A Multivariable Convolutional Neural Network for Forecasting Synoptic-Scale Sea Surface Temperature Anomalies in the South China Sea

Abstract

The sea surface temperature anomaly (SSTA) plays a key role in climate change and extreme weather processes. Usually, SSTA forecast methods consist of numerical and conventional statistical models, the former can be seriously influenced by the uncertainty of physical parameterization schemes, the nonlinearity of ocean dynamic processes, and the nonrobustness of numerical discretization algorithms. Recently, deep learning has been explored to address forecast issues in the field of oceanography. However, existing deep learning models for ocean forecasting are mainly site-specific, which were designed for forecasting on a single point or for an independent variable. Moreover, few special deep learning networks have been developed to deal with SSTA field forecasts under typhoon conditions. In this study, a multivariable convolutional neural network (MCNN) is proposed, which can be applied for synoptic-scale SSTA forecasting in the South China Sea. In addition to the SSTA itself, the surface wind speed and the surface current velocity are regarded as input variables for the prediction networks, effectively reflecting the influences of both local atmospheric dynamic forcing and nonlocal oceanic thermal advection. Experimental results demonstrate that MCNN exhibits better performance than single-variable convolutional neural network (SCNN), especially for the SSTA forecast during the typhoon passage. While forecast results deteriorate rapidly in the SCNN during the passage of a typhoon, forecast errors in the MCNN can be effectively restrained to slowly increase over the forecast time due to the introduction of the surface wind speed in this network.

Sea Surface Temperature and Marine Heat Wave Predictions in the South China Sea: A 3D U-Net Deep Learning Model Integrating Multi-Source Data

“…This model exhibited strong performance in both normal and extreme weather conditions. Recently, Miao et al [39] also reached similar conclusions. Based on a multivariate CNN model, they used SSTA, wind speeds, and surface current velocity as input variables to predict SSTA, achieving more accurate forecasts.…”