Explicit wave overtopping formula for mound breakwaters with crown walls using CLASH neural network-derived data

Based on the Crest Level Assessment of Coastal Structures (CLASH) Neural Network Overtopping prediction method, a new 16-parameter overtopping estimator (Q6) was developed for conventional mound breakwaters with crown walls, both with and without toe berms. Q6 was built up using the overtopping estimations given by the CLASH Neural Network and checked using the CLASH database. Q6 was compared to other conventional overtopping formulas, and the Q6 obtained the lowest prediction errors. Q6 provides overtopping predictions similar to the CLASH Neural Network for conventional mound breakwaters but using only six explanatory dimensionless variables (Rc=Hm0; Ir; Rc=h;Gc=Hm0; Ac=Rc, and a toe berm variable based on Rc=h) and two reduction factors (g f and g b ). Q6 describes explicit relationships between input variables and overtopping discharge, and hence it facilitates use in engineering design to identify costeffective solutions and to quantify the influence of variations in wave and structural parameters.The authors are grateful for financial support from the Spanish Ministerio de Economia y Competitividad (Grant BIA2012-33967). The first author was funded through the FPU program (Formacion del Profesorado Universitario, Grant AP2010-4366) by the Spanish Ministerio de Educacion, Cultura y Deporte. The authors also thank Debra Westall for revising the manuscript.Molines, J.; Medina, JR. (2016). Explicit wave overtopping formula for mound breakwaters with crown walls using CLASH neural network-derived data. Journal of Waterway Port Coastal and Ocean Engineering. 142(3). https://doi.org/10.1061/(ASCE)WW.1943-5460.0000322142