An in-depth conversation with Dr. Thomas Roetger
“Whatever happens in the sky affects the world — and the future of aviation.”
The loudest cars aren’t the fastest. The fastest aircraft don’t leave the biggest imprint. Not every cloud is natural — nor as innocent as we think. High above us, white lines form, disappear, or slowly evolve into thin cirrus sheets. Some of them matter far more than we imagine. Since the beginning of human curiosity, trying to understand the universe has been one of our most fundamental instincts — and one of science’s main reasons for existing. At ENV-ISA, we believe that scientific knowledge should reach wider audiences with clarity, simplicity and purpose. That is the essence of our “Science for Life” mission. Driven by this idea, we are launching Env·iTalks, a new interview series where we bring scientific expertise into public light, one conversation at a time. This first episode opens with a topic often seen, rarely understood: contrails.
Could you briefly explain what exactly a contrail is?
Contrails are the white traces that aircraft sometimes leave behind them. The exhaust gases contain a lot of water vapour, and as it’s very cold up there in the atmosphere at cruise altitude—typically 10 to 12 kilometers—the water vapour from the aircraft engine exhaust freezes and forms these white lines, which are actually an accumulation of ice crystals. But in order to form crystals, you need condensation nuclei, which are small particles either already in the air or coming out of the aircraft exhaust.
The water molecules condense around these nuclei to form the crystals that we see in the sky as white trails behind the aircraft. Sometimes there’s nothing visible after an aircraft. Sometimes you see a short white line that immediately disappears. And sometimes, what we call persistent contrails is a longer line that over time becomes broader and more fluffy, eventually transforming into a real cloud.
Meteorologists call this type of long-shaped formation “cirrus clouds”. So actually, contrails can transform into cirrus clouds—especially when generated in conditions of quite humid air, where they tend to persist and have the time to get transformed into these cirrus clouds.
If I understand correctly, contrails are mainly traces of vapor interacting with cold air, but there are also particles involved, correct?
The majority of the particles serve as condensation nuclei for the formation of ice crystals at high altitude. The soot particles, for instance, are a combustion product of the aircraft engines. Soot is essentially what we call black carbon. It can be either pure carbon or complex molecules that contain a lot of carbon, sometimes combined with a few hydrogen atoms. In fact, it’s a mixture of many different molecules, but they all have a high carbon content, representing unburnt or partially burnt products of the combustion process in the aircraft engine.
Since we’re talking about combustion products, how does contrail formation relate to engine power in different flight phases — and how does this connect to health risks?
Actually, contrails form in the cruise part of the flight, typically at 10 to 12 kilometers altitude above the ground, with the very cold ambient temperatures around –40°C or below that are necessary for contrails to form —. Regarding combustion and the soot particles you mentioned, yes, they can also have harmful effects on health, which is well known. But we have to distinguish between different phases of the flight.
When an aircraft is close to the ground — during takeoff, initial climb, and just before landing — the combustion products, exhaust gases, and soot particles are emitted near ground level and become part of the air that people living near airports breathe. In that context, the health effects are quite significant, especially because these emissions contain particles of many different sizes. What comes out of the aircraft exhaust is to a large extent ultrafine particles, which are extremely small, typically between 10 and 100 nanometers (millionths of millimeters) in diameter. Because they are so small, they can easily penetrate deep into the lungs when inhaled, posing particular health risks. Of course, at high altitude, it’s a different situation. It’s generally considered relatively unlikely that these particles would return to ground level in significant amounts, although some researchers are currently studying to which extent they could gradually contribute to ultrafine particle concentrations near the surface.
When we talk about these particles, sometimes we refer to them simply as ‘particles,’ and other times as soot or black carbon. Why do these different names exist?
When we talk about aircraft exhaust, we have both gaseous substances that are pollutants — for example, nitrogen oxides and carbon monoxide, which are harmful to health — and solid particles. The solid particles emitted from aircraft exhaust are mainly soot or black carbon. These are relatively large molecules with a high carbon content, sometimes combined with a small number of hydrogen atoms.
One of the most well-known soot molecules is the so-called ‘fullerene,’ often nicknamed the ‘football molecule’, because it resembles a football. If you imagine a football made up of hexagons and pentagons — like the stitched leather panels — you can picture a molecule where each corner represents a carbon atom. In this structure, there are 60 corners, so you have 60 carbon atoms connected together in the shape of a football. This is a typical component of soot: unburnt carbon resulting from hydrocarbon fuel that has not been completely oxidized during combustion.
Of course, when discussing health effects, it’s important to consider the chemical nature of these particles and the gaseous components. But when we talk about contrail formation, the key property is simply the presence of solid particles onto which water molecules in the engine exhaust can adhere, forming ice particles.
