The Global Warming Reduction Potential of Night-Trains

New report: Night trains can cut 3% of total EU greenhouse gas emissions

According to opinion polls, 7 out of 10 Europeans would be willing to take the night train instead of the plane if the offer seemed reasonable to them. Back-on-Track, a European network of night train initiatives, has used this as a basis to examine air passenger numbers in the EU in 2019 to see which air connections could be replaced by night train connections. Distances up to 1500 km as well as distances up to 3000 km were considered with different scenarios. Overall, up to 32 % of passengers could switch to night trains if there were an attractive offer. This would reduce emissions from air traffic by 26 %. In order to create such an offer, up to 2500 more night trains would be required, as well as a considerable improvement in the framework conditions, in particular a reduction in track access charges.

Why this report?

The potential of night trains has been investigated in a number of recent studies, but none of these studies answered the question of how much greenhouse gas emissions could be avoided by shifting passenger traffic from air to rail in a best-case scenario where all barriers are removed. The studies either investigated the potential of a predefined network or the examination of passenger potential passenger numbers were based on some given constraints.

And when climate effects were calculated, the non-CO₂ radiative forcing of aviations was either ignored or more cautious assumptions were used. In particular, knowledge about the effect of water vapour has only consolidated in recent years.

How did we calculate aviation emissions?

We wanted to base our calculations on the latest findings and to do this we first had to supplement the most reliable database, the EEA emissions data, with the best available calculation value for the greenhouse effect of contrails, which makes a significant difference. Since the effect of water vapour weakens quickly, it makes a big difference whether one wants to calculate the current global warming (with GWP*) or the effect over a period of 100 years (with GWP100). We have opted for the former. Not so much because a saving of 1.7% sounds less exciting than 3% – even 1.7% would be an impressive figure. But because the question is how to combat current global warming, not how to predict the climate in 2122, and here GWP* provides the better answers.

Adding the radiative forcing of contrails and other non-CO₂ greenhouse gases based on GWP* triples the CO₂ value to a share of almost 12% of total EU greenhouse gas emissions (not the CO₂e value, which includes greenhouse gases other than water vapour).

Even the CO₂e emissions share from aviation of more than 5% of total emissions (inserted in Figure 1 for information) may come as a surprise, as the aviation industry keeps referring to two figures – either the share of CO₂ emissions from domestic aviation (not including flights between two EU Member States) which which is about 1% of national emissions, or the share of aviation in global CO2 emissions, which is between 2% and 3%.

A serious problem

Emissions from international aviation (e those from international navigation/shipping) have been excluded from nationally determined contributions (NDC) so they are not part of binding agreements. For this reason, they are usually not included in transport sector emissions in statistics. This is misleading, as flights between two EU Member States also count as international aviation by definition. So about 80 % of all emissions from aviation do not appear in transport statistics – We have corrected this in Figure 2.

This makes more obvious, why emissions from aviation have become a serious problem. Unlike in road traffic non-fossil sources of energy are not even in sight. And unlike road traffic where emissions fell for some time and then stagnated, emissions from aviation have steadily increased. Aviation emissions are the main reason why transport is the only sector where emissions have not decreased in the last 30 years, but have increased by almost 40% – of course, if you don’t exclude international aviation. We have extended the development of aviation emissions according to the WEM forecast to show: If we do not act, aviation emissions will be the single largest contributor to climate change by 2040, ahead of the energy supply and industry sectors

To tackle this problem the European parliament recently discussed slowly increasing the share of so called sustainable aviation fuels – carbon-based fuels derived from biomass. That is generally to be welcomed, as these fuels are carbon neutral, more effective but also much more expensive. However they cannot solve the problem of water vapour, so a significant global warming effect remains.

Other solutions are batteries, which make the planes too heavy to travel longer distances, solar cells, which require huge wingspans that slow the planes down, or hydrogen, the most reasonable solution so far, but only if the water vapour from the fuel cells is emitted pre-condensed. However, all these other solutions are still far from being ready for the market.

Why night trains are an obvious solution

Night trains do unfortunately not yet play a role in this discussion.

Although a commonplace is that trains are better for the environment and especially for the climate than planes, the debate cites very different dimensions of this relationship. As the TRAN Chair of the European Parliament, Karima Delli, said at this year’s Back-on-Track conference, “some say emissions are six times higher, some say as much as ten”. Austrian railways, by the way, say 50 times. While all these ratios may be valid for something, we need to define what we want to know here: The valid ratio for the EU energy mix, for night trains with high load factors, for well-to-wheel emissions (i.e. including emissions for fuel production and transport) for the reference year 2019, for a per-km value and including radiative forcing of non-CO2 greenhouse gases. The correct ratio is then 1:28, which means that an average night train in the EU will cause 3.6% of the emissions of air transport on the same route in 2019. We have visualised this ratio in Figure 3 and compared it with other modes of transport.

However, night trains are the only obvious solution to replace the aeroplane on routes of more than 500 km (perhaps 700 km in France) by land transport, where high-speed trains can no longer compete, at least not in terms of travel time.

As shown in Figure 4, night trains become an attractive alternative to air travel above 500 km, when high-speed trains and routes can no longer compete with the total journey times of aircraft. Up to 1500 km, they can use conventional tracks and RIC-compatible standard rolling stock. High-speed night trains running on existing high-speed lines can cover even further distances of up to 3000 km.

