Towards Spatio-temporal Sea Surface Temperature Forecasting via Dynamic Personalized Graph Network

Sea surface temperature (SST) is uniquely important to the Earth’s atmosphere since its dynamics are a major force in shaping local and global climate and profoundly affect our ecosystems. Accurate forecasting of SST brings significant economic and social implications, for example, better preparation for extreme weather such as severe droughts or tropical cyclones months ahead. However, such a task faces unique challenges due to the intrinsic complexity and uncertainty of ocean systems. Recently, deep learning techniques, such as graphical neural networks (GNN), have been applied to address this task. While such techniques achieve certain levels of success, they often have significant limitations in exploring dynamic spatio-temporal dependencies between signals. To solve this problem, this paper proposes a novel graph convolution network architecture with static and dynamic learning layers for SST forecasting. Specifically, two adaptive adjacency matrices are firstly constructed to respectively model the stable long-term and short-term evolutionary patterns hidden in the multivariate SST signals. Then, a personalized convolution layer is designed to fuse these information. The developed network can be learned in an end-to-end manner. Our experiments on real SST datasets demonstrate the state-of-the-art performances of the proposed approach on the forecasting task.