Artificial intelligence for route planning has limited the warming contrails to more than 100 years

A trial involving thousands of flights between the US and Europe found that planes produce fewer contrails if they follow artificial intelligence-recommended flight paths to reduce the impact of global warming.

Condensation caused by soot particles produced by aircraft engines is thought to cause more warming than the carbon dioxide emitted by aircraft. The research also showed that certain ice-rich regions of the upper atmosphere are more likely to form contrails when a plane flies over them, and that AI can use detailed weather forecasts to predict where those regions will be.

Small-scale trials have been conducted to show that aircraft diverted over these regions will produce less contrails, but the practice has yet to be applied to large-scale commercial flights.

Now, Dinesh Sanekommu at Google and colleagues used an AI contrail-forecasting tool to provide routing advice in a randomized control test of more than 2,400 real American Airlines flights.

The trial involved eastbound flights from the US to Europe and lasted approximately 17 weeks, from January to May 2025. The direction was only one-way because these flights took place at night, a time when contrails have been found to have a more pronounced warming effect. During the day, contrails can have a cooling effect by reflecting sunlight back into space.

Each flight route between two cities was randomly assigned to one of two groups. For the first group, air traffic controllers had the option in their flight planning software to select an AI-optimized low-footprint route, but for the second, no alternative was suggested.

Although controllers in the first group always had the option of choosing a route with low condensation curves, only 112 of the 1,232 flights in that group ended up choosing an alternative route due to operational concerns such as cost or safety, Sanekommu says.

According to an AI analysis of satellite imagery of flight paths, flights that flew the contrail-optimized route suggested to air traffic controllers had 62 percent less visible contrails. When all flights that had the opportunity to use the contrail-optimized route are included, the effective overall reduction in contrail formation was 11.6 percent compared to the control group.

“It validated the thesis that if we can figure out how to safely and properly integrate into the flight planning process, then this is a scalable route to consider contrail avoidance for many flights,” says Sanekommu.

The team estimates that the warming effect of the flights was reduced by 13.7 percent in the entire group considering the proposed route and by 69.3 percent for the flights that flew the optimized route. There was no statistically significant difference in fuel consumption between the groups.

“That’s probably the best you can do, at least with the tools we have at this point,” he says Edward Gryspeerdt at Imperial College London. “It suggests that it’s possible. The 62 percent reduction in satellite-observed contrails that they see probably didn’t happen by accident.”

But it’s unclear how much the 11.6 percent can be improved in real-world operations, due to the complexity of flight planning, Gryspeerdt says. “You can’t necessarily just scale it to a 60% reduction in contrails for every flight everywhere, but even a 10% reduction in contrails is still a non-negligible effect.”

topics: