Numerical modelling of the equilibrium profile in Valencia (Spain)

In this paper, mathematical models for obtaining the parameter A of the potential function, which describes the equilibrium beach profile, were generated. The aim of the research was to analyse the most important variables related to the equilibrium beach profile, so that the models are used only those variables that are easy to obtain. On the other hand, the sensitivity of the models must be taken into account, since small changes in variables can cause significant deviations in the models results. For this reason, a detailed research of the sensitivity of the studied models in relation to the different variables has been performed. The parameters considered for the evaluation and selection of the optimal model are: the volume error, the Pearson coefficient (R2), the relative mean square error (MSE/Var), the mean error (ϵ) and the relative percentage error (δ) and the model stability under changes in the input variables of up to 20% by a Monte Carlo simulation. The best models have been validated using data corresponding to beaches from outside the area where the dataset has been collected.