Abstract
Abstract. Soil loss and sediment transport in Mediterranean areas are driven by complex non-linear processes which have been only partially understood. Distributed models can be very helpful tools for understanding the catchment-scale phenomena which lead to soil erosion and sediment transport. In this study, a modelling approach is proposed to reproduce and evaluate erosion and sediment yield processes in a Mediterranean catchment (Rambla del Poyo, Valencia, Spain). Due to the lack of sediment transport records for model calibration and validation, a detailed description of the alluvial stratigraphy infilling a check dam that drains a 12.9 km2 sub-catchment was used as indirect information of sediment yield data. These dam infill sediments showed evidences of at least 15 depositional events (floods) over the time period 1990–2009. The TETIS model, a distributed conceptual hydrological and sediment model, was coupled to the Sediment Trap Efficiency for Small Ponds (STEP) model for reproducing reservoir retention, and it was calibrated and validated using the sedimentation volume estimated for the depositional units associated with discrete runoff events. The results show relatively low net erosion rates compared to other Mediterranean catchments (0.136 Mg ha−1 yr−1), probably due to the extensive outcrops of limestone bedrock, thin soils and rather homogeneous vegetation cover. The simulated sediment production and transport rates offer model satisfactory results, further supported by in-site palaeohydrological evidences and spatial validation using additional check dams, showing the great potential of the presented data assimilation methodology for the quantitative analysis of sediment dynamics in ungauged Mediterranean basins.
Highlights
Modelling sediment yield is a complex task due to the nonlinearity of natural processes intervening at slope and basin scale (Schumm and Lichty, 1965; Coulthard et al, 1998; Roering et al, 1999)
Implementation considered the total volume retained in the “main check dam” for calibration and spatial validation, and volumes associated with individual flood layers for temporal validation
Detailed alluvial stratigraphy was analysed in two parallel trenches across the dam infill
Summary
Modelling sediment yield is a complex task due to the nonlinearity of natural processes intervening at slope and basin scale (Schumm and Lichty, 1965; Coulthard et al, 1998; Roering et al, 1999). Together with a better understanding of hydrodynamic processes involved in the surface runoff, sediment production and sediment transport, have stimulated the development of physically based and distributed parameter models (e.g. WEPP, EUROSEM and LISEM). The reliability of such sediment yield models depends on a robust calibration and/or validation process that, at ungauged catchments, as it is the case of most of small basins around the world, may limit a broad use of such models. Different authors have used the sediment volume accumulated in lakes and reservoirs as an indirect validation method for modelling sediment yield at regional scale (Van Rompaey et al, 2003; Grauso et al, 2008). Some examples are Geiger (1957), Ackermann and Corinth (1962), Rohel (1962), Farnham et al (1966), Callander and Duder (1979), Jolly (1982), Le Roux and Roos (1982), Duck and McManus (1993), Avendano Salas et al (1995, 1997) and Verstraeten et al (2003)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
