The Hamburg-Hanover High-Speed Line
A new high-speed line (NBS) between Hamburg and Hanover has received the approval of the government, and will go up for a Bundestag vote shortly. The line has been proposed and planned in various forms since the 1990s, the older Y-Trasse plan including a branch to Bremen in a Y formation, but the current project omits Bremen. The idea of building this line is good and long overdue, but unfortunately everything about it, including the cost, the desired speed, and the main public concerns, betray incompetence, of the kind that gave up on building any infrastructure and is entirely reactive, much like in the United States.
The route, in some of the flattest land in Germany, is a largely straight new high-speed rail line. Going north from Hanover to Hamburg, it departs somewhat south of Celle, and rejoins the line just outside Hamburg’s city limits in Meckelfeld. The route appears to be 107 km of new mainline route, not including other connections adding a few kilometers, chiefly from Celle to the north. An interactive map can be found here; the map below is static, from Wikipedia, and the selected route is the pink one.
For about 110 km in easy topography, the projected cost is 6.7 billion € per a presentation from two weeks ago, which is about twice as high as the average cost of tunnel-free German NBSes so far. It is nearly as high as the cost of the Stuttgart-Ulm NBS, which is 51% in tunnel.
And despite the very high cost, the standards are rather low. The top speed is intended to be 250 km/h, not 300 km/h. The travel time savings is only 20 minutes: trip times are to be reduced from 79 minutes today to 59 minutes. Using a top speed of 250 km/h, the current capabilities of ICE 3/Velaro trains, and the existing top speeds of the approaches to Hamburg and Hanover, I’ve found that the nonstop trip time should be 46 minutes, which means the planned timetable padding is 28%. Timetable padding in Germany is so extensive that trains today could do Hamburg-Hanover in 63 minutes.
As a result, the project isn’t really sold as a Hamburg-Hanover high-speed line. Instead, the presentation above speaks of great trip time benefits to the intermediate towns with local stops, Soltau (population: 22,000) and Bergen (population: 17,000). More importantly, it talks about capacity, as the Hamburg-Hanover line is one of the busiest in Germany.
As a capacity reliever, a high-speed line is a sound decision, but then why is it scheduled with such lax timetabling? It’s not about fitting into a Takt with hourly trip times, first of all because if the top speed were 300 km/h and the padding were the 7% of Switzerland, the Netherlands, and Sweden then the trip time would be 45 minutes, and second of all because Hamburg is at the extremity of the country and therefore it’s not meaningfully an intercity knot that must be reached on the hour.
Worse, the line is built with the possibility of freight service. Normal service is designed to be passenger-only, but in case of disruptions on the classical line, the line is designed to be freight-ready. This is stupid: it’s much cheaper to invest in reliability than to build a dual-use high-speed passenger and freight line, and the one country in the world with both a solid high-speed rail network and high freight rail usage, China, doesn’t do this. (Italy builds its high-speed lines with freight-friendly standards and has high construction costs, even though its construction costs in general, e.g. for metro lines or electrification, are rather low.)
Pedestrian Observations 
The “benefit” of building the line to freight standards is that in case of a disruption on the classical line, they can cancel/delay some passenger trains on the high speed line so that freight trains can be diverted there. (There aren’t many slots for slow freight trains in between trains running at 250 km/h…)
How does this pass any cost/benefit analysis? Probably they don’t believe the line can be built any cheaper?
Supposedly, adding freight traffic is also a frequently used trick to boost the cost benefit analysis numbers. Wendlingen Ulm apparently claimed there would be 17 daily freight trains. In reality there was a single train in 2.5 years. https://www.bw24.de/baden-wuerttemberg/nur-ein-einziger-gueterzug-hat-in-zweieinhalb-jahren-diese-milliardenschwere-bahnlinie-befahren-93697829.html
The Barcelona – Perpignan line was designed to carry freight as well, which made extra sense because it removed the need to transfer the load from iberian gauge to international gauge trains. The new line was a financial failure and it needed to be rescued by the public goverment, which in order to see some relevant use of the line has put a 90% discount on the fee to use the international tunnel. Also, the new line often required two locomotives to pull because of the steeper slopes. Looks like for freight operators, lowering cost is more important than speed by itself?
My impression is that apart from Germany, rail freight mode share is quite disastrous in the EU, so not too surprising that Perpignan-Barcelona flopped. But it’s not really the freight-compatible NBS that are saving Germany.
The mode share of rain for freight in Spain is pretty disastrous, though in the last few years there’s a lot of investment in the Mediterranean corridor to make it compatible with the international gauge. Sometimes at the cost of the local passenger trains though. We’ll see the final result in some years, I’m seeing some things that don’t make me very optimistic though.
The latest NATO agreements allow spending on transport to count as defense spending (see a complex agreement that I’m not really qualified to read for details). If a war would break out NATO would take over all rail lines to get troops and equipment to the front (wherever that is).
The above is a stretch though, and in normal peace time should not apply at all.
The thing is that construction of this line is so expensive you could actually build two lines, one freight and one passenger for the same price. If Germany is still able to build at those costs…
(But tbc 30M€/km is “normal” nowadays but that does represent considerable cost inflation already)
Hamburg may be at the extremity of the German network, but it’s the hub for traffic to/from Denmark (both mainland and via the Fehmarn belt), no?
The line will pass between Soltau and Munster, population 15 000 and seat of Germany’s largest military base. Soltau’s Bundestag representative is Lars Klingbeil, SPD co-leader since 2021, Federal Finance Minister and Vice-Chancellor since May 2025. He used to oppose this project. What happened?
Is it really so bad to build a line mainly for capacity relief?
Also, given the flat terrain I don’t believe adding freight capacity substantially adds to the cost.
It’s not bad to build a line for capacity relief – the Tokaido Shinkansen and the LGV Sud-Est were both built for capacity relief on the classical lines, too. What’s bad is that the average speed on this new line is 150 km/h, which is less than what the Shinkansen achieved in 1965 (the Hikari averaged 170 km/h), and much less than that exact alignment can achieve today with non-DB levels of timetable padding (46 minutes times 1.07 is 183 km/h, 42 minutes times 1.07 is 199 km/h).
ICE currently link Hanover and Hamburg Hbf in 76′ on the 184 km route via Uelzen. Their average speed of 145 km/h is respectable given the 200 km/h speed limit. If average speed on the more direct NBS is only going to be 150 km/h, one understands why they extol the benefits for the intermediate stops.
The intermediate stops are also the most important part of getting the line built. So talking about them shows the Germans are learning from the mistakes of others (HS2). Good.
HS2 opening has been delayed until 2035. Is it too late to learn from the Germans, HS1 or the LGV Atlantique? There are only 17 000 Vendomois but with riders driving up to 30 km to reach it, Vendome-Villiers TGV has a decent ridership in a 50 000 inhabitants basin. Why can’t they put a station in Brackley, population 16 000 and less than 15 km from Banbury (54 000), Buckingham (16 000) and Silverstone racetrack?
Something being a silly political compromise doesn’t make it ‘the most important part of getting the line built’.
If you don’t do the “silly” political compromises you either have excessive costs or stuff doesn’t happen at all.
The disruption of these projects for local people along the route is huge. They have to meaningfully benefit.
No they are not. Trying to appease everyone is how you kill a scheme – see Northern Powerhouse Rail.
Excessive costs can only be partially attributed to not bowing down to silly demands at best.
@Weifeng, the service that serves the smaller intermediate places could be hourly as per your suggestion below or it could be half hourly serving one of the intermediate stops per half hour, that would be less bad for the overall service but still give them a benefit.
I do think sometimes there is a giant middle ground that the politicians/rail industry still manage to miss.
Something might be an necessary evil, it still isn’t ‘the most important’.
When HS2 Ltd sought advice from SNCF they said avoid building Haute Picardies.
Avoid building Haute-Picardie doesn’t mean avoid building anything.
HS2 must skirt population centers like Wendover (7500 inhabitants), Aylesbury (80 000), Brackley (16 000), Kenilworth (23 000) and Coventry (345 000). This is uniquely challenging if you have nothing to offer to the inconvenienced residents.
French and Spanish high speed lines had the luxury of passing at a comfortable distance from similar towns. The LGVs surrounding Paris cross vast expanses of flat agricultural land with around 50 inhabitants per square km but even that landscape is no guarantee of smooth sailing. The LGV Atlantique faced significant opposition in rural Loir-et-Cher. Pierre Fauchon, one of the local senators, let Paris know that if the line was moved west of Vendome (17 000) instead of closer to Blois (47 000) and a TGV station was promised he might turn the things around. Many saw him as a traitor but Vendome-TGV, in a more desirable location than Haute-Picardie, has found a clientele of long distance commuters. In 2019, a plaque was placed in the station to commemorate the late senator.
And that was bad advice the French gave.
The intermediate stops won’t be served by ICE but rather by regional trains. A similar thing exists on the Nürnberg-Munich line and on the Stuttgart-Ulm line. It was even proposed for the Nürnberg-Erfurt line with a stop “near” Ilmenau but it was not put into practice because the potential stop is in the middle of nowhere…
@Herbert, a fast regional train that goes to the nearest major city is a reasonable compromise for smaller places with less than 100k people.
Those places are still benefiting meaningfully from the project which is the important thing.