Night trains could save the greenhouse gas emissions of air travel in just a few years if policy makers acted now. We have to admit, however, that EU railway policy is not the easiest field in which to act. In order to measure whether the effort is worthwhile, one should know what result can be expected in return. That is why we have calculated the achievable greenhouse gas reduction.

We were able to base our report on the 2021 study by our french group Oui au train de nuit, which examined EU passenger data to determine the volume of passengers that could be shifted to night trains.

Unlike the present report, our 2021 study suggested that distances between 1500 and 3000 km could be covered by conventional rolling stock running for more than one day – albeit with only a fraction of the passenger volumes assumed for distances below 1500 km. While this is still a viable option, high-speed night trains could of course keep travel times competitive with air travel, so it can be assumed that a higher proportion of passengers will transfer. For this reason, high-speed trains are the means of choice in a best-case scenario in terms of avoidable emissions.

What is in the best case scenario?

We assumed a capable rail infrastructure on TEN-T lines prepared for at least 160 km/h max. speed and equipped with ETCS, the completion of construction projects (like Fehmarnbelt or Brenner tunnel, Madrid-Lisbon or Warsaw-Tallinn), the opening of high-speed routes for technically suitable night trains, modern rolling stock with substantial improvements of privacy, security and comfort, an average occupancy rate of 80% like for other means of transport that require booking, track access and station charges at marginal cost, a VAT exemption for cross-border journeys and platforms allowing the purchase of tickets from start to finish for all rail routes with a best price guarantee. These are a lot of wishes, we know that. It is a best-case scenario, but none of it is unachievable.

What is the answer to the question?

To learn more about our calculations, please read the report, which also answers some frequently asked questions we gathered when discussing our paper with other organisations working on transport and environmental issues. You can also download our dataset. The table contains all assumptions and methods, some of which can be changed to check their influence on the result. But we won’t keep you in suspense any longer, here is the answer to the question of how much greenhouse gas emissions could be avoided by shifting passenger traffic from air to rail:

Frequently asked questions

1) The network does not have the capacity, particularly looking at congested main stations:

True right now, it would not have the capacity for all the lines needed to reach the calculated goal of 3%. But apart for some exceptions we would not need more than the realisation of reinforcement projects which are currently under way and could be finished by 2035. Night trains are a very effective way to increase rail capacity, as they are flexible to fill gaps and you can add padding at night. We estimated that with our maximum scenario, high capacity and occupancy rate they would in average add not more than 3 trains per hour per direction to one of the 12 main gateways. This is 15% of the total capacity. We think this is not impossible.

Regarding high speed tracks, this would imply political decisions on the best way to use them over night: For cargo traffic or maintenance (are they really doing works every night?), or for high speed passenger traffic? With our study politicians can now decide which usage could do more to protect our climate. 

Regarding congestion in the main stations, night trains can do with minor stations without loosing too much passenger volume: Frankfurt Süd, Wien Westbahnhof, København Ny Elleberg … Of course stopping in the main station would be ideal for changing to day trains. But these minor stations are still much easier to reach than airports. 

2) The rail industry does not have the capacity to build all these trains

Our best-case scenario would require a lot of new rolling stock, indeed. Spread over a time frame of 12 years our estimated need for rolling stock would add 14% to the current market volume. This is a lot, but not impossible. Five percent growth have been predicted for the rail industry anyway. It also depends on what you need. For 200km/h carriages the capacity is there right now. This might not be true for high speed trains but at least Alstom currently has spare capacity in Germany – it wants to lay off people. But mainly it depends on the demand. If Tesla has extra demand they build a new plant. Rail manufacturers will do the same, but only if clear political decisions are taken which will create the demand: By lowering track access charges, speeding up ETCS rollout, opening the high-speed lines and sorting out rolling stock financing. 

3) Using opinion poll data is not suitable to estimate shift potential

No. The real potential is in the end a political question: How do you set market conditions that allow trains to be attractive? This will ultimately change the chosen means of transport. Recent studies estimated the potential by looking at old perceptions of night trains. This method did lead to predictions that have recently been outnumbered. There is a conceivable shift in preference by travellers, most night trains are currently booked out. And current preference rates might further increase once the new generation of rolling stock with capsule beds as recently presented by ÖBB is available. These studies were very valuable non the less, as they show which constraints have to be lifted to increase potential.Provide a better product and you might attract a new market: Once these constraints are lifted, we end up with the hypothetical general preference, for which these opinion polls provide the best data available. And even if you don’t want to believe in more demand: Even under given circumstances there is proven demand for at least 15 more routes in Europe. Let’s make a start with those, and a start with the rolling stock to run those, and see where we get.

4) The result of opinion polls are very dependent on how you ask.

This is true in general, but the consumer preference of 7 in 10 form the poll on which we relied, is in line with other polls. At a recent Civey poll for Germany with a slightly different question (focused on availability, not on reasonable price) also showed that 7 in 10 would decide in favour of night trains  So we are confident that this number is quite close to real preferences. 

5) The potential estimation is imprecise

True. We can be more precise once we look at each possible train in the future. This will decrease potential here and there. On the other hand we did not include flights under 100.000 passengers, this might increase numbers. Also, some flights might disappear completely due to a good alternative, this effect was also not yet considered. So we are confident, that the magnitude will not change too much. 

6) The study does not aim to provide a realistic plan

We aim to define a potential, not to provide a prediction. This is why our calculation is only looking on a best-case scenario. But even a best case-scenario is attainable – the estimated potential is technically possible. The question is, whether it is politically conceivable. Here, we share some scepticism.