1.College of Geographical Sciences,Qinghai Normal University,Xining 810008,China
2.Qinghai Provincial Key Laboratory of Physical Geography and Environmental Process,Xining 810008,China
3.Key Laboratory of Water Cycle and Related Land Surface Processes,Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences,Beijing 100101,China
4.School of National Safety and Emergency Management,Qinghai Normal University,Xining 810008,China
5.State Key Laboratory of Ecological Safety and Sustainable Development in Arid Lands,Dunhuang Gobi and Desert Ecological and Environmental Research Station,Northwest Institute of Eco-Environment and Resources,Chinese Academy of Sciences,Lanzhou 730000,China
6.Key Laboratory of Mountain Hazards and Earth Surface Processes,Institute of Mountain Hazards and Environment,Chinese Academy of Sciences and Ministry of Water Resources,Chengdu 610041,China
Objective To accurately trace sediment sources of water conservancy facilities such as rivers and reservoirs in alpine regions under combined wind-water erosion environments by using multiple composite fingerprint methods. Methods In the Shagou River Basin, a tributary of the Longyangxia Reservoir on the Yellow River, soil samples were collected from aeolian sand and fluvial sediment source areas, along with fresh sediment samples at the basin outlet. Forty elemental fingerprint factors were analyzed using X-ray fluorescence spectroscopy. Three fingerprint methods were used to analyze sediment sources, including the multi-group fingerprint factor method, the machine learning optimal composite fingerprint method, and the Walling⋅C optimal composite fingerprint method. Results For fingerprint factor screening, the CT-KW-DFA method achieved a cumulative discrimination rate of 82.40% using discriminant function analysis (DFA), while the CT-RF-DFA method reached 100%, demonstrating a 17.60% improvement in discrimination capacity over the CT-KW-DFA method. The CT-RF-DFA method better distinguished sediment source regions. The multi-group fingerprint factor method indicated that aeolian sediment contributed 53.40% while fluvial sediment contributed 46.60%. The machine learning optimal composite fingerprint method revealed that aeolian sediment contributed 63.00%, and fluvial sediment contributed 37.00%. The Walling⋅C optimal composite fingerprint method revealed that aeolian sediment contributed 50.11% and fluvial sediment contributed 49.89%. The average contribution rates across the three methods were 55.50% for aeolian sediment and 44.50% for fluvial sediment. The sediment sources revealed by the multi-group fingerprint factor method were closest to the average of the three methods. In the machine learning optimal composite fingerprint method, the Bayesian model demonstrated good convergence and excellent fitting performance. In the Walling⋅C optimal composite fingerprint method, the goodness-of-fit of the Walling⋅C multivariate mixing model was 94.50%. Conclusion The computational processes of all three composite fingerprint methods perform well in tracing sediment sources in alpine river regions. All three methods indicate that in the Shagou River Basin, aeolian processes contribute a higher proportion of sediment than fluvial processes. The combined effects of seasonal aeolian activities and changes in river ice conditions are the dominant factors controlling sediment transport. This study is important for revealing sediment sources under combined wind-water erosion in alpine regions, and provides technical support for the erosion prevention and control of water conservancy facilities such as rivers and reservoirs in alpine regions.
SUNX H, TIANL Q, FANGH W, et al. Changes in global fluvial sediment concentrations and fluxes between 1985 and 2020[J].Nature Sustainability,2025,8:142-151.
[2]
YANGH, LIUD, THOMPSONJ R, et al. Combating sediment surge in Tibetan rivers[J].Science,2024,384(6696):e630.
YUANJ, XUQ X, DONGB J, et al. Sediment deposition of Three Gorges Reservoir and its impact on the reservoir area in recent 20 years[J].Journal of Lake Sciences,2023,35(2):632-641.
[5]
GHADERI DEHKORDIN, KHALEDI DARVISHANA, ZAREM R, et al. Temporal changes in the average contribution of land uses in sediment yield using the 137Cs method and geochemical tracers[J].Water,2025,17(1):e73.
[6]
LIUG H, SCHMALZB, ZHANGQ, et al. Assessing effects of land use and land cover changes on hydrological processes and sediment yield in the Xunwu River watershed, Jiangxi Province, China[J].Frontiers of Earth Science,2022,16:819-833.
[7]
ANACHEJ A A, WENDLANDE C, OLIVEIRAP T S, et al. Runoff and soil erosion plot-scale studies under natural rainfall: A meta-analysis of the Brazilian experience[J].Catena,2017,152:29-39.
[8]
EVRARDO, CHABOCHEP A, RAMONR, et al. A global review of sediment source fingerprinting research incorporating fallout radiocesium(137Cs)[J].Geomorphology,2020,362:e107103.
[9]
MINELLAJ P G, MERTENG H, SCHLESNERA, et al. Combining sediment source tracing techniques with traditional monitoring: The "Arvorezinha catchment" experience[J].Hydrological Processes,2022,36(9):e14665.
[10]
WALLINGD E, PEARTM R, OLDFIELDF, et al. Suspended sediment sources identified by magnetic measurements[J].Nature,1979,281:110-113.
[11]
COLLINSA L, WALLINGD E, LEEKSG J L. Source type ascription for fluvial suspended sediment based on a quantitative composite fingerprinting technique[J].Catena,1997,29(1):1-27.
