Physical model study on impacts of landslide generated wave action on embankment dams

The master thesis "Physical model study on impacts of landslide generated wave action on embankment dams" has been written in collaboration with the Department of Hydraulic and Environmental Engineering at NTNU and the Norwegian Water Resources and Energy Directorate (NVE). The thesis focuses on studying the effects of landslide generated waves on embankment dams in order to estimate safe freeboards. A landslide falling into a reservoir produces impulsive waves than can have catastrophic consequences to the reservoir sidewalls, the dam body and, in case of overtopping, endanger human activities downstream of the dam. Therefore, the investigation of landslide hazard and risk on reservoir banks is of the utmost importance. The tests were conducted at the scaled model of the Viddal reservoir (Norway) and slide velocity, wave height from 10 stations, overtopping height on the dam crest and overtopping volume have been collected. In total 17 tests were conducted varying the following parameters: slide volume (2, 4 and 6 blocks), freeboard (30 mm and 80 mm) and dam slope (1:1.6, 1:2 and 1:2.4); in addition for the two last tests the dam roughness was varied by gluing stones on the upstream dam slope. The impacts of these parameters on the run-up height and the overtopping volume have been studied; moreover, the processes of wave propagation and run-up have been discussed. The run-up height and the overtopping volume monitored during the tests have then been compared with the results calculated with the Heller s numerical method. The calculation method uses 15 governing parameters to compute the overtopping volume and the run-up height caused by subaerial landslide generated waves. The parameters can easily be measured in the model. Moreover, the method enables to calculate the desired results for both 2D and 3D reservoir geometry. The collected results show a strong correlation between the overtopping volume and the slide volume. It was also noticed the great influence of the freeboard on the overtopping size. Even though only two tests were performed, the impact of the upstream dam slope roughness was found to be significant. On the other hand, the influence of the dam slope is minor, causing only slight changes in the overtopping volume. The results obtained using the Heller s method show large discrepancies between observed and simulated data. The main reason is the method assumption of an idealized reservoir geometry which strongly differ from the model shape. It is thus concluded that the Heller s method cannot be applied to the Viddal reservoir and reservoirs with similar geometry, in order to determine a safe freeboard