Abrupt collapse of Arctic sea ice may come with little warning

Climate change often appears gradual, but abrupt shifts may occur with little warning. Recent research reveals that sensitive subsystems – such as sea ice, permafrost, and rainforests – can suddenly tip into new states.

These changes pose serious risks to ecosystems, water resources, and human populations.

“Abrupt shifts” refer to rapid, large changes within a decade or less. These include sudden snow loss on the Tibetan Plateau, permafrost thaw, or sea ice collapse. Once triggered, these events may unfold faster than society can adapt.

How scientists found fast changes

To detect these changes, the researchers applied Canny edge detection, a method from computer vision, to climate model data. The team scanned 82 environmental variables across 57 CMIP6 climate models.

The goal was to identify sharp changes in variables like sea surface salinity, snow cover, and carbon storage.

Each model simulated a future where carbon dioxide rose by one percent per year for 140 years, reaching four times the pre-industrial level.

Many models show sudden changes

The study from Utrecht University found that vulnerable regions included the Arctic, the Tibetan Plateau, and the North Atlantic subpolar gyre. In contrast, the Indian, West African, and South American monsoons remained relatively stable across most models.

Subsystems showing more consistent abrupt changes included Arctic summer and winter sea ice, Antarctic sea ice, the North Atlantic subpolar gyre, land permafrost, and snow cover over the Tibetan Plateau.

Even systems not officially recognized as tipping elements experienced abrupt changes in several models, suggesting wider potential for sudden climate shifts than previously thought.

Fast cooling in the North Atlantic

Multiple climate models are detecting early signs of abrupt shifts in major Earth systems. In the North Atlantic subpolar gyre, surface waters are freshening and cooling, weakening ocean convection and disrupting currents like the AMOC.

Twenty-four models showed abrupt temperature or salinity changes here – sometimes at just 0.5°C above preindustrial levels.

Over the Tibetan Plateau, 12 models showed rapid changes in snow-related variables. Upwelling shortwave radiation – a stand-in for albedo – shifted in 11 models, alongside sharp declines in snow depth and coverage.

These changes could foreshadow glacier retreat, though most models don’t fully simulate glaciers.

Permafrost and forest changes

Permafrost in northern Asia may also thaw faster than expected. Nineteen models showed abrupt losses of frozen soil moisture across areas totaling over one million square kilometers (386,000 square miles), suggesting thawing could happen regionally even if localized shifts appear minor.

Even forests may not be immune. While slow to respond and harder to model, the Amazon and boreal forests showed sudden changes in vegetation indicators like leaf area index and carbon fluxes in up to five models, hinting at potential tipping points in these ecosystems.

Abrupt sea ice changes

Sea ice showed some of the clearest abrupt shifts. Arctic summer ice, often seen as stable, experienced large abrupt shifts in 17 models. Winter sea ice showed abrupt losses in 21 models.

Antarctic sea ice had fewer shifts, but 11 models showed area losses greater than 0.5 million square kilometers (193,000 square miles).

Barents sea ice, above Scandinavia and Russia, showed abrupt shifts in 43 models. Eighteen of them had large-area losses, beginning as early as 0.5°C above preindustrial levels. Two models had losses over 0.5 million square kilometers (193,000 square miles).

More heat means more Arctic sea ice shifts

The more a model simulated warming, the more abrupt shifts it showed. At 1.5°C, matching the Paris Agreement limit, six of ten subsystems already showed sudden changes. This included the Arctic, North Atlantic, and Tibetan Plateau.

Models with higher climate sensitivity reached tipping conditions sooner. Shifts often happened between 0.5°C and 5°C, depending on the system.

Barents sea ice showed changes at lower warming than Antarctic sea ice or Arctic winter sea ice.

Some abrupt changes may be missed

Edge detection works well for fast transitions but misses slow-moving collapses. The models used do not simulate all systems fully, such as Greenland’s ice sheet. So actual risks may be higher.

Also, fast warming in the models may mask slower but real-world patterns. The current CO₂ increase is slower than in the simulation, so future studies must adjust for that.

Even at 1.5°C of warming, many systems show abrupt shifts in multiple models. Monsoon systems may resist change, but polar and high-altitude regions respond fast.

This study does not predict precise tipping points, but it reveals just how close we might be. The next big change in our climate might not come gradually. It may hit all at once.

The study is published in the journal AGU Advances.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–