[12]
COLLINSA L, BLACKWELLM, BOECKXP, et al. Sediment source fingerprinting: Benchmarking recent outputs, remaining challenges and emerging themes[J].Journal of Soils and Sediments,2020,20(12):4160-4193.
ZHOUH P, ZHOUH W, CHENY D. Sediment source fingerprinting: Progress in uncertainty analysis[J].Acta Pedologica Sinica,2019,56(6):1279-1289.
[17]
BALTIA, HAMDIA, ABIDS, et al. Enhanced fingerprint classification through modified PCA with SVD and invariant moments[J].Frontiers in Artificial Intelligence,2024,7:e1433494.
[18]
MALHOTRAK, LAMBAJ, SRIVASTAVAP, et al. Fingerprinting suspended sediment sources in an urbanized watershed[J].Water,2018,10(11):e1573.
[19]
GOVANE, JCKSONA L, INGERR, et al. Simmr:A package for fitting Stable Isotope Mixing Models in R[J].Journal of Statistical Software,2023,5(2):1-27.
[20]
PALAZÓNL, NAVASA. Variability in source sediment contributions by applying different statistic test for a Pyrenean catchment[J].Journal of Environmental Management,2017,194:42-53.
[21]
PALAZÓNL, LATORREB, GASPARL, et al. Comparing catchment sediment fingerprinting procedures using an auto-evaluation approach with virtual sample mixtures[J].Science of the Total Environment,2015,532:456-466.
NOSRATIK, FATHIZ, COLLINSA L. Fingerprinting sub-basin spatial suspended sediment sources by combining geochemical tracers and weathering indices[J].Environmental Science and Pollution Research International,2019,26(27):28401-28414.
[24]
CHENH H, YANGJ, HAOZ J, et al. Research on indoor positioning method based on LoRa-improved fingerprint localization algorithm[J].Scientific Reports,2023,13(1):e13981.
[25]
MESHRAMS G, MESHRAMC, POURHOSSEINIF A, et al. A multi-layer perceptron(MLP)-fire fly algorithm(FFA)-based model for sediment prediction[J].Soft Computing,2022,26:911-920.
[26]
DERAKHSHAN-BABAEIF, NOSRATIK, FIENERP, et al. Source fingerprinting sediment loss from sub-catchments and topographic zones using geochemical tracers and weathering indices[J].Journal of Hydrology,2024,633:e131019.
[27]
MALHOTRAK, ZHENGJ Y, ABEBEA, et al. Application of sediment fingerprinting to apportion sediment sources: Using machine learning models[J].Journal of the Asabe,2023,66(5):1205-1221.
ZHANGJ Q, YINM F, LIUY, et al. Influence of fingerprinting properties and estimation models on sediment source discrimination via composite fingerprinting approach[J].Science of Soil and Water Conservation,2022,20(6):143-150.
CHENY, WEIX P, ZHANGA G, et al. Analysis of sediment source in small watershed of karst depression based on composite fingerprint method[J].Journal of Soil and Water Conservation,2020,34(5):131-144.
ZHOUM, LINJ H, HUANGY H, et al. Using composite fingerprints to qualify sediment source in watershed with intensive exploitation on red soil region[J].Journal of Soil and Water Conservation,2019,33(1):20-24.
DUP F, HUANGD H, QINW, et al. Sediment sources in a small watershed located in the black soil region of northeast China based on different models and various fingerprints[J].Journal of Soil and Water Conservation,2020,34(1):84-91.
TONGS C, LIG R, LIJ F, et al. Soil basic characteristics and wind erosion in degraded barren patches of alpine meadow in the source of Yellow River[J].Research of Soil and Water Conservation,2023,30(4):10-17.
[40]
YANGZ Z, NIW K, NIUF J, et al. Spatiotemporal distribution characteristics and influencing factors of freeze-thaw erosion in the Qinghai-Tibet Plateau[J].Remote Sensing,2024,16(9):e1629.
[41]
MEINSHAUSESNN, BUHLMANNP.Stability Selection[J].Journal of theroyal statistical society series B:Statistical methodology,2010,72(4):417-473.
[42]
YANGY, ZHANGJ, WANGX C, et al. Deep structure and geothermal resource effects of the Gonghe basin revealed by 3D magnetotelluric[J].Geothermal Energy,2024,12(1):e6.
DONGM, YANP, WANGX X, et al. Sediments characteristics and environment significance of climbing dunes in the Duoshizai, Gonghe basin[J].Journal of Arid Land Resources and Environment,2022,36(5):79-86.
[45]
ZHUB C, CAOL X, YANGS, et al. Spatiotemporal variations in fingerprinting sediment sources in a watershed disturbed by construction[J].Land,2024,13(8):e1314.
[46]
BREIMANL. Random forests[J].Machine Learning,2001,45(1):5-32.
ZHAOT Y, WANGZ B, WUY Y, et al. A review of studies on sediment sources of small catchments using composite fingerprinting[J].Research of Soil and Water Conservation,2020,27(2):377-382.
NIUB C, WUZ l, CHENQ, et al. Uncertainty analysis of the sediment sources resulting from multiple composite fingerprints method[J].Science of Soil and Water Conservation,2025,23(3):23-32.
[51]
NIUB, LIUB, ZHANGX J, et al. Evaluation of five different sediment fingerprinting approaches for estimating sediment source contributions in an arid region[J].Geoderma,2022,427:e116131.