Catchment-wide variations and biogeochemical time lags in soil fatty acid carbon isotope composition for different land uses: Implications for sediment source classification

• Published in: Organic geochemistry 2020-08, Vol.146, p.104048, Article 104048
• Main Authors: Upadhayay, Hari Ram, Griepentrog, Marco, Bodé, Samuel, Bajracharya, Roshan Man, Cornelis, Wim, Collins, Adrian L., Boeckx, Pascal
• Format: Article
• Language: English
• Subjects: Cluster analysis; Headwater; Land terraces; Long-chain fatty acids; Sediment fingerprinting
• ISSN: 0146-6380; 1873-5290
• DOI: 10.1016/j.orggeochem.2020.104048

•Variations in fatty acid δ13C values for land uses were assessed at catchment scale.•Highly variable δ13C-FAs values reflected vegetation composition and change.•Paired δ13C values of fatty acids and bulk carbon detected C4 crops legacy effects.•A priori land use classification for sediment sources was questionable.•Sediment source classification required combining functionally similar land uses. Stable carbon isotope values of fatty acids (>20 C-atoms) (δ13C-FAs) have been increasingly used to apportion sediment sources using isotope mixing models. Understanding the variation of δ13C-FAs within catchment land uses is crucial for correct classification of sediment sources but it has rarely been evaluated. Here, we assessed the variability of δ13C values of FAs (C22–32) in soils under land uses within a catchment (23 km2) in the mid-hills of Nepal. High δ13C-FA variability in agricultural terraces (−32.8‰ to −20.2‰) and mixed forest (−37.3‰ to −22.3‰) was observed. This is due to differences in inputs of FAs of different chain-lengths with variable δ13C-FAs values by C3 and C4 crops as well as from farmyard manure in agricultural terraces. The pine plants succession in the forest and differences in FA preservation explained the observed variation in δ13C-FA values in a mixed forest. A pine forest transect that had undergone land cover change about 40 years ago showed high variability of δ13C-FA values strongly linked to legacy effects of previous C4 crops. Importantly, soils from the border of a land use and fresh surface deposits mostly increased the within-land use variability of δ13C-FAs values. Overall, δ13C-FA values have low potential to function as a robust tracer for high resolution discrimination of land uses and the validity of a priori sediment source classification is most questionable in catchments that have undergone land cover change. Cluster analysis was promising for identifying and combining functionally similar land uses to define more meaningful sediment sources at the catchment scale.