摘要
Accurate identification of sediment sources and erosion hotspots is crucial for implementing targeted soil and water conservation measures.This study analyzed the temporal dynamics of sediment source contributions during seven rainfall events(A_(1)–A_(7))that generated both runoff and suspended sediment at the outlet of the Hantaichuan watershed,located in the wind–water erosion crisscross region of the northern Loess Plateau,China.A total of 82 suspended sediment samples were collected at the Xiangshawan hydrological station during 2021 and 2022.Additionally,145 soil samples were collected from three primary potential sources:sandy soil,loess soil,and Pisha sandstone.Mid-infrared(MIR)spectroscopy and geochemical fingerprinting were used to trace and quantify sediment sources.For MIRbased source apportionment,twelve predictive models were developed using partial least squares regression(PLSR)and support vector machine regression(SVMR)in combination with six spectral preprocessing techniques.The PLSR model with first-derivative Savitzky–Golay(SGD1)preprocessing achieved the best performance.Source apportionment results indicated that Pisha sandstone was the dominant sediment source(44.00%–72.23%),followed by sandy soil(14.23%–37.00%)and loess soil(10.01%–24.00%).Among the analyzed rainfall events,the contribution of Pisha sandstone was lowest in the small flow magnitude event A3(57.53%)and highest in the large flow magnitude event A6(63.48%),suggesting that sediment source composition was strongly controlled by rainfall event magnitude.Both MIR spectroscopic and geochemical fingerprinting methods showed high consistency,confirming MIR spectroscopy as a reliable,non-destructive,and cost-effective alternative for rapid sediment source apportionment and for supporting erosion control prioritization in highly erodible catchments.
基金
supported by the National Key Scientific Research Project(Grant No.2023YFC3209303)
jointly funded by the National Natural Sciences Foundation of China(Grant Nos.42577406 and 42077076)
the National Natural Science Foundation of China and Yellow River Water Science Research Joint Fund(Grant No.U2243211).
