Wave height prediction at the Caspian Sea using a data-driven model and ensemble-based data assimilation methods

摘要

There are successful experiences with the application of ANN and ensemble-based datanassimilation methods in the field of flood forecasting and estuary flow. In the present work, thencombination of dynamic Artificial Neural Network and Ensemble Kalman Filter (EnKF) is applied onnwind-wave data. ANN is used for the time propagation mechanism that governs the timenevolution of the system state. The system state consists of the significant wave height that isnaffected by wind speed and wind direction. The relevant inputs are selected by analysing thenAverage Mutual Information. By help of the observations, the EnKF will correct the output of thenANN to find the best estimate of the wave height. A combination of ANN with EnKF acts as annoutput correction scheme. To deal with the time-delayed states, the extended state vector isntaken and the dynamic equation of the extended state vector is used in EnKF. Application of thenproposed scheme is examined by using five-month hourly buoy measurement at the Caspian Seanand several model runs with different assimilation–forecast cycles.The coefficient of performancenand root mean square error are used to access performance of the method.