A novel approach of combining isotopic and geochemical signatures to differentiate the sources of sediments and particulate nutrients from different land uses

摘要

Determining the source of sediments and associated nutrients from terrestrial to aquatic environments is critical for managing the detrimental impacts of soil erosion and loss of nutrients from terrestrial into aquatic environment. However, tracing the source of particulate nutrients from different land uses has not been adequately carried out due to methodological difficulties in separating sources, particularly in the Great Barrier Reef (GBR) catchment. The objective of this study was to develop a method to differentiate the sources of particulate nutrients from soils collected from different land uses (combination of beef and dairy grazing, sugarcane, forest and banana) using both geochemical and isotopic signatures. In order to select a discriminative group of signatures, all soil samples collected from each of the land use areas were fractionated to 63 mu m size fraction and were analysed for both isotopic (delta C-13, delta N-15) and acid extractable geochemical properties (e.g. Zn, Pt and S). Considering the fact that the outcome of tracing models often depends on the type and robustness of the methods used, here we have employed a stable isotope mixing model (SIAR) to evaluate if the suite of selected elements could be used to estimate the relative contribution of different sources for a series of five virtually created sediment mixtures. For the five groups of virtual sediments, the SIAR model provided close estimates to the contribution values of sediment sources with the Mean Absolute Error (MAE) varying from 0.30 to 2.88%. Results from this study show for the first time that the combined use of isotopic and geochemical signatures enable the SIAR model to provide an accurate estimation of source apportionment where a variety of land uses needs to be investigated and shows promise as a valuable new sediment and particulate nutrient tracing tool. (C) 2018 Elsevier B.V. All rights reserved.