The RER Paris Needs

Since the 1960s, Paris has gradually built itself to have a 5-line regional rail network connecting the city and its suburbs, with more than a billion riders a year. Unfortunately, investment has been slow in the last 20 years; the fifth line, the RER E, is being extended to the west, but other problems are not being fixed through more investment. Some regional rail lines remain disconnected from the system, including one of the city’s six intercity rail terminals, Gare Montparnasse. While east-west capacity is being augmented through the RER E extension, north-south traffic is jammed and yet is not slated to receive any relief, despite past studies.

Taking everything together, this is what Paris needs to do to complete the conversion of all commuter rail in Ile-de-France to RER standards:

Full-size image can be seen here; warning: 71 MB.

Dashed lines are new tunnels to be built. Most of the dashed green line is the under-construction RER E extension from Saint-Lazare to La Défense and points west. The remainder, between Les Halles and Auber/Saint-Lazare, is a new tunnel that should be built, giving away the extension to the RER D instead. With a full line extended, the RER D could take over the entire SNCF-run part of the RER A while also continuing west to Mantes-la-Jolie as is planned for the RER E extension, so the RER A can gain the Transilien L branches to the southwest with a short curve from La Défense to Puteaux.

In addition, what is now the shared RER B and D tunnel between Gare du Nord and Les Halles should be four-tracked; the stations at both ends thankfully already have separate platform tracks for the RER B and D, and in 2003 a somewhat disruptive plan to four-track the tunnel was estimated to cost 700 million. Since the RER D tracks are to continue west, the new dual track tunnel should continue south across the river and connect to Montparnasse, creating the RER F; the RER F should take over the current northern branches of the RER B to form a southwest-northeast line, while the current southern branches of the RER B should be connected to what are now the northern branch of the RER D and the branches of Transilien H.

The RER C and E should be broken and recombined using a short four-track tunnel across the river, creating northeast-southwest and northwest-southeast trunk lines. Today, the RER C misses the Paris CBD and has an awkward connection to the RER A; with this recombination, the connections would still require a lot of walking at transfer stations but they’d exist and passengers would get solid two-seat rides.

Finally, a handful of outer-urban and suburban fixes would be useful: a few infill stations, depicted with gray filling; using all four tracks on the RER F trunk line to Aulnay (currently the RER B) to make it easier to run express trains to Charles-de-Gaulle; building a short suburban tunnel through Chaville to connect the RER F and A branch; continuing T3 to form a full circle using the Petite Ceinture in lieu of the awkward RER C branch today; constructing an infill station at the RER E/F junction in Meudon.

Excluding the ongoing RER E extension, the total length of new tunnel in city center is 8 km of two-track tunnel and about 1.5 km of four-track tunnel. This would set non-Anglosphere world records in construction costs per kilometer, just as the RER A did; costs in the 400-500 million per km can be expected given the complexity of tunneling under so many older Métro lines, so the system would cost around 5 billion, perhaps reaching 6-7 billion with the extra suburban tunnels and infill stations.

The map doesn’t go to the edge of Ile-de-France, or else it would be even bigger, but the plan should be to connect every Transilien line to this system, even ones in faraway exurbs. Frequency to the exurbs need not be very high – today they get hourly service off-peak, and half-hourly service in the future should be plenty to small towns on the edge of Ile-de-France; of course, closer-in suburbs as well as major secondary centers like Meaux and Evry should get much higher frequency, and the trunks should get a train every 3-4 minutes even off-peak.

The point of this exercise is that Paris has already done the hardest parts. The RER A and B exist, and Châtelet-Les Halles was dug at enormous expense in the 1970s. Even at the high per-km costs of connections underneath the center of Paris, the tunnels Paris needs to build in the next 10-15 years are low-hanging fruits for completing the project of connecting the entire region through a unified RER network.


  1. Nilo

    I enjoyed this crayon. The basic intent is that ever RER crosses through the center of Paris with its own devoted two tracks. How many of the transfers are capable of being a cross transfer? It seems like B-F and C-E should be since they involve new construction. It seems like the extant A-D transfer isn’t cross platform, would it be possible to make the RER D platforms at Auber cross platform from the RER A ones?

    • Alon Levy

      A-D has no way of being cross-platform, Auber is so constrained that even an ordinary Métro transfer quality would be difficult, and it might end up being as bad as a Métro-RER transfer. But Les Halles has non-terrible transfers that are not cross-platform – I transferred wrong-way between the RER A and B for a few weeks and it was fine.

  2. RVA_Exile

    Paris is getting multiple circumferential metro lines as we speak; surely the Grand Paris Express will reduce demand on central segments (for a city that is not growing that fast)?

    Is there precedent for a station/transfer onto an existing bridge with the vertical separation of the Viaduc de Meudon and the RER C in the valley below?

    The T6 is already tunneled at some expense between Viroflay and Viroflay-Rive Droite for a connection between these two outer branches.

    • Herbert

      The distance between stops on (older lines of the) Paris metro is famously tiny… you need a system with bigger distances between stations for the outer suburbs…

      • michaelrjames

        The M15 orbital line under construction will have 36 stations spanning its 75km for average spacing of 2.08km, about 3 times the existing Metro station spacing, and comparable to the current RER average of 2.3km.

      • RVA_Exile

        If this is in reference to the new circumferential metro lines, the Grand Paris Express will be designed for speed and be just as fast as the RER lines if not faster, much like the “express” Line 14 through Central Paris today.

    • RVA_Exile

      Just noticed that this map severs the current Line L, with North-South service through La Défense on the way to Saint-Lazare.
      La Défense is big enough to need its own radial network and a lot of people are commuting from inner suburbs to La Défense (Asnières, Bécon, Clichy, Levallois). Not everyone is going to Paris 8.

      • michaelrjames

        Won’t a lot of those users be served by M15 which will have stations at Nanterre-la Folie, La Defense, Bècon-les-Bruyéres, Bois-Colombes, Les-Agnettes and Les-Gresillions (=Asniéres)? ie. the suburbs north (and south, not named here) of La Defense and including Levallois-Perret & Clichy just across the river from these stations.