When we talk about the environmental impact of aviation, we generally focus on carbon dioxide. But there’s also this category called ‘non-CO₂ effects.’ I realise that the public sometimes struggles to fully grasp what ‘non-CO₂‘ means. Many people think that aircraft operators would simply have to minimise CO₂ emissions, and everything is fine — but that’s not the case. There are other kinds of molecules and effects, different from carbon dioxide. Could you explain a bit what we mean by this ‘non-CO₂’ category?
The climate change we are concerned about today is driven by the greenhouse effect, or more precisely, by the reinforcement of this effect. Sunlight reaches the Earth and warms it, but part of the heat becomes trapped in the atmosphere, unable to fully radiate back into space. Several substances contribute to this greenhouse effect, and CO₂ is one of them. CO₂ allows sunlight to reach the Earth but absorbs infrared radiation, preventing it from escaping back into space.
However, there are other substances besides CO₂ that have similar effects. The so-called ‘Kyoto gases’ other than CO2, whose concentrations must be reduced following international agreements, are not generated in aircraft engines, but a few additional factors contribute to aviation’s greenhouse effect in addition to CO₂ emissions.
One of these is nitrogen oxides, which lead to the formation of ozone — itself a greenhouse gas. Another significant factor is contrails. From our daily experience, we know that clouds trap heat. On a cloudy day, less sunlight reaches the surface, but on a cloudy night, temperatures don’t go down so much because the heat is trapped. Persistent contrails, which can evolve into cirrus clouds, essentially act like clouds.
Meteorologists might say this is a somewhat simplified description, but contrails do behave similarly: they trap heat during the night and can have a slight cooling effect during the day. However, overall, their warming effect is stronger than their cooling effect. That’s why contrails are considered one of the non-CO₂ contributions of aviation to climate change
I came across a NASA source on which mentions that soot particles are emitted particularly during takeoff, climb, and cruise. But I couldn’t fully understand if contrails form only once the aircraft reaches around 10 to 12 kilometers altitude, why are soot emissions so significant during takeoff and climb? Is there any specific explanation for the higher soot production in these phases? Is there any specific information about soot emissions during the takeoff phase?
When you really go into the details of these effects, it’s important to clarify certain aspects to avoid confusion. As I mentioned earlier, the impact of soot particles in the lower atmosphere is mainly a health issue, rather than a climate issue.
Now, regarding the statement that emissions are particularly high during takeoff — you have to consider the full context of the article. At takeoff, the engine operates at full power, which means the combustion is actually more complete compared to low power settings. Soot emissions are often higher at lower power regimes, such as during approach or even taxiing, because combustion efficiency decreases at these settings. In those phases, the combustion process is less complete, producing more soot.
Of course, during takeoff, the engine operates at maximum thrust, so you burn more fuel overall, which still results in a significant amount of soot being emitted, even if combustion efficiency is high. During cruise, the engines typically run at 60 to 70% of full thrust, but there is still a continuous emission of soot particles, which, in the cold upper atmosphere, serve as condensation nuclei for ice particles and contribute to contrail formation.
So, when the NASA article mentions that soot emissions are particularly high at takeoff, in a sense, that’s true — but it’s important to distinguish between the amount of fuel burned and the combustion efficiency, as well as to consider what specific effects these emissions have.
This is exactly why I’m asking. We already understand CO₂’s impact on climate change, but here I’m referring specifically to the direct effects on human health. Since takeoff is the moment when an aircraft is closest to human habitats, that’s what prompted my question. So, when we look at the issue from a health perspective, would it be accurate to say that the components of contrails — or rather, the particles emitted that can lead to contrails — may have a more direct health impact than carbon dioxide, especially for people living near airports?
Carbon dioxide is not a poisonous substance; it has no direct effect on human health. Contrails require soot particles as condensation nuclei, but the soot particle itself is extremely small compared to the resulting ice crystal. In fact, I would estimate that one of these ice crystals is more than 99% ice, with only a tiny fraction consisting of the original particle.
However, at low altitudes, close to the ground, it’s the soot particles themselves that pose a health risk, as they can be inhaled directly in their particulate form. These soot particles at low altitudes are not involved in contrail formation, which occurs only in a very cold atmosphere at high altitudes.
I understand now that soot is a common factor, but at low altitudes, this isn’t related to contrail formation, is that right?
Right, these are two different aspects: the health risks associated with small particles, and their role in contrail formation. Compared to other sources of soot — such as old-fashioned heating systems or car exhaust — the soot particles emitted by aircraft engines are particularly ultrafine.
