Description:

Modern causal discovery took shape in the 1950s and 1960s, meaning the field is not even 70 years old. In that short time, however, it has become more concrete and has revolutionized many scientific fields. Despite the revolution in other fields, atmospheric and oceanic science, and in general earth science, has only recently applied causal discovery techniques. This has led to some groundbreaking work, such as showing the directionality of feedbacks in atmosphere-sea ice interactions, revealing tropospheric and stratospheric pathways between the Madden Julian Oscillation and the North Atlantic Oscillation, and evidencing what drives tropical cyclone rapid intensification, to name a few applications.

Current applications of causal discovery to atmospheric and oceanic science are the intended focus of this session. The session will start with an invited review of literature and history of causal discovery in atmospheric and oceanic science to set the scene and facilitate subsequent talks. We invite talks on the whole spectrum of causal discovery, from new ways of causal reasoning incorporating linear, nonlinear, or machine learning-based causal frameworks, to new scientific discoveries based on applying causal discovery methodologies to pressing problems in atmospheric or oceanic sciences.