        • Yom Sen

          Bécon la Bruyère will be served by M15 but not Courbevoie, Asnières, Clichy-Levallois and Pont-Cardinet. Courbevoie, Asnières, Clichy and Levallois are all cities with more than 60k people and densities around 20’000/km2 so multiple stops make sense. I think they have a lot of jobs as well, Clichy-Levallois for example is at walking distance from L’Oréal and Bic headquarters. Asnières and Pont-Cardinet would also allow connections to the yellow line in Alon’s plan. I don’t think it’s worth connecting current line L to RER A instead of going to Saint-Lazare.

          • michaelrjames

            (Oops, posted incorrectly, second attempt.)
            Bécon-les-Bruyères is Asnières, and it is also metres from Courbevoie. Les-Gresillions is also Asnières–or literally the other side of the road–and at the other end relative to Bécon, and Les-Agnettes is in the middle of Asnières, so it is actually better provisioned than the L. Of course not directly to St-Lazare but M15 has transfers to M13 and RER-C1 at two of those stations.

          • Yom Sen

            So what? Are we really debating about getting rid of a transit line with 12 trains/h at rush hours in the middle of a very dense area (around 20000/km2) and connecting 2 of the largest CBDs in Europe? Asnières station is 1km from Bécon and further away from the other stations. Courbevoie, only served by this branch, has 4M passengers/year even though it is at walking distance from La Défense, Bécon La Bruyère is at 10M, Asnières at 21M et Clichly-Levallois at 30M, 15th busiest station in France almost at par with Lyon Part-Dieu (La Défense is at 62M)
            Note as well that connection in La Défense with M15 will not be ideal since they are not able to build a station close the the Grande Arche, see article below in French. To go from Versailles or Saint-Cloud to the north of Paris a connection to M15 in Bécon la Bruyère will be probably more practical.

          • michaelrjames

            All of that doesn’t mean the eventual configuration won’t provide substitute services. For example, Asniéres has 4 stations on M15 which could absorb all the current pax at those stations you list. And all of those four M15 stations serve directly adjoining suburbs. The main oddity is Clichy-Levallois which has those high ridership numbers because Levallois is a transit desert, and of course it is right on the border of the two suburbs–in fact only ≈800m from M13-Marie-de-Clichy however given M13 crowding I don’t imagine the plan is for pax to transfer to it. I assume they have plans to remedy whatever is lost.

            Anyway, I’ve just exitted the rabbit-hole because it is only Alon who is closing L. In fact his map still retains Clichy-Levallois and the Saint-Nom-la-Bretèche line, so I’m not sure how we got to this point. As far as I can see there are no official plans to close it? This actually makes more sense of the slightly odd looking M15 route because it is not duplicating that line and serves those surrounding suburbs plus Asniéres.

          • Alon Levy

            In fact his map still retains Clichy-Levallois and the Saint-Nom-la-Bretèche line

            Seriously: their, not his.

      • fjod

        Yes, this is a bit of a downgrade for that whole area; the stations at La Garenne – Colombes, Les Vallées and Courbevoie go from having really quite good service now (10 and 15 minute intervals) to what I assume is no service under this plan (meaning bus/tram rides to La Défense or Bois-Colombes). M15 of course doesn’t hugely help, because these journeys aren’t mostly circumferential. I’d be interested to hear the rationale for this, especially considering how much development/intensification potential this area has.

        • Alon Levy

          I have this crayon idea to extend M3 to Bécon and Les Vallées… Slower than Transilien L, but it runs to more than just Saint-Lazare and it comes every 3 minutes, not 15.

          • seb

            A shuttle from Saint-Lazare to La Defense and Nanterre would easily solve this. I guess, the current Transilien L platform at La Defense would still be there. There is enough track capacity (Pont d’Asnieres is 6-track) and removing the Transilien J from Saint Lazare frees platform space. The current 15 minute frequency comes from giving L riders from farther suburbs a faster way to the center. This will no longer be an issue. I think I would also add an infill stop at Courbevoie Sport.

          • Yom Sen

            I just had a look at the schedule, that is a bit weird, each line (Paris to Saint-Nom and Paris to Versailles) has all different stops except La Defense and Saint-Cloud. If you compare to late evening trains that do all stops, the gain is only 2 or 3 min for 4 skipped stations. We could easily double the frequency at 7.5 min off peak and 5 min at peak
            PSL to Nanterre-Université is simpler with 1 train doing all stops every 10 min off peak and 1 every 6-7 min at peak.
            Based on Google Maps, Pont d’Asnières is even 10-track.

          • michaelrjames

            It’s only a tad over 1km from M3-terminus at Pt-de-Levallois-Bécon to Bécon-les-Bruyères (M15) so one imagines it would eventually be done. M3 is one of only two Metro lines (excluding the two inner-circumferentials 2 & 6) that don’t connect to M15. M10 in Boulogne-Billancourt is the other one and it, and nearby M9 terminus, are also quite close and so they may ultimately be joined, is a reasonable presumption–ie. to join M15-St-Cloud and M15-Pont-de-Sèvres, respectively.

          • michaelrjames

            Oops. M15 already will connect to M9 at Pont-de-Sèvres. I had made the incorrect assumption that M15 was still on the Sèvres side but this is the point where M15 crosses the river from leftbank to rightbank.

    • Alon Levy

      The city’s residential growth is negative at this point, but I imagine there’s commercial growth around the center – housing is not being demolished, it’s just being built at sub-New York rates.

      And M15 and M16 are okay for getting people from various inner suburbs to La Défense and Saint-Denis, but a lot of other trips are completely unserved. Ongoing commercial construction around Porte ce Clichy, for example, requires this connection of Transilien J to the RER that I’m proposing, and not GPX.