In the environmental committee of the United Nations’ International Civil Aviation Organization (ICAO), a standard for particle or particulate matter emissions was established in 2016, which is relatively recent. One of the reasons for this delay was the difficulty in accurately measuring both the mass and especially the number of these very small particles, simply because sufficiently precise measurement instruments did not exist before.
When you read about urban air pollution, you usually encounter PM10, referring to particulate matter smaller than 10 micrometers (thousandths of millimeters), which is relatively large. Sometimes PM2.5 — particles smaller than 2.5 micrometers — is also mentioned. But you rarely see information on ultrafine particles, categorized as PM0.1 — particles smaller than 0.1 micrometer, or 100 nanometers. The majority of particles from aircraft exhaust fall into this ultrafine category, which is small enough to easily penetrate deep into the lungs when inhaled.
The first step was simply being able to measure these particles, and once that became possible, certification standards could be established. Newer engines are now being developed to minimize ultrafine particle emissions. But this is still a relatively recent development, as more than ten years ago, we simply didn’t have the capability to measure them accurately
Contrails can look very different. What explains these variations, and how do atmospheric conditions influence their formation and evolution?
I’ll leave the detailed explanation to the meteorologists and climate scientists, but generally speaking, you have short-lived contrails that are not really a concern since they disappear quickly — you see the aircraft in the sky and within ten minutes they’re gone. Then, you have the more persistent ones. These start as straight lines, but over time they broaden, become somewhat fluffy, and remain as wide, relatively thin stripes of cirrus clouds.
The formation and lifetime of contrails depend greatly on the humidity and temperature of the air. Ice condensation occurs if the air is already supersaturated with water vapor. Of course, there’s always some natural condensation of water vapor in the atmosphere, even without aircraft. At very high altitudes, where it’s extremely cold, the amount of water vapor needed for condensation is quite small. Even at low concentrations, condensation can occur when temperatures are low enough, but it still requires condensation nuclei.
In weather forecasts, humidity is always expressed as a percentage. When we say 100% humidity, it means that the air holds as much water vapor as it can before condensation occurs. However, the absolute amount of water vapor corresponding to 100% humidity depends on the temperature: at lower temperatures, 100% humidity represents a much smaller quantity of water vapor.
Does this mean that contrails are more of a concern in cities or in regions with higher humidity?
This isn’t really related to cities because we’re talking about 10 kilometers altitude, far above ground level. But yes, if the surrounding humidity is already high, and an aircraft flies through it, then the additional water vapor from the aircraft exhaust, combined with the soot particles, can trigger condensation. Without the aircraft, the existing water vapor might not have been sufficient to condense, but the aircraft provides the necessary nuclei, leading to contrail formation.
Is there a direct relationship between humidity and global warming, and is there scientific evidence supporting this?
Humidity is a weather parameter that changes constantly, sometimes even from hour to hour. But I understand what you mean. Yes, global warming and climate change tend to increase average humidity levels over time and across many regions. So you’re absolutely right that, in the long term, higher global temperatures will likely increase the probability of contrail formation. However, I don’t know how quickly this effect develops or how significant the increase will be. But the general trend is correct.
From a public perspective, can someone tell – just by looking – whether a contrail is likely to be climate-significant?
Absolutely. If a contrail remains thin and disappears after a few minutes, it’s harmless. If it persists for many minutes, extends over kilometers, becomes fluffy, and starts to show wavy patterns like small clouds, then the likelihood of persistence is much higher. In regions with dense air traffic, you often see many such trails.
However, the presence of humidity on the ground doesn’t always reflect the situation at higher altitudes. Humidity layers at contrail-forming altitudes can be quite thin, often less than 2,000 feet, or about 600 meters. One of the objectives of the BeCoM project is to improve the prediction of these contrail formation zones. Ideally, pilots or air traffic controllers could adjust the flight level by a few hundred meters to avoid these zones. Tests with single aircraft have shown that this principle works. The challenge is that, with current instruments, it’s not yet reliable enough to predict the precise altitude where contrail formation occurs. If you’re off by a few hundred meters, you might inadvertently fly directly into the zone you’re trying to avoid. This is why, even though the principle works, it’s not yet reliable as a standard operational procedure.
I find this possibility really interesting. To make such a solution viable and feasible, we obviously need more information. But let’s say we do have accurate data about weather conditions along the flight path, would this information need to be integrated operationally by air traffic control and pilots? Because so far, most efforts seem to focus on engines and fuels. But now, there will also be new players involved in the system, right?
