Quantifying provenance of reservoir sediment using multiple composite fingerprints in an arid region experiencing both wind and water erosion

Authors: NIU, BAICHENG - Chinese Academy Of Sciences; QU, JIANJUN - Chinese Academy Of Sciences; Zhang, Xunchang; LIU, BENLI - Chinese Academy Of Sciences; TAN, LIHAI - Chinese Academy Of Sciences; AN, ZHISHAN - Chinese Academy Of Sciences

Submitted to: Geomorphology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/10/2019
Publication Date: 2/13/2019
Citation: Niu, B., Qu, J., Zhang, X.J., Liu, B., Tan, L., An, Z. 2019. Quantifying provenance of reservoir sediment using multiple composite fingerprints in an arid region experiencing both wind and water erosion. Geomorphology. 332:112-121. https://doi.org/10.1016/j.geomorph.2019.02.011.
DOI: https://doi.org/10.1016/j.geomorph.2019.02.011

Interpretive Summary: Quantitative information of sediment sources is useful for developing good protection measures that effectively control soil erosion in a watershed. In this study, a multiple composite fingerprints method was tested and used for quantifying the sediment provenance in the Danghe river watershed and for further providing effective sediment control strategies for the watershed in an arid region experiencing composite wind and water erosion. To accommodate the wide ranges of sediment size distributions resulted from the combined water and wind erosion, sediment contributions from each source were estimated for each of six particle size groups. Surface samples were taken from three geomorphic source areas of dunes, gobi, and upstream mountains, and sediment samples from the watercourses near the reservoir. Samples were analyzed for 30 elements using a dispersive X-ray fluorescence spectrometer. The elements were statistically screened, tested for conservativeness, and range checked. All non-conflict composite fingerprints were used to analytically solve linear mixing models for the three sources. The results showed that the proportional contributions from each source to reservoir sediment were well estimated with an approach of the multiple composite fingerprints, showing 49.88% from upstream mountains, 28.42% from gobi, and 21.70% from dunes with the relative contributions in each size group followed a similar trend. The largest fraction of the sediment was from the particle size group of 0.063-0.1 mm, accounting for about 37.09%, for which 5.10, 10.17, and 21.82% were from the dunes, gobi, and upstream mountains. . Results showed that selective erosion existed in both erosions, especially in wind erosion, and suggested multiple particle size tracking be used in cases of severe selective erosion to improve the accuracy of estimation. Also, the contribution rate per unit area was the largest from dunes, and thus controlling wind-blown sand directly to the river by stabilizing the dunes was recommended to reduce the reservoir siltation. This work would be helpful to local conservation specialists to develop conservation plans and reduce sediment load to the reservoir.

Technical Abstract: Quantitative information of sediment sources is useful for developing good protection measures that effectively control soil erosion in a watershed. In this study, a multiple composite fingerprints method was tested and used for quantifying the sediment provenance in the Danghe river watershed and for further providing effective sediment control strategies for the watershed in an arid region experiencing composite wind and water erosion. To accommodate the wide ranges of sediment size distributions resulted from the combined water and wind erosion, sediment contributions from each source were estimated for each of six particle size groups. Surface samples were taken from three geomorphic source areas of dunes, gobi, and upstream mountains, and sediment samples from the watercourses near the reservoir. Samples were analyzed for 30 elements using a dispersive X-ray fluorescence spectrometer. The elements were statistically screened, tested for conservativeness, and range checked. All non-conflict composite fingerprints were used to analytically solve linear mixing models for the three sources. The results showed that the proportional contributions from each source to reservoir sediment were well estimated with an approach of the multiple composite fingerprints, showing 49.88% from upstream mountains, 28.42% from gobi, and 21.70% from dunes with the relative contributions in each size group followed a similar trend. The largest fraction of the sediment was from the particle size group of 0.063-0.1 mm, accounting for about 37.09%, for which 5.10, 10.17, and 21.82% were from the dunes, gobi, and upstream mountains. . Results showed that selective erosion existed in both erosions, especially in wind erosion, and suggested multiple particle size tracking be used in cases of severe selective erosion to improve the accuracy of estimation. Also, the contribution rate per unit area was the largest from dunes, and thus controlling wind-blown sand directly to the river by stabilizing the dunes was recommended to reduce the reservoir siltation.