  3. michaelrjames

    Not bad.
    But as you admit, rather extravagant. The tunnel required from Montparnasse thru to link to D and St Lazare is massive.

    costs in the €400-500 million per km can be expected given the complexity of tunneling under so many older Métro lines

    Is it really related to being under the Metro (or anything)? Surely once you opt for deep tunnel (like RER or London CrossRail) the cost is pretty consistent, and the most expensive part (I learnt this from you) are any deep stations and especially transfer stations? As far as the tunnel itself, it is not so much the depth but the size required for the heavy trains? In fact in Paris deep tunneling is more to avoid all the other stuff–the sewers, water, telcoms etc, under the city since the Metro is mostly very shallow.

    Maybe you’ll be alive when they finally get around to building something like this, but I suspect I won’t be. I might be one of those Permanent Parisians in Pere Lachaise! (Yes, on my bucket list is to kick the bucket in Paris! Then be interred within proximity of Colette et al.)
    As RVA_Exile points out, the next €30bn or so of works will be consumed by the GPX plan. It will change the pax flow but how much it might reduce pressure on the RER is to be seen.

    • michaelrjames

      Looking at the central bits I’ve realised that many of those switches you’ve made would be extremely expensive and disruptive. In the past I have suggested just the disruption to D (to build your Nord-Chatelet tunnel) would be a considerable factor in reluctance to do it; it would require D to terminate at Nord and from the south at Chatelet, for at least 2 years? And your mix-and-matching of NE-B to SW-Montparnasse, and some of the others, will involve very complex stitching of existing tunnels to new ones, almost certainly the two partners are at different levels, requiring closing the line(s) for the work.
      You’ve abandoned C1, ie. the bit that crosses the river over Ile-des-cygnes? And with its remaining northern bits you’ve created a complicated branching-looping structure of the sort you criticize the (southern) line C.
      BTW, Herbert that bit of C1 reused part of the old PC railway. If they were going to reuse more of it, they would have used it for T3 instead of building it on the Boulevards des Marechaux but I am sure there were multiple reasons both for not using the PC and for building the tramway. For one thing the traverses across the Seine don’t exist anymore, except for that one used by C1. The original Auteuil bridge-viaduct (aka Pont du Jour) that carried it was a massive affair that is no longer there–I think it was replaced by Pont du Garigliano; I’ve never read the history of why it disappeared (flood damage or demolition?).

      Anyway, Alon’s plan is quite something. All mainline rail stations are connected to at least one other and all 6 lines pass thru the centre with four serving the Auber-Haussman-St-Lazare locus and four serving Chatelet-Les Halles, and two serving both. Three serve Gare du Nord (as they currently do), two serve Gare de Lyon (ditto)
      There is a pleasing symmetry and aesthetics about it. It is quite beautiful. That kind of thing appeals to the French but not enough to overcome the obstacles in the near or medium term. But for me it has been firmly established as the equivalent of an earworm.

    • Alon Levy

      Yes, the cost is mainly the stations, but these stations are extremely expensive, like the €450 million RER E La Défense station. I haven’t seen a detailed breakdown but I imagine the €700 million in 2003 terms for quadrupling the RER B+D tunnel is mostly not two single-track tubes but rather connections to the platforms at Les Halles and Gare du Nord; of course if a new station has to be built then the costs go into the billions, as the link in the post says (and this is why I’m pushing the cheaper but more disruptive option).

  4. Alex B

    Would be interesting to use the abandoned section of the RER C to extend the T3b south to the 16th and potentially connect to T3a somewhere. Or let it be a standalone shuttle line?

    • michaelrjames

      That’s not a bad idea, except of course there are no official plans to abandon that segment of C1 (are there? or does Alon know something …). It would be nice for completionists! And for historicists since it recreates the PC network by re-using part of it. Incidentally the norther segment of this PC line was the Auteuil-St-Lazare line which was the first passenger Metro type service in 1854, even earlier than London’s Metropolitan line (1863) so there is deep history in that track.

      • Yom Sen

        There are plans to extend T3b to Porte Dauphine in 2023 and RER C is using PC only until the next stop Avenue Henri-Martin. The following part was a junction to connect Saint-Lazare and Invalides the 2 main stations for the Compagnie de l’Ouest. A tramway between Porte de Clichy and Champ-de-Mars would be nice, probably more adapted than RER C with better frequencies. Duplicating tramway between Porte de Clichy and Porte Dauphine would be a bit akward so maybe it’s better to extend T3b and just use RER C tracks from Porte Dauphine.
        No, no plan to abandon this segment, we’re just dreaming here, Alon’s plan is unfortunately very unlikely…

        • RVA_Exile

          Could just do a T3a/b surface line somewhat like Alex B is suggesting, from Porte Dauphine to Pont du Garigliano to complete the loop, leaving the RER C tranquille in the tunnel
          From Porte Dauphine along Bd Flandrin-Sandeau-Augier-Beauséjour-Montmorency-Exelmans. RoW still exists for most of it

        • michaelrjames

          RER C is using PC only until the next stop Avenue Henri-Martin.

          Sorry, not being difficult but I just don’t understand that. It appears to follow the PC route all the way from Kennedy-Radio-France to Porte-de-Clichy, eg. it runs under the full length of Bd Pereire.

          Re T3b: yes, it looks like eventually they will fill the gap (Dauphine to Garigliano) and no reason not to, in that the Boulevards-des-Marechaux are as wide as elsewhere. I suppose it serves fewer people because one side has the bois, and perhaps fewer destinations.

          • michaelrjames

            OK, I understand now but think you are being a bit too literal. You’re saying only the very circumferential parts are the PC–as implied by that drawing and the caption indicating the PC in bold. But the spur (not bold) from Henry-Martin to Kennedy-Radio-France and across the island to leftbank was part of the same PC network, and that spur served one of the investors in the syndicate that built the PC, namely the Ouest company. True, they only created it later in the century for PC connections to their Champs du Mars and Invalides stations.

