Determination of different types of sediments in a river reservoir and computation of their volumes

Paper deals with sediments settled in the Elbe reservoir Les Království (King s Forest) between 1914 and 2005. This reservoir is laid on upper Elbe closing the mountain watershed of Krkonoše (Giant Mountains) with the source of the river. To this profile, watershed can be taken as mountain and foothill area. The sediments, before approximately 1950, were non-polluted by heavy metals and specific organic matter because of the undisturbed natural environment. After that time, the conditions of sedimentation became worse. With the industrial boom and ploughing of the grazing land away there, the amount of sediments settled in the reservoir increased much times and high pollution appeared. To this point, the badly treated sewage waters and washing outs of the soil with pesticides and herbicides from the new high-sloped cropland contributed for many years. With this, the amount of sediments increased up to 21% of the total volume of the reservoir, i.e., approximately, to 2 mil per cubic m. It was decided in 1984 that the sediments have to be removed from reservoir. Up to this time, the quality of sediments was followed in the surface layers only and this quality was applied to all layers of sediments. Thus the wrong idea appeared that all sediments are polluted. On the contrary, the authors suppose that the old layers are not polluted and, in the case of cleaning of the reservoir, these need not be stored in a protected dump. Differentiating the sediments to polluted and non-polluted would bring a considerable increase in economy of cleaning. Thus the authors looked for a method to distinguish the sediment components one another. Among others, the method of the radar search of sediments appeared to be promising enough to describe the sediment layers from the actual bottom down to the original bedrock. To this aim, a new radar antenna working in two regimes was applied. In shallow parts of the reservoir, it was towed in cross profiles on water surface while in the deep ones it was moved along the bottom escorted by a diver. Using this, we were able to distinguish 5 components of sediment (i.e. the non-consolidated and cohesive sediment, the sandy loam, sandy silt and the gravel), to fix their positions and to estimate, in accordance to their apparent grain distribution, if they could be polluted or not. Moreover the mathematical method just presented here was worked out to compute volumes of different sediment components directly from radar signals. This method enables to compute the sediment volumes in relatively straight parts of the valley and the meandering ones. Thus the authors have presented a method enabling the estimate of component volumes on one side and to fix their positions to take samples for quality analyses on the other side.