@dralaindumas
The Chilterns are not Vendome. A better comparison would be how LGV Sud-Est does NOT stop at or near Montereau or Sens. There’s the perfectly decent Chiltern main line serving the town centres with plenty of scope for capacity improvements (longer trains) if needed; and removing London – Birmingham inter-city from Chiltern’s remit means it can more effectively serve the Chilterns. There is absolutely no reason to divert passengers to an alternative routing that requires car-based rail heading.
LGV Sud-Est has no intermediate station inside of the first bifurcation towards Dijon. If it had it wouldn’t be able to operate a (max) 14tph timetable it does today or the intended 16tph timetable post ETCS. This is what HS2 should be emulating. Maintaining HS2’s ability to eventually run 18tph is of paramount importance. GB needs a Dutch-style timetabling philosophy with absolute zero-tolerance of timetable irregularity. Vendome-Villiers has half a dozen calls per day each direction – such an operating model has no place on a trunk line this close to London. That HS2 should have a Vendome-Villiers can only be suggested by someone who doesn’t know how to run a railway.
HS2 gives Coventry 2tph empty fast trains to London devoid of Birmingham passengers and 4tph empty trains to Birmingham (2 of them from Northampton) devoid of London passengers, and half-hourly direct connectivity to Milton Keynes and Watford Junction which the current WCML timetable structure does not provide. That’s significant capacity and connectivity improvements.
Kenilworth is on the Leamington Spa – Coventry – Nuneaton axis. Fastest Intercity services being migrated to HS2 means the existing railway can have a simple half-hourly rhythm, prioritising connectivity over raw journey time, with places like Kenilworth connecting into Swiss/Dutch style interchange nodes at Leamington Spa, Coventry and Nuneaton.
Yes, Sens would be a better comparison, and the Chiltern lines do a decent job. The rest of your post is mistaken.
The TGV did stop in Sens, joining the LGV towards Marseille in St Florentin. The loss-making service ended in 2011 because the local councils would rather subsidize new TER trains. Scheduling was not a problem since, outside of peak hours, the LGV uses about 50% of its capacity.
Your error is to put the technical aspect above the political one. Shinji Sogo is regarded as the father of the Shinkansen. He had nothing to do with the technical development but Sogo’s “persuasion technology” is in short supply. Germany has the engineering and financial resources but the persuasion shortage is the reason why the Hanover-Hamburg HSR, sorely needed for years, still may not be built.
This is the point the SNCF president made when he visited Vendome for the late senator remembrance. He acknowledged that SNCF did not plan or need the station but that they needed people like Fauchon who could find an acceptable compromise. The compromise found in the UK, cutting the Euston, Manchester and East Midlands segments seems absurd. Your world beating dream is putting the cart before the horse. HS2 would only need 18 trains per hour if high speed tracks reached Leeds, Manchester and beyond but inefficiency, delays, cost overruns and unpopularity have imposed drastic cuts. Most people are being persuaded that HSR is a bad idea and the project only goes on to prevent further international embarrassment.
Though it has bigger fish to fry than Birmingham, the Tokaido Shinkansen serves Atami and Maibara whose population is below 40 000. It only started to run 16 trains an hour in 2020, 56 years after the line was completed, 45 years after the Shinkansen reached Hakata.
Public opinion is fickle. In Villiers, the small village hosting the Vendome TGV station, SNCF had to promise to return to agricultural use the quarry it was creating. At the quarry closure, the locals had changed their mind and the hole in the ground was turned into a 16 hectare water sports facility. Crossrail was for years in the news for its delays and cost overruns but this was quickly forgotten once trains filled up. Britain’s High Speed Rail problem is that the embarrassment will not cease at HS2 opening. The Old Oak Common-Curzon Street service will not be popular.
You can see how busy the trains are on the Chiltern Website (https://timetables.chilternrailways.co.uk/#/timetables/3529/Table%202) – and the only service that is busy all the way from Birmingham to London is the 09:13 service on a Saturday morning. On the return trip a handful of the evening departures from Marylebone to Birmingham are busy the whole day, in part due to a lack of rolling stock meaning that the 16:37 runs with only 4 cars, and in part because there is no evening peak on the Chiltern Mainline.
As these issues are all around cost unless HS2 has lower fares than the classic intercity route today it is unlikely to make much difference for more than a handful of customers.
These benefits are extremely light compared to the benefits of a station or the service improvements that the Seine River towns saw.
So does Erlangen just outside Nuremberg, it still has useful service.
Fine, but then there needs to be meaningful benefits for the Chiltern towns, i.e. full electrification and using the time gains and the speeding up of the local services from Marylebone to High Wycombe so that all of the classic Birmingham trains can call at High Wycombe and Haddenham and Thame Parkway on a half hourly clockface. The passing tracks at Beaconsfield should likely also be re-instated.
There also needs to be a meaningfully faster express service from Aylesbury to central London that runs at least hourly and takes at most 35-40 minutes either via Princes Risborough or via Amersham – this includes the track improvements to deliver such a service.
@dralaindumas
HS2 responded to the political brief of developing a reverse A network from London to Manchester Leeds – this brief came from the Coalition government in the early 2010s after a political review of options which included reverse S, reverse E and reverse A.
During early stages of scheme development the major political constraint is HM Treasury. The overwhelming political driver is to contain project scope. Adding a Chiltern stop would add costs and reduce overall scheme benefits, and the numbers would be clear to see and No. 11 Downing Street would sure have vetoed it. The drip-drip inclusion of more and more mitigation measures causing unit costs to rise was not well understood at the time – No. 11 did not effectively manage their own MPs and their own party’s councillors. The government chose a consenting route that left HS2 Ltd vulnerable and without a single voice from the government.
HS2 Ltd is not like SNCF – a state-owned company with a high degree of operating independence, a type of organisation that could lobby at a disinterested government. HS2 Ltd is but an extension of the civil service and in hindsight a very ‘green’ organisation. It is absolutely not equipped or empowered to second guess or navigate Conservative Party internal divisions. Since 2015 until the last election we had a revolving door of Transport Secretaries, HS2 Ministers, 7 Chancellors and 5 Prime Ministers. There was no consistent political direction on how to manage scope control vs unit cost control. Promotion of HS2 was purely up to ministers and not HS2 Ltd – everyone in the technosphere understood the capacity dimension but ministers insisted on selling the speed angle.
In hindsight, committing to Chiltern electrification and timetable improvements may have softened some of the Chiltern opposition to HS2 but that was completely outside of HS2 Ltd’s very tightly defined remit from ministers, and there was no way Treasury under No. 11 would have endorsed this approach at early stage scheme development.
This all assumes that the desire to control project scope is the main contributor to exploding unit costs – this is far from proven.
Euston will be built at HS2 will carry 10tph. All 10tph will be Paris – Lyon levels of busy. People have forgotten about Edinburgh tram’s saga – 2 years after Euston opening they will forget about HS2’s saga just the same.
@ Weifeng Jiang. Thanks for the explanation. It seems HS2 was a top down enterprise with the UK government sending HS2 Ltd on a difficult mission, convincing the Chiltern that they should accept years of heavy construction and some noise for the privilege of watching trains pass at record speed. SNCF did almost the same thing 40 years earlier but cows are the main train spotters on the LGV Sud-Est. SNCF wrapped up its land acquisition phase in 3 months. After adjustment for inflation, LGV Sud-Est land cost per km were 2.5% of HS2’s.
Another difference is that the TGV was a ground up endeavor with SNCF starting to use the term in 1965, testing a gas turbine prototype, developing the rolling stock with Alsthom, designing the first LGVs, discussing financing with international investors and intimately associated with operations and the setting of track access charges. A large number of actors, from local councils to foreign and private entities have later been involved. This includes the EU, Luxembourg for the LGV Est, Switzerland for Rhin-Rhone. The scope may or may not include renovation of existing stations, creation of new ones, electrification of connecting lines. No element is more important or a panacea but the actors who negotiated their financial contribution keep each other in check. The Brittany LGV calendar was : public consultations in 1994, 1995, 1997 and 2000 leading to the selection of a 1000 m wide area, accurate routing in 2005, new public consultations in 2006, Declaration d’Utilite Publique in 2007, financing protocol in 2009, contract covering construction and 25 years of operation by Eiffage in 2011, works from July 2012 to fall 2016, opening in July 2017 on time. In 2007 Euros, cost ended up 2% below the DUP estimate.
With the smoother extension to Leith, the Edinburgh tram saga has been forgotten by the man in the street. The decision makers may not have a short memory. A long construction phase costing way over 1 billion Pounds and a ridership around 1/3 of predicted one will give second thoughts to anyone planning a tram line in the UK.
Come on man, even if you stopped every single train the station is only going to cost £100m or so at most and the hit is a tiny 5-6 minutes per service. And you don’t need to offer a service level that strong.
Plus in terms of construction cost you can subtract off at least some of the mitigations off the total which adds up to billions if not tens of billions and there is the additional ticket revenue as well that would likely amount to £25-100m/year plus parking fees etc.
The real reason it didn’t happen isn’t Downing Street – although it is possible in the abstract that Osbourne liked the idea of it being fast – it is that the train people don’t like stops at smaller places so it was never seriously proposed or suggested.
And yes it is probably true that due to city placement/branching in Britain that you would need to stop a fair percentage of trains at any given extra stop, most likely at least 1/4 or so in order to provide reasonable access to other destinations – but that doesn’t justify doing absolutely nothing at all – or four tracking north of Amersham and having a change at Birmingham Interchange.
