Eliminating contrails from flying could be incredibly cheap
Could we halve aviation's climate impact at a fraction of the cost of sustainable aviation fuels?
Eliminating CO2 emissions from flying is going to be expensive, regardless of the solution the world adopts.1
But aviation also contributes to global warming through its non-CO2 effects. Those are mostly “contrails”, which I’ll explain in more detail soon. Getting rid of those could be incredibly cheap. So cheap that it’s difficult to understand why we don’t just go ahead and fix it.
On a recent podcast, I spoke to Ian McKay, CEO of Orca Sciences, about this. One of their portfolio projects is Contrails.org. Their solution to eliminating contrails is to accurately forecast and model the atmospheric conditions that generate them, and reroute planes so that they avoid these “contrail-forming” parts of the atmosphere.
This is a solution that I hadn’t really paid much attention to, and most people are unaware of. So I thought I’d do a deep dive on contrails; explore how this solution might work; and whether it’s really that cheap.
To pre-empt the critics: this solution does not mean the aviation industry can ignore the CO2 impacts of flying. Tackling contrails would not absolve them of responsibility for finding low-carbon alternatives to jet fuel. It’s not a substitute, but an addition. Currently, their non-CO2 impacts are not measured or reported, so bringing more attention to contrails means they’re taking full responsibility for their climate impact, which is not the case at the moment.
What are contrails?
When you see a plane in the sky, you might see a small, white cloud-like trail behind it. Those are contrails (short for “condensation trails”).
Water vapour, soot and other particles (basically pollutants) are emitted from the back of jet engines. Water droplets can condense around these particles, and because it’s pretty cold up there, they can freeze to form ice crystals. Sometimes these white lines are very faint and hard to see. But in some cases, they can form “cirrus clouds”: wispy ones that form at high altitudes.
These contrails can have both cooling and warming impacts. I’ve sketched this out in the schematic below.
Some sunlight can reflect off of them, rather than passing through to the surface, which has a cooling effect.
Most sunlight, though, does pass through, and outgoing irradiation then gets trapped by the cirrus clouds. This has a warming effect, which tends to be larger than the cooling one, so on net, contrails cause warming.
Since someone asked about this over email: the fact that there is no sunlight at night, and less during winter, there is less to “reflect” off the top of cirrus clouds. That means the cooling effect is weaker at night, and in winter, and the net warming effect stronger. This means avoiding contrails in winter and at night has an even stronger impact on reducing warming.2
What impact do they have on global warming?
It’s common to want to compare them to CO2 emissions, but it’s first worth emphasising how different the contributions are in terms of intensity and persistence.
Contrails have a strong “effective radiative forcing” effect. This basically measures the net change in energy flow at the top of the atmosphere: and that change in energy flow dictates how much warming is needed at the surface to offset it. But, this warming effective is very short-lived. If we were to stop contrails today, the warming effect would disappear within a day or so.
Think of it like a very brief but strong pulse of energy.
CO2, on the other hand, has a smaller effect on radiative forcing, but once you emit it, it stays there for centuries or more.
I thought this diagram from Contrails.org makes this point clearly. This article by them explains the comparison in much more detail.
When we think about the climate impacts of aviation, then, most of the warming from CO2 emissions is not due to the emissions this year, but the cumulative effect (which persists) over the past 70 years. But for contrails, the warming impact is only really from those created very recently (hours to days); the small temperature response decays over months to a few years.
You might have heard people say that “more than half of the warming caused by aviation comes from non-CO2 sources”. A big part of that is contrails. But this does not mean that for any given flight, half of the warming is coming from contrails and the other half from burning jet fuel.
This apparent ‘half-half’ balance is a coincidence of timing: the cumulative CO₂ effect built up over decades happens to be of similar order to the instantaneous contrail effect for that year. In the chart below you can see the effective radiative forcing caused by CO2 and contrails in 2019. Again, the CO2 emitted in 2019 is just a small part of the warming. Most of comes from emissions built over decades, that stay there. It just so happens that this cumulative amount of warming is not that different from the instantaneous, short-lived impact of contrails in 2019. Eventually more and more CO2 emissions will accumulate, and the share coming from contrails will shrink in relative terms.
