Like any economic sector, aviation has an impact on the climate and contributes to ongoing global warming.
This contribution is largely dominated by in-flight emissions from aircraft: it is due to carbon dioxide (CO2) emissions – in 2018, they represented 2.5% of global CO2 emissions from fossil fuels – but also to a number of so-called “non-CO2” effects.
These “non-CO2” effects are related to nitrogen oxides (NOx), water vapour, and particulate matter – also known as aerosols when these particles are airborne – emitted during the combustion of paraffin by engines.
The non-CO₂ effects of aviation
NOx has no direct effect on the climate, but it does have an impact on ozone (which it helps to increase short-term concentration) and methane (which it helps to decrease medium-term concentration). Ozone (O3) and methane (CH4) are both greenhouse gases.
Under current conditions, it is commonly accepted that the warming effect of increased ozone due to aviation outweighs the cooling effect of decreased methane. The resulting effect depends not only on the amount of NOx emitted, which depends on the aircraft models, but also on the altitude, latitude and season of flight.
Water vapour emitted by aircraft is a greenhouse gas that tends to warm the climate when emitted at altitude. But, above all, it is the source of contrails when atmospheric conditions allow their formation – and the occasional transformation of these into cirrus clouds.
In the latter case, we speak of “contrail-induced cirrus clouds”; the ice that makes up the cirrus clouds would not have condensed without the passage of the aircraft or it would have done so only later.
The warming greenhouse effect of aircraft contrails and cirrus clouds outweighs their cooling albedo effect. But here again, there are very large variations in the total effect depending on atmospheric conditions, latitude, time of flight and season.
Finally, the particles emitted by aircraft can modify the optical properties of the different types of clouds they encounter before being carried away by rain or deposited on the surface. These effects are, however, very complex and still very uncertain, so that even their warming or cooling character remains unknown.
Different time scales
There are several different time scales associated with these different effects.
CO2 is a long-lived gas in the atmosphere. When one ton of fossil CO2 is emitted into the atmosphere, about 30% of the initial increase in concentration persists after 100 years and a further 20% after 400 years.
Conversely, NOx, particles and water vapour disappear from the atmosphere after a few weeks if they are emitted at the cruising altitude of aircraft. The same applies to their induced effects on ozone or clouds.
On the other hand, the effects of NOx emitted by aviation on methane materialise on an intermediate time scale since the lifetime of methane in the atmosphere following a disturbance is 12 years.
Beyond the time scales associated with the lifetimes of chemical species, we must also consider those of the climate system itself. A disturbance of energy introduced into the climate system, even over a short period of time, has a lasting impact on the climate, as the ocean absorbs this additional energy before gradually releasing it back into the atmosphere.
The radiative forcing index is a bad idea for accounting for emissions
Climate change is usually measured by the concept of “radiative forcing”: this quantity measures the radiative imbalance of the planet due to past emissions and is expressed in relation to a reference period generally set at 1850, a time when industrial activities were still low.
For CO2, which has a long lifetime in the atmosphere, and to a lesser extent methane, this incorporates past emissions that have a lasting impact on atmospheric concentrations. For short-lived pollutants, only the most recent emissions matter, as the oldest emissions no longer exert a radiative forcing.
For the aviation sector, according to current knowledge, non-CO2 effects are responsible for a positive radiative forcing that tends to warm the climate. The ratio of total radiative forcing to the radiative forcing due to CO2 is called the Radiative Forcing Index (or RFI).
Some carbon footprint calculators use the RFI as a multiplicative factor of CO2 emissions to take into account non-CO2 effects and thus “convert” CO2 emissions into “CO2-equivalent”. However, we believe that this does not make much sense.
To convince ourselves of this, we can perform the following thought experiment: let’s assume that before the Covid-19 crisis, in 2019, non-CO2 effects are responsible for twice the radiative forcing of CO2, which corresponds to an RFI of 3 (i.e., (2 + 1)/1). One ton of CO2 emitted by aviation would therefore correspond to 3 tons of “CO2-equivalent”.
