Chinese researchers have built an artificial intelligence (AI) inference and prediction system for the Indonesian Throughflow (ITF), which can make valid ocean current predictions seven months in advance.

The ITF serves as a main branch of the global heat/salt conveyor belt by transferring the upper ocean waters from the Pacific to the Indian Ocean through the Indonesian seas. Researchers had previously lacked ideal ways to predict the ITF, due to the model biases in the current numerical simulation systems.

In this new study, Chinese researchers from the Institute of Oceanology, Chinese Academy of Sciences and Nanjing University of Information Science & Technology used satellite data and AI models based on deep learning to construct the inference and prediction system for the ITF.

The researchers used sea surface heights between the Indian and Pacific Ocean basins to design their AI model and trained the model with oceanic data sets.

They then inputted satellite data from 1993 to 2021 into the system to reproduce the ITF during this period. The results were highly consistent with internationally acknowledged ITF field observation data. Meanwhile, the AI system can also make a valid prediction seven months in advance.

The researchers have reported the system in the journal Frontiers in Marine Science. They said that the system may provide a new tool for studying ocean circulation and climate change in the Indo-Pacific Ocean and ease the pressure of real-time oceanographic observation.