  5. anonymouse

    Chatelet has 7 platform tracks. How do you intend to go about fitting 4 lines through there? Is the station wide enough to rearrange the platforms?

    • Alon Levy

      The station is extremely wide. The RER A and B platforms are 17 meters wide, and even at rush hour they’re not full – the limiting factor is vertical access points and capacity on the train, not the platforms themselves. I think the RER D platforms are narrower, but still wide enough you can turn the center track into two tracks without too much pain.

  6. Lee Ratner

    Sometimes I think that we transit policy experts and transit fans are able to come up with good plans on paper but we aren’t burdened by a real actual budget or politics. When you don’t have to deal with these things, it is very easy to come up with great plans. Once you run into the problems of budget and that different politicians, civil servants, and other citizens have their own ideas on what is good and bad transportation policy, it becomes a lot harder.

    • Alon Levy

      Okay, but the two useless lines of GPX – M17 and M18 – cost in the vicinity of what I’m proposing for this. The cost per km of M17 and M18 is of course much lower because they’re suburban tunnels, but they’re long, whereas the point of RER and S-Bahn tunneling is only tunnel the central few kilometers at high per-km cost and low per-rider cost.

      • michaelrjames

        That is fairly shocking. Am I wrong again in having assumed most of these would be reusing existing track? For example I assumed M17 was just taking the place of the current RER-B3 (CDG) which itself will be replaced by the Airport Express. M17 has all the same stations listed and thus I assumed it would be a very cheapo line to create.

        • Alon Levy

          No, M17 is an entirely greenfield line. See map here:

          It’s schematic at first but if you zoom in enough it becomes geographically accurate and makes it clear the line runs parallel to the RER and only uses the RER right-of-way around Parc des Expositions.

          The CDG Express is built as a combination of greenfield and existing. The Gare de l’Est platforms preexist, but the plan is for a tunnel to the approach to Gare du Nord, then the ROW toward the airport (I think Transilien K, not the RER B), then a connection to the LGV Interconnexion Est. I think the airport platforms involve a new tunnel as well but don’t quote me on that.

          • Mikel

            (quoting in case this comment ends up in the wrong thread)

            The CDG Express is built as a combination of greenfield and existing. The Gare de l’Est platforms preexist, but the plan is for a tunnel to the approach to Gare du Nord, then the ROW toward the airport (I think Transilien K, not the RER B), then a connection to the LGV Interconnexion Est. I think the airport platforms involve a new tunnel as well but don’t quote me on that.

            The CDG Express seems kind of overkill given RER B and M17?

            Also, according to that map M14 and M17 both terminate at Saint-Denis Pleyel. Would it make sense to through-run them so that the Saint-Lazare area gets a one-seat ride to CDG? Or would the interlining with M16 mess up the frequencies?

          • Yom Sen

            CDG Express will connect the airport only to Gare de l’Est while RER B connects it to Gare du Nord but also many other stations in Paris and southern and northern suburbs. It will be much more expensive and managed privately more or less like Orlyval before its bankruptcy so it will probably have the same fate…
            I still don’t get why are they building M17. Its main purpose seems to be connecting Saint-Denis-La-Plaine and Le Bourget to CDG even though they are already among the places with the best connection with CDG through RER B. Second purpose would be to serve Le Bourget airport (the airport is only for private jets, but the area is also residential) and Triangle de Gonesse which is was to be developped with Europacity. Le Bourget aéroport could be served by a short M7 extension and Europacity project has been abandonned in 2019. RER B frequency is not very good but we could increase it. Same than for Saint-Lazare – La Défense, poor frequency is due to having several services with different stop patterns. Frequency at rush hours, the only one constrained by capacity is actually quite good at 1 train every 6 min, the problem is off peak when only 1 train every 15 min serves intermediate stops.

          • michaelrjames

            RER-B3 (the spur from Aulnay to CDG) is being discontinued when the CDG-Express starts up, because they want to run more trains on the main line (B5) which is being extended and upgraded. The new airport train is intended to be express so M17 is to serve those stations currently served by B3 (Villepinte & Parc des Expositions); I assumed it would be taking over the B3 tracks but Alon says it will have something new. I guess its other function is to connect the new mega-station of Pleyel-St Denis to these sites and CDG, to take pressure off Gare-du-Nord/de l’Est.
            At any rate it has been delayed yet again, out to 2025.

          • Yom Sen

            No, CDG Express will use the tracks for Paris-Laon and a short new line along the LGV Interconnexion Est. It is not replacing RER B to go to the airport, it’s just a private project, trains won’t be accessible to Navigo pass holders so it would be a big scandal if the RER section is removed. I have never heard of such plan anyway. What would make sense is that they remove the direct trains to better serve intermediate stations.
            Yes for M17, linking to Pleyel is probably the main function. However, it would make more sense if it was a M14 extension like suggested by Mikel.

          • michaelrjames

            Yom Sen:

            It is not replacing RER B to go to the airport, it’s just a private project, trains won’t be accessible to Navigo pass holders so it would be a big scandal if the RER section is removed. I have never heard of such plan anyway.

            You’re on the ground (apparently) and seem clued in so it’s almost enough to overcome my (scientific) scepticism. … but the mantra behind the CDG-Express is “to alleviate the saturation of the RER B line”; by this they mean the commuter part (not the airport part which is not saturated) and this won’t happen if B3 keeps running even if the trains are empty, and as I understood it, it’s that the airport RER trains occupy slots (from Nord to Aulnay, IIRC it is one out of 2 or 3 trains) that prevent more suburban commuter trains from the east, and increase crowding of both those B5 trains and the B3 trains. The airport trains don’t carry enough pax (ie. to/from airport) to justify their taking this space but of course the service still needs to be frequent or it’s useless. Sounds like they haven’t really resolved these conflicts.
            Re “just a private project”, well not really, as expected in France:

            It will be built and operated by a 50:50 joint venture between SNCF and Paris Aéroport.[6] Hello Paris, a Keolis and RATP Group joint venture, will operate the line for 15 years from January 2024 with a fleet of Alstom Coradia Liners.[7]

            Last I read, the €1.7bn cost is to be funded by the government–ie. not even a PPP–which explains its continual postponement. So it may be run as a franchise, which is partly to avoid complications re high charges and Navigo etc. The problems are that if B3 is still there, first, who will use the CDG-Express especially at the very high cost; the kind of people for whom that cost doesn’t matter are not the kind to use public transit. Not to mention that it terminates at Gare de l’Est and most use the RER-B for other, more convenient, stops that are closer to their final destination. Secondly, if B3 still runs then what ROW and stations does CDG-Express use, ie. within the airport? Is there space and time for them to share the terminal stations? If you are forced to get some shuttle bus to connect to CDG-Express on the perimeter then it is even less attractive.

