Skip to main content
ARS Home » Plains Area » El Reno, Oklahoma » Grazinglands Research Laboratory » Agroclimate and Natural Resources Research » Research » Publications at this Location » Publication #369194

Research Project: Uncertainty of Future Water Availability Due to Climate Change and Impacts on the Long Term Sustainability and Resilience of Agricultural Lands in the Southern Great Plains

Location: Agroclimate and Natural Resources Research

Title: Evaluation of five different sediment fingerprinting approaches for estimating sediment source contributions in an arid region

Author
item NIU, BAICHENG - Chinese Academy Of Sciences
item Zhang, Xunchang
item LIU, BENLI - Chinese Academy Of Sciences
item QU, JIANJUN - Chinese Academy Of Sciences
item LIU, BING - China Institute Of Water Resources

Submitted to: Geoderma
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/23/2022
Publication Date: 9/1/2022
Citation: Niu, B., Zhang, X.J., Liu, B., Qu, J., Liu, B. 2022. Evaluation of five different sediment fingerprinting approaches for estimating sediment source contributions in an arid region. Geoderma. 427:116131. https://doi.org/10.1016/j.geoderma.2022.116131.
DOI: https://doi.org/10.1016/j.geoderma.2022.116131

Interpretive Summary: Fingerprinting methods are widely used to quantify sediment provenance at a watershed scale. However, different fingerprinting methods often yield different results in the same watershed. Thus, the selection of an appropriate fingerprinting approach is of great importance to obtain reliable estimation. The objectives of this study are to (1) evaluate the performance of five fingerprinting approaches, (2) compare the efficiency of analytical and various numerical solutions, and (3) investigate the effects of the number of composite fingerprints as well as the tracer number in a single composite fingerprint on the estimation accuracy. Source samples were collected from three geomorphic areas of dune, gobi, and mountains in an arid watershed, and sediment samples near a reservoir. Results showed either of the four approaches could provide satisfactory estimation of source contributions except the distance weighting method. However, the sum of absolute relative errors indicated that the multiple composite fingerprints (MCF) method using algebraic solutions, which is the simplest in computation, provided the best estimation, making it useful if a large number of composite fingerprints (CFs) is available. A single (optimal) composite fingerprint (SCF) or a small number of CFs should be solved using Monte Carlo (MC) simulation to increase sample size and representativeness of the selected tracers in order to obtain reliable estimation. For a moderate number of CFs, optimization using tracer mean concentration, which can generate similar estimation accuracy to MC simulation but is much less computationally intensive, can be used in lieu of MC simulation. Overall results suggested the MCF approach were superior to the SCF approach. Generally, increasing the number of CFs tended to improve the estimated accuracy. These conclusions should be tested further under different geographical and erosional conditions. The findings will be useful to soil erosion scientists and soil conservationists to identify the provenance of sediment sources.

Technical Abstract: Fingerprinting methods are widely used to quantify sediment provenance at a watershed scale. However, different fingerprinting methods often yield different results in the same watershed. Thus, the selection of an appropriate fingerprinting approach is of great importance to obtain reliable estimation. The objectives of this study are to (1) evaluate the performance of five fingerprinting approaches, (2) compare the efficiency of analytical and various numerical solutions, and (3) investigate the effects of the number of composite fingerprints as well as the tracer number in a single composite fingerprint on the estimation accuracy. Source samples were collected from three geomorphic areas of dune, gobi, and mountains in an arid watershed, and sediment samples near a reservoir. Results showed either of the four approaches could provide satisfactory estimation of source contributions except the distance weighting method. However, the sum of absolute relative errors indicated that the multiple composite fingerprints (MCF) method using analytical solutions, which is the simplest in computation, provided the best estimation, advocating its use if a large number of composite fingerprints (CFs) is available. A single (optimal) composite fingerprint (SCF) or a small number of CFs should be solved using Monte Carlo (MC) simulation to increase sample size and representativeness of the selected tracers in order to obtain reliable estimation. For a moderate number of CFs, optimization using tracer mean concentration, which can generate similar estimation accuracy to MC simulation but is much less computationally intensive, can be used in lieu of MC simulation. Overall results suggested the MCF approach were superior to the SCF approach. Generally, increasing the number of CFs tended to improve the estimated accuracy. There was no clear relationship between the number of tracers in a CF and the estimation accuracy, as the optimal number of tracers is situation dependent. Actually, a minimal number of tracers in a CF is preferred in order to minimize the estimation uncertainty. Results also showed a CF with higher discriminant ability may not necessarily translate to better estimation due to some degrees of non-conservativeness of the tracers used, again advocating the use of MCF methods or multiple approaches that tend to average out potential estimation errors. These conclusions should be tested further under different geographical and erosional conditions.