Author(s): Klajdi Sotiri, Stephan Hilgert, Stephan FUCHS
Keywords: Reservoir; Sedimentation; Sediment classification; Sediment properties; Hydro-acoustics;
Abstract: The most direct impact of dam construction is disruption of riverine systems and their biochemical cycles. A common byproduct of this disruption is the accumulation of sediments upstream of the dam. The accumulated sediment is often rich in nutrients, organic matter and sometimes pollutants. The increased concentration of these components in the sediment has a direct influence on the water quality of the reservoir thus, the reservoir management. Consequently the necessity emerges, not only to collect data about the amount but also about the quality of accumulated sediment. In many cases the limiting factor for traditional lakebed single point sampling is the extensive area of the impoundments, as high costs are generated. We implemented a coupled approach of sediment point measurements (core and grab sampling) with hydro-acoustic sediment parameters, in order to map sediment characteristics. Six different reservoirs (Vossoroca, Capivari, Passauna Reservoirs in Brazil, Urft- and Große Dhünn- (GDTS) Reservoir in Germany and Phalaborwa Barrage in South Africa) were investigated between 2011 and 2018. These reservoirs are located in different geographic and climatic zones and are characterized by different watershed land use, morphometry and purposes. The siltation rate varies strongly from nearly no accumulated sediment to extreme storage capacity loss. For the hydro-acoustic survey a single beam echo-sounder (EA 400, Kongsberg) with two frequencies (200 & 38 kHz) was used. Over 70 sediment core samples and 30 sediment grab samples were taken as groundtruthing for the hydro-acoustic measurements. Significant correlations were observed between geo-chemical parameters (LOI 550°C) as well as physical parameters (grain size distribution and density) with a selection of hydro-acoustic parameters. All six reservoirs were covered with dense measurement transects allowing for a sediment classification of the entire lakebed surface. Regression analyses were performed for the derivation of empiric models. By applying the models to the obtained hydro-acoustic spatial information, the detailed type of lakebed was derived. The results show that the method is transferable, independent from the location, catchment properties, morphology and trophic state of the waterbody. Within certain ranges of uncertainty, the acoustic classification can be used to create maps of sediment types. These serve as a basis for any type of sediment related measure to be implemented e.g. dredging.