The unusually warm sea surface temperature events known as marine heatwaves (MHWs) have a profound impact on marine ecosystems. Accurate prediction of extreme MHWs has significant scientific and financial worth. However, existing methods still have certain limitations, especially in the most extreme MHWs. In this study, to address these issues, based on the physical nature of MHWs, we created a novel deep learning neural network that is capable of accurate 10-day MHW forecasting. Our framework significantly improves the forecast ability of extreme MHWs through two specially designed modules inspired by numerical models: a coupler and a probabilistic data argumentation. The coupler simulates the driving effect of atmosphere on MHWs while the probabilistic data argumentation approaches significantly boost the forecast ability of extreme MHWs based on the idea of ensemble forecast. Compared with traditional numerical prediction, our framework has significantly higher accuracy and requires fewer computational resources. What's more, explainable AI methods show that wind forcing is the primary driver of MHW evolution and reveal its relation with air-sea heat exchange. Overall, our model provides a framework for understanding MHWs' driving processes and operational forecasts in the future.

Following my previous post "New Tools at the Horizon", one question was twirling in my mind: why the stock market is forecastable, but the forecasts are not affordable? The forecastability of the market is an evidence, because if it were not – being it just a random walk – there would not be the possibility to have an output from the neural networks that manage the forecast process. For a neural network to work, there must be some sort of structure inside tha data that can be used to produce the forecast/diagnosis. And this hidden structure is present indeed inside the market data, otherwise r.Virgeel would be totally blind and dumb. This is a sample chart of a blind network: not structure is evaluated and the output is just an array of zero values.