FSU Researchers Can Now Predict Winter Weather Months Ahead

If you’ve ever been caught off guard by a brutal cold snap with only a few days’ warning, a new study from Florida State University could change that experience dramatically. FSU researchers have developed a method to accurately forecast winter weather conditions months before winter even begins — a leap that could reshape how industries and communities brace for extreme cold.

The study, published in the Journal of Geophysical Research: Atmospheres, centers on the stratospheric polar vortex, or SPV — a powerful band of high-altitude winds that encircles the Arctic during winter and acts as a containment barrier for frigid polar air. When that vortex is strong, bitter cold stays locked near the poles. When it weakens or becomes unstable, Arctic air can surge deep into North America and Eurasia, triggering the kind of historic cold events that have repeatedly caught populations unprepared.

“When the SPV is strong, that cold air tends to stay in the Arctic. When it is weak, cold air is more likely to spill southward into North America and Eurasia,” lead author Michael Secor, a recent doctoral graduate in meteorology from FSU’s Department of Earth, Ocean and Atmospheric Science, said in a news release. “The further in advance we can accurately predict the vortex, the further in advance we can help people and organizations prepare for weather conditions that affect agriculture, water management, energy use and public health.”

Until now, scientists could only reliably forecast SPV behavior about two weeks into the future. That window has long been a frustrating ceiling for meteorologists — accurate enough for immediate planning, but far too narrow for sectors that need to make infrastructure, crop or energy decisions weeks or months ahead of a major cold event.

A New Way of Looking at an Old Problem

Rather than continuing to refine short-term, day-to-day SPV forecasts, Secor approached the problem from a fundamentally different angle. Instead of asking what the vortex will do tomorrow, he asked what patterns govern its behavior across an entire year — and whether those broader patterns are more predictable than scientists assumed.

“Rather than trying to forecast the day-to-day evolution of the vortex, we start with the idea that its broader behavior over the course of the year may be more predictable,” Secor added. “We then use climate patterns such as the El Niño-Southern Oscillation, or ENSO, a temperature-based, recurring pattern in the Pacific Ocean known to influence the vortex, to predict those parameters in advance of winter. From there, we can work backward to reconstruct how the vortex will behave day to day, with an accuracy exceeding the current forecasting systems used by weather agencies.”

ENSO — which alternates between a warm phase called El Niño and a cold phase called La Niña — is a well-established driver of global weather. El Niño tends to bring cold, rainy conditions to the southern United States and suppresses Atlantic hurricane activity, while La Niña generates opposite effects. By incorporating ENSO data and other large-scale climate signals before winter arrives, Secor’s method can effectively work backward from those patterns to reconstruct how the polar vortex is likely to evolve, day by day, throughout the coming season.

Co-author Ming Cai, a professor in FSU’s Department of Earth, Ocean and Atmospheric Science, noted the implications extend beyond any single winter.

“This work shows that a large portion of subseasonal-to-seasonal variability is not random but embedded in the annual evolution of the climate system,” Cai said in the news release.

That insight is significant. It challenges the long-held assumption that weather beyond two weeks is essentially chaotic and unpredictable. Instead, the research suggests that much of what looks like randomness in winter weather is actually structured — driven by recurring, detectable patterns in the climate system that scientists can learn to read well before a single snowflake falls.

Why It Matters

The practical stakes are high. Farmers making planting and harvest decisions, utility companies managing energy demand, water resource managers anticipating drought or freeze events, and public health officials preparing for cold-related illness outbreaks all operate on timelines that far exceed the current two-week forecast window. A forecasting method that works months ahead doesn’t just improve the accuracy of weather apps — it could meaningfully reduce economic losses and protect lives.

The research is also notable because it may improve predictions of other climate phenomena with strong yearly cycles, including ENSO itself. If the methodology generalizes, it could open new doors in seasonal climate science far beyond winter storm forecasting.

Tallahassee’s record-breaking snowfall in January 2025 offered a vivid real-world reminder of what’s at stake when extreme winter events catch communities off guard. Events like that one — driven in part by SPV disruptions — underscore why extending forecast accuracy from two weeks to several months is more than an academic exercise.

Recognition and What’s Next

The paper earned an Editors’ Highlight from the American Geophysical Union, a distinction awarded to fewer than 2% of all papers published across the AGU’s journal portfolio. For Secor, who completed the work as part of his doctoral dissertation, the recognition carries personal weight.

“Publishing my dissertation work feels like reaching an important milestone in a journey that began with a fascination with weather at a young age,” Secor said. “It has made me reflect on how fortunate I have been to not only have this opportunity, but also to have people in my life who encouraged my scientific interest both early on and through my doctoral studies.”

Cai, who advised Secor throughout the project, highlighted the originality of the approach.

“Michael’s dissertation research, which represents a significant contribution for someone at this stage of his career, reflects not only his technical expertise but also the ability to rethink a long-standing problem from a fundamentally different perspective,” Cai said.

Source: Florida State University