But as it stands today, we could get rid of around half of the warming impact — maybe slightly less — from aviation, if we were to tackle contrails. The impact would be almost immediate.
How, then, do contrails stack up in terms of total warming? They contribute roughly 2% to the world’s effective radiative forcing; tackling them would reduce that by a similar amount.3
What this comparison should make extremely clear is that reducing contrails does not mean we don’t also need to tackle CO2 emissions from aviation. Ultimately that is the persistent driver of long-term temperature change. What tackling contrails now would do is slightly reduce the rate of warming (and therefore do something reduce the risks of nearer-term feedbacks that could affect the release of CO2 from natural systems, and also affect long-term temperature change). It is not an excuse or a substitute for finding a way to decarbonise jet fuel.
Only a few percent of flights cause most of the warming
One crucial reason why eliminating contrails could be so cost-effective is that a very small percentage of flights create the majority of the impact. This means we don’t need to divert or shift the trajectory of all the world’s flights; only a few percent of them.
In the chart below, you can see the breakdown of the warming effect across the world’s flights.4 On the left-hand side, we have the share of flights, and on the right, their collective contribution to the total warming impact of contrails.
Just 3% of flights generate 80% of the warming. A further 14% generate 29%.
You might notice that this sums to 109%. But this is because some flights generate a cooling effect of 9%. Put them together and we get 100%.
Most flights — three-quarters of them — barely generate contrails at all and cause no warming or cooling.
Some sources cite slightly different numbers for this “80% warming effect”. For example, Contrails.org cite 5% of flights. I’ve seen others quote 2%.5 But the point remains the same: a few percent of the flights completely dominate the climate impact.
There are ways to dramatically reduce them
So, how can we get rid of these contrails?
Contrails with a strong warming impact mostly form in thin regions of the atmosphere, which are cold and humid. If planes fly through these zones of atmosphere, contrails are much more likely to form.
The solution, then, is for some planes to take a short detour to avoid them. You can see this in the schematic below.
How would we know which planes to re-route and by how much?
Using detailed weather forecasts, satellite images, and flight plans, scientists can identify where these zones will be far in advance and work with flight planners to find a way to reroute flights crossing these zones to avoid them. These forecasts and models are what Contrails.org do.
Google also launched “Project Contrails” which uses Artificial Intelligence (AI) to build models that can do this.
Of course, this wouldn’t work if these planes had to do a severe detour. People would not be happy about a longer flight time. And, the extra fuel that would need to be burned to go the extra distance would eventually cancel out the climate benefits from getting rid of the contrails.
The proposed detours typically result in a 1% shift (and again, this is only for a small percentage of flights). That means increasing fuel use and flight time by around 1%. So if your flight is three hours long, it’s only adding an extra two minutes. For a 10-hour flight, six minutes. This seems socially acceptable to me; most people would barely notice.
Stopping warming from contrails could be incredibly cheap
The fact that the warming impact is skewed towards such a small share of flights dramatically reduces the costs.
What are the costs associated with implementing this?
There are operational costs associated with weather prediction, modelling, and integration into flight planning. Especially with the integration of AI, this is probably not that expensive. A bit more costly is the extra jet fuel that’s needed for rerouted planes.
When I spoke with Ian McKay, he suggested the additional cost would be around $5 per flight. I think he meant this as $5 spread across the entire flight (not per passenger). This is also the figure they give on Contrails.org.6 I also think that in this assumption, the costs are spread evenly across the entire airline fleet (regardless of whether they’re rerouted or not). For the small share of rerouted flights alone, the “per flight” cost would be higher.
That’s incredibly low. Spreading that over 100 passengers, and each is paying just 5 cents extra.
Other studies have reported higher costs, although they’re still incredibly cheap.
