A Space-Time Partial Differential Equation Based Physics-Guided Neural Network for Sea Surface Temperature Prediction

Abstract

Sea surface temperature (SST) prediction has attracted increasing attention, due to its crucial role in understanding the Earth’s climate and ocean system. Existing SST prediction methods are typically based on either physics-based numerical methods or data-driven methods. Physics-based numerical methods rely on marine physics equations and have stable and explicable outputs, while data-driven methods are flexible in adapting to data and are capable of detecting unexpected patterns. We believe that these two types of method are complementary to each other, and their combination can potentially achieve better performances. In this paper, a space-time partial differential equation (PDE) is employed to form a novel physics-based deep learning framework, named the space-time PDE-guided neural network (STPDE-Net), to predict daily SST. Comprehensive experiments for SST prediction were conducted, and the results proved that our method could outperform the traditional finite-difference forecast method and several state-of-the-art deep learning and physics-guided deep learning methods.

Spatiotemporal Fusion Prediction of Sea Surface Temperatures Based on the Graph Convolutional Neural and Long Short-Term Memory Networks

“…The average values of the nodes of the entire graph can be realized by left multiplying ÃX by the inverse matrix D−1 of the degree matrix. This can be represented by Equation (7).…”

“…Through SST prediction, we can effectively understand ocean dynamics, which will enable us to adequately prepare for such challenges. However, the existing numerical prediction models require a profound understanding of and the ability to replicate the physical evolution of SST [7] to build a complicated model; these rely heavily on the accuracy of initial parameters, and it is difficult to capture the complex physical evolution accurately, which limits the development and accuracy of SST prediction. Along with the continuous advancement of deep learning technology, the data-driven modeling strategy has emerged as a powerful complement to numerical prediction models.…”