Probabilistic Machine Learning of the Natural Variability of Climate

Because of natural or internal variability, the behavior of processes ranging from unresolved small-scale physical and dynamical processes to the response of the climate system at the largest scales is probabilistic rather than denterministic. Indeed, it is also the case that while climate models are skilful at predicting the response of the climate system to external forcing, they are less skilful when it comes to predicting natural variability. A variety of probabilistic machine learning techniques ranging from Reservoir Computing to Generative Adversarial Networks to Bayesian Neural Networks are considered in the context of modeling natural variability. At the large scales, these models are seen to improve upon the Linear Inverse Modeling (LIM) approach which has itself been sometimes thought of as capturing the bulk of the predictable component of natural variability.