Dike design needs to focus more on the subsoil structure and variations over the entire length of a dike. The soil structure determines the risk of piping and varies more than is assumed in the present standards. This may represent a risk for the dike, but it can also be used to make local dike upgrades more efficient.
That is the conclusion stated by Wim Kanning in his thesis ‘The weakest link – Spatial variability in the piping mechanism of dikes’. His research looked at how soil variability can be included in dike design to reduce the risk of piping. Wim Kanning received his doctorate on Friday, 14 December at Delft University of Technology.
Piping is an important failure mechanism in flood defences but there has been a surge in scientific and practical interest in the phenomenon in recent years. Piping starts when there is a difference in water levels, resulting in the uplift of the blanket layer on the inside of the dike. A groundwater flow then washes away the sand below the dike. The result is a pipe which undermines the dike, leading to a possible failure.
Wim Kanning explains that the risk of piping is primarily determined, in addition to the difference in the water level and the width of the dike, by two factors: ‘It depends on the permeability of the subsoil and the size of the particles. The more permeable the soil, the greater the risk of piping. The larger the particles, the higher the resistance to the water and the lower the risk of piping. The variations in these parameters are not adequately represented in the present dike standards.’
In addition, Wim Kanning’s research has shown that the structure of the subsurface can vary more than has been assumed until now: ‘ That has implications for longer dikes in particular. At present, indicative cross-sections are used for entire dike sections, but they are not valid for every location. That approach is inadequate when it comes to describing piping. This cuts two ways: it can represent a risk, but it may also mean that less upgrade work may be needed than we think. So accurate calculations could lead to major cost savings.’
Wim Kanning’s doctorate presented two models for improving the calculations: ‘ The first model is a simple design model. I developed it after analysing a large number of historical dike breaches. The second model describes the formation of piping in different soil structures. The second model is particularly promising when it comes to modelling piping and coming up with a dike design that takes the real risk of piping into account .’
[mappress]
Press Release, December 28, 2012