We might have chosen to publish this at a different time: many of us are on edge following the recent accident at LaGuardia. The piece was already in the pipeline but, more importantly, the stress flight teams work under is an important part of the story, and it’s not a bad thing to have the reasons for the apparent reluctance outlined here flickering on our screens.
Data scientist Hannah Ritchie, senior researcher at the University of Oxford, flagged research last fall showing that contrails, which are believed to contribute about 1% to 2% of human-caused global warming, could be cut by up to 80% cheaply, and with immediate benefits. In that piece she noted that no matter what we do, cutting airline CO₂ emissions will be expensive, with benefits far in the future. Contrails’ contribution to climate change is greater than half that of CO₂ airline emissions, and controlling contrails could cut aviation’s contribution to climate-change by 35-40%.
Factoid: Contrails are estimated to cause the same amount of global warming each year as all the CO₂ emitted by airplanes between 1940 and 2018.
For reference, 80% of contrail warming is generated by 3% of flights, while 29% comes from 14% of flights. Seventy-six percent of flights generate few contrails, and 7% of flights provide a 9% cooling offset, which returns us to 100%. There are different versions of these numbers, but they don’t vary by much. Contrails are short-lived, hanging in the sky for minutes or hours, while CO₂ persists for centuries, which presents a “unique opportunity for climate mitigation.”
Contrails are a collaborative effort. Under certain atmospheric conditions, dependent on hour, season, cloud cover and reflection, water vapors emitted in flight may form cirrus clouds that trap heat in the lower atmosphere. Using accurate predictions of where such conditions are likely to form, airlines can modify routes to avoid suspect areas, as they do for turbulence and thunder storms. (Those involved in this work are engaging machine-learning in hopes of improving results.)
Last week Ritchie highlighted a new study recently conducted by Google and American Airlines, including 2,400 international flights identified as having “substantial” climate effect. This randomized study was designed to test the efficacy of contrail avoidance efforts led by airline dispatchers themselves.
Prior studies have shown pilots and dispatchers manually modifying routes based on contrail forecasts can be effective, but could not be brought to the scale needed for real mitigation, and identifying areas where contrails were most likely to form demands long manual hours. Additionally, prior studies have relied on modeling, so this collaboration used automated satellite verification that allowed a blind assessment at a scale formerly unattainable.
In this study Google’s AI contrail forecasts were incorporated into the flight-planning software and dispatchers were given the option, with no incentives, to give the tradition plan or the avoidance plan to the pilots of the 1,232 planes in the treatment group.
Since the study was intended to test the effectiveness of dispatchers controlling the process, the headline showed contrails created in the treatment group to be 12% below those of the study group. But dispatchers had issued avoidance plans to just 15% of the 1,232 flights, and only 60% of the pilots successfully executed the avoidance plans. Among the 112 flights that followed the avoidance plans, contrails were 62% lower. The authors call their results robust, with all subgroups of the treatment group showing statistically significant reductions in contrails, and with an insignificant, 1%, change in fuel usage. (One tranche even showed a counterintuitive reduction.) They believe their methodology sets a new standard for evaluating the usefulness of contrail avoidance trails, as long as they stand up to peer-review at large scale.
The authors discuss the low take-up rates for the avoidance plans mentioned above, and believe they could be raised through better communication. In follow-up interviews, they found that flight crews and dispatchers often preferred not to make the mid-flight ascents and descents required by the plan, even though they are safe, conducted in cooperation with flight controllers, and used to avoid turbulence and heavy weather. Dispatchers ranked safety and efficiency higher than contrail avoidance, and were less likely to use the avoidance plan when airports were busy, or when managing turbulence. (There were also some technical features that could not be activated.)
Software included only top down images of the flight, which made it difficult for pilots and dispatchers to understand ascents and descents, and the authors believe could have been better managed with vertical images. Even small divergence from the plan can put planes back in contrail regions, so the minor adjustments made by the pilots also likely affected the outcomes.
The authors suggest future collaborations between multiple airlines and Air Navigation Service Providers could resolve many of these issues.
A 50% blend of sustainable and traditional aviation fuel has been shown to reduce soot emissions by 50% to 70%, but in 2024 SAFs made up just 0.3% of all aviation fuel, and that is expected to rise to just 2% to 5% by 2030. Alternative engine technology, being installed in new airlines, has the ability to reduce emissions by as much as 70%, but probably will not be used by entire fleets for decades.
The avoidance detours increase flight time by about 1%, so two minutes on a three-hour flight, and lift average fleet-fuel costs by $20 a flight, or one or two bucks per ton of CO₂ equivalent warming.
Knock-on benefit: Contrails are a real magnet for conspiracy theorists. Wouldn’t it be great to reduce those opportunities by more than half?
More Info...
Subscribe >
More Info...
