Author(s): M. Klosch; B. Blamauer; H. Habersack
Linked Author(s): Mario Kloesch
Keywords: No Keywords
Abstract: Recently great progress has been made in simulating riverbank erosion by fully coupling its main processes, fluvial erosion and mass failure, and hence allowing them to interact during the simulation. The timing and geometry of failures codetermine the overall bank retreat during the simulated flood event. This work aims to gain appropriate field data for verification of existing modeling approaches. At a restored section of the Drava River, Austria, a time-lapse camera was installed, taking pictures of the front side of a cohesive riverbank at user-defined time intervals. Additionally the bank face was surveyed seven times using terrestrial photogrammetry. A water level gauge was installed few meters downstream of the investigated riverbank. By combining the picture information with the digital elevation model from photogrammetry or with the measured water level height, geometric data during flood events was obtained. The volumes of single failures were calculated using the digital elevation models from photogrammetry and additional information from camera survey during the flood event. The timings of the failures have been expressed by relating them to their position on the hydrograph of the corresponding flow event. According to recent findings in fully-coupled simulations, the observed riverbank erosion proceeds as a series of failures distributed throughout the flow event combined with fluvial erosion. If only surveying the riverbank before and after the event, adjacent bank failures may be mistaken as one large failure, maybe resulting in wrong model calibration. The observed limited longitudinal extent of single failures questions the two-dimensional representation of riverbank geometry in bank erosion models performed so far, and suggests consideration of the limited longitudinal extent of single failures.
Year: 2009