Landslide-generated Impulse Waves in Reservoirs: Basics and Computation

Abstract

Landslide-generated impulse waves are typically caused by landslides, rockfalls, shore instabilities, snow avalanches or glacier calvings in oceans, bays, lakes or reservoirs. They are particularly relevant for the Alpine environment because of steep valley sides, possible large slide masses and impact velocities and the great number of reservoirs. In this manual, a state-of-the-art on the impulse wave generation and its effects on dams are presented including a computational procedure. Based on this method, engineers or natural scientists may predict the hazards originating from impulse waves efficiently and economically. The 1st edition of this manual was published in 2009. This 2nd edition includes both updates of existing and new computational approaches for additional hydraulic processes. The introduction in Chapter 1 contains background information on the topic and compares the available methods dealing with landslide-generated impulse waves. The method presented in this manual is based on generally applicable equations derived from hydraulic model tests. Chapter 2 introduces basic principles of the water wave theory. The computational procedure is presented in Chapter 3 and shown in Figure 3-1. It is based on the findings of impulse wave generation and its effects on dams. The computational procedure (Figure 3-1) includes two steps: in Step 1 the generally applicable equations are applied according to Chapter 3, whereas in Step 2 the effects not contained in Step 1 such as the effective instead of the idealised reservoir geometry are considered according to Chapter 4. In Step 1, the mass movement is modelled as a granular slide. To analyse the effect of impulse waves on dams, the wave height, amplitude, period and length are important. These are computed with the equations of Heller and Hager (2010) and Evers et al. (2019) as a function of the slide parameters. Two extreme cases for estimating the wave parameters are considered: (a) laterally constricted (2D) and (b) free radial propagation of the impulse waves (3D). The wave generation in both (a) and (b) depend on the identical parameters, whereas these for the wave propagation are not identical. Once the necessary wave parameters in front of the dam are determined, the run-up height and the overtopping volume may be computed according to Evers and Boes (2019) and Kobel et al. (2017), respectively. Potential overland flow on horizontal shorelines is covered by the equations of Fuchs and Hager (2015). The force effects on dams are computed using the method of Ramsden (1996). This method is first applied as if the dam would be vertical since the horizontal force component is independent from the dam inclination. The additional vertical force component for inclined dams is then computed assuming static wave pressure. If an impulse wave partially overtops a dam, only a partial water pressure has to be considered resulting in a reduction method. Once the results from Step 1 are available, the effects of the geometrical differences to the idealised extreme cases (a) and (b) have to be quantified in Step 2 according to Chapter 4. These differences may result from the prototype reservoir geometry differing from the idealised 2D or 3D geometries, or by the non-granular mass characteristics. The impulse wave parameters may considerably differ due to these effects. The presented method of Ruffini et al. (2019) allows for impulse wave height estimation in intermediate reservoir geometries between 2D and 3D. Approaches for the assessment of edge wave propagation along the shoreline perpendicular to the slide impact direction include equations of Heller and Spinneken (2015) and McFall and Fritz (2017). Moreover, the extent of underwater landslide deposits is covered by the equations of Fuchs et al. (2013). Step 2 is also required if the computational tool is applied, because these include only the generally applicable equations from Step 1. Finally, Sections 4.6 and 4.7 contain a sensitivity analysis and some reservoir safety aspects. Chapter 5 includes four computation examples and the application instructions for the computational tool. In Chapter 6 the conclusions are presented. Although the computational results, such as the run-up height, seem to be exact, it should be kept in mind that the present method results in estimations. Safety allowances for all planned actions have to be considered. Predictions that are more exact may emerge from a prototype-specific model test or numerical simulations.