Subscribe >
Flight paths, contrails, and climate change
We might have chosen to publish this at a different time: many of us are on edge following the recent accident at LaGuardia. The piece was already in the pipeline but, more importantly, the stress flight teams work under is an important part of the story, and it’s not a bad thing to have the reasons for the apparent reluctance outlined here flickering on our screens.
Data scientist Hannah Ritchie, senior researcher at the University of Oxford, flagged research last fall showing that contrails, which are believed to contribute about 1% to 2% of human-caused global warming, could be cut by up to 80% cheaply, and with immediate benefits. In that piece she noted that no matter what we do, cutting airline CO₂ emissions will be expensive, with benefits far in the future. Contrails’ contribution to climate change is greater than half that of CO₂ airline emissions, and controlling contrails could cut aviation’s contribution to climate-change by 35-40%.
Factoid: Contrails are estimated to cause the same amount of global warming each year as all the CO₂ emitted by airplanes between 1940 and 2018.
For reference, 80% of contrail warming is generated by 3% of flights, while 29% comes from 14% of flights. Seventy-six percent of flights generate few contrails, and 7% of flights provide a 9% cooling offset, which returns us to 100%. There are different versions of these numbers, but they don’t vary by much. Contrails are short-lived, hanging in the sky for minutes or hours, while CO₂ persists for centuries, which presents a “unique opportunity for climate mitigation.”
Contrails are a collaborative effort. Under certain atmospheric conditions, dependent on hour, season, cloud cover and reflection, water vapors emitted in flight may form cirrus clouds that trap heat in the lower atmosphere. Using accurate predictions of where such conditions are likely to form, airlines can modify routes to avoid suspect areas, as they do for turbulence and thunder storms. (Those involved in this work are engaging machine-learning in hopes of improving results.)
Last week Ritchie highlighted a new study recently conducted by Google and American Airlines, including 2,400 international flights identified as having “substantial” climate effect. This randomized study was designed to test the efficacy of contrail avoidance efforts led by airline dispatchers themselves.
Prior studies have shown pilots and dispatchers manually modifying routes based on contrail forecasts can be effective, but could not be brought to the scale needed for real mitigation, and identifying areas where contrails were most likely to form demands long manual hours. Additionally, prior studies have relied on modeling, so this collaboration used automated satellite verification that allowed a blind assessment at a scale formerly unattainable.
In this study Google’s AI contrail forecasts were incorporated into the flight-planning software and dispatchers were given the option, with no incentives, to give the tradition plan or the avoidance plan to the pilots of the 1,232 planes in the treatment group.
Since the study was intended to test the effectiveness of dispatchers controlling the process, the headline showed contrails created in the treatment group to be 12% below those of the study group. But dispatchers had issued avoidance plans to just 15% of the 1,232 flights, and only 60% of the pilots successfully executed the avoidance plans. Among the 112 flights that followed the avoidance plans, contrails were 62% lower. The authors call their results robust, with all subgroups of the treatment group showing statistically significant reductions in contrails, and with an insignificant, 1%, change in fuel usage. (One tranche even showed a counterintuitive reduction.) They believe their methodology sets a new standard for evaluating the usefulness of contrail avoidance trails, as long as they stand up to peer-review at large scale.
The authors discuss the low take-up rates for the avoidance plans mentioned above, and believe they could be raised through better communication. In follow-up interviews, they found that flight crews and dispatchers often preferred not to make the mid-flight ascents and descents required by the plan, even though they are safe, conducted in cooperation with flight controllers, and used to avoid turbulence and heavy weather. Dispatchers ranked safety and efficiency higher than contrail avoidance, and were less likely to use the avoidance plan when airports were busy, or when managing turbulence. (There were also some technical features that could not be activated.)
Software included only top down images of the flight, which made it difficult for pilots and dispatchers to understand ascents and descents, and the authors believe could have been better managed with vertical images. Even small divergence from the plan can put planes back in contrail regions, so the minor adjustments made by the pilots also likely affected the outcomes.
The authors suggest future collaborations between multiple airlines and Air Navigation Service Providers could resolve many of these issues.
A 50% blend of sustainable and traditional aviation fuel has been shown to reduce soot emissions by 50% to 70%, but in 2024 SAFs made up just 0.3% of all aviation fuel, and that is expected to rise to just 2% to 5% by 2030. Alternative engine technology, being installed in new airlines, has the ability to reduce emissions by as much as 70%, but probably will not be used by entire fleets for decades.
The avoidance detours increase flight time by about 1%, so two minutes on a three-hour flight, and lift average fleet-fuel costs by $20 a flight, or one or two bucks per ton of CO₂ equivalent warming.
Knock-on benefit: Contrails are a real magnet for conspiracy theorists. Wouldn’t it be great to reduce those opportunities by more than half?