If we are fair in order to electrify the Chiltern lines (and potentially connecting lines such as Didcot-Aynho, Leamington-Coventry, Reading-Basingstoke, Aylesbury-Princes Risborough, Oxford to Milton Keynes and Birmingham Snow Hill to Kidderminster) would have cost serious money in total and would add a lot of scope. At £5m/km in todays money we are looking at – what £2bn – I mean its less than the mitigations, and would add value to the railway overall, but its real cash too. And the upgrades for the line into Aylesbury to speed that up would cost real money on top as well. Certainly £100-200m via Princes Risborough and maybe more via Amersham depending on what you did.
This does assume that people care a lot about saving half an hour off their journey times against the current classic express service that is currently the strongest classic service in the world.
At HS2 service frequencies on the trunk route, to maintain capacity either everything stops or nothing stops. For 18tph all to make a Chiltern Parkway stop means another Old Oak Common sized station. Oh and to add a massive car park because it’s a parkway station in the middle of an AONB would be really environmentally sensitive wouldn’t it!
Again, this all assumes Chiltern opposition is the direct cause of unit cost escalations. You and Alain simply latched onto the one thing you have a superficial understanding of and decided that must be the panacea.
Oh, and your obsession with Belgium throwing away their entire domestic rail operating conventions just so London – Amsterdam passengers can transit through a couple of minutes quicker is really politically astute isn’t it.
With what’s coming out from Radio 4’s Derailed – what’s clear is that HS2 Ltd was too dogmatic in the 360-400km/h design speeds (the business case only ever assumed 330) and the very large turning radii (>8000km) that made the route too restrictive and too difficult to avoid otherwise avoidable obstacles, and because the alignment was under-designed at Hybrid Bill stage this wasn’t apparent until post power detailed design. The Hybrid Bill process was not effective in allowing efficient dealing with challenges and oppositions (and HS2 Ltd actually caved unnecessarily in many cases). Some (very) poor judgments from HS2 Ltd and political leadership led to the cost escalations. The Chiltern dimension is but a contributing factor, not the root cause.
In hindsight, for the double digit £bn HS2 cost overrun, assuming that could effectively have been avoided, then low single digit £bn Chiltern electrification and timetable improvement would just have been a rounding error in the overall programme cost. What the government should have done was to have given HS2 Ltd sufficient protection, but absolutely insisted on scope control and unit cost control, and developed a small number of complimentary classic railway upgrade schemes that would have cost a small fraction of HS2’s cost overrun.
This implies the exact opposite of what you appear to believe. If your pet line is “currently the strongest classic service in the world” (whatever that might mean) then it is also showing strongest evidence in the world that demand exceeds demand, especially given the utterly ruinous UK rail ticket pricing regime whose express design is to suppress “excess” demand.
If you stopped e.g. half the trains in Buckinghamshire and half at Stoke on Trent then the total cost would be approximately one train path per hour. And if the initial service level is really that tight then built a second high speed line on the East which could be done TGV style for TGV costs given Kings Cross has lots of capacity.
If you do the stop north of Aylesbury it isn’t actually in the AONB or anywhere close to it.
You also really don’t need an ‘Old Oak Common’ sized station. Haddenham and Thame Parkway with 4 tracks, twice as long platforms and a second or perhaps third footbridge linked to the platforms with ramps would be fine.
There are multiple factors, one is the excessive design speed and the other is the excessive mitigations due to the lack of stops.
I actually suggested that Belgium buys some EMUs and be a bit more ruthless about stopping times so it can run its domestic services faster – which would also allow the international services to be faster as there would be less contention.
@dralaindumas, the Edinburgh tram managed 12 million passengers in 2024 with 23 stops, per stop that is higher than Croydon Trams 20.0 million a year with 39 stops. And the Newhaven extension only opened in June 2023. So I think the Edinburgh tram is now doing quite well.
I think though we probably all agree really on ridership, once it gets to Euston I am sure HS2 will have respectable ridership, but while it is terminating at Old Oak Common out of the city I don’t think it will be as successful. Certainly before the Euston extension the 7.5 million or so extra passengers you would get from a stop at Stoke and Buckinghamshire and from more services to Birmingham Interchange are likely to be helpful to increase that number, especially as those passengers will be pretty much purely additive as the journey time improvements from those places are much more meaningful.
Even in terms of fast service to Glasgow in order to justify it on a busy line you need more passengers than you can plausibly get out of a London-Preston-Glasgow service. Better to add the additional stops to justify it in the south and the midlands where the population density is higher than north of Crewe where in the catchment area of the stations it is lower.
@dralaindumas
First consultation in 1994 to spade in the ground in 2012 – that’s nearly 20 years of ‘talking’ – that’s about twice as long as HS2’s talking stage (10 years give or take). In hindsight HS2’s talking stage was too short, but HS2’s development was in a rare period where both the main parties were pro infrastructure investment (and genuinely so), and everyone thought that any greater length of talking would risk losing that cross-party support. In fact it was a miracle that the post Brexit Conservative Party happened to be led by someone (for all the wrong reasons) who liked big infrastructure, but all the Brexit faithfuls that came afterwards were libertarian small-state anti-infrastructure fundamentals.
Back to LGV Bretagne – when during the 18-year period did land purchase happen? And any land safeguarding before that? In Britain the received wisdom is that once you’ve announced land safeguarding you don’t want to hang around, because you’d condemn people in planning blight – basically if your house is identified as being potentially eventually compulsory purchased then you’ll struggle to sell you house, so the government should buy your house as soon as practically possible.
Another thing that seemed to have gone wrong was the Hybrid Bill was based on an outline design only. Accurate routing and detailed design were contractors’ responsibility, but the bill gave them very limited flexibility (something like a 200m wide corridor) so if an obstacle came up from surveys the only way to overcome that obstacle was an expensive one. The Hybrid Bill only amounted to an outline planning permission – HS2 Ltd still had to apply for about 1000 individual planning permissions from hostile councils – HS2 Ltd was essentially held over a barrel and had to buy its way out. All of that was failure of the political class.
I don’t think you need to have a 10 year talking period. You just need to actually listen to people on the ground. You will need some sort of public consultation and also to go and visit all the local communities along the route and hear what they say and be prepared to spend real money to keep them happy. Probably with some cheese and wine and canapés for the initial meetings.
As part of the talking phase you will need to make sure each and every one of community affected by the construction benefits in some way, either with a community project of some description that you fund or with reasonable access to a station or with some other explicitly tied meaningful infrastructure improvement.
All in all 2-3 years of talking and genuinely listening is surely sufficient.
It sounds to me like you have identified two more mistakes with the HS2 process that I wasn’t aware of.
@ Weifeng Jiang. The nearly 20 years of “talking” were probably related to the 1992 Bianco decree requiring this dialogue as soon as large projects are considered. The LGV Sud-Est, built in the same animal husbandry environment as the LGV Bretagne, had followed the opposite approach. The March 1974 governmental approval was kept secret and SNCF discreetly surveyed the land before the March 1976 Declaration d’Utilite Publique aka DUP.
In Bretagne, some farmers complained of a 10 year period during which they could not obtain any construction permits. Altogether 23 km2 of land were acquired for the 182 km LGV and 35 km connecting lines. Six km2 returned to agricultural or other uses afterwards. In addition, Eiffage, the builder acquired 9,2 km2 for preservation purposes to compensate for HS2 ecological impact, and temporarily rented land for storage purposes. This is considerably less than the 70 km2 acquired for HS2 phase 1 and 2. It seems 3 to 4% of the affected agricultural exploitations closed for good with their remaining land redistributed to neighbors. The area potentially affected by the redistribution known as remembrement was about 450 km2. This region relatively large properties were often divided in barely continuous pieces and the LGV was the catalyst behind a mostly but not always satisfactory attempt at rationalization. In contrast with HS2, built up areas were barely affected. The closest one is 170 m from the line in Torce, population 1270, where, in the only report I could find, the mayor and council thanked Eiffage for the small wood and pond gifted to the commune.
@dralaindumas, it’s also about the fact that the A41 or A418 which have been closed on occasion between Aylesbury and Bicester/Thame are going past much larger villages and towns than in France so the disruption is much greater and for many more people than when roads serving one village of a thousand people were cut off.
There’s probably as many people living in the villages between Aylesbury and Thame inclusive as one of the Seine river towns. And Bicester alone is 40k like a Seine river town too.
People also commute from Aylesbury to Oxford. There is no equivalent in France near the TGV lines.
@ Matthew Hutton. I agree. The only HS2 equivalent in France would be the LGV Cote d’Azur along the difficult route chosen by the politicians, and it is going nowhere.
The Croydon tram has a declining ridership but, adjusted for inflation, only cost about a third of the shorter Edinburgh tram line. Edinburgh tram was a major investment for the city council (2400 Pounds/capita) and phase 1 construction was more painful for the neighbors than expected. To my surprise, when trying to go from Waverley to the Airport, I was sold a bus ticket. The ride was fine. A quick search showed this was a city-owned bus. This looked to me like a self inflicted wound.
HS2 is another major investment causing years of disruption. The 66.6 billions announced in January 2024 for Old Oak Common to Curzon Street represent about 1168 Pounds per capita in England. Denying the region a couple of stations because, decades from now, the line may get congested makes me wonder whether sado-masochism is becoming mainstream in the UK.
