Comparison between different wave runup level prediction formulas based on the wave breaking type

In this study, the performance of some models for predicting the wave runup level is investigated. To do so, they are compared with each other over a database of 1390 field and laboratory data points. This comparison has actually two aims: The first one is to investigate the models using all data points and the second one is to consider the influence of wave breaking type on the accuracy of the formulas. The latter goal is achieved by dividing the data points based on the type of wave breaking using the Iribarren number. It is also important to mention that most of the models used here for comparison have been optimized by Power et al. in 2019 so that more accurate outcomes would be obtained by them. These models depend on certain wave and beach parameters, including wave height, wave length, wave period, and seabed slope. The results of the comparison have been demonstrated using several statistical characteristics (e.g. RMSE, R2 , etc.) so that a good understanding could be obtained from the behavior of each formula. At first, when the type of wave breaking is not considered in the comparison and all data points are used, the optimized formula derived from both studies of Holman and Atkins et al. seems to be the most accurate one with the least prediction error. Then, when the data points are divided into groups based on the wave breaking type (i.e. spilling and plunging), different outcomes are achieved. The optimized formula proposed by Poate et al. has almost the best performance in the case of spilling type and the formula of Power et al. is the most accurate one when considering the plunging type, showing that the type of wave breaking plays an important role in the accuracy of wave runup level prediction formulas.