            I agree that in some ways extending M14 to span M17’s proposed route, makes some sense. However, metro cars don’t work so well serving airports–eg. no provision for luggage compared to the big trains of RER with their overhead racks and more space–and it seems a long route for most Parisians, and M14 is bound to be very busy. Travellers are prone to inertia, and staying on the one train even if it is a longer ride rather than change to a potentially more direct/faster route at Pleyel. Keeping it separate and terminating at Pleyel forces us to change there, and thus disperse the pax load across more lines (and RER lines more able to handle luggage). However I just found out that M17 proposed route now extends beyond Pleyel to: Les Gresillons, Colombes, La Garenne-Colombes and Nanterre-La-Folie. Is this to satisfy the masters of the universe at La Defense?

            If you are correct, then I can’t see CDG-Express ever getting built because the business case seems flawed. OTOH, to me it still makes some sense that if it does get built, to have M17 replace RER-B3. Alleviates B(5) issues and it still provides a RATP (Navigo) means of getting to the airport even if it means a trek to Pleyel first, or RER-B to Le Bourget-RER with M17 transfer. I also see that in its current route M17 has transfers to 12 Metro lines so could have fairly high utility and a more assured 2-seat ride to most people’s final destination. And now enough air travellers might be tempted by the more direct, faster and reliable Express? It seems seriously excessive to have three rail services to CDG, so forgive me but I return to my scepticism–or perhaps, confusion.

          • Yom Sen

            Well, to be honest I’m not on the ground and most of my knowledge is based on what I read on French forums…
            You’re right for the “private” aspect, I rather meant it was supposed to be managed separately and 100% self funded. From what I understood, the € 1.7B is a kind of advance that will be paid back by a tax on flight tickets, but the ambiguities on this is what makes the project controversial.
            I think the (non-official…) mantra for CDG Express is rather creating a connection to CDG that avoids mixing up with people from La Courneuve or Aulnay-sous-Bois.
            I agree with all your criticism regarding CDG Express. Unfortunately it looks like it is already too well advanced, I think they are already working on Gare de l’Est Porte de la Chapelle connection.
            For RER B CDG branch, I would get your arguments if it was to build a brand new line but not removing an existing line (same than for line L Saint-Lazare – La Défense actually…). M17 plan is not to reuse the RER B infrastructure and at this point M17 project is too well advanced so it’s either cancel or keep the plan as it is. M17 will duplicate access to CDG from Le Bourget and La Plaine, CDG Express will duplicate it for the Gare du Nord area, but RER B will still provide direct access to major francilian hubs like Aulnay sous Bois, Chatelet, Saint-Michel, Denfert, Antony (and Orlyval) and Massy that neither M17 nor CDG Express will.
            Note also that on RER B north the busiest stations are on the common trunk and that Sevran-Beaudettes on B3 branch has more traffic than Sevran-Livry on the B5. Having more capacity for the B5 branch will not be that useful. But you’re right, since we will not need the direct trains Gare du Nord – CDG (only off-peak and only 4 min faster than omnibus trains) we will be able with the same number of trains (12/h) to have a off-peak frequency of 5 min instead of 7.5 min on the common part and 10 min instead of 15 min on each branch.

  7. Herbert

    No word on reusing the mostly still extant old ring railway ROW?

    After all, the Ring (which, if I need to remind readers was NOT fully available again until 2002 due to partition) is the busiest part of the Berlin S-Bahn…

    • RVA_Exile

      See above comments on Grand Paris Express. With the Line 15, Lines 2/6 and Tramway 3a/3b, along with other partial circumferentials like T4, Line 18, etc., that would be overkill, even for a crayon exercise like this.

    • michaelrjames

      Further to my earlier post (somewhere above) about the PC losing its river crossings, the original bridge was demolished in 1962 because it had too low clearance for river traffic and too many pylons in the river. The new bridge, Pont du Garigliano, is a modern steel bridge with a single central pylon in the river and the highest clearance of any Paris bridge, for which reason apparently it is popular for suicide jumpers!
      From 1863 to 1962, the two-level stone viaduc d’Auteuil (or viaduc du Point-du-Jour) ran across the river at this point, carrying Petite Ceinture trains on its upper level and car traffic on its lower level. When car transport increased on the Boulevards des Maréchaux, and since the arches of the existing structure were too low to be passed under by river traffic, it was decided to replace it. Construction began in 1963 under the architect Davy and the engineer Thenault, and the new structure was inaugurated on 1 September 1966. The present structure is named after General Juin’s victory at the Battle of Garigliano in Italy in 1944.

    • Eric2

      The real question is why did they build brand new trams along the length of the circumferential line rather than just rebuilding the line to provide a much faster ride…

    • Alon Levy

      It’s unclear, but this is part of the plan! Paris has built an extremely busy tramway closely parallel to most of the Petite Ceinture – reusing the Petite Ceinture would have been problematic because the radial Métro lines all have stops at the intersections with the Boulevards des Maréchaux, where T3 runs, but generally not at the intersections with the Petite Ceinture, which the Métro never tried to integrate with and which lost frequent service by the 1910s. But there’s a gap in the T3 circle in the west, and there, a takeover of the Petite Ceinture would be great, since the Métro does have stop at the intersection points.