By ‘talking’ I obviously mean the period for modelling, alignment selection, and detail design before you are able to put spades in the ground, not literal talking.
HS2 Ltd could not do any geotechnical survey before Hybrid Bill powers were granted. It might be that there should have been a two-stage process – where the initial powers gave a say 1000m wide corridor where they could do the detailed surveys. That would have provided an opportunity for HS2 Ltd to say ‘the ground conditions are more challenging than expected – minister we can contain project cost and keep construction footprint to a narrower corridor if we don’t sink the line into a cutting for aesthetic reasons and don’t build as long a green tunnel’ or ‘we have discovered these obstacles and we may be able to lower design standards to 320km/h with a tighter turning radius’, then got construction powers on a much narrower corrido with sufficiently detailed design and contractors don’t then have to price in a lot of risk.
Keeping approvals secret would never fly in Britain. You can manage Chiltern opposition with far more above-board means.
A densely populated country requiring 3 tiers of rail services where the top tier necessarily has to pass close to people without stopping is a problem statement France doesn’t encounter, so France does not offer ready solutions the UK can apply. Aylesbury is some 60km from Central London as the crow flies. There is not a SINGLE LGV station 60km from the centre of Paris as a parkway station to bring commuters into Paris. France does not offer a precedent to what you guys are proposing. The Amsterdam – Rotterdam HSL doesn’t stop at Zoetermeer.
Even a 2-platform station with through tracks would have the same footprint as Old Oak Common. In order for deceleration trains not to slow down non-stop trains behind you need deceleration loops about 10km long in order for a stopping train to diverge at 230km/h one planning headway ahead of the next train. The mitigation is worse than the original impact. Creating lots of rail-heading car trips cannot be good for the environment. If you build it outside the AONB in Aylesbury then nobody in the AONB that you are trying to placate will drive backwards to catch the train so you’ve built a complete white elephant.
If you stop half the trains at Chiltern Parkway then every odd train will have to be overtaken by an even train in order take the next odd path – that would require an absolutely on-time operation with no trains running in the wrong order. You have suddenly created something far worse than Welwyn Viaduct. It’s simply unworkable.
France might just get away with the airlines-on-rails model where something like Vendome causes no harm. Britain is ‘clockface metro’ territory where something like Chiltern Parkway would destroy the entire operational principle of HS2. Just like the Swiss sometimes have a blind spot and struggle to understand the Dutch approach, it seems the French have a blind spot and struggle to comprehend that clockface metro land exists outside of their borders.
Weifeng Jiang, you accuse France of having a blind spot in regard to clock face scheduling while demonstrating an almost planet size spot of your own. There is more to this planet than France and The Netherlands. Matthew already mentioned Japanese and Taiwanese HSR stations close to the capital, but let’s not forget Ebbsfleet at 37 km from St Pancras. HS2’s mistake was trying to reinvent the wheel. Do not repeat it. Ten km passing loops are not necessary. Deutschlandtakt, JR Central, JR East (and SNCF) use passing loops in the 800 to 1300 m range.
The high population density found along HS2 meant that many people were going to be inconvenienced. HS2 argued that the line was necessary for the good for the country but the Chiltern people said they would love to help but that they were very worried for the bat colonies. Instead of asking for altruism, HS2 needed to find a larger group of local beneficiaries for example by opening a few strategically placed parkway stations and/or diverting some Chiltern Railways trains on HS2 as done on HS1. Congestion would not be a problem in the near future, and congestion charges would take care of it later on.
@Weifeng, South East has a population density of 492/km2, that is high by international standards, but it isn’t as high as Taiwan (656), the Netherlands (541) or South Korea (530), India which is building HSR now is barely lower at 488. Japan is lower at 340, but I cannot believe it isn’t significantly higher along the Tokaido and Sanyo routes.
If you go for the solution where all trains stop at Birmingham Interchange for Coventry and East Birmingham and half the trains in Buckinghamshire and half in Stoke then broadly you stop all services for half an hour in Buckinghamshire and then have a half hour gap where no trains stop and vice versa. It costs you ~one train path as obviously the first Stoke train has to wait for the Aylesbury train to depart.
You don’t do anything mad like stop every other train at a station unless you simply four track the whole line.
Japan and Taiwan have high speed rail stations less than 60km out of Tokyo/Taipei.
Who cares what the footprint in area is? It’s agricultural land that costs £10k/acre.
In terms of North Buckinghamshire to London traffic I would expect a similar ridership into London as Haddenham and Thame Parkway currently gets, i.e. 800k a year, plus on top of that you would get significant ridership towards the north – likely at least the same again.
Rotterdam Centraal to Hoofddorp is only 48km, and even though typically the high speed services don’t stop at Hoofddorp it is certainly less than 60km from Schiphol Airport to Rotterdam where they do all stop.
And south of Rotterdam at least some of the HSL trains do call at Breda and Noorderkempen on their way to Antwerpen Centraal.
No trains for 30 minutes then a train every 3 minutes for the next 30 minutes? What are you on and where can I get some?
‘Who cares what the footprint in area is?’ You proposed this as a mitigation to HS2’s environmental impact through the Chilterns. Do you not see your own hypocrisy?
@Weifeng Jiang, Haddenham and Thame Parkway only has an hourly fast service off peak and an offset slower train as well. It still gets 800k riders a year.
The whole point of stopping ~30 minutes worth of trains is so that you can get direct service to each northern destination on an hourly basis.
Get back to me when Montereau gets a parkway station with a train every 3 minutes for half an hour then we’ll talk.
@Weifeng Jiang, you could have a parkway and serve 3-4 destinations in the north or whatever and run the other trains fast if you wanted. That is also an option in play.
As it is there are destinations in the north that you have the capacity to serve that are weaker (like perhaps Chester or Blackpool or Blackburn) that you can serve with the first train (or the first timetabled train, the first one can be pick up or drop off only – and also you can in a rural station have the Spanish solution so the doors don’t even open to board when you don’t want to enable that) in the block to Buckinghamshire/Stoke.
Do not forget also that the extremely conservative 1.5 million riders for a Buckinghamshire station is a lot more than the beetroot stations in France or Germany.
This is how a sensible planning process deals with objections.
https://nltimes.nl/2025/07/26/council-state-green-lights-plan-upgrade-amsterdam-alkmaar-railway
From your article:
So local people will benefit from the project.
It’s not like HS2 where between London and Crewe there is an average of 130km between stops.
Planned improvements to the Amsterdam—Alkmaar line are several orders of magnitude removed from HS2.
Yes, clearly selling projects in terms of service upgrades and community benefits is good. Yes, justifying the need for property takings is good. Yes, not being derailed by a small number of negatively effected landowners is good. But come on! This is routine, medium-scale package of upgrades to a routine existing regional rail line.
@dralaindumas
You keep moving the goalpost and changing your argument. Stop it.
Your original thesis was that an SNCF approach would have led to a station in the Chilterns.
LGV Est, Sud Est and Nord all go through areas of similar population density as the Chilterns 60km from Central Paris. None of these places have LGV stations. This fact alone is enough to debunk your claim.
Given the choice between open fields and serving small towns 50 to 100 km from Paris like Sens, Creil, Chateaudun or Beauvais, SNCF chose the open fields for LGV Sud-Est destined to serve distant markets like Switzerland, the Med, Italy and Catalunya, for the LGV Atlantique serving the coast from Brittany to the Basque region, for the LGV Nord where TGVs were expected to run to London, the English midlands, The Netherlands and the lower Rhine region, and for the LGV Est going towards Luxembourg, Southern Germany and Austria. SNCF is not regretting this choice. The Yonne and Aube Departements would like to build a TGV station on the LGV Sud-Est but SNCF is not going to contemplate the idea until a second Paris-Lyon LGV is available.
But note that the open field choice is not available in England. The LGV Sud-Est, Atlantique and Nord go through areas of much lower density than HS2 or HS1. The would be LGV Normandie along the lower Seine basin is another story. If you want to know what SNCF would have done with HS2 look no further than HS1 which was built with French influence and standards. You must be flexible to address local concerns. The Chilterns are the wrong place for the fastest highspeed train on the planet. HSR stations at a short distance from the metropolis are not a heresy. The Shinkansen, Taiwan and Indonesia HSR and Segovia in Spain exemplify the practice and California wants to do it too.
@Weifeng Jiang, which towns in France are at a similar distance from Paris as Aylesbury and have a similarly light set of benefits? Especially towns with a similar hinterland that has places the size of Thame, Bicester and Oxford cut off from the town the size of Aylesbury during construction.
And further from London, what about Coventry or Stoke? Both of which in France would get TGV service in both directions to their central station – as likely would Leamington Spa at least.
@dralaindumas
You are quite wrong. LGVs Sud Est, Est and Nord go through almost the exact same population density as the Chilterns at the same radius from Central Paris / Central London.
https://www.tomforth.co.uk/circlepopulations/
You can do the work yourself. I’m not going to debate facts with you.
HS1 serves a completely different corridor characteristics and is not comparable. HS1’s long-distance market potential is 4tph vs HS2’s 18tph. Gravesend was unusual in that it was at the end of a very long and slow railway that could only operate stopping services, whereas the Chiltern has a good 160km/h main line with limited stop services – the appropriate type of railway for 60km from London. Chiltern has a good main line (yes it could do with electrification and timetabling improvement but that’s it), whereas Gravesend was in desperate need for a regional main line and HS1 happened to have the spare capacity to acts as Gravesend’s posh main line.
