LanTu: Dynamics-Enhanced Deep Learning for Eddy-Resolving Ocean Forecasting

TL;DR

LanTu is a novel deep learning system that incorporates ocean dynamics to improve mesoscale eddy forecasting, outperforming traditional models in accuracy and lead time for regional ocean predictions.

Contribution

This paper introduces LanTu, a dynamics-enhanced deep learning model that integrates multiscale physical constraints for improved eddy-resolving ocean forecasting.

Findings

LanTu outperforms existing numerical and AI-based systems in key ocean variables.

LanTu achieves forecast lead times of over 10 days.

Incorporating physical constraints improves deep learning forecast accuracy.

Abstract

Mesoscale eddies dominate the spatiotemporal multiscale variability of the ocean, and their impact on the energy cascade of the global ocean cannot be ignored. Eddy-resolving ocean forecasting is providing more reliable protection for fisheries and navigational safety, but also presents significant scientific challenges and high computational costs for traditional numerical models. Artificial intelligence (AI)-based weather and ocean forecasting systems are becoming powerful tools that balance forecast performance with computational efficiency. However, the complex multiscale features in the ocean dynamical system make AI models still face many challenges in mesoscale eddy forecasting (especially regional modelling). Here, we develop LanTu, a regional eddy-resolving ocean forecasting system based on dynamics-enhanced deep learning. We incorporate cross-scale interactions into LanTu and construct multiscale physical constraint for optimising LanTu guided by knowledge of eddy dynamics in order to improve the forecasting skill of LanTu for mesoscale evolution. The results show that LanTu outperforms the existing advanced operational numerical ocean forecasting system (NOFS) and AI-based ocean forecasting system (AI-OFS) in temperature, salinity, sea level anomaly and current prediction, with a lead time of more than 10 days. Our study highlights that dynamics-enhanced deep learning (LanTu) can be a powerful paradigm for eddy-resolving ocean forecasting.