Combining Coupled Modelling and Machine Learning to Constrain Antarctica's Uncertain Future

While the latest technical summary from the Intergovernmental Panel for Climate Change anticipates several tens of centimeters of sea level rise by the end of the 21st century, they also state that "In a low-likelihood, high-impact storyline and a high CO2 emissions scenario, ice-sheet processes characterized by deep uncertainty could drive global mean sea level rise up to about 5 m by 2150". Such a high increase in sea level would be globally catastrophic, but how likely is it?

Our consortium brings together Finland's leading experts in glaciology, statistics, high performance computing and machine learning to address this question. We will do this by simulating the complex networks of interactions between the Antarctic Ice Sheet and it's environment. We will combine computer models for these interactions with efficient machine learning techniques to estimate the possible range of Antarctica's global sea level contribution.