Spatial variability of geophysical parameters relevant for the analysis of failure mechanisms of dikes within the depositional environments of the Rhine Meuse delta, the Netherlands

Abstract

The dikes in the Netherlands play a significant role in the flood defence. This is of large importance because the majority of the area in the Netherlands is susceptible to floods. Therefore, maintaining and improving the strength of the dikes to prevent failure is a priority. The probability of occurrence of failure mechanisms of a dike is related to the texture, composition, and stratigraphy of the subsurface below a dike, besides the strength of the dike itself. The Holocene subsurface of the Rhine-Meuse delta is highly heterogeneous, due to many avulsions of the distributaries of the Rhine that have left a complex network of channel belt deposits in the subsurface. Therefore it is often difficult to obtain a good estimate of the value and variation of the subsurface model parameters relevant for the failure mechanisms of dikes. The research parameters in this study are the thickness of the confining layer, the cone tip resistance, the sleeve friction ratio, and the median grain size. To obtain insight in the spatial variation of these parameters, the variation of the subsurface parameters must be considered in a geomorphological context. This may provide useful knowledge to validate the model parameters and uncertainties. This study aims to find if the spatial variation of the relevant subsurface parameters considered in the depositional context of the deposit improves the characterization of these subsurface parameters and improves the estimation of the probability of piping and liquefaction. A large database of CPTs and borehole logs was made, which provided data on lithology, median grain size, cone tip resistance, and sleeve friction (ratio). For every lithological layer, the depositional unit and the subsurface parameters of interest were defined. In this context, the spatial variability within the depositional unit of the subsurface parameters could be determined. It was found that the thickness of the confining layer overlying channel belt deposits is related to the age of the channel belt. The thickness of the confining layer generally increases with age of the channel belt, because of continuous aggradation in the study area. The rate of the thickening of the confining layer was found to be related to the rate of aggradation in the study area, which is forced by the creation of accommodation space, and sedimentation rate. The uncertainty within the results was too large to use this to validate a model parameter, but the relationship could be used to locate channel belts that could form a risk for piping. The cone tip resistance versus the channel belt generations could be used to define which channel belt generations most probably do not contain soil layers that are sensitive for liquefaction, as the ranges of the cone tip resistance of some channel belt generations do not contain values that are indicative of this type of soil. This methodology could be used in other parts of the Rhine-Meuse delta, but the results are strictly limited to the research area of which the results are determined, as the grain size of the channel belt sand bodies varies throughout the Rhine-Meuse delta. For the sleeve friction (ratio) and the median grain size, the spread within the depositional units was too large to be used to validate model parameters for the calculation of risk of piping and liquefaction. The grain size of channel belts varies on a smaller level than between different channel belt generations, and therefore scaling down de units in which the spatial variability of the median grain size is determined could be the subject of further research. Considering the spatial variation of the thickness of the confining layer, the cone tip resistance, the sleeve friction (ratio), and the median grain size in the context of depositional units do not provide characteristic values for a certain model parameter within the area of study. The added value to the determination of the risk for piping and liquefaction of considering the subsurface parameters, primarily the thickness of the confining layer and the cone tip resistance, in the context of their depositional and geological context is to address channel belt areas that impose a higher chance of piping and or liquefaction problems than others. This would help to determine locations where more subsurface research is needed.