Author(s): Stephan Hilgert
Keywords: Sediment; GraviProbe; Siltation; Reservoir; Hydro-acoustic;
Abstract: Reservoir siltation threatens the operation and ecological state of many reservoirs worldwide. Due to land use change and more frequent erosive rain events expected in the future, the sediment load to reservoirs is likely to increase. Sediment accumulation in reservoirs and related problems cause already more than 20 billion USD of follow-up costs per year globally. In many cases, the siltation rate, the type of sediment and the spatial distribution within the reservoir are unknown. This makes an appropriate management of the reservoir impossible. Independent of the use of, e.g. hydropower generation or drinking water supply, the exact storage capacity, surface-volume curve, predicted future scenarios as well as the sediment distribution in the reservoir represent crucial information. In order to obtain this type of information for reservoirs, especially if no or little historic information is available, we developed a methodology to assess the existing sediment magnitude as well as the type of sediment. We combined a dual-frequency hydro-acoustic (38 & 200 kHz) sediment classification approach with rheological measurements using a dynamic penetrometer. The measurements were conducted on a medium-sized Brazilian reservoir with a maximum depth of 17 m, which is used for drinking water supply. A set of significant correlations between acoustic backscatter parameters and physical sediment properties (undrained shear strength and cone penetration resistance) allow for a detailed characterization of the sediment composition (grain size, density). The acoustic “hardness” of the sediment reached a Pearson correlation coefficient of R = 0.7 with the cone penetration resistance obtained from the penetrometer. In addition, the sediment magnitude was derived. The use of the penetrometer for ground truthing helped to identify the influence of gas voids in the sediment, which strongly changes the acoustic response. Based on the regressions between hydro-acoustic and physical/rheological parameters, the local characteristics were extrapolated over the reservoir surface by conducting hydro-acoustic surveys. Next to the assessment of the lake bed type, a precise estimation of the sediment mass can be performed. Therefore, the developed methodology produces essential information for the management of reservoirs, including the planning of measures like dredging or a life-time assessment of the usable reservoir volume.