Deep Learning-Derived Long-Term Dataset of Internal Waves in the Northern South China Sea from Satellite Imagery

Abstract

Abstract. Internal waves (IWs) are an important ocean process in transmitting energy between multiscale ocean dynamics, making them a crucial oceanic phenomenon. The South China Sea (SCS) is renowned for its frequent large-amplitude IW activities, emphasizing the importance of collecting and analyzing extensive observational data. In this study, we present a comprehensive IW dataset covering the northern SCS covering 112.40–121.32° E and 18.32–23.19° N, spanning from 2000 to 2022 with a 250 m spatial resolution. The IW dataset comprises 3085 high-resolution MODIS true-color IW images paired with precise IW position information extracted from 15830 MODIS images using advanced deep learning techniques. IWs in the northern SCS are divided into four regions based on extracted IW spatial distributions, facilitating detailed analyses of IW characteristics, including spatial and temporal distributions across both the entire northern SCS and its sub-regions. Notably, we uncover typical "double-peak" distributions corresponding to the lunar day, underscoring IWs' close relationship with tides. Furthermore, we identify two IW-free silence regions attributed to underwater topography influences, indicating varied IW characteristics across regions and suggesting underlying mechanisms warrant further investigation. The constructed dataset holds significant potential for applications in studying IW-environment interactions, developing monitoring and prediction models, validating and enhancing numerical simulations, and serving as an educational resource to foster awareness and interest in IW research.