Carbon, Nitrogen, and Corresponding Stable Isotope Signatures Reveal Channel Banks as Major Sediment Sources in a Tropical Agricultural Watershed
Uncertainties about the applicability of δ13C and δ15N as tracers of sediment sources in tropical river basins highlight the need for more in-depth investigations of these isotopes. This study therefore assessed the effectiveness of δ13C and δ15N signatures in discriminating sediment sources in an agricultural catchment in Northeast Brazil. Three potential sediment sources were sampled as follows: unpaved roads, sugarcane cultivation, and channel banks. Suspended and riverbed sediments were used as target sediments. Source and sediment samples were sieved to two particle size fractions: <63 and <32μm. The isotopes were evaluated using conservativeness tests, Kruskal–Wallis, linear discriminant analysis, and virtual mixtures. Our results indicated that δ13C and δ15N together are effective tracers for modeling sediment sources, providing significant detail on sediment delivery patterns in a tropical catchment under intensive land use. Both fractions showed no significant differences in conservativeness or source apportionment. However, the <63μm fraction yielded more robust discrimination potential and model estimates. Therefore, future studies in other catchments under similar conditions could focus on a single fraction, preferably the fraction <63μm, optimizing effort without compromising scientific robustness. Channel banks contributed the majority of sediment in both size fractions, indicating that agricultural expansion into riparian zones—resulting in the absence or inadequate type of vegetation cover—has accelerated erosion. This underscores the urgent need to restore riparian forests and protect these vulnerable areas, while also emphasizing the importance of developing innovative, interdisciplinary approaches to effectively manage and integrate riparian vegetation into landscape planning
| Item Type | Article |
|---|---|
| Open Access | Gold |
| Keywords | Bayesian modeling, Catchment management, Particle size, Sediment fingerprinting, Stable isotopes |
| Teams | Farming Footprints and Adaptations |
| Project | Resilient Farming Futures (WP2): Detecting agroecosystem ‘resilience’ using novel data science methods, Resilient Farming Futures |
| Date Deposited | 23 Jan 2026 10:50 |
| Last Modified | 23 Jan 2026 10:50 |


