There is a shift in the approach for designing coastal structures. In the past, dikes were designed on the probability of exceedance of an incoming wave during storm conditions. In the near future, the design criteria will be the probability of flooding of the hinterland. In order to determine this flooding probability, the strength of the dike has to be known. This thesis focuses on the erosion of the grass sod during overtopping wave volumes. Several tests have been performed in the last few years with the wave overtopping simulator. During this thesis, some of these locations have been tested for the strength of the grass sod with a newly developed method. This new method, called the sod pulling method, tests the actual strength of the grass by lifting the grass sod out of the top layer. However, this method makes at least two cuts in the sod in order to attach the pull frame. So a methodology has to be developed to calculate the strength of an intact grass sod from the measured data. The measured forces needed to lift the grass sod can be rewritten into critical grass normal stresses for an intact sod, which can be done with a practical method developed in this thesis. In this method a shape factor is introduced to compensate for the influence of the cuts made in the sod before testing. Because of the heterogeneity of the grass sod, the individual test results are not the most important parameters. In order to determine the representative strength of the grassed slope during wave overtopping conditions, the strength of the weakest sections in the grass sod have to be determined. This can be done by assuming a normal distribution for the strength of the grass, where the 2.5% tail value will be used as the governing strength of the sod. In order to determine the parameters of the normal distribution, at least 30 tests are needed to be done as condition 2 test with the 20 by 20 centimetres frame size under saturated conditions, at random locations on the bottom half of the slope. The square root of the calculated critical grass normal stress is one of the input parameters in the critical velocity formula, which also uses the pore water pressure, the relative turbulence intensity and the density of the water. When the critical velocity resulting from this formula is compared with the determined critical velocity during the wave overtopping simulations, there is good correspondence between the values for the tested locations in this thesis. So the sod pulling test can provide results that are reliable enough to determine the critical velocity of a dike section.