Imagine an interpreter in a large hall who hears all conversations at once and, based on them, understands what will be talked about in an hour. That's how the CAS-Canglong neural network works: it looks at pressure, wind, and heat across the entire Earth and predicts ocean temperature three months ahead.
Previously, making a forecast for such a period was very difficult. The atmosphere changes quickly, while the ocean slowly accumulates heat. Physical models require enormous computers and often make mistakes. The new program learned from historical weather records and now makes forecasts in seconds.
In southern oceans, the forecast matches reality 90% of the time for one month ahead and 70% for three months. This means that if the program says the water is warming, it most likely will. Such forecasts can help farmers, energy companies, and emergency services prepare for trouble in advance.
The model's name, Canglong, refers to a mythical dragon from ancient Chinese mythology that controls rain. The model was trained on data from 1959 onward; over that period, ocean temperature in the tropics has risen by about 0.5 °C.
In Kim Stanley Robinson's novel 'The Ministry for the Future', heat in India becomes a trigger for global change. Accurate ocean temperature forecasts months ahead could provide time to prepare for such events.