Combined Pullout Tests And Wave Overtopping Simulations On Three Species-Rich Grass Covered Dikes In The Netherlands

Abstract

Grass cover erosion by wave overtopping is a major failure mechanism for earthen dikes. Grass cover erosion resistance (represented by the critical velocity, U_c) can be assessed using full-scale, destructive tests with the wave overtopping simulator (WOS) in combination with the erosion model ‘cumulative overload method’ (COM). Although these tests provide valuable information, they are relatively expensive and time consuming. Therefore, a small-scale grass pullout test (GPT), which translates the vertical pullout force of a grass sod to a U_c, may be an attractive alternative. In this paper, based on comparative testing with the WOS and GPT on three species-rich grass covers on dikes in the Netherlands, we assess the correspondence between the U_c obtained with the WOS and GPT for species-rich grass covers. Ultimately, by also including historical tests, we aim to get more insight into the suitability of the GPT for quantitative erodibility assessment of grass covers in general.

During testing with the WOS, no failure was observed for the tested grass cover sections. Therefore only a lower bound of the U_c could be derived. Results indicate that the estimated U_c with the GPT is around the lower bound values obtained with the WOS. This is in line with previous tests on conventional grass covers, for which the GPT provides a more conservative estimate of the U_c than the WOS. Adaptations in the translation from pullout force to U_c may correct for the negative bias and hence improve the reliability of the GPT for quantitative erodibility assessment. Several suggestions to adapt the GPT in future work are provided in this paper. These adaptations should be tested and validated for conventional as well as species-rich grass covers to guarantee the general applicability of the method.