A new study published in Communications Earth & Environment gives possible insight into the underprediction of sea ice depletion and the formation of Arctic cyclones. Led by Steven Cavallo, a professor in the School of Meteorology at the University of Oklahoma, the study could lead to more accurate weather and climate models and better forecasting of Arctic cyclones.

Since 1979, Arctic sea ice extent, a term that describes the area of the Arctic Ocean covered by ice, has been reduced by 40% in the late summer months. Global climate models have consistently underpredicted that reduction. The study examines what Cavallo and his co-authors call “very rapid sea ice loss events,” or VRILEs. The sea ice decline since 1979 is the sum of numerous VRILEs that individually occur in periods of 5 to 18 days.

Cavallo’s publication suggests that Arctic cyclones are at least partially to blame. Arctic cyclones are weather phenomena that are tricky to predict and even trickier to incorporate into models. Though the exact mechanisms of why these cyclones may accelerate the ice loss are not fully understood, Cavallo suggests two theories. The first is the interaction of turbulent seas with ice.

“If the winds get strong and the ice is thin enough, [the cyclone] can create waves that break the larger ice floes. Breaking them up into smaller ice floes accelerates the melting; it can happen at a really fast time scale,” he said.

The second theory is that upwelling, the mixing of warmer water below the sea’s surface with cooler waters at the surface, increases temperatures that help melt the younger, thinner ice from below in a short amount of time.

Observations of these events and their effects are difficult. Ships avoid forecasted storms, and a plane could not fly into an Arctic cyclone close enough to the ocean’s surface to collect data on upwelling or wave–ice interactions.

Cavallo says they’ve discovered that cyclones have to be in the right place to make such a drastic difference to the sea ice extent, needing to occur over an area of thin ice that is usually no more than a year old.

The research also suggests a connection between Arctic cyclones and tropopause polar vortices, or circulation in the upper troposphere over the polar regions.

Cavallo said tropopause polar vortices are sometimes present for months before an Arctic cyclone forms, while Arctic cyclones are usually only predicted several days in advance. Because the vortices are present so far ahead of a cyclone, they could lead to better forecasts of cyclones. This would benefit residents in areas such as Alaska, northern Canada, and Greenland and aid the shipping industry, which has made increased use of the Arctic as ice continues to recede.

“Now that we think these processes are occurring, the question is how do we get that information into the models so that we can get better predictions,” said Cavallo. “It’s a hard task.”

Cavallo says that the broader scientific community is still unsure when the Arctic will become ice-free, but that looming lack of ice could significantly impact large-scale atmospheric dynamics throughout the Northern Hemisphere.

“We’re still trying to figure out exactly how sea ice changes will affect any of the extreme weather that is happening right now.”