You entire premise that Chilternites are being unusually hard done by is false. They are getting no more unfair treatment than the good people of Melun, Meaux, Chateaux Thierry or Compiegne.
In fact LGV Sud Est users a former railway alignment through Villecresne, on an urban corridor inaccessible by rail (Valenton – Brie-Comte-Robert). HS2 Ltd could only dream of the robustness the French planning system could employ.
The elevation of Chiltenites is plain weird. It’s a small number of enormously privileged people who are prepared to take the whole country to ransom. They used their power and influence and spread misinformation in the national media. They are a bunch of hypocrites who have no problems with driving their 4X4s on the M40 and are just about to open the South East Aylesbury Link Road that has just as wide a construction footprint as HS2.
@Weifeng Jiang. I spent time on the TGVs and on Google-earth and could not to find a 2000 inhabitants French village exposed to the TGV noise. Meanwhile HS2 path is visibly in direct contact with densely built neighborhoods. My premise is not that the Chilternites are unusually hard done. Knowing the land and its language much better than I do, they can speak for themselves and do not need my advocacy. I think they are unlikely to be cowed by your “bunch of privileged hypocrites” label. I am just trying to point out the fallacy of HS2 Ltd approach.
The Interstate program could cut through poor and densely populated neighborhoods all over US towns because it was supported by a large constituency. With exits every few miles, US highways benefit a lot of people. What HS2 is trying to do is akin to building a 100 mile stretch of highway devoid of exits through a rich region.
We agree on other points. SNCF benefiting from disused corridors preserved for decades (the path between Paris and Massy-Palaiseau being the best example) contrasts with the UK’s habit of building on abandoned railbed. Good services by Chiltern Railways do not help HS2’s cause. Segovia was unlikely to protest HSR passing at 2 km from the town when the new AVE station, at 68 km from Chamartín, saves 75 minutes over conventional rail.
@Weifeng Jiang, this isn’t true.
In contrast in France:
The French figures are significantly lower.
@Matthew Hutton
You are better than being this brazenly disingenuous. HSR construction has an impact radius of 3km at best.
The road closures on the A418, A41 etc have a much larger impact area than that. Certainly it’s difficult to argue that e.g. Thame or Bicester aren’t affected. And they are a lot more than 3km away from the line.
And what does the 3km radius show?
The populations affected by HS2 peak around 38740 in West Ruislip, 19027 in Cosne Valley western slopes, 10563 in Amersham, 28028 in Mandeville, 53818 in Aylesbury, 11609 in Brackley, 12665 in Leamington, 32943 in Kenilworth and 20535 in Litchfield.
The populations affected by LGV Sud-Est peak around 35014 in Savigny, 14644 in Montereau, 13932 in Sens. Not only HS2 numbers are significantly higher but they are an addition with HS2 bringing WCML noise to the areas. LGV Sud-Est was a subtraction, taking traffic out of the PLM line passing through these towns and affecting a larger number of inhabitants. The true peaks were in Le Creusot (3873), Montchanin (4514), Cluny (4586), Loche (7429), Montanay 8067, Cailloux 10237. Le Creusot/Montchanin and Loche (Macon) got a TGV station, Montanay and Cailloux at 17 and 13 km from Lyon Part-Dieu can use the TGV, Aylesbury and Kenilworth got the middle finger.
And people who live in Aylesbury and work in Oxford is 100% a thing.
How is any of this relevant to Hamburg – Hanover?
@Weifeng, the general global practice is to build intermediate stops for all but the smallest settlements as there is a wide swathe of disruption caused by the construction.
That the Germans are doing so as part of this project is good. It is also something the Americans on e.g. the North East Corridor and any other countries attempting to build high speed rail should take into account.
That doesn’t count because I’m sure Real Englishmen(tm) drive everywhere. Even the Unreal ones. Or is it UnEnglish. Everyyyyyyyyyyybody uses roads. That people on a train are NOT using the roads doesn’t count.
Only in the imagination of BANANAs and NIMBYs. I’m sure there were screeches about the chickens going dry and the cows stopping laying. And the soy milk at Costa curdling.
I lived blocks and blocks away from the construction of Interstate 78. It was annoying going through or the detours around the actual construction site but we were blissfully unaware of it blocks and blocks away.
So you were affected by the construction.
Of course when the road was finished you also got to use it to speed up your journeys so you benefited from the project.
It’s pity your mommy didn’t teach you that occasionally there may be minor inconveniences.
This is splitting hairs.
Coventry’s prize is a 125mph service with substantially more capacity for itself in a clockface timetable and vastly improved regional connectivity. This is a far superior railway product to that seen at Le Creusot and Macon-Loche. Macon Ville – Lyon Part Dieu service level is shite; Coventry – Birmingham is not.
https://www.economist.com/britain/2025/07/23/the-peril-of-trying-to-please-people
@Weifeng, France’s 10th biggest city, Lille, gets TGV service in multiple directions. Saitama, Japan’s 10th largest city has all Shinkansen’s from Tokyo to the north stopping at one of its stations Ōmiya. Germany’s 10th largest city, Essen, has around 10 ICE/Eurostar services an hour stopping in its main station going in all sorts of directions.
Coventry currently has fairly bad service especially going north – it should get better treatment.
I do actually agree that the policy is reasonable – though some compromise might have been wise – and certainly Reeves should have been prepared to talk to the farmers. However actually the farmers are pretty fucked off about it and there are quite a lot of signs complaining about it on farms along the motorways I have been on recently – and there have been numerous other protests by farmers.
There’s no signs of this with HS2 in my experience.
This was bad compromise, not that compromise wasn’t necessary.
I don’t think I disagree actually. You’re correct in that building it for 300 km/h and the resulting quicker trips to Hamburg would be useful. Most passengers will actually want to go or come from there.
It’s just that the current situation when it comes to on-time performance and reliability is so incredibly unsatisfactory I suppose I view solving that as a huge win on its own. Add to that the opportunity for better headways on the regional trains serving Lüneburg and the wasted potential for more speed just doesn’t seem so pressing to me, at least not right now…
Oh and FWIW I think they are planning on allowing at least some freight trains on the line during the day; and probably large amounts during the night time window of 22 pm to 6 am. The 10 planned sidings for overtakings and notably the grade-separated connections to the Maschen marshaling yard and at Celle don’t make sense otherwise. The line to Maschen is exclusively used for freight and the Celle link provides access to the marshaling yard in Lehrte east of Hanover and also towards Czechia.
Another topic: Have you noticed they’re planning on building the line really close to the former (un)loading facility of the Bergen-Belsen concentration camp? In case that doesn’t fly they propose a tunneled alternative in that area. Would love your thoughts on that issue.
Hamburg isn’t Tokyo and Hanover isn’t Yokohama. Hamburg isn’t Paris either. Just because it’s busy, doesn’t mean it has capacity problems.
Click through to the presentation, p. 3: they’re projecting 396 trains per day in 2030. Hanover is a small city producing small people, but that line is used by trains connecting Hamburg with the entirety of the former West Germany. (Similarly, the busiest German intercity line, Frankfurt-Mannheim, has almost the traffic density of the LGV Sud-Est, because it connects Frankfurt and many cities to its north with much of Southern Germany.)
They could predict a bazillion umpteen hours a day. I got as far as confirming, with the Wikipedia article on metropolitan agglomerations in Germany, that indeed, Hamburg is not Tokyo and Hanover is not Yokohama. And that neither of them are Paris. I don’t do this for a living and 396 is overly optimistic. Perhaps someone who is attempting to do this for a living could, I dunno, assume Germans get Japanese levels of urges to take high speed rail trips and at a round trip per year figure out how many 1,000 passenger trains per hour that is.
Alternately, in an apparently bravura performance, the Japanese can get more than 12 trains an hour on their high speed lines. 16 trains an hour for 16 hour service day is 256. Someone is being very optimistic.
Keep in mind that the commuter train from the suburb 10, 15 kilometers out doesn’t need tracks halfway between the two. Someone’s numbers are seriously off.
16 trains an hour per direction for 16 hours a day would be 512 trains a day.
As it stands Great Western runs 385 passenger trains a day into and out of London Paddington for example, 444 including empty services to the depot.
And there are no massively huge cities in the Great Western network.
again, Hamburg isn’t Tokyo. neither is Berlin or Paris. There’s London in the Great Western Network. Hamburg isn’t London either.
Hamburg isn’t the analog of London but of Bristol. The analog of London is the combination of Hanover, Frankfurt, the entire Rhine-Ruhr, Mannheim, Stuttgart, Nuremberg, and Munich.
It took 15 seconds to find
https://en.wikipedia.org/wiki/Metropolitan_regions_in_Germany
Hamburg is one of the busiest ports in the world. The Maschen marshalling yard is the second biggest in the world. Hamburg Port handles almost double the amount of containers Tokyo Port does https://en.m.wikipedia.org/wiki/List_of_busiest_container_ports
Alon, to be fair including local service there are 500 trains a day through Bristol temple meads, plus the extras through Bristol Parkway.
And Hamburg HBF is Europe’s second busiest station so I am not sure a comparison with a major London or major Paris terminal (albeit not them all) is illegitimate.
re. Hamburg vs Tokyo. Gotta be careful with incomplete statistics. When talking about Tokyo you have to include the totals of the Keihin region, which includes Yokohama- then the numbers are at least as much as Hamburg or a bit more.
