photo by Paulo Brando
A growing body of research is examining disruptive and abrupt shifts in weather extremes. Contributing to this work, the research team has developed a novel method to identify so-called weather whiplash events.
The team defines a weather whiplash event as a long-lived (4 or more consecutive days), continental-scale pattern in the upper-level circulation of the atmosphere that shifts abruptly (over 1-3 days) to a substantially different pattern, bringing a stark end to persistent weather conditions throughout the region.
This definition eliminates the possibility of misidentifying sharp, localized weather changes caused by features like fronts, discrete disturbances like tropical storms, and changes in low-level winds (e.g., shifting from onshore to offshore or from downslope to upslope).
To identify weather whiplash events, the team is using an AI pattern recognition tool called Self-Organizing Maps (SOMs). Unlike methods used in previous studies, this new approach does not rely on measurements or simulations of precipitation or temperature, so it avoids uncertainties introduced by instrument error, local influences, and deficiencies in models.
This tool allows us to analyze real-world data to identify weather whiplash events in the past. Then, the team applies the same approach to modeled simulations of past weather to see how well those models capture whiplash events, and to future modeled projections to investigate how the frequency of weather whiplash events changes under different emissions scenarios.
The research team is currently applying this method to three large (120° longitude) sections of the northern hemisphere—the northeast Pacific Ocean/North America, the North Atlantic Ocean/Europe, and Asia.
This work is supported by Woodwell Climate’s Fund for Climate Solutions.