Revealing the Impact of Global Heating on North Atlantic Circulation Using Transparent Machine Learning

The North Atlantic ocean is key to climate through its role in heat transport and storage. Climate models suggest that the circulation weakening but physical drivers of this change are poorly constrained. Here, root mechanisms revealed with explicitly transparent machine learning (ML) method Tracking global Heating Ocean Regimes (THOR). Addressing fundamental question existence dynamical coherent regions, THOR identifies these their link distinct currents such as formation regions deep water masses, location Gulf Stream Current. Beyond a black box approach, engineered elucidate source predictive skill rooted understanding. A labeled data set using an interpretable equation transform k-means application model data, allowing theoretical inference. multilayer perceptron then trained, explaining combination layerwise relevance propagation theory. With abrupt CO2 quadrupling, weakens due shift northward eastward If increased 1% yearly, similar weaker patterns emerge influenced by natural variability. scalable applicable range only depth, dynamic sea level wind stress, could accelerate analysis dissemination data. constitutes step toward trustworthy ML called for within oceanography beyond, predictions physically tractable.