  8. Eric2

    The way you rescue RER C from its indirectness and lack of transfers and access to important destinations is nice.

    Do you really think Paris is going to abandon the giant investment it is making right now in extending the RER E west?

    You don’t think it is worthwhile to have a north-south line through La Defense so all the commuters can avoid the center city, or are you relying on the future Metro for that?

    • michaelrjames

      The 8km tunnel for westward extension of RER-E is reassigned to be part of the new line that has RER-D east of Chatelet. It’s the green hashed line except for the few most eastward kilometers connecting Auber to Chatelet (ie. the new bit proposed by Alon to connect to D instead of E).

    • Alon Levy

      I’m not proposing abandoning the RER E plan, just connecting it to the RER D with a short additional tunnel from Les Halles to Auber.

      I think a north-south line via La Défense is great, which is why I keep saying M15 is a great investment, alongside the M14 extensions and to some extent M16. The stop spacing on M15 is going to be pretty wide, unlike on M1-13, so it should be useful for suburban service.

  9. RERF

    What’s wrong with the ideas floated for an RER F that simply is a new tunnel between Montparnasse and St Lazare? Doesn’t this cover grounds for every commuter rail entry point having a tunnel?

    • Alon Levy

      Montparnasse and Saint-Lazare were owned by the same company, and their Transilien lines both point west. Key suburbs like Versailles and Mantes-la-Jolie are connected to both, so that most reasonable branches of a Montparnasse-Saint-Lazare tunnel would self-intersect in the west.

      • fjod

        Your plans already remedy the worst aspects of this problem though, as RER A gets the western lines out of Saint-Lazare (to Versailles etc) and RER E one of the lines to Mantes (via Poissy).

      • Mikel

        Does the rather awkward shape of the current RER C have a similar origin? Or is it accidental?

        • Alon Levy

          It’s similar, yes. The RER C was done on the cheap – a 1 km connection between Invalides and Gare d’Orsay. These had not originally been run by the same railroad, but the suburban lines both pointed south-ish, except for that awkward connection to the Petite Ceinture that leads to the lowest-ridership intra-mural RER stations.

          • Mikel

            I see… it gets close to the city centre without actually touching it. Along with the missed RER A connection, it feels similar to C-5 in Madrid.

            As an aside, has an infill station at Quai Sant-Bernard ever been seriosly considered? At least on paper it looks like low-hanging fruit: the university campus has some 30,000 students and the tracks already run on an open trench. Fixing the missed connection to M7 would be considerably more expensive, though.

          • Yom Sen

            Paris is a polycentric city. Saint-Michel-Notre-Dame is actually just in front of Notre-Dame in Ile de la Cité which was the historical gaulish town and is traditionally considered as the centre of the city. You could argue as well that Musée d’Orsay and Invalides which are not far from Assemblée nationale, Louvre, Grand-Palais and Champs-Elysées are quite central. But yes, it avoids the financial and main jobs center between Saint-Lazare and Les Halles.
            RER C has already too many stops which makes it slower than other RER lines. On the other hand, it’s missing so many important locations like Jussieu campus or Pont des Arts in front of Louvre. Even Tour Eiffel and Champ de Mars are at more than 500m from the station that is named after them. If I had to draw a crayon map, I would remove Pont de l’Alma and Musée d’Orsay stations from RER C and build a tramway on surface from Champ-de-Mars to Austerlitz that would serve many more locations and would be much easier to use for tourists.

      • RERF

        What if you tried RER F with a tunnel between Montparnasse and Garde du Nord (Montparnasse-Invalides-St. Lazare-Gare du Nord), such that it takes on the branch of the C that you gave to the E. This would remove the need for line-swapping of the B, D, E..

  10. df1982

    I like the plan in principle, especially because it solves a lot of the annoying features of RER C (especially the fact that it avoids interchanging with A). But the big problem with your plan is that it seems, on paper at least, to involve a lot of switching around of lines, and breaking up existing or under construction tracks to then follow new paths. Wouldn’t this be extremely disruptive to existing passengers (and way more so than your plans for NY, for instance)? Are there ways to achieve the same goals without re-drawing the lines as radically as you do?

    Also, if all commuters trains go through the tunnels, what happens to the existing terminals? Are there enough intercity trains to use all six of them? To be honest, I think leaving the outer-suburban commuter trains terminating in Paris is not such a bad thing. Does a commuter travelling 60km+ really need a one-seat ride?

    • Eric2

      I think the switching around is a minimal concern. For every destination that loses a direct ride due to switching, another destination gains a direct ride.

      One could imagine a similar plan without switching around, in which a tunnel connects Saint-Lazare to Montparnasse, and another connects Lyon to Tour Eiffel. But this would have a lot more missed connections between parallel lines. Much worse for travel overall.

      If the existing terminals are too large for their traffic, so what? That does no harm. Maybe traffic will expand eventually. Maybe some of the tracks can be redeveloped as real estate (though you want to be super-cautious here, and anyway the real estate value is limited in Paris where you can’t build skyscrapers).

      • df1982

        By “disruptive” I was referring more to the initial construction, in that it would take crucial segments of the RER out of a commission while tunnels are re-routed and stitched together. Sometimes it is worth the pain to have a better result in the long run, I’m just dubious that that is the case here.

        • Eric2

          OK. You have a point.

          How about RER F being a north->south tunnel from Saint-Lazare->Saint Michel->Montparnasse (to allow for RER B transfer). The RER C north branch can either remain unchanged, or feed into the new tunnel like Alon suggested.

          And four-tracking the RER B/D segment between Nord and CLH, of course.

          This allows for through-running all Transilien lines. There are some missed transfers (D-F, also the existing C-A,D,E missed transfers, though if the north branch of C is given to F then it gains transfers to A and E). But the overall cost should be maybe half of Alon’s plan, with much less rider disruption. Also less political disruption, I can’t imagine the people responsible for RER E (political or technical) being happy about their career accomplishment being cut up and used for “better” lines.