This paper modelled over 84,000 flights and found that the additional cost of operations and fuel burn for rerouting increased costs by around $1.1 million.7 By my calculations, that’s around $10 to $15 per flight.
We can do a very basic back-of-the-envelope calculation to sense-check this. The total fuel cost of flying from Barcelona to Berlin is probably around $2,000.8 If the flight burned 1% extra fuel due to rerouting, the extra cost for the flight would be around $20. Add the operational costs of the forecasting, and this could be $30 to $40. Then spread across all flights, not just the rerouted ones, and this falls back down to the $5 to $10 range again.
Transport & Environment (T&E) estimates that the cost could range from $2 to $5 per passenger (or hundreds of dollars per flight).9 They do note that they make very conservative assumptions, and therefore find costs that are 3 to 10 times higher than those from other sources.
For a flight in Europe, such as from Barcelona to Berlin, the cost would be €1.20 ($1.88) per passenger. A Transatlantic ticket would be more expensive, around €3.90 for a trip from Paris to New York. Given that an economy ticket from Paris to New York probably costs around €350 to €400, this would increase the cost by around 1%.
Perhaps, then, the best estimate is somewhere in the middle: around 50 cents per passenger.
Translating this into the cost per tonne of carbon dioxide equivalent — the “carbon abatement” cost — shows how cheap this is compared to many other climate solutions. It’s probably in the range of a few dollars per tonne CO2e. Contrails.org estimates that it’s slightly below $1 per tonne.
Switching to “sustainable aviation fuel” currently has an estimated cost in the range of hundreds of dollars per tonne of CO2e avoided.10 Rather than a flight ticket being 1% more expensive, it would be more than double the price. Eliminating contrails is therefore hundreds of times cheaper and can be scaled much more quickly than replacing the entire aircraft fleet or its fuel source.11
Why aren’t we doing more to eliminate contrails?
When I asked Ian McKay why airlines were not doing more, he gave two main reasons.
The first is that even if the cost per flight is low, the total cost across their entire fleet adds up. Let’s take a quick example for British Airways. They operate around 300,000 flights per year. If we reroute 2% of those to avoid contrails, and rerouting increases fuel burn by around 2% (I’m being deliberately harsh here), then I estimate that the additional fuel costs are in the range of $1.2 to $2 million per year.12 Let’s say that the operational costs of forecasting and modelling adds another 50%. That takes us to around $2.5 to $3 million.
In 2024, British Airways had an operating profit of around $2.7 billion. Contrail avoidance would therefore be just 0.1% of its operating profits.
But I’m not convinced that this cost factor is the main reason. They could pass this cost on to consumers; flight prices vary by a lot more than a few dollars for a variety of factors. They could either make a huge deal of the fact that they’re dramatically cutting their climate impact, and get “PR” buy-in from consumers for that. Or they could keep quiet, and most consumers would never notice the difference in cost.
The second — which seems more likely — is that, currently, most people are unaware of the climate impact of contrails. In that sense, airlines can basically ignore it and pretend they don’t exist. By trying to tackle them, they’d only draw more attention. People would then be aware that the climate impact of aviation is even higher than they thought.
I still think that the airline that steps up and commits to eliminating contrails — possibly even claiming to have halved its climate impact — would be well-received by many customers. I would see it as reputational gain, rather than a risk.
Nonetheless, there are no signs that the aviation industry itself is going to step up. This is where government policy could step in.
Rather than an airline leading by example, a country or region could. In a more pro-climate political environment, the United States could have led this effort domestically, mandating that internal flights eliminate their contrails. More likely is the European Union. It has already been making some progress in this direction — not by mandating that airlines pay for contrail avoidance — but by simply reporting these climate impacts in the first place. Earlier this year, its trading system regulations were updated to require airlines to monitor and report non-CO2 impacts. That sounds basic, but it is not the standard across most of the world; these impacts are usually not included. Unsurprisingly, it has received pushback from the aviation industry, with them asking for these reports to be voluntary.
