Predicting sea surface temperatures with coupled reservoir computers

Abstract. Sea surface temperature (SST) is a key factor in understanding the greater climate of Earth, and it also an important variable when making weather predictions. Methods machine learning have become ever more present data-driven science engineering, including areas for Earth science. Here, we propose efficient framework that allows us to make global SST forecasts using coupled reservoir computer method specialized this domain, allowing template regions accommodate irregular coastlines. Reservoir computing especially good forecasting spatiotemporally complex dynamical systems, as that, despite many randomly selected weights, highly accurate easy train. Our approach provides benefit simple computationally model able predict SSTs across entire Earth's oceans. The results are demonstrated generally follow actual dynamics system over period several weeks.