of mineral grains in the Red and Mekong Rivers, Vietnam: Provenance and exhumation implications for Southeast Asia. Geochemistry, Geophys. Geosystems 7, Q10005 (2006).
Bernet, M., Brandon, M. T., Garver, J. I. & Molitor, B. R. Fundamentals of detrital zircon fission-track analysis for provenance and exhumation studies with examples from the European Alps. Spec. Pap. Geol. Soc. Am. 378, 25–36 (2004).
Hu, Z. B. et al. The linking of the upper-middle and lower reaches of the Yellow River as a result of fluvial entrenchment. Quat. Sci. Rev. 166, 324–338 (2017).
Stevens, T. et al. Genetic linkage between the Yellow River, the Mu Us desert and the Chinese Loess Plateau. Quat. Sci. Rev. 78, 355–368 (2013).
Malusà, M. G., Zattin, M., Andò, S., Garzanti, E. & Vezzoli, G. Focused erosion in the Alps constrained by fission-track ages on detrital apatites. Geol. Soc. Spec. Publ. 324, 141–152 (2009).
Lin, X. et al. The uplift history of the Haiyuan-Liupan Shan region northeast of the present Tibetan plateau: Integrated constraint from Stratigraphy and thermochronology. J. Geol. 119, 372–393 (2011).
Peng, H. et al. Thermochronological constraints on the Meso-Cenozoic tectonic evolution of the Haiyuan-Liupanshan region, northeastern Tibetan Plateau. J. Asian Earth Sci. 183, 103966 (2019).
Zheng, D. et al. Rapid exhumation at ~ 8 Ma on the Liupan Shan thrust fault from apatite fission-track thermochronology: Implications for growth of the northeastern Tibetan Plateau margin. Earth Planet. Sci. Lett. 248, 198–208 (2006).
Jiao, X. et al. Sourcing of the Oligocene to Pliocene sediments of the Ningnan Basin: Evidence for Tibetan Plateau growth and local faulting unravelled by detrital apatite fission-track and U–Pb double dating. Basin Res 36, 1–23 (2024).
Lease, R. O. et al. Middle Miocene reorganization of deformation along the northeastern Tibetan Plateau. Geology 39, 359–362 (2011).
Wang, X. et al. Cenozoic pulsed deformation history of northeastern Tibetan Plateau reconstructed from fission-track thermochronology. Tectonophysics 672–673, 212–227 (2016).
Wang, Y., Zhang, J., Qi, W. & Guo, S. Exhumation history of the Xining Basin since the Mesozoic and its tectonic significance. Acta Geol. Sin. English Ed. 89, 145–162 (2015).
Li, B. et al. Cenozoic cooling history of the North Qilian Shan, northern Tibetan Plateau, and the initiation of the Haiyuan fault: Constraints from apatite- and zircon-fission track thermochronology. Tectonophysics 751, 109–124 (2019).
Zhang, J., Wang, Y., Zhang, B. & Zhao, H. Evolution of the NE Qinghai-Tibetan Plateau, constrained by the apatite fission track ages of the mountain ranges around the Xining Basin in NW China. J. Asian Earth Sci. 97, 10–23 (2015).
Wang, X. et al. Eocene to Pliocene exhumation history of the Tianshui-Huicheng region determined by Apatite fission track thermochronology: Implications for evolution of the northeastern Tibetan Plateau margin. J. Asian Earth Sci. 42, 97–110 (2011).
Wang, X. et al. Cenozoic morphotectonic evolution of the northeasternmost Tibetan Plateau: Evidence from detrital thermochronology. Glob. Planet. Change 225, 104131 (2023).
Peng, H. et al. Long-term and multiple stage exhumation of the Ordos Basin, western North China Craton: Insights from seismic reflection, borehole and geochronological data. Earth-Science Rev 238, 104349 (2023).
Zattin, M. & Wang, X. Exhumation of the western Qinling mountain range and the building of the northeastern margin of the Tibetan Plateau. J. Asian Earth Sci. 177, 307–313 (2019).
Duvall, A. R. et al. Low-temperature thermochronometry along the Kunlun and Haiyuan Faults, NE Tibetan Plateau: Evidence for kinematic change during late-stage orogenesis. Tectonics 32, 1190–1211 (2013).
Vermeesch, P. On the visualisation of detrital age distributions. Chem. Geol. 312–313, 190–194 (2012).
Brandon, M. T. Probability Density Plot for Fission-Track. Radiat. Meas. 26, 663–676 (1996).
Vermeesch, P. Multi-sample comparison of detrital age distributions. Chem. Geol. 341, 140–146 (2013).
Thomson, K. D., Stockli, D. F. & Fildani, A. Anthropogenic impact on sediment transfer in the upper Missouri River catchment detected by detrital zircon analysis. Bull. Geol. Soc. Am. 134, 2485–2502 (2022).
Brandt, S. A. Classification of geomorphological effects downstream of dams. Catena 40, 375–401 (2000).
Grant, G. E., Schmidt, J. C. & Lewis, S. L. A geological framework for interpreting downstream effects of dams on rivers. in Water Science and Application (eds O’Conno, J. E., Grant, G. E.) 7 209–225 https://doi.org/10.1029/007WS13 (2013).
Glotzbach, C., Van Der Beek, P., Carcaillet, J. & Delunel, R. Deciphering the driving forces of erosion rates on millennial to million-year timescales in glacially impacted landscapes: An example from the Western Alps. J. Geophys. Res. Earth Surf. 118, 1491–1515 (2013).