          • df1982

            How about the D continues from Les Halles to St. Lazare and takes over the St. Lazare Transilien lines, while a new underground line is built from Montparnasse to Nord, and then takes over the northern branches of the D?

            It solves the D/B merger, and avoids Alon’s criticism of a St. Lazare-Montparnasse line (that it doesn’t hit the centre of Paris). And I think it would resolve a lot of existing missed connections. Albeit C would still be the RER’s red-headed step child, and you lose the Lyon-Nord connection.

          • Alon Levy

            Saint-Lazare-Montparnasse 100% hits the center of Paris. Saint-Lazare has way more jobs around it than Les Halles. The problem is that a Saint-Lazare-Montparnasse connection results in teardrop-shaped lines, rather than in true cross-city service.

          • df1982

            OK, I’ve always considered Les Halles to be the centre of Paris – it’s indisputably the main transport interchange – and a St. Lazare-Montparnasse line would either have to miss it or make a significant dogleg to connect to it. Other than that I agree teardrop shaped lines are best avoided if possible.

          • michaelrjames

            The business (ie. finance) district is concentrated in the area surrounding the junctions of the 1, 2, 8 & 9 arrondissements. A lot of this lies between Haussmann-St-Lazare (RER-E), Auber (RER-A) and Chatelet-les-Halles (RER-A, B, D). Auber is ≈2.1km by street to Chatelet-les-Halles or 1.7km as the crow flies. In this zone, east or south-east of Auber-Opera are Credit Lyonnais HQ, BNP-Paribas HQ, la Bourse (stock exchange), the Banque de France (central bank) and the old Bourse de Commerce (the round building at the west end of the garden/park above Halles, originally for grain trading, then the Paris Chamber of Commerce and in June this year to become the new home of the Pinault art collection). Over time the business district has moved more west as the Halles – Beaubourg area has become more entertainment-shopping-arts. Of course a lot has move even further west by about 8km to La Defense and hence the quick link by RER-A and soon the wester extension of RER-E.

            In any case, I believe the prime aim of Alon in those mix-and-matches of the different lines is to provide cross-city thru-routes like NW to SW, NE to SW etc. As someone pointed out, the only real loss is the RER-D direct link from Gare du Nord to Gare de Lyon which does seem a pity. But I wonder with all the new and expensive links between tunnels that one couldn’t have one’s cake and eat it. Run some lines as branches–eg. retain D’s current route to Gare de Lyon alternating with its new route (RER-F) to Montparnasse. Likewise the new western tunnel (as E western extension) could swap between Gare de Lyon and Montparnasse and for that matter the old E route of Gare du Nord. I suppose it loses a certain purity and simplicity but would gain a lot for users?

          • df1982

            I agree that the main purpose is to provide cross-city through-routes, which I think is achieved by the Les Halles-St. Lazare and Montparnasse-Nord lines I proposed, with less disruption than Alon’s plan. What you’re suggesting would involve a lot of interlining and reverse-branching, which given the capacity crunch on the RER would probably not be such a great idea. Lyon-Nord travellers would just have to wear the extra interchange, I guess.

          • michaelrjames

            But my point was that the connections for the routes already exist. I’m just saying that if one goes to the expense of making/changing those new connections a la Alon’s scheme, may as well keep the old links functional for this kind of flexibility.
            I suspect the reality is that they won’t sacrifice the Nord to Lyon link.

  11. seb

    Couldn’t one use the already existing connection at Viroflay so to only tunnel the RER F, leave the RER A above ground?

    Also how difficult is it to 4-track the LGV between Chatillion and Massy with some in-fill stations for the RER F to relieve the southern RER B and retire the Robinson branch with its requirement for the train car length?

  12. Yom Sen

    Great plan.
    Main drawback would be lost direct connection between Gare de Lyon and Gare du Nord.
    RER C has a weird shape it’s true, but it is sometimes useful as it works as a metro line within Paris serving neighborhoods and routes ignored by metro. For example, I have often travelled on RER C between Maison de la Radio (10 min walk from the closest station, Ranelagh) and Saint-Michel, in your plan that would involve 2 changes in Champ-de-Mars and Concorde.
    However, it’s more than compensated by new direct connections. That would also provide useful capacity for Gare de Lyon, Gare du Nord and Montparnasse so we could get rid of poorly connected stations of Bercy and Vaugirard and use Gare du Nord for the new CDG-Express.
    Why are you modifying RER A and B as well? You would have already a big bang on RER C, D and E, having some stability on RER A and B would be nice I suppose.
    It would be great as well if the tunnel Montparnasse – Gare du Nord (and possibly Gare de Lyon – Saint-Lazare) could be used for some TGV.

    • Alon Levy

      You’re right about the RER A and B. I made these changes to clean up the map a bit, but it’s 100% feasible to keep the RER B as is (and use the extra capacity to take over Transilien K and add service) and have the RER F instead take over Transilien H and the northern branch of the RER D.

      For the TGV, I could crayon even further and propose to build a Paris Gare Centrale underneath Gare du Nord and Gare de l’Est, with tunneled connections south toward Montparnasse and Gare de Lyon for LGV service. I suspect the Philippe cabinet might even like the idea for the excuse to redevelop the railyards and do urban renewal around Gare du Nord…

      • df1982

        I’ve always thought that building over the trenches to Gare du Nord, Gare de l’Est and St. Lazare would be a good opportunity to provide some parkland in a part of Paris that is desperately in need of it (all the major parks are to the south, except for the Buttes Chaumont). But apartments would probably be more likely.

    • Mike

      Yes I noted the loss of the Gare du Nord and the Gare de Lyon direct connection by RER D. Currently well used as Eurostar arrives at the former and many trains to provincial France, Switzerland and the south coast as well as the sleeper to northern Italy leave from the latter. I’ve personally used this link many times.