Japan isn’t Germany. Germany runs around 60 000 trains a day and Japan run 26 000.
It’s quite important for regional, intercity and freight rail. And no freight can’t just run at night if all slots are full. You are also forgetting the freight from the baltic ports and from Denmark/Sweden.
Also there are already 320 train a day currently between Hamburg and Hannover.
Frankfurt HBF sees more trains than Canada in a day. Frankfurt is smaller than the major Canadian cities. Germany has vastly different System than any other major country. It’s more comparable with Switzerland, Austria, Czechia.
In addition is Hamburg the connection between Scandinavia and most of the rest of Europe. Also some 3 millions Germans living north of it.
Once again, be careful with source data- the Japan number is only for JR- you have to include the 16 major private railways (大手民鉄) that operate in the metro areas (some are intercity too, i.e. Kintetsu) to get a better idea of total nationwide operational density. Foreign observers tend to ignore the private railways (or just don’t know about them) and have the preconception that railways in Japan= JR only.
Hamburg is the third largest port in Europe and the legacy line is used by freight trains. There is a capacity issue because passengers cannot ride on the freight trains, and slots taken by passenger trains are slots that freight trains cannot use.
Freight isn’t sentient and doesn’t care if it travels in the dead of night either. If it’s arrived in the port of Hamburg on a slow boat from China it doesn’t matter if it the train leaves after midnight. It was a slow boat from China because the stuff that has to be there now now NOW NOW NOW arrived on an airplane.
A few moments of Wikipedia and YouTube…. It’s as many TEUs as the Port of New York and New Jersey. I leave it up to you to wander railfan videos of the freight trains passing the platform at Roselle Park New Jersey. There is one railfan who annotates how many cars and how long the train is, within the video. Many many more than the typical European freight train. Which might help explain 400 a day. Makes me wonder how Rotterdam and Antwerp cope.
You would think someone who wants to make a living doing this would explain those things more thoroughly. There might be the problem that many passenger train railfans apparently think stuff arrives on the store shelves by butterfly and the garbage departs the same way.
On it’s own, that can actually be a red flag for high speed running, as it can indicate weak soil. This isn’t the case, as far as I know, for the coarse glacial sediments in the Lüneburger Heide, but it is a problem for Ottawa-Quebec high-speed rail. I’ve been compiling a set of posts on the problem (click my user link). You can get trains on flat land where a critical speed for sound generation is below 300 km/h and sometimes as low at 200 km/h.
Germany has apparently set quite conservative limits to ensure the trains do not exceed 50% of the critical speed of the soil. Again, this shouldn’t be an issue over coarse glacial till, but it certainly is a reason for some significant subsoil remediation or viaduct construction on HS2 or some Chinese HSR lines.
Please come to NC & VA and convince them that SEHSR from Richmond to Raleigh does not need to be designed to freight-friendly standards either (nor does HSR need to stop in La Crosse, VA, 2020 population 614).
Never mind this, I’m trying to explain the same thing to Connecticut and Amtrak re the bridge replacements.
If the bridge gets stuck … in the position to allow for river traffic.. that not only stops service to small former fishing villages but between New York and Boston. Someday, far in the future, high speed trains will be ruining the bucolic charms of the Connecticut Turnpike, There will still be trains toddling between the former fishing villages. Where they can send the trains of garbage, sludge or recycleables out. Using the same locomotives that brought stuff in. …Beer from the InBev plant in Newark New Jersey comes to mind. I’m imagining a distributor in the general vicinity of New London, close to the consumption in the casinos.
You can replace the bridge with one with a 3.5% ruling grade, not a 1% one. (The freight trains, besides being few in number, are long, which means the effective grade is less than this since it’s averaged over the train’s length and the bridges are much shorter than the freight trains.)
Fight for the bridges and not the stations I think 😜.
There’s also quite a lot of precedent for high grades.
Zero freight, and off-topic except for Alon (sorry!), but FYI Switzerland is building a new direct line (replacing a fun circuitous historic route with a switchback direction reversal at Chambrelien) between Neuchâtel and La Chaux-de-Fonds. The new line is nearly all in tunnel, nearly all single track with one intermediate crossing-point surface station between the tunnels, will support 15 minute directional headways, and — check this out! — features sustained 5% grade in tunnel with 130kmh operation. Brake test! Two 5.7km tunnels, just 1.9km between them.
Search terms are “ligne directe” “Neuchâtel” “Chaux-de-Fonds” — there’s a pile of information out there.
Nice presentation from alternative announcement: https://www.ne.ch/medias/Documents/23/20230907_Pr%C3%A9sentation_LigneDirecte.pdf. (Also see route in regional infrastructure context)
That said, last month the project schedule has been slipped two years out to 2041 due to the decision to reduce construction impacts in the mid-point agricultural valley and construct the two tunnels sequentially, evacuating all spoils via rail and barge through the tunnels to a single attack point at the base.
Only the Swiss can build what is essentially a subway line between two towns of 45 and 37 thousand inhabitants.
@Alain, https://www.networkrail.co.uk/running-the-railway/our-routes/western/mid-cornwall-metro/ is fairly similar.
Thanks Matthew. The Cornwall project is a worthwhile but more modest endeavor : A platform is restored, signaling will be improved and a 400 m passing loop created, allowing hourly service.
I had other comparisons in mind. Two stations in Swiss cities totaling less than 90 000 inhabitants will be linked by two 5,7 km tunnels while the direct connections between Boston, Manhattan, Manchester and Glasgow main stations located 1.7, 1.5, 1.5 and 0.5 km apart, or the 1 km HS2-HS1 link, will still be missing.
They will let people who don’t live in La Crosse use the station. It’s halfway-ish between Richmond and Raleigh. They will let those people go other places too, like Washington D.C. and Charlotte. Or Atlanta. Or New York.
You probably don’t need a stop between Richmond and Raleigh as the places are pretty small – and they are small enough that the French probably wouldn’t. However it is pretty normal to have one.
Virginia isn’t France.
That a new high-speed line is being approved in Germany is itself a miracle.
Germany has its constitutional make up and a localised planning and legal framework. That does HSLs having stations that ought not be there and going around environmental or political obstacles making the alignment slower. If built the line will still represent a considerable time saving over the existing line and have capacity for 10+ tph non-stop high-speed trains. That will have to be good enough. Technocracy and politics will have to meet somewhere, and that somewhere is different in Germany and in China.
In a takt world – one corridor saving 15 minutes isn’t enough to move an entire node. Hamburg may have to stay as a 00/30 node in order for Hamburg – Berlin and Hamburg – Rhine-Ruhr still to work. In any case, I don’t think Germany should bother with nodal concentration. A future ICE network should adopt the Dutch model of piling on frequency. Even in the evenings the Dutch prioritise giving Utrecht – Gouda a 15-minute frequency over ensuring both sets of trains leave as close to 00/30 as possible. The ICE network has too many trains to make the Swiss-style takt work. The Dutch approach is still kind of a takt in that it’s still highly structured with good symmetry. In a German context, specifying journey times to meet nodal times is irrelevant (or should be irrelevant).
I’m not worried about current journey time estimations or whatever padding that may be in the timings. The important bit is that the line is built. Once Germany has delivered the necessary capacity and brought infrastructure back to acceptable standards, then timetabling approach and journey time calculation conventions can change.
Just to illustrate why I think ’00/30′ nodes are irrelevant in (most of) Germany. In a world where all planned HSLs are built (including Hannover – Bielefeld), you’d be looking at this sort of service level from Hamburg towards Hannover
This is far too many services to not be spread evenly around the clockface. It feels like a lot of trains but only in line with the current number of trains Rome sends through and past Florence.
I wonder whether you add in a 1p2h service from Copenhagen to Berlin and perhaps from Copenhagen to Amsterdam and/or Brussels.
Copenhagen to Berlin may well be more than 1tp2h, but that wouldn’t be anywhere near Hamburg – Hanover. Such a service should use the Bad Kleinen curve.
Could Amsterdam – Bad Bentheim be an international path for more than 1tp2h (the current Amsterdam – Berlin service)? Maybe.
Unless the “trains” are a single car or perhaps two there aren’t enough people in Greater Hamburg.
https://en.wikipedia.org/wiki/Hamburg_Metropolitan_Region
I wasn’t aware Rome was sending out single cars.
Hamburg isn’t Rome. It’s not Paris and it’s definitely not Tokyo. It’s not New York or New Haven either. And definitely not the Center of the Universe, Stamford. In very very round numbers the French or the Japanese take one – singular – round trip on high speed rail every year. Across broad networks not some obscure stub end. I leave it up to you to do the arithmetic. it’s not 16 trains an hour.
Stop feeding the troll. Simply ignore. Don’t reply.
@Alain
Only the Swiss can build what is essentially a subway line
Not exactly. The line is single track with a passing station in the middle. That means the 15 min headway mentioned is probably close to its max capacity, since a train in one direction must arrive at the end of the line before another in the other direction departs. Every subway I am aware of is double track and can sustain 24+tph per direction, not 4tph. Still a very impressive project for Switzerland, or anywhere.
