The border section of the Salzach between the mouth of the Saalach and the mouth into the Inn is in a bed erosion process. This deep erosion, which was deliberately initiated as a result of regulatory measures at the end of the 19th century, is now causing the river bed to cut into the gravelly subsoil. This ersoion has now reached a depth of up to 4 m in the area below the mouth of the Saalach. A slowing down of this deepening tendency is not in sight.
On the contrary, due to the special conditions of the river bed (see project rehabilitation-river-salzach", a massive acceleration of the deepening process can be observed, as the most recent river bed recordings after the 100-yearly flood in 2002 on the Salzach River prove. All solution variants developed in the Salzach Water Management Framework Study to stop the process of deepening the Lower Salzach contain so-called river bed riprap strips (structures with the same bottom made of coarse stones for stabilising the river bed with about twice the gradient compared to the unsecured intermediate sections) as structural elements in combination with a bank-parallel widening of the Salzach river bed. The rolling strips have lengths of 0.8 to 3 km.
The physical model test served to develop guidelines for the dimensioning and practical implementation of riprap strips to safeguard the hydraulic engineering and bed morphological function.
CUSTOMER: Office of the Salzburg State Government
The model investigations are essentially intended to provide information on the following questions:
- Optimization of the function and design of the river bed riprap strips, taking into account the latest research results on the topic of surface stabilization of river beds, including consideration of practical implementation possibilities (construction process, existing material for river bed stabilization).
- Proof of stability of the construction measures including connections in headwater and tailwater.
- Investigation of the effects on the river morphology (self-development of the river following the riprap strips, influence on alternating bars, risk of relocation of diversion points into the tributary water in the area of the riprap strips).
With an assumed layer thickness of 0.5 m and river bed widths between 100 and 170 m, the material requirement per kilometer of riprap strip is between 50,000 and 85,000 m³. For such large quantities of material for bed stabilizing should not make to great demands with regard to grain composition (e.g. single grain material), as they are otherwise very difficult and very expensive to obtain. In the model test, it was therefore also necessary to determine what range of grain compositions is permissible for the riprap material. Furthermore, it was to be examined whether a design for the river bed protection is possible which reduces the material requirement or demand (alternative river bed protection possibilities on the basis of self-acting armour layer, loose rip rap, resolution of the riprap strips in sleeper sequence, structured and flat ramps).
The model investigations were divided into two parts. On the one hand, guidelines for the dimensioning of the bed protection were developed on the basis of a sectional model and stability proof was provided. On the other hand, the design of the connecting sections upstream and downstream of a riprap strip and the general effects of the measures on the river morphology were investigated in a full model. The sectional model was carried out in a tilting glass channel with a width of 54 cm. The area of the full model was 5 x 40 m. Both the sectional and the full model were carried out on a scale of 1 to 50.
Sectional model
The following variants were examined in the sectional model:
Single-layer riprap (stone size 3 - 22 kg) on existing bed material with a slope of 1.5 ‰ - This solution was considered problematic with regard to the safety of the stabilization effect and overload capacity and would have to be made of larger stones and two layers.
Coarsening of the surface layer - The resistance of the riverbed could be increased with a corresponding coarsening of the top layer of the bottom material to such an extent that a stable surface layer (armour layer) up to the design discharge was formed. It showed a slow destruction process under overload, but requires a very high amount of material.
Loose riprap with an occupancy density of 40 % and a slope of 1.5 ‰ - The loose riprap has the advantage that part of the bed remains uncovered and therefore the material requirement is lower. But the necessary even distribution of the stones on the bed causes greater effort in the execution of the method.
Left image: riprap
Center image: coarsening of the surface layer
Right image: loose riprap
Strips of riprap (bed sills) at the beginning (top) and after design discharge (bottom)
View through the glass wall into the model
Flat structured ramp with a slope of 10 ‰ - The ramp consisted of crossbars at intervals of 10 to 15 m, which were finished from stones between 2.5 and 5 tons. The basins between the crossbars were covered with stones between 300 and 600 kg. Since the material requirement and thus the area secured can be considerably reduced by the larger gradient, this securing method was also investigated in the full model in a straight section under 3-dimensional flow conditions.
Model of a structured ramp in the glass channel
Full model
The structured ramp with a gradient of 10 ‰ investigated in the full model had a height of 1.5 m and a river bed width of 85 m. The stones of the crossbars and basins were selected analogous to the stone sizes optimized in the sectional model and bedded on an appropriate filter layer. Adjacent to the foot of the ramp was a tailwater-bed protection consisting of one dense layer, of basin stones. The length of the tailwater-bed protection corresponded to the ramp length. Towards the end of the tailwater protection the density of the stone layer was reduced to zero. In addition, the banks were protected against undercutting by a 3 m wide stone layer. Due to the accelerated current reaching into the headwater, a wedge-shaped stone deposit was required at the crown of the ramp to a depth of about 2 m below the headwater level. The structured ramp designed in this way proved to be stable until the design flood.
Structured ramp in the full model (view from above)
A likewise investigated structured ramp with interrupted crossbars for the purpose of generating a meandering flow on the ramp at low water discharge did not bring the desired success.
Structured ramp with interrupted crossbars for initiating a pendulum flow on the ramp
The knowledge gained from these experiments is not only important for the status of the Salzach, but also in general, and can be transferred to other water bodies in the Alpine and pre-Alpine regions that are in a state of erosion in order to achieve and sustainably secure both the water management and the ecological function in the sense of the EU Water Framework Directive.