Atmospheric Rivers as Interacting Elements of the Earth System: A Complexity Science Perspective

The critical role of atmospheric rivers (ARs) in the global water cycle, along with their intensification under global warming, underscores the urgency of understanding and predicting their dynamics and impacts at both regional and global scales. Despite significant advances, this endeavor remains challenging because ARs lie at the interface of weather and climate. These synoptic-scale systems produce short-term, localized impacts while shaping long-term global patterns of moisture, wind, and precipitation. AR genesis and evolution emerge from interactions within the coupled ocean-atmosphere system, while AR-induced precipitation can lead to natural disasters through land-atmosphere interactions. By transporting vast amounts of moisture over great distances, ARs establish teleconnections that influence weather across thousands of kilometers. At the same time, their activity is shaped by large-scale climate phenomena such as the El Niño–Southern Oscillation and the Pacific Decadal Oscillation. Advancing AR science, therefore, requires treating ARs as integral components of the Earth system and unraveling their interactions across a broad range of spatial and temporal scales.

In this talk, we present and discuss the paradigm of complexity science and the exciting opportunities it offers for advancing AR science. Building on a solid foundation of dynamical systems, stochastic climate theory, and network theory, complexity science integrates nonlinearities, feedbacks, and uncertainties into the study of ARs. By employing novel methods such as event synchronization, climate networks, and probabilistic causation, complexity science provides powerful tools to investigate non-local interactions, uncover hidden dynamics, and refine impact attribution in AR research. To ensure the robustness of findings, complexity science integrates null models, hypothesis testing, confidence bounds, and sensitivity analyses. Emerging research avenues, such as AR networks, community detection, low-order modeling, and tipping dynamics, can now be explored through the lens of complexity science. By establishing a rigorous theoretical and methodological foundation, complexity science paves the way for innovative research on AR dynamics, impacts, and prediction.