Time-Series Graph Network for Sea Surface Temperature Prediction

Abstract Sea surface temperature (SST) is an important indicator for balancing surface energy and measuring sea heat. Various effects caused by the sea temperature field significantly affect human activities to a large extent. It is important to predict the sea surface temperature efficiently and accurately. Existing SST prediction methods regard each sensor as an independent individual, without considering the structure or topology of the sensor network and ignoring the connectivity of the sensor network. To address this problem, this study investigates the SST prediction problem from the perspective of graph learning. We propose a time-series graph network (TSGN) that can jointly capture graph-based spatial correlation and temporal dynamics. TSGN uses a long short-term memory network to aggregate the features of time series data and establishes a graph neural network model to complete the SST prediction task. Experiments using SST data from the Pacific Northwest from 2001 to 2005 show that this method is more efficient and accurate when dealing with data containing time-series information and is superior to the existing SST prediction methods.

[1]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[2]  Yong Liu,et al.  Contextualized Point-of-Interest Recommendation , 2020, IJCAI.

[3]  Ah Chung Tsoi,et al.  The Graph Neural Network Model , 2009, IEEE Transactions on Neural Networks.

[4]  Ye Yuan,et al.  Big graph classification frameworks based on Extreme Learning Machine , 2019, Neurocomputing.

[5]  Anthony J Bishara,et al.  Testing the significance of a correlation with nonnormal data: comparison of Pearson, Spearman, transformation, and resampling approaches. , 2012, Psychological methods.

[6]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[7]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[8]  Chao Zhang,et al.  Explainable time-frequency convolutional neural network for microseismic waveform classification , 2021, Inf. Sci..

[9]  V. M. Mendoza,et al.  Numerical experiments on the prediction of sea surface temperature anomalies in the Gulf of Mexico , 1997 .

[10]  Enrique López Droguett,et al.  Prediction of sea surface temperature in the tropical Atlantic by support vector machines , 2013, Comput. Stat. Data Anal..

[11]  P. Nobre,et al.  Statistical prediction of sea‐surface temperature over the tropical Atlantic , 2004 .

[12]  Zhang Li-xu,et al.  The change trend of happened frequency and the R/S forecast of frequently happened year for red tide in China , 2010 .

[13]  Yunxiang Zhang,et al.  Middle Pleistocene mammalian fauna of Shanyangzhai cave in Qinhuangdao area, China and its zoogeographical significance , 2010 .

[14]  Ghim Ping Ong,et al.  GraphSAGE-Based Traffic Speed Forecasting for Segment Network With Sparse Data , 2020, IEEE Transactions on Intelligent Transportation Systems.

[15]  Alessio Micheli,et al.  Neural Network for Graphs: A Contextual Constructive Approach , 2009, IEEE Transactions on Neural Networks.

[16]  S. Aparna,et al.  Prediction of daily sea surface temperature using artificial neural networks , 2018 .

[17]  Gerhard Nahler,et al.  Pearson Correlation Coefficient , 2020, Definitions.

[18]  Wojciech Zaremba,et al.  Recurrent Neural Network Regularization , 2014, ArXiv.

[19]  Panos Kalnis,et al.  GCN-MF: Disease-Gene Association Identification By Graph Convolutional Networks and Matrix Factorization , 2019, KDD.

[20]  Junyu Dong,et al.  Prediction of Sea Surface Temperature Using Long Short-Term Memory , 2017, IEEE Geoscience and Remote Sensing Letters.

[21]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[22]  Davide Bacciu,et al.  Contextual Graph Markov Model: A Deep and Generative Approach to Graph Processing , 2018, ICML.

[23]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[24]  Philip S. Yu,et al.  A Comprehensive Survey on Graph Neural Networks , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[25]  Jure Leskovec,et al.  Inductive Representation Learning on Large Graphs , 2017, NIPS.

[26]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[27]  Kenli Li,et al.  Gated Residual Recurrent Graph Neural Networks for Traffic Prediction , 2019, AAAI.

[28]  F. Scarselli,et al.  A new model for learning in graph domains , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[29]  Hao Liu,et al.  GNEA: A Graph Neural Network with ELM Aggregator for Brain Network Classification , 2020, Complex..

[30]  Joan Bruna,et al.  Spectral Networks and Locally Connected Networks on Graphs , 2013, ICLR.