Not exactly, but not necessarily inferior. Subways are generally double track and good for 24 tph but in 2024, not far from Neuchâtel, on a 5.9 km metro hampered by a single track segment, the small (30.7 x 2.4 m) Lausanne M2 trains carried 36 million, much more than Los Angeles heavy rail (22.5 m over B + D lines totaling 27 km and using 137 x 3.05 m trains). 2025 ridership could reach 40 million. M1, a 7.8 km light rail line, carried 15 million, more than most US light rail lines and several full size US metro lines, on a single track with headways down to 5′ at peak hours.
The Swiss have shown that one track that you can afford is better than 2 that you can’t. I may add that the gigantic but largely empty US subway stations are not attractive.
The 2. Something km section of Nürnberg subway between Ziegelstein and the Airport is single track and is used in practice for 10 minute headways (infrastructure might support 6 minute headways but VAG wants a “legible” timetable and wiggle room in case of delays)
This isn’t particularly related to this post, but have you read “Order without Design” by Alain Bertraud and/or “Planet of Cities” by Shlomo Angel? I’m right-leaning, but strongly support public transport and better urbanism because it’s simply the only economically viable approach for long-term success. You’re one of the only online urbanists I’ve seen address the economic side of urbanism (not just transit costs, but production versus consumption and the like), so I’d be interested to hear if you’ve read either of those books (which, together, I would describe as the Bible of economic urbanism), and if so, what you thought of them. I’m particularly interested in your thoughts on Bertraud’s approach to the question of centralization versus localization and top-down versus bottom-up in urban planning.
I haven’t read the books but I’m familiar with both (they’re Marron colleagues) through their articles and other projects, like the Atlas of Urban Expansion. I think of the Atlas as less a bible of economic urbanism and more a good descriptive dataset useful for developing countries.
I recall writing an analysis of Bertaud’s “ideas” in comments on Alon’s blog, but of course can’t find it now.
I have read his book and found it not very helpful, indeed somewhat confused and misleading to naive readers. I agree with Nolan Gray’s remark: Alain Bertaud has gained a reputation as a “real-life Indiana Jones of urban planning.” But there’s also a Forrest Gump-like quality to Bertaud’s lengthy resume.
The book is an attempt to claim, or even prove, that markets, particularly jobs markets, create cities of the best fit for function. Not planning. Even though in almost every example he gives he describes a core of careful urban planning! In his chapter on Paris (and Barcelona), the former Parisian puts himself through hoops until he grudgingly concedes that planning of the extreme form of Haussmann, has worked! Kinda undermining the rest of his book’s contentions. I’d also point to beyond Haussmann, for example in the Metro and RER (and now tramways). Paris’s was planned as a system and implemented by the state, and it shows when comparing to London and NYC where a whole bunch of lines were independently developed by private interests (“market-driven”) and only patched together as a working network much later by the state. On housing you can read Yonah Freemark on how socialist Paris has done brownfield redevelopment for mixed residential so much better than the Anglosphere (eg. https://projections.pubpub.org/pub/3kq6u3x4/release/1).
I tracked down the long discussion on some of these issues, and copy some bits relevant to Bertaud below. I found his arguments about street areas etc totally weird and counter logical esp. relating to Paris v suburbs.
The point was that La Defense was carefully and totally planned with its podium above the freeways and streets, and Metro & RER lines making very fast connections to the old CBD in intramuros Paris. In fact one could compare it directly to London’s Canary Wharf which was a terrific idea to create a secondary CBD but which was initially left to private property interests and then needed rescuing. Because it was done in the Thatcher era it was done largely according to Bertaud’s precept of “leave it to the market”. But the transit was a mess with the toy railway, Dockland Lightrail, progressively and expensively upgraded over the years but still inadequate until finally a fortune was spent on proper Metro (Jubilee line extension). Even that was a bit weird because it took a very long scenic route (IIRC it crosses the Thames 3 times!), and really only in the last few years (30+ years later) does the Elizabeth line/Crossrail provide rapid access between the new and old CBDs. No accident that La Defense is Europe’s largest financial district today.
and:
As its title “Streets as public spaces and drivers of urban prosperity” implies, the UN report is not really using street space as an ersatz for density. It is rather trying to make the case that street space and connectivity cause prosperity as defined by its City Prosperity Index.
However economists tell us that productivity is the ultimate driver of prosperity. Showing a good correlation between its Productivity Index and other components (Infrastructure Development, Environment Sustainability, Quality of Life, Social Inclusion) of its City Prosperity Index, the UN graphs confirm standard economic theory as opposed to its own theory. In Cities with Prosperity Index between 0.7 and 0.799, Auckland and Moscow have the lowest Street and Connectivity Index and the highest Productivity. In the 0.6 to 0.699 CPI chart, Bangkok has the lowest Street and Connectivity and the highest Productivity. Same thing with Yerevan in the 0.5 to 0.599 chart. Ignoring its charts, the UN report states that “Cities that have a high productivity index are also cities that have reduced traffic congestion and improved walkability through better street connectivity.” Its first chapters correctly argue that the quality of the street design is important but the report simply overlooks the fact that prosperous urban metropoles have other options than the streets when it comes to connectivity.
I don’t think I argued that of the 9th but of the urban-renewed parts of the 1st (Les Halles and what is now Mandela Park). I’m fine with keeping the Conciergerie and other historic sites, but if it’s 1970s urban renewal, it’s fine to replace it with something taller. Same thing with Jussieu, which somehow has historic protections despite being widely loathed by urbanists and architects.
The 70s when the ugliest style buildings were built all over the world. The only compensation is that they were so unattractive and also poorly finished that many got a deserved premature termination.
The Jussieu campus is a “fine” example and should have been demolished especially the hideous central high-rise which was closed for years as they removed asbestos from it. But it is a huge site and I suppose the logistics were/are terrible. It is heritage listed because it is one of the biggest single brutalist complexes in the world (is Barbican bigger?).However there are plenty of other brutalist examples in Paris, some that are more acceptable (eg. Communist Party HQ by Oscar Niemeyer at Pl Colonel Fabien).
London actually has better interchanges between its lines than a lot of newer state built as a single system networks (Shanghai and Singapore come to mind) – and there are only a handful of true missed connections.
The most obvious is the lack of connection between the Piccadilly Line and the Central line – and while they cross 3 times it is difficult to argue you should add three extra stops – one perhaps at Park Royal is surely sufficient.
The Bakerloo at Kenton/Northwick Park makes sense now to allow more NW London-NW London journeys, but it isn’t particularly useful for accessing the city centre and wouldn’t have made sense at construction time.
Then the other one is Marylebone and the Circle Line, but is Baker Street really further than the RER platforms at Gare du Nord?
There’s a good point. Some of Paris’s interchanges are so long they are more hostile environments to navigate in than London’s Out of Station Interchanges. Give me Euston – Euston Square any day over the never ending passageways of Montparnasse-Bienvenue.
That is obviously because Gare Montparnasse was relocated about 400m southward in the 1960s. Previously I’m not sure there was any correspondance between the M4 at the northern end (the original station) and the other Metro lines further south. You’re referring to the very long moving footpath/travelator connecting the two. This is where one of the few examples of a high-speed travellator was installed. It had segments, a bit like a regular escalator, that slid over each other at the ends so the speed at the ends was reduced. Standard speed is 3km/h and they initially ran at 12km/h but had to reduce to 9km/h; however enough people kept falling over at the ends that it was discontinued. I’m hoping they will install a maglev version.
Anyway, with Paris’s long correspondances you have to ask if you’d rather be up at street level in the rain or snow etc. and having to do some street navigation rather than a tunnel that takes you precisely where you need/want to go. If you really don’t like them you can always devise a route that avoids them.
Also I am sure the number of correspondances on the Paris Metro, either absolute or per km, is much greater than for London. It actually makes the system extremely utile. The long ones provide connections that London simply doesn’t.
@Michael, if you want to complain about the connections between e.g the north London line and the radials you are on much stronger footing. There Paris is a lot better.
Duh. Making money is why people are in cities. If aren’t any jobs there aren’t going to be many people. Duh.
He doesn’t argue that “markets create cities of the best fit for function, not planning”, he argues that planning that fails to take markets into account are setting themselves up for failure. From the very beginning of the first chapter:
He’s not saying we need to abolish planning and replace it; au contraire, he’s saying that planning needs to acknowledge and consider economic and market issues. I don’t necessarily agree with some of his conclusions in the final chapter on how city planning should work, but leaving aside that chapter, the book makes many points in relation to issues frequently overlooked by urban planners which are of great importance.
About Paris:
There’s a famous quote from economist Thomas Sowell: “There are no solutions. There are only trade-offs.” Bertaud’s point here is that the land planning policy that Paris has decided on is one that prioritizes historical preservation over affordability. He’s not saying that decision is wrong, he’s clarifying the implicit tradeoff. He’s also very clear about the need for centrally-planned urban transportation networks, so I’m unsure why you’re acting as if he’s anti-transit.
On Paris and Barcelona:
He specifically contrasts Haussmann with the type of urban planning he criticizes:
Where did I say that Bertaud was “anti-transit”? I can’t even remember what he wrote about that.
The issue of building heights is not “maintaining Paris city center as it looked at the time of the Impressionists” though it is one of the outcomes. It is about maintaining the original rationale about those building heights which in fact were established in the 17th century. And human needs or preferences for air and light haven’t changed. In fact the regulations did change after the Impressionists, during the Belle Epoque, allowing higher building but with steeper setbacks to still meet those requirements. The fact is that the Anglosphere, particularly London and NYC have established related regulations but then trash them in the name of various justifications but in reality purely for the developers. You know, selling air rights etc. And yet they still don’t achieve higher densities than Paris! And they certainly don’t achieve anything approaching the aesthetics of Paris. Nor the good urbanism at street level; NYC recent development is slowly destroying what they had (read Michael Sorkin). Lose, lose.
