Purely satellite data–driven deep learning forecast of complicated tropical instability waves

Forecasting fields of oceanic phenomena has long been dependent on physical equation–based numerical models. The challenge is that many natural processes need to be considered for understanding complicated phenomena. In contrast, rules of the processes are already embedded in the time-series observation itself. Thus, inspired by largely available satellite remote sensing data and the advance of deep learning technology, we developed a purely satellite data–driven deep learning model for forecasting the sea surface temperature evolution associated with a typical phenomenon: a tropical instability wave. During the testing period of 9 years (2010–2019), our model accurately and efficiently forecasts the sea surface temperature field. This study demonstrates the strong potential of the satellite data–driven deep learning model as an alternative to traditional numerical models for forecasting oceanic phenomena. Deluges of satellite-derived ocean products not only provide an unprecedented golden opportunity for in-depth research but also demonstrate the urgent need to develop useful methods to explore time-series observation. Sea surface temperature (SST) is one of the ocean products that has the most extended history and a critical parameter to help people understand scientific questions in physical/biological oceanography and atmosphere-ocean interaction. Since becoming relatively easy to measure from space with high accuracy, SST has been widely used to reveal various critical oceanic phenomena, e.g., tropical instability wave (TIW) (1). Traditional statistical analyses in previous studies have limitations in model complexity in addressing events that are complicated by nature.