SNCF carbon emission numbers

sncf-connect.com shows "CO2 émis pour cet itinéraire" on its "Détail de l'itinéraire" pages.

Looking at, for example, Barcelona to Lyon on some random date, I see:

(1) a 5h31 TGV journey (with a change at Montpellier): 1.8 kg.

(2) an 8h48 journey using Spanish trains to Girona then Port Bou, then French trains to Narbonne then Nimes, and finally an hour on a TGV to Lyon: 4.8 kg.

I believe this is all electrified, right? And they aren't likely to be running diesels-under-the-wires, UK-style, are they?

So how can they end up with an extra 3 kg of CO2 for the journey using slower trains?

Do they offset (some of) the emissions from TGV service but not regional services, perhaps?

It’s unclear if the figures are per person - if so, lower capacity on the slower service might mean the total gets spread among fewer passengers

Indeed, emission factors are calculated by passenger-kilometer for different categories of trains (TGV, TER etc) and, for TER trains, SNCF take into account the share of electric traction and the load factors at the level of regional operations but not at the granular level of each train service.

This means that for journeys on TER trains, even when they run electric at full occupancy, the emission factor used is somehow penalized by the mere fact that other TER trains run on Diesel with low loads. However in carbon accounting it is always preferable to use penalizing hypotheses rather than too optimistic ones.

On the subject of carbon offsetting, it should never be used to present an artificially lower emission factor, this goes against GHG Protocol standards.

All the methodology is available here: https://www.sncf-voyageurs.com/medi...yageurs_methodologiegenerale-infoges_2024.pdf

SHD said:

All the methodology is available here: https://www.sncf-voyageurs.com/medi...yageurs_methodologiegenerale-infoges_2024.pdf

Click to expand...

It’s an interesting document which highlights how many factors need estimating to get the final figure. My French isn’t great but it looks like they have different emissions factors for trains running to Switzerland and the Netherlands but not explicitly to Spain - another confounding issue might be the different carbon intensity of the French vs Spanish electricity supply

SHD said:

All the methodology is available here: https://www.sncf-voyageurs.com/medi...yageurs_methodologiegenerale-infoges_2024.pdf

Click to expand...

Thanks, that's very useful.

Edited to add:

Two key numbers from that document:

Emissions (kg CO2 equiv.) per passenger-kilometer: TGV 2.0, TER 19.4.

That's clearly wrong if the TER were electric and running on the same track with the same stopping pattern as the TGV but at lower speed.

Consider Marseille-Nice. According to the table on page 7, that is 4.3 kg of CO2 on a TER. But if instead I took the TGV, and travelled via Paris - i.e. Marseille-Paris-Nice, it would total only 3.9 kg!

endecotp said:

Thanks, that's very useful.

Edited to add:

Two key numbers from that document:

Emissions (kg CO2 equiv.) per passenger-kilometer: TGV 2.0, TER 19.4.

That's clearly wrong if the TER were electric and running on the same track with the same stopping pattern as the TGV but at lower speed.

Consider Marseille-Nice. According to the table on page 7, that is 4.3 kg of CO2 on a TER. But if instead I took the TGV, and travelled via Paris - i.e. Marseille-Paris-Nice, it would total only 3.9 kg!

Click to expand...

It's not "wrong" per se, it's just that they are using averages per train type, not per route. In reality, some routes will have lower emissions (e.g. fully electrified routes with high average load factors), while others have worse (e.g. diesel routes with low load factors). Even individual services will differ, as the higher the actual loading, the lower the emissions per passenger-km. For a long time, UK national rail reported emissions per passenger-km that were consistently too high, as they consistently used out of date carbon intensity figures for electricity. For fully electrified routes, this was then compounded due to averaging with diesel routes.

It does seem like an odd choice to have all TER grouped together, rather than varying it by region or by fuel type (electric vs diesel), but that could well be due to data limitations.

A further factor impacting the figures could well be that TER services stop and accelerate more frequently, and that could lead to an average energy consumption that is higher than on a TGV service, especially if the TER stock is (or was) older on average than the TGV stock.

== Doublepost prevention - post automatically merged: 16 Jul 2026 ==

endecotp said:

Emissions (kg CO2 equiv.) per passenger-kilometer: TGV 2.0, TER 19.4.

Click to expand...

Presumably these figures are per 100 passenger-km (or something similar), rather than per passenger-km?

Trainbike46 said:

Presumably these figures are per 100 passenger-km (or something similar), rather than per passenger-km?

Click to expand...

Yes, sorry, they are g/km, just a factor of 1000 out!

== Doublepost prevention - post automatically merged: 16 Jul 2026 ==

Trainbike46 said:

A further factor impacting the figures could well be that TER services stop and accelerate more frequently, and that could lead to an average energy consumption that is higher than on a TGV service, especially if the TER stock is (or was) older on average than the TGV stock.

Click to expand...

I'd love to see that quantified. (Do electric TERs have regenerative breaking?)

I used the example of Marseille-Nice because the TGV and the TER go at about the same speed and have fairly similar stopping patterns. Yet looking right now, it claims that a TGV with 4 stops taking 2h38 uses 0.9kg CO2, while a TER with 6 stops taking 2h40 takes 5.5kg.

endecotp said:

Yes, sorry, they are g/km, just a factor of 1000 out!

== Doublepost prevention - post automatically merged: 16 Jul 2026 ==



I'd love to see that quantified. (Do electric TERs have regenerative breaking?)

I used the example of Marseille-Nice because the TGV and the TER go at about the same speed and have fairly similar stopping patterns. Yet looking right now, it claims that a TGV with 4 stops taking 2h38 uses 0.9kg CO2, while a TER with 6 stops taking 2h40 takes 5.5kg.

Click to expand...

It's not a claim, it's the result of a model, which has its own limits. Trainbike46's explanations above are excellent.

Remember that those are figures per passenger-kilometer.

For TGV (and Ouigo), SNCF Voyageurs has very precise data about the number of passenger-kilometers on their services, notwithstanding edge cases such as fare evaders and disruptions. Typical TGV journeys include significant sections of high-speed running, where passenger-kilometers clock up quickly. It is obviously possible to buy a Marseille-Nice TGV ticket, and to travel on a TGV between those stations, but this is not a representative journey of TGV operations.

For all the faults of French railways, it should not be forgotten that the total number of passenger-kilometers per year in France is comparable or equal to Germany's, while the number of long-distance train-kilometers is roughly two times lower (the French network has low utilization rates), and this is mostly due to the fast-and-long-distance TGV operations.

Compare https://ec.europa.eu/eurostat/datab...d/default/table?lang=en&category=rail.rail_pa and https://ec.europa.eu/eurostat/datab...h/default/table?lang=en&category=rail.rail_tf (filter on passenger trains). I use 2019 data because for some reason Germany has ceased providing its statistics to Eurostat after 2021.