Super-resolution of subsurface temperature field from remote sensing observations based on machine learning

Hua Su, An Wang, Tianyi Zhang, Tian Qin, Xiaoping Du, Xiao-Hai Yan

Abstract: Highlights•A new deep neural network approach to reconstruct subsurface temperature.•Super-resolution of subsurface temperature from 1° to 0.25° from remote sensing.•CNN outperformed LightGBM in the case of big training samples.•Higher-resolution data support for global ocean warming and internal variability study.