Air traffic control has always been a partner in the air transport system. But practically speaking, it would mean that air traffic management optimization tools would need to include a new parameter indicating contrail formation zones. They would then need to keep these areas free of aircraft — which is easier said than done.
First of all, even if we could determine these zones precisely, they are not perfectly horizontal. After 100 kilometers, for example, a pilot might need to adjust altitude again because the contrail formation zone could be slightly tilted. And of course, aircraft are not flying alone; traffic control must manage many flights in the same airspace. If you block one flight level, you have to squeeze the remaining aircraft into fewer available levels, which complicates the task for air traffic control.
Now, I’d like to focus on the main approaches being developed to address contrail-related challenges. As you mentioned, projects like BeCoM are working on weather prediction, but beyond that, what other key areas are currently being explored to manage contrail-related aspects?
If prediction methods become accurate enough, weather forecasters, together with air traffic control and pilots, will be able to identify contrail formation zones. Air traffic control could then optimize flight trajectories to avoid these zones. This approach has already been tested in some trial campaigns.
However, one reason why airlines and pilots are not entirely enthusiastic is that avoiding contrail zones often means flying at altitudes that are not aerodynamically optimal. Flying away from the most fuel-efficient altitude results in higher fuel consumption, which raises both operational costs and CO₂ emissions. Still, general predictions suggest that by avoiding contrail formation, the climate benefit would outweigh the small additional fuel consumption and CO₂ generation, leading to a net positive effect at relatively low cost.
The challenge is convincing pilots and airline financial departments that it’s worth doing. We need more data to clearly show the cost-benefit balance. In congested airspaces like Europe, this argument is less striking because aircraft rarely fly at their optimal altitude anyway. But in less congested areas, such as the North Atlantic or routes between Europe and Asia, where many flights follow similar paths with less conflicting traffic, pilots can typically select the most fuel-efficient altitudes. In such cases, if air traffic control were to deny a requested altitude for contrail avoidance, it would create a conflict between environmental benefits and operational efficiency.
Does this mean contrail avoidance might be more feasible on long-distance routes?
Yes, this could be manageable, especially in uncongested airspace. On the other hand, we would need a much clearer understanding of the actual costs in terms of additional fuel and CO₂ compared to the benefit. And today, since there’s no penalty for contrail formation, people ask: Why should I do it? So, I don’t think there is a standard miracle answer. It clearly requires more research to be sure what we’re talking about.
Turning to solutions now: from your perspective, what should be prioritised, and which actions are most critical – when addressing soot-particle effects?
Coming back to the health problem, which is distinct from the contrail issue — as I mentioned, certification standards for particulate matter emissions from aircraft engines have been existing for less than 10 years now. These standards are putting pressure on engine manufacturers to develop engines that emit less soot.
This also links to another subject: sustainable aviation fuels. Although they originate from different sources, they are very similar to conventional fuel derived from crude oil. But the main difference is that they generate much less soot when burned. So, better engines and sustainable or cleaner fuels would both reduce the generation of soot particles, which would benefit the health of people living near airports and also reduce the likelihood of contrail formation.
Some people even suggest developing logistics concepts where sustainable fuels are reserved for flights that are likely to pass through contrail formation zones, because the benefit would be strongest there. Yeah, this is complicated… because one of the main advantages of sustainable fuels is that you normally don’t need to distinguish between them. Pilots don’t need to care whether they have conventional fuel or a blend of sustainable fuel in the tanks; it doesn’t affect the aircraft’s performance — or even slightly improves it. And of course, you can mix them at wide proportions. You don’t need a second set of tanks and pipes at the airport fuel farms.
But if you want to reserve sustainable fuel for certain flights, then you would need a second set of tanks and pipes, which would be quite a significant investment for airports and fuel companies. But again, this is one of the many pathways being studied, and perhaps, if you think far enough ahead before building a new tank, this could become a solution.
Climate change is such a big challenge that it’s always dangerous to say we want to find the one miracle solution. We need a combination of all these pathways. Depending on conditions — in some world regions, at some airports, with some airlines — one solution might be most efficient, and somewhere else, another. But we really need a combination of all of them. And in research, which is the foundation, all these pathways should be followed because you never know which solution in practical life will be the best and most successful. But we need to try them all.
That’s a powerful way to conclude! This has been a truly rich and valuable conversation, connecting many aspects the public doesn’t always have the chance to explore. Hearing this from you — a scientist with broad expertise — really helps frame the topic more holistically. As you know, bringing science to life starts with understanding, so thank you again for this meaningful contribution.
My pleasure. I really enjoyed our conversation — your questions made it easier to connect these technical aspects across different fields.
(Interview by Efsun G. Özgencer)