Using Neural Network parameterizations of the Nonlinear Energy Transfer for Application in Wave Models

Abstract

LONG-TERM GOALSConsidering the crucial importance of nonlinear interaction S nl for the development of third generation wave models the long-term goal of this work is to improve accuracy of calculating nonlinear interaction S nl in wind wave models, and hence improving wave prediction in general. OBJECTIVESThe objective of this work is to develop a computationally cheap yet accurate approximation for S nl . APPROACHThe approach is based on the neural networks (NN) technique. It is used to accelerate the calculations and improve the accuracy of the parameterization of nonlinear interaction S nl . The nonlinear interaction source term can be considered as a nonlinear mapping between a source term S nl and a spectrum Fwhere T in is the exact nonlinear operator given by the full Bolzmann interaction integral [1]. This algorithm is a factor 10 4 too expensive for use in operational wave models. We intend to use the NN technique to produce a cheap and accurate alternative approach.