At the peak of the Covid-19 crisis in spring 2020, aviation activity was reduced by a factor of 4. CO2 emissions from aviation then fell drastically from their 2019 level, but this did not lead to a decrease in radiative forcing from CO2, as its concentration in the atmosphere continued to grow.
The radiative forcing from non-CO2 effects, on the contrary, decreased in line with the decrease in traffic (even by more than a factor of 4, as the methane effects of past emissions persist over time). During Covid-19, taking into account the same radiative forcing for CO2 – which increases only slightly from one year to the next – but a forcing divided by 4 for the “non-CO2” effects to take into account the reduction in traffic in 2020, we arrive at an RFI of 1.5 (i.e. (2/4 + 1)/1). One ton of CO2 emitted by aviation would therefore correspond to only 1.5 tons of CO2-equivalent instead of the 3 tons of CO2-equivalent before Covid.
This is nonsense because flights in spring 2020 will of course have the same climate impact as the same flights in spring 2019! So there is no reason why flights in 2020 should “count” half as much as those in 2019.
The fundamental reason why RFI is not appropriate as a multiplier is that radiative forcing accumulates the effects of past emissions, whereas we wish to compare the climate effects of current emissions, either to allocate an aviation user’s fair share of emissions, or to assess different technological or operational options that could be implemented in the future.
Which climate metric to use?
Fortunately, there are climate change metrics that can be used to estimate the future climate impact of a flight today, despite the short time scale of radiative forcing from non-CO2 effects.
In particular, the Global Warming Potential (GWP), which measures the radiative impact over a future period, typically 100 years, of emissions that occur at a point in time. The GWP of one kg of pollutant (such as CH4 or NOx) can then be compared with that of one kg of CO2, and the concept can easily be extended to aircraft trails.
Another metric, the Global Temperature change Potential (GTP), is defined in a similar way, but based on the change in average surface temperature at a certain time (50 or 100 years) after an emission pulse.
These metrics lead to much lower CO2 multiplication factors than the RFI unless time frames much shorter than 50 years are chosen. The choice of time frame is a political choice that can have important implications. A short time frame neglects the substantial part of the CO2 warming that occurs beyond the time frame. Choosing a long time frame may minimise the short-term effectiveness of warming reduction solutions based on non-CO2 effects.
To illustrate the importance of the metric used to estimate the total climate impact of aviation or when calculating the carbon footprint, let us compare the RFI of aviation with the multiplicative factors associated with these other metrics.
According to a recent estimate, the various radiative forcings from aviation result in an RFI of 2.9. If the GWP at a time horizon of 100 years is used to calculate the equivalent emissions of the different aviation disturbances, a multiplication factor of 1.7 is derived. For the TMP with a time horizon of 100 years, the multiplication factor is only 1.1.
This shows that the choice of one metric rather than another is crucial for calculating the multiplication factor.
A decision aid
Some of the solutions envisaged to reduce the climate impacts of aviation have the dual advantage of reducing both CO2 and non-CO2 effects.
In this case, the metrics simply serve to quantify the net gain for the climate. In contrast, other solutions require a trade-off between the CO2 and non-CO2 effects of aviation. For example, engine manufacturers know how to reduce NOx emissions but often at the expense of CO2 emissions. A partially or fully decarbonised fuel, such as hydrogen, could lead to greater non-CO2 effects.
Strategies for modifying aircraft trajectories to reduce the effects of contrails or NOx can also be envisaged, but at the cost of increased fuel consumption and therefore CO2 emissions. In these cases, it is relevant to compare the different effects (CO2 and non-CO2) with several adapted metrics to understand which effect prevails over the other and at what timescale (20, 50 and 100 years for example) and to be able to take the best possible decisions.
As far as we know, the non-CO2 effects of aviation have, on the whole, a warming effect on the climate. It is therefore relevant to try to reduce them in order to reduce the overall warming impact of aviation.
It is also important to ensure that the technologies being developed to decarbonise aviation do not induce excessive non-CO2 effects. Each solution must be examined and its impacts assessed using the most appropriate climate metrics, without forgetting to take into account other possible impacts (air quality, noise, biodiversity, etc.).