Rogue Waves in the Drake Passage: Unpredictable Hazard

Didenkulova, Ekaterina; Talipova, Tatiana; Pelinovsky, Efim

doi:10.1007/978-3-030-78927-5_7

Cited by 1 publication

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typically, research focuses on linear superposition in finite-bandwidth seas (12), wave breaking (13), and wave-wave interactions in weakly nonlinear seas (14,15) or through the modulational instability (16). Apart from these universal mechanisms, there are also countless possible interactions with localized features such as non-uniform topography (17), wave-current interactions like in the Agulhas (18) or the Antarctic Circumpolar Current (19), or crossing sea states at high crossing angles affecting wave breaking (20). We call this set of mechanisms the physical effects Φ.…”

Section: A Causal Graph For Rogue Wave Generationmentioning

confidence: 99%

Machine-guided discovery of a real-world rogue wave model

Häfner,

Gemmrich,

Jochum

2023

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Big data and large-scale machine learning have had a profound impact on science and engineering, particularly in fields focused on forecasting and prediction. Yet, it is still not clear how we can use the superior pattern-matching abilities of machine learning models for scientific discovery. This is because the goals of machine learning and science are generally not aligned. In addition to being accurate, scientific theories must also be causally consistent with the underlying physical process and allow for human analysis, reasoning, and manipulation to advance the field. In this paper, we present a case study on discovering a symbolic model for oceanic rogue waves from data using causal analysis, deep learning, parsimony-guided model selection, and symbolic regression. We train an artificial neural network on causal features from an extensive dataset of observations from wave buoys, while selecting for predictive performance and causal invariance. We apply symbolic regression to distill this black-box model into a mathematical equation that retains the neural network’s predictive capabilities, while allowing for interpretation in the context of existing wave theory. The resulting model reproduces known behavior, generates well-calibrated probabilities, and achieves better predictive scores on unseen data than current theory. This showcases how machine learning can facilitate inductive scientific discovery and paves the way for more accurate rogue wave forecasting.

show abstract