Revealing the impact of climate change on North Atlantic circulation using transparent machine learning
Abstract
The North Atlantic ocean is key to climate through its role in heat transport and storage. Climate models suggest that the circulation is weakening but the physical drivers of this change are poorly constrained. Here, the root mechanisms are revealed with the explicitly transparent machine learning (ML) method Tracking global Heating with Ocean Regimes (THOR, figure below), applied to the North Atlantic ocean. Addressing the fundamental question of the existence of dynamical coherent regions, THOR identifies these and their link to distinct currents and mechanisms such as the formation regions of deep water masses, and the location of the Gulf Stream and North Atlantic Current. Beyond a black box approach, THOR is engineered to elucidate its source of predictive skill rooted in physical understanding. A labeled data set is engineered using an explicitly interpretable equation transform and k-means application to model data, allowing theoretical inference. A multilayer perceptron is then trained, explaining its skill using a combination of layerwise relevance propagation and theory. With abrupt CO2 quadrupling, the circulation weakens due to a shift in deep water formation regions, a northward shift of the Gulf Stream and an eastward shift in the North Atlantic Current. If CO2 is increased 1% yearly, similar but weaker patterns emerge influenced by natural variability. THOR uses the ocean depth, dynamic sea level and wind stress as inputs, and is readily applicable to other models with a portable and containerized code. THOR can be run on select CMIP6 data in the cloud, and is seen as a potential blueprint for accelerated analysis. Trustworthy ML is called for within oceanography and beyond, and THORs' transparency is a step towards this as its predictions are physically tractable. Figure caption: Sketch of THOR workflow. Method to identify dynamical regimes that are indicative of dynamics contributing to the AMOC variability. THOR is engineered for interpretability and explainability of ML predictive skill for transparent, and as such to move towards trustworthy ML.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.A14C..06S