Modeling sediment sources and yields in a Pyrenean catchment draining to a large reservoir (Ésera River, Ebro Basin)

Abstract

The study aimed to use the Soil and Water Assessment Tool (SWAT) model to simulate erosion processes in an alpine–prealpine catchment in order to provide data and information that may be relevant for managers so as to minimize reservoir siltation and water quality degradation. The main objective was to assess sediment production across the catchment and sediment supply to the main reservoir. The Barasona reservoir catchment (1,509 km2) is located in the Central Spanish Pyrenees, in the Ebro Basin. This catchment was selected for the case study given the regional significance of the Barasona reservoir and its siltation problems. The catchment has a mountain climate, with strong altitudinal and north–south gradients. The catchment is characterized by heterogeneous topography and lithology, resulting in a varied mosaic of slopes, soil types, and land covers. The Jueu karst system and two small headwater reservoirs were parameterized and calibrated in the model. The SWAT model sediment calibration for the catchment was based on a prior monthly hydrologic calibration, and the model validation was based on the sediment depositional history of the Barasona reservoir. The simulation period (2003–2006) and the validation period (1993–2002) produced average sediment yields to the reservoir of 643,000 and 575,000 t year−1, respectively. Large variations in sediment production were found between the subcatchments in the Barasona catchment due to differences in rock outcrops, land cover, and slope gradient. Sediment loss in the Jueu karst system was 15,500 t and the two small headwater reservoirs retained 31,200 and 50,300 t. Sediment production in relation to precipitation showed high temporal variability, with specific sediment yields to the Barasona reservoir ranging from 2.74 to 8.25 t ha−1 year−1. Strong lithological control was observed for sediment production in the subcatchments. The main sediment sources were located in the badlands developed on marls in the middle part of the catchment (internal depressions). The proposed model has proved useful for identifying areas where significant erosion processes take place in large alpine–prealpine catchments at a regional level and also for assessing discharge losses by the karst system and the sedimentary role of the small reservoirs. The information obtained through this research will be of interest in assessing the spatial distribution of sediment sources and areas of high sediment yield, which will be useful to establish criteria for remediation strategies.