Progress will undoubtedly be met with initial resistance, but I still think that regulatory policy seems like the most likely path to widespread implementation.
What would help a lot is increasing public awareness of the existence of contrails, their climate impacts, and how inexpensive it could be to eliminate them. There is a general understanding that decarbonising aviation is expensive, and this often means the aviation industry gets more of a free ride. But this is based on replacing jet fuel. If people were aware that it could cut a huge chunk of its footprint at a fraction of the cost, they might be more demanding.
Eliminating a few percent of the world’s warming is a big deal when the costs are so small. It seems insane to me that such a cheap solution is sitting there, completely untapped.
This could be substituting jet fuel for an alternative such as green hydrogen or biofuels.
But some suggest that it could be cheaper to keep burning jet fuel and try to capture — and securely store — an equivalent amount of CO2 directly.
Their question went further, asking if having some additional warming in winter is actually beneficial as it reduces risks such as cold-related deaths.
This could be true if the impacts were local. However, the warming that results is both global, and lasts over the long-term (even if the immediate forcing is short-lived, as we’ll come on to).
Lee, D. S., Fahey, D. W., Skowron, A., Allen, M. R., Burkhardt, U., Chen, Q., ... & Wilcox, L. J. (2021). The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018. Atmospheric Environment.
This is based on data published in the Transport and Environment (T&E) 2024 Report: Contrail avoidance: aviation’s climate opportunity of the decade.
Teoh, R., Engberg, Z., Schumann, U., Voigt, C., Shapiro, M., Rohs, S., & Stettler, M. E. (2024). Global aviation contrail climate effects from 2019 to 2021. Atmospheric Chemistry and Physics.
Here they say:
“Better yet, properly implemented, contrail management is low-cost: studies show a fleet-average fuel cost of roughly $5.00 per flight, or less than $1 per tonne of CO2 equivalent warming avoided.”
Agarwal, A., Meijer, V. R., Eastham, S. D., Speth, R. L., & Barrett, S. R. (2022). Reanalysis-driven simulations may overestimate persistent contrail formation by 100%–250%. Environmental Research Letters.
This assumes burning around 3,000 litres of fuel, weighing around 2.5 tonnes.
The cost per tonne is around $900.
That gives a total cost of around $2250.
Again, these costs are distributed across all flights, not just those that are rerouted.
Watson, M. J., Machado, P. G., Da Silva, A. V., Saltar, Y., Ribeiro, C. O., Nascimento, C. A. O. D., & Dowling, A. W. (2024). Sustainable aviation fuel technologies, costs, emissions, policies, and markets: A critical review. Journal of Cleaner Production.
Here’s an ugly, but useful graph from the UK Government’s cost-benefit report on SAFs.
This is based on fuel costs ranging from $600 to $1000 per tonne.
Another advantage of talking about the scientific reality of contrails is to preempt the lunatics who spread the “chemtrails” conspiracy theories… 😎
Very interesting, two thoughts come to mind:
1. Is the deviation purely horizontal, or can the plane avoid making contrails by changing altitude? Avoiding contrails in 3D may be a lot easier than avoiding them in 2D. However, the modelling and forecasting would be more complex.
2. In crowded areas such as over Europe, planes follow strict routes from way point to way point. There is no scope to deviate from the route. That keeps air traffic control relatively simple, and safe, and allows it to be split across national boundaries. The planes avoid other planes in the 3D space. To allow regular route deviation would require much more sophisticated ATC systems and specifically algorithms.
There have been proposals for making ATC dynamic and forward looking, so each plane is given a unique flight path and this is checked to avoid all other flight paths. In effect, the planes avoid each other in both 3D space and in Time. So the plane from London to Milan may be in the exact same space as the plane from Barcelona to Berlin, but 60 seconds later.
I think in crowded airspace, contrail avoidance would need more advanced ATC; but equally, contrail avoidance would be built into the ATC system.