Iterative Integration Of Deep Learning In Hybrid Earth Surface System Modelling

This article proposes a hybrid Earth Surface System Modelling (ESSM)-Deep Learning (DL) framework to provide customized, scalable and accurate solutions to ESSM tasks. The framework is designed to intelligently guide the entire modelling lifecycle, and can self-renew and continuously improve its performance by incorporating user feedback from practical applications.