A domestic research team developed an artificial intelligence (AI) model that learns 160-plus years of global ocean data and predicts ocean conditions for the next 200 days in a matter of seconds. Because it enables long-term ocean forecasting with far fewer computing resources than conventional supercomputer-based numerical models, it is drawing attention as a foundational technology to improve responses to climate change and the accuracy of long-range climate prediction.
The Korea Institute of Science and Technology (KIST) Climate and Carbon Cycle Research Center team led by Doctors Kang Dae-hyeon and Kim Jeong-hwan said on the 26th it developed an AI-based global three-dimensional ocean prediction model, "KIST-Ocean." The findings were published in the international journal Science Advances.
As climate change increases the frequency of extreme weather such as heat waves, downpours, droughts and typhoons, the importance of long-term climate prediction is growing. In particular, the ocean, which covers about 70% of Earth's surface, stores and circulates vast amounts of heat and carbon and plays a key role in climate variability such as El Niño and La Niña.
However, existing ocean prediction models must compute complex physical equations, requiring massive supercomputing resources and long run times. This has limited rapid forecasting and the analysis of diverse climate change scenarios.
To overcome these limitations, the team pre-trained on global large-ensemble simulation data from 1850 to 2014 generated by the Community Earth System Model 2 (CESM2), and then fine-tuned a Deep Learning-based AI model, KIST-Ocean, to reflect ocean characteristics using ocean reanalysis data from 1982 to 2013.
KIST-Ocean uses 62 ocean variables—including not only sea surface temperature but also water temperature, currents and salinity—to predict the three-dimensional ocean state down to a depth of 600 meters at five-day intervals. It is also designed to incorporate six surface boundary conditions transmitted from the atmosphere to the ocean, including wind stress, longwave and shortwave radiation, and latent and sensible heat fluxes.
The team said KIST-Ocean completed pre-training in about 33 hours and fine-tuning in about 2.4 hours using just one NVIDIA A100 GPU, and generated a 200-day global ocean forecast in 6 to 7 seconds, sharply reducing computation time and expense compared with conventional numerical models.
In virtual experiments verifying how accurately it reproduces real ocean physics, KIST-Ocean produced ocean waves and upwelling/downwelling driven by atmospheric changes consistent with established ocean physics theory. In the 2015 super El Niño case, it also successfully reproduced the rise in sea surface temperatures in the equatorial Pacific and changes in internal ocean heat distribution.
The team said the results show KIST-Ocean goes beyond merely learning past data and can effectively capture the complex physical interactions between the atmosphere and the ocean.
In experiments that input actual atmospheric conditions, KIST-Ocean achieved higher accuracy in 200-day forecasts than the "persistence forecast," which assumes the current ocean state remains unchanged. For sea surface temperature prediction, it outperformed the North American Multi-Model Ensemble (NMME) for up to six months.
The team plans to advance the current ocean-only model into an AI-based Earth system model integrating the atmosphere, ocean, sea ice and land. They expect this will enable not only long-term climate prediction but also faster and lower-expense analysis of diverse climate change scenarios.
Kang Dae-hyeon said, "This study shows that AI can not only deliver outstanding computational efficiency but also realistically reproduce the complex physical relationships between the atmosphere and the ocean," adding, "It will lay the groundwork for developing AI-based Earth system models and strengthening capabilities to respond to the climate crisis."