A DEEP LEARNING MODEL TO PREDICT SHORELINE CHANGE

Abstract

As coastal population increases, so does the risk for social and economic losses under a changing climate. To assess future changes, much progress has been made towards developing shoreline numerical models, although producing reliable shoreline change predictions remains a challenge (Montaño 2020). Here we present a Deep Learning (DL) model to predict long-term shoreline evolution due to waves and large-scale atmospheric patterns. The model is based on two types of Artificial Neural Networks: Long-Short Term Memory (LSTM) networks and Convolutional Neural Networks (CNN).