Amidon, W. H., Burbank, W. & Gehrels, G. E. Construction of detrital mineral populations: insights from mixing of U-Pb zircon ages in Himalayan rivers. Basin Res. 17, 463–485 (2005).
Saylor, J. E., Knowles, J. N., Horton, B. K., Nie, J. & Mora, A. Mixing of source populations recorded in detrital zircon U-Pb age spectra of modern river sands. J. Geol. 121, 17–33 (2013).
Malusà, M. G., Resentini, A. & Garzanti, E. Hydraulic sorting and mineral fertility bias in detrital geochronology. Gondwana Res. 31, 1–19 (2016).
Chen, H. et al. Apatite fission-track thermochronological constraints on the pattern of late Mesozoic-Cenozoic uplift and exhumation of the Qinling Orogen, central China. J. Asian Earth Sci. 114, 649–673 (2015).
Zhang, Y. et al. Testing the cenozoic lower crustal flow beneath the Qinling Orogen, northeastern Tibetan Plateau. J. Struct. Geol. 165, 104747 (2022).
Lin, X., Jolivet, M. & Cheng, F. Spatiotemporal Evolution of Central Qilian Shan (Northwest China) Constrained by Fission-Track Ages of Detrital Grains from the Huangshui River. Minerals 13, 890 (2023).
Pang, H. et al. Mineralogy and geochemistry of modern Yellow River sediments: Implications for weathering and provenance. Chem. Geol. 488, 76–86 (2018).
Lin, A., Yang, Z., Sun, Z. & Yang, T. How and when did the Yellow River develop its square bend? Geology 29, 951–954 (2001).
Wang, X. et al. Cenozoic tectonic uplift history of Western Qinling: Evidence from sedimentary and fission-track data. J. Earth Sci. 24, 491–505 (2013).
Reiners, P. W. & Brandon, M. T. Using thermochronology to understand orogenic erosion. Annu. Rev. Earth Planet. Sci. 34, 419–466 (2006).
Carter, A. & Moss, S. J. Combined detrital-zircon fission-track and U-Pb dating: A new approach to understanding hinterland evolution. Geology 27, 235–238 (1999).
Zuza, A. V. et al. Tectonic evolution of the Qilian Shan: An early Paleozoic orogen reactivated in the Cenozoic. Bull. Geol. Soc. Am. 130, 881–925 (2018).
Olivetti, V. et al. Variations and provenance trends in the Oligocene-early Miocene glaciomarine sediments of the Central Ross Sea, Antarctica (DSDP Site 270). Glob. Planet. Change 221, 104042 (2023).
Zattin, M., Talarico, F. M. & Sandroni, S. Integrated provenance and detrital thermochronology studies on the ANDRILL AND-2A drill core: Late Oligocene-Early Miocene exhumation of the Transantarctic Mountains (southern Victoria Land, Antarctica). Terra Nov 22, 361–368 (2010).
Gleadow, A. J. W. & Duddy, I. R. A natural long-term track annealing experiment for apatite. Nucl. Tracks 5, 169–174 (1981).
Hurford, A. J. Standardization of fission track dating calibration: Recommendation by the Fission Track Working Group of the I.U.G.S. Subcommission on Geochronology. Chem. Geol. Isot. Geosci. Sect. 80, 171–178 (1990).
Galbraith, R. F. On statistical estimation in fission track dating. J. Int. Assoc. Math. Geol. 16, 653–669 (1984).
Liang, H. et al. Sedimentary basin evolution and its implications for outward expansion of the northeastern Tibetan Plateau: Insights from the Tongxin Basin, China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 575, 110460 (2021).
Guo, C. et al. Apatite Geo-Thermochronology and Geochemistry Constrain Oligocene-Miocene Growth and Geodynamics of the Northeastern Tibetan Plateau. Geophys. Res. Lett. 52, e2024GL113157 (2025).
Acknowledgements
This work was funded by the National Natural Science Foundation of China (grant no. 42230815), the Open Fund of the State Key Laboratory of Continental Evolution and Early Life (grant no. 24CD01), and the Italian Ministry of University and Research (MUR) through the project titled ‘The Geosciences for Sustainable Development’ (Budget MUR – Dipartimenti di Eccellenza 2023 to 2027; Project ID C93C23002690001), XJ acknowledges the China Scholarship Council (grant no. 202006440023). We thank two anonymous reviewers and the editor, Dr. Maria Laura Balestrieri, for their constructive comments and helpful suggestions that improved this work. We appreciate Dr. Silvia Cattò for her continuous support in the preparation of the samples and the guidance in the thermochronology lab. Thanks Feng Xiaolin and Zhou Yang for their contribution in the field work.
Author information
Authors and Affiliations
Contributions
Xiaoqin Jiao: Writing – original draft, Writing – review & editing, Methodology, Investigation, Formal analysis, Data curation, Conceptualization; Valerio Olivetti: Writing – original draft, Writing – review & editing, Methodology, Formal analysis, Conceptualization, Supervision; Jianqiang Wang: Writing – review & editing, Methodology, Investigation, Supervision, Funding acquisition; Heng Peng: Writing – review & editing, Investigation; Massimiliano Zattin: Writing – review & editing, Methodology, Formal analysis, Conceptualization, Supervision, Funding acquisition.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Earth & Environment thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editors: Maria Laura Balestrieri and Joe Aslin. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Jiao, X., Olivetti, V., Wang, J. et al. Detrital fission-tack analysis determines the signal of anthropogenic infrastructure in upper Yellow River sediment transfer. Commun Earth Environ 7, 380 (2026). https://doi.org/10.1038/s43247-026-03540-w
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s43247-026-03540-w