      • fjod

        And Thalys, and northbound TGV to Lille and the north, and services to Gare de l’Est. This does feel like the biggest connection you actually want RER to serve directly, especially given the luggage and infrequent travellers. And as you say, it’s very well used.

        • RVA_Exile

          To the tune of “Don’t cry for me Argentina…”
          Don’t cry for the lost connection, the truth is the RER D was never very good in the first place
          The RER D is so infrequent, with such a bizarre schedule and random station dwells or terminating trains at Gare du Nord or Châtelet, that it is usually faster today to take the RER B one station to Châtelet-les-Halles, cross-platform transfer to the RER A, and get to Gare de Lyon that way faster than waiting for an RER D that seems as hypothetical as A line service in New York sometimes.

          • fjod

            I don’t really follow your argument. RER D is bad due to capacity problems in the shared tunnel, but its route is very useful. Alon’s plan solves the capacity problems by constructing a second tunnel, so RER D would no longer be bad.

            Besides, the sorts of passengers who make this journey don’t mind a 5-minute wait for a direct train. They don’t know there’s a cross-platform transfer at Châtelet, and will disproportionately have luggage, children, language problems etc., all of which make them assign a significant penalty to this transfer.

  13. fjod

    I’ve just noticed that the Versailles-Massy-Juvisy branch(es) of RER C have been struck off. What would happen to these under your plan; I assume splitting into their own circumferential line? Though this line is geographically circumferential, I wonder if it really is functionally circumferential – are passengers in fact using it to connect with the nearest radial?

  14. SB

    List of Alon’s Crayons for tracking purposes
    New York:
    Washington DC:

  15. Eric2

    Looking at Alon’s map:
    Why should the D go to La Defense when the A already goes there? Here are two alternative plans that avoid this duplication and also avoid the need to mess with the E:
    1) Let the E continue to La Defense as planned, and have the D take over Saint Lazare, and leave the C unchanged.
    2) let the E continue to La Defense as planned, route the D from Saint Lazare south to Invalides, and take over the C branch as Alon planned for the the E to do.

    • michaelrjames

      Why should the D go to La Defense when the A already goes there?

      The western extension to RER-E (the part that Alon rearranges to join to the SE part of RER-D) is specifically to relieve the stress on RER-A which carries >300m pax pa. The only duplication is within Paris. Once it reaches La Defense RER-E will take over the 55km long A3 branch of RER-A (currently terminates at Cergy-la-Haut, being extended to Mantes-la-Jolie .. halfway to the coast!).
      However, it does duplicate the La Defense-St-Lazare-Chatelet-Forum-GdLyon route so one wonders if swapping with the new Montparnasse RER wouldn’t provide better choices, both for western-E commuters and all who pass thru La Defense. Though remember that the current arrangement gives them a choice of Gare du Nord or Gare de Lyon. The counter-argument is that all of them retain the ability to transfer at Chatelet for any of these (Montparnasse, Lyon, Austerlitz, Nord) however the counter-counter argument is that giving a choice at La Defense will take pressure off the Chatelet-Forum hotspot. That’s why I floated the idea of running trains to both destinations, (ie. reverse branching I suppose?) so that those travellers from way out in western Ile de France can choose a one-seat ride to either Gare de Lyon or Montparnasse (or for that matter, the link that has already been built, the E route to Gare de Nord). I know there is an impact on frequency but this is not so critical because all trains pass thru Chatelet-Forum so travellers have the choice of taking any train and transferring, or waiting for the one-seat-ride train.

      • Yom Sen

        From what I can see on Wikipedia, RER E is not planned to take over any branch from RER A, only the transilien to Mantes-la-Jolie via Poissy.

        A similar version of Alon’s plan but less disruptive and with less tunnels to build could be:

        RER C: Kept as it is except that the North-West branch would go in the 2km tunnel through Saint-Lazare and Concorde instead of Petite Ceinture. The branch to Versailles would still be part of RER C and you could also add another branch from Saint-Lazare (for example trains to Ermont via Argenteuil) but not the whole line J. That would be 22 trains/h at rush hours between Musée d’Orsay and BFM with current schedules (14 to Saint-Lazare, 8 to Versailles)
        RER D: The new line to Montparnasse would become a branch. With current schedules we would have 24 trains/h between Saint-Denis and Les Halles, 12 on the Gare de Lyon branch and 12 on the Montparnasse branch (excluding Paris-Dreux that go too far to be integrated in RER). Part of Transilien from Gare du Nord could be integrated in RER D, I would let the longer branches still terminating at Gare du Nord.
        RER A, B, E: no change

        It would give less capacity improvement, but would still fix the most critical part (between Gare du Nord and Les Halles). It would integrate into RER most of the stations in petite couronne. It would create fast connections between Gare du Nord and Montparnasse and between Austerlitz and Saint-Lazare while keeping the direct connections Gare du Nord – Gare de Lyon and Gare du Nord – La Défense.

        • michaelrjames

          OK, RER-E would only duplicate part of RER-A3 though very approx. it looks like about half the 55km? At any rate it is designed to relieve the stress on RER-A which it will achieve partly by that relief on the A3 branch–though the new stretch to Mantes-la-Jolie will pick up new pax–and giving those coming to La Defense etc from Paris the option of A or E (as well as other existing routes like M1 and future M15, M17).
          “RER D: The new line to Montparnasse would become a branch.”

          Yes, there appears a logical convergence as it was I suggested earlier, that would retain the D link between Nord-Lyon. The more I think about it, the more this new tunnel (Montparnasse to Nord) becomes compelling. In the past I was a bit cool on Alon’s lamentation about the missing tunnel Nord-Chatelet for D, not because it wouldn’t be useful but because of other priorities. However I think it becomes more interesting, more “sellable” as part of this double duty of serving Montparnasse (and as you say, making it feasible for D to takeover some more Transilien lines out of Gare du Nord).

    • Alon Levy

      I grab tiles from Open Street Maps, paste them into one file, and then draw over it in Inkscape. I can share the code I use for it, which should get me banned from ever touching a computer again.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.