Protecting the Alpine Rhine Valley from flooding

Author: Renata Barradas Gutiérrez

The Rhine, one of the main rivers in Europe, sources from the Swiss Alps in the canton of Grisons. The Alpine Rhine Valley extends over 90 km along the Rhine from its source in Switzerland via Liechtenstein to Austria. The Valley has a history of devastating flood events that date back to the 11th century. Today, around 300,000 people live in the lower Rhine Valley and numerous companies, including Leica Geosystems, flourish in this area. Due to the intense population and major economic activities in the Rhine Valley, damage potential from major flood events is estimated at EUR 10 billion.

To protect people, settlements, and as economic activities in the Valley, more room for flood runoff and water retention needs to be given to the Alpine Rhine. Therefore, the flood protection project “Rhein – Erholung und Sicherheit” (“Rhine - Recreation and Safety”) – or short Rhesi - seeks to increase the flow capacity of the Alpine Rhine from 3,100 m³/s to at least 4,300 m³/s on the international stretch between kilometre 65 at the junction of the tributary river Ill and km 91, where the Alpine Rhine discharges into Lake Constance. The project costs, funded equally by Austria and Switzerland, are currently estimated at EUR 1 billion.

“To achieve the requested level of flood protection, the channel geometry of the Alpine Rhine needs to be altered to enhance flood protection along the project perimeter. In the Rhesi project, a very modern approach has been chosen: instead of raising the river’s levees to take account of the elevated discharge of 4,300 m³/s, the required flow section will be created by increasing the river width from currently 60 – 70 m up to several hundred meters in the future. The river channel, which has at present a very technical shape due to diverse river restoration measures during the last 150 years, will by this means retrieve a near-natural state with conditions that mimic the state of the river system before human intervention,” explains Florian Hinkelammert-Zens, environmental engineer at the Laboratory of Hydraulics, Hydrology and Glaciology (VAW) at the Swiss Federal Institute of Technology Zurich (ETH).

To evaluate the effects of the projected measures and to check the hydraulic calculations and assumptions of the Rhesi project, VAW of ETH Zurich has been commissioned with hybrid model experiments on behalf of the International Rhine Regulation (IRR) body. These investigations consist of two main parts: 1) experiments in a physical hydraulic model and 2) accompanying numerical simulations.

“Two key project sections are replicated consecutively at a scale of 1:50 in extensive hydraulic models. For each section, a flow length of approximately 5 km is replicated (around 110 m in model scale) with watercourse widths ranging from 250 m to 350 m (around 8 m in model scale),” says Hinkelammert-Zens. “At the same time, numerical computer models of the project were created to provide and evaluate the boundary conditions of the hydraulic models, to validate the results and to carry out sensitivity analyses.”

As a result, these two hydraulics models are among the biggest models of alpine rivers ever built, with average dimensions of 110 x 9 m. Both are located in an old factory building in Dornbirn, Austria, where ETH Zurich designed a water circuit with a discharge of 400 l/s. The system consists of a high-level tank, inlet and outlet basins, a water return line in the basement and a deep tank, from which the water is pumped back into the high-level tank (max. 400 l/s).

3D terrain modelling for flood modelling

“During a flood event, a riverbed is subject to significant changes due to high water discharges and flow velocities. Hence, sediment can be deposited at several locations, leading to rising water levels, or can be eroded, e.g. around bridge piers or along the river banks. Both scenarios can be dangerous and have a negative effect on flood protection. To replicate these morphological changes, the hydraulic models are equipped with movable riverbeds.” says Hinkelammert-Zens.

To observe the impact of different sediment loads and various scenarios, a large number of scientific experiments with varying parameters (e.g. water discharge and sediment load) are conducted. By means of a laser scanner, the model topography is measured before and after every single experiment. The acquired data is then used to create terrain models which serve as a basis for the determination of areas where sedimentation and erosion occur in the riverbed.

From data capture to actionable data

Right: 3D terrain model of a section of the Alpine Rhine (viewed in flow direction) /Left: Movable riverbed in the hydraulic model after the conclusion of an experiment

To capture the topographical data before and after each experiment, the research team of ETH Zurich relies on a Leica ScanStation P20, Leica Geosystems targets and a Leica TS02 total station to geo-reference the laser scans with 15 reference points. The ScanStation P20 is mounted on a mobile tripod and deployed on four scanning positions to capture the whole model. With a scanning height of approx. 2.7 m - to minimise shadowing effects if viewing angles are too steep and to avoid dead angles – and a resolution of 3 x 3 mm at a radial distance of 10 m to the device, high-quality data with very low noise can be obtained.

After each experiment, the data is imported into Leica Cyclone 3D point cloud processing software to register the data and merge the point clouds. At this point, an area of 4000 m² is represented with approximately 250 million points. The point cloud is then ‘trimmed’ using polygons to cut-off the data points outside of the model boundaries. The remaining data points are then transformed into grid cells with a cell size representing 50 cm x 50 cm in real life. Finally, the topographical data is converted into the Swiss National Coordinate System.

Right: visualisation of the observed changes in the riverbed in the hydraulic model after evaluating the laser scan (red: erosion on the outside of the curve, blue: sedimentation on the inside of the curve, viewed in flow direction)/ Left: laser scan in the experiment hall (viewed in flow direction)

“The 3D point clouds are used to create grid datasets with approximately 15 million grid cells with a resolution of 0.5 x 0.5 m, each of them representing one distinct point of time during the experiments. This data is then further processed in geo-information systems in order to create surface views as well as longitudinal and lateral profiles of the mobile riverbed. This enables us to compare different points in time of the experiment with each other,” explains Hinkelammert-Zens.

The referenced grid dataset can be used in GIS applications for various evaluations, including:

• Surface views: The grid values of the scan made at the beginning of the experiment are deducted from those made at the end of the experiment. In this way, the ETH team creates a view where the relative differences in the height of the model riverbed are visible.

• Transverse profiles: The team creates cross profiles at certain positions, extracting grid values to create lateral profiles. Using the scans before and after the tests, the experts can visualise the observed changes and compare them to the project goals.

• Longitudinal profiles: The extracted cross profiles are averaged for the longitudinal profile. By comparing the averaged riverbed elevations before and after the experiments and by observing the changes in nature, the team of experts can validate the hydraulic model.

Intermediate results and future steps

The investigations by VAW of ETH Zurich already led to significant inputs for the further development of the Rhesi project. At first, the model was calibrated via the replication of past flood events. During this process, the water levels and riverbed topography obtained in the hydraulic model were compared to data captured during those events in full-scale. After successful completion of this step, the hydraulic model was adapted to the future shape of the river, as projected in Rhesi. Since then, various long-term scenarios and high flood events have been simulated to investigate the effects of the Rhesi project on the river morphology and water levels.

As the investigations are still ongoing, only intermediate results can be cited. Up to today, the results show that the assumptions and projections of the Rhesi project were correct and are a solid basis for the elaboration of future project stages with greater detail. The hybrid model experiments will continue until summer 2022, exploring answers to the following technical questions:

• Where will gravel banks be positioned?
• Where will depressions resp. scours occur and what will be their maximum depth?
• How deep must the river banks be protected against erosion and scouring?
• How can bridge piers be secured against erosion and scouring?
• What is the amount of driftwood clocked at bridges during flood events? What will be the effect on the water levels?

The findings of these scientific experiments, supported by reality capture technology from Leica Geosystems, are the basis to ensure sustainable river planning and assure that the Rhesi flood protection project is technically and economically viable. This integrated flood risk management approach will significantly reduce flood risks and improve the ecological and recreational value of the Alpine Rhine in the international stretch.