Articles | Volume 5, issue 1
Research article
17 Feb 2017
Research article |  | 17 Feb 2017

Modelling a century of soil redistribution processes and carbon delivery from small watersheds using a multi-class sediment transport model

Florian Wilken, Peter Fiener, and Kristof Van Oost

Abstract. Over the last few decades, soil erosion and carbon redistribution modelling has received a lot of attention due to large uncertainties and conflicting results. For a physically based representation of event dynamics, coupled soil and carbon erosion models have been developed. However, there is a lack of research utilizing models which physically represent preferential erosion and transport of different carbon fractions (i.e. mineral bound carbon, carbon encapsulated by aggregates and particulate organic carbon). Furthermore, most of the models that have a high temporal resolution are applied to relatively short time series (< 10 yr−1), which might not cover the episodic nature of soil erosion. We applied the event-based multi-class sediment transport (MCST) model to a 100-year time series of rainfall observation. The study area was a small agricultural catchment (3 ha) located in the Belgium loess belt about 15 km southwest of Leuven, with a rolling topography of slopes up to 14 %. Our modelling analysis indicates (i) that interrill erosion is a selective process which entrains primary particles, while (ii) rill erosion is non-selective and entrains aggregates, (iii) that particulate organic matter is predominantly encapsulated in aggregates, and (iv) that the export enrichment in carbon is highest during events dominated by interrill erosion and decreases with event size.

Short summary
This study presents a model that accounts for preferential erosion and transport of sediment and soil organic carbon in agricultural landscapes. We applied the model to a small catchment in Belgium for a period of 100 years. After a thorough model evaluation, these simulations shows that sediment and carbon export are highly episodic and that the temporal variability is largely influenced by selective erosion and deposition.