All the rest is just blather. One could say that true market forces in fact have determined Paris format much more than London or NYC. It is Parisians that put a stop to Pompidou’s attempts to emulate American development modes (combined with his early death) whereas in the Anglosphere it is property developers and their proxy politicians who determine it, on basis of pseudo market forces which in fact they force on their cities. The madness of unfettered “market” forces is catering to rich foreigners rather than their own working native populations, the very people who make the city function. This is not happening in Paris.
Contrary to that opinion the trend over the past decades has been the reverse. Today Paris has a higher share of affordable housing and social mix than before, obviously due to socialist government since 2001, and probably irreversible. Here’s Yonah Freemark in 2019:
And Fergus O’Sullivan (CityLab 2022):
I know that most of the Anglosphere media would have one believe otherwise but it isn’t true. Paris is not becoming a gentrified haven of BoBos (any more than it ever was and most of that happened under Haussmann). Further, this concerns only intramuros Paris. The fact is if you want to prioritise personal space then your choice is clear: just a few short kilometres away it is available (in the Petite Coronne) and with excellent transit back to your intramural job (and soon elsewhere in the PC too, ie. M15 & GPX). The fact is that Paris provides for many more people (≈2.2m) to reside in its highly regulated space than Manhattan (≈1.6m) and at least 4x more than in the equivalent London core. Further, in the latter cases it increasingly is just the very privileged who can afford it (“gentrified” is not adequate, they are becoming the preserve of the very wealthy).
It isn’t about historical preservation. That’s just Bertaud’s prejudice. As I have repeatedly stated on this blog, I would apply the same rules to any modern city development anywhere in the world. As would some other planners like Michael Sorkin. The objective is to create amenable living conditions and wonderful urbanity. It does this while providing high residential densities not surpassed anywhere else (or rarely, for such a large zone as Paris).
Today, Paris proper is steadily shedding working-class residents, because the large majority of housing that isn’t social is only affordable to the middle class, and the share of housing that is social is capped by the social budget and landlords’ ability to take losses. This is constantly pointed out in French media; UK/US media reporting on non-English-speaking countries borders on fake news, at every level (we’re talking Financial Times graphs that basically lie about how people in France and Germany vote in order to tell a story), and you should not rely on it.
….Because like every place else in the first world, it’s been shedding working class jobs?
And I though being working class was a state of mind not income dependent.
https://pedestrianobservations.com/2025/01/25/quick-note-rural-drivers-arent-being-oppressed/#comment-164063
This is clearly what Michael wants. Preservation of Paris he went to as a young man without minorities, tourists and other undesirables, effortlessly superior to the hated country country he was raised to respect from Australia.
That Paris’s nimbyism is have really damaging effects on the 5th Republic’s political stability as the white ghettos of Picardy and Pays-de-Calais put FN in reach of power. Ile de France is pre-tax as dominant as London but has a smaller share of the French population.
And don’t get me started on how the French state has fucked up Lyon and Marseilles. UK at least has the excuse that no policy could make geographically isolated Steampunk cities work. Lyon and Marseilles have better logistics access than Paris, yet manage to be way poorer because the French state fucked up transit, development etc.
And no the Toulouse growth doesn’t compensate for it “better than Bristol” isn’t that much of achievement for a relatively small part of the state with good background geography (logistics access plus nice weather).
Borners, the France you are talking about doesn’t exist. How could Paris fucked up transit in Lyon when its advice was ignored ? RATP’s engineering branch suggested a deep small bore metro and Lyon went for a superficial full size one (cars are 2.89 m wide). Lyon did its own thing, with automation based on VAL software but without platform screen doors. Its own engineering department founded in 1968 quickly diversified into modern tramways. It was involved in the Nantes and Grenoble projects at the forefront of French tram revival. It is now a large international consultancy known as Egis Rail. Annually, Lyon’s metro carries about 220 million when the more conventional Marseille one stagnates at 71 million. Marseille metro, whose trains are closely based on Paris MP73, suffers from low frequency and short opening hours because drivers have a cost. It is now quickly moving towards driverless operation.
The big question in Lyon is the expansion towards the wealthy 5th arrondissement. No-one is looking at Paris for an answer. The locals would prefer an extension of the metro whose tires are well adapted to the hills. The Greens in power at town hall are tramway fans. They are rapidly expanding Lyon’s network in the working class eastern suburbs. They point out that the tram has been popular (annual ridership of 112 million is almost 3 times as much as Manchester’s larger network) and that the expected daily ridership of around 55000 is in the tramway range. They think trams can be adapted to a 8% slope in a 1500 m long tunnel between a bridge on the Rhone and the first station in the hills.
I lived for years in Marseille and often visits. Marseille has its problems, unrelated to Paris, and good sides that can’t be found in Paris. As for white ghettos, I don’t think they exist anymore in France. Immigration pressure is visible in Picardie, Pas-de-Calais and elsewhere, in Paris as well as the countryside. France is near a Greek-style budgetary crisis and, yes, the RN is close to power but I don’t think it is a question of Paris nimbyism or Paris vs the rest. Everyone is on the same boat.
Analyses by Centre for Cities and the Resolution Foundation suggest France is less spatially unequal than the UK.
https://www.centreforcities.org/wp-content/uploads/2016/09/16-09-21-Competing-with-the-continent.pdf
https://economy2030.resolutionfoundation.org/wp-content/uploads/2023/09/A-tale-of-two-cities-p2-Manchester.pdf
Greater Manchester is 35% less productive than Greater London whereas Lyon is only 20% less productive than Paris.
Objectively, Lyon, Bordeaux and Toulouse have far better urban transit than comparable British cities. That French regions are rapidly implementing higher frequency clock-face regional timetables suggests French regions have some sort of financial muscles that British regions almost certainly wouldn’t.
@ dralaindumas @ Matthew Hutton
The 4th level ping pong was getting silly…
Back onto Hamburg – Hanover. Hamburg is at the end of a large country where there’s always the next aviation market up for grabs.
There’s a very big range of what constitutes ‘global norm’ and a Hamburg – Hanover distance with no intermediate stops would still be comfortably within global norms. Soltau and Bergen *are* the smallest of places.
Here’s the thing, places like Soltau and Bergen will always have conventional railway alternatives, whereas once you’ve built a compromised high-speed line you’ll never build an incrementally better alternative. Once you’ve settled for a slower alignment and reduced line capacity, you risk locking in Hamburg – Frankfurt as permanently on the wrong side of 3 hours, and Hamburg – Stuttgart permanently on the wrong side of 4 hours. Those are the journey time ranges where a small journey time improvement drives a lot of mode shift from air.
Yes you can use legislative means to shift the rail/air mode shift curve. Spoiling the small number of people for whom there are more suitable alternatives while using the strong arm of the law on those who won’t have alternatives is politically inherently unfair.
I don’t know why railfans think all the trains have to pass between all the platforms. The train that stops in Wideplacealongtheoldtracks can be a diversion off the very high speed line. Think of it as passing siding that isn’t tightly placed next to the main tracks. The local train to EastWideplace, Wideplacedorf and Utopia can arrange themselves around it’s once or twice an hour arrival and departure. Or they can get Wideplace Parkway and the bus can be timed to meet the trains.
What you are proposing sits comfortably between global norms but not within German NBS ones. Are Germans are ready for it ? A 300 km/h 4-track line was one of the options under consideration but it wasn’t chosen. The most popular Alpha-E alternative, a third track along the existing line, was the worse in many respects. The route that will be presented to the Bundestag is a compromise. Serving Bergen and Soltau is an important component. Unlike the Chiltern towns, these places are poorly connected to their metropolis. Bergen still has the tracks but they are only used for historical and freight trains. From Soltau, it takes 2 hours and a connection, when Hanover is only 80 km away on the highway.
I would leave the fairness argument aside because the concept can be twisted in so many ways. Mein Kampf, which led to the Bergen-Belsen concentration camp, is a several hundred pages long rant about how unfairly Germans have been treated by the Jews, their Western neighbors, their own government, and the Slavs in the Austro-Hungarian empire. I would rather concentrate on what works or doesn’t work in terms of transportation and urbanism.
You accuse me of not going beyond Europe and now you refuse to go beyond Germany. Even within Germany, Erfurt – Bamberg is not far from Hamburg – Hanover distance, so even domestically there are precedents.
You are the one who invoked the concept of fairness with discussing Chiltenites vs the rest of the country. It’s pretty squalid to invoke the Godwin law.
I don’t refuse to go beyond Germany. I am already there. I just wondered whether the Germans were willing to go beyond Germany. Germany, Belgium and the Netherlands intersperse high speed and conventional rail, resulting in modest average speed and modest HSR ridership. The Hanover-Hamburg project approved by the government belongs to this philosophy: 250 km/h is the bare minimum in high speed rail and compatibility with freight is in the German tradition.
In the rest of the world, namely Japan, South Korea, China, Taiwan, France, Spain and Italy, high speed rail offers high average speeds. I think Belgian and Dutch HSR routes are most reasonable given the size of the countries and their population density. In larger Germany the question is whether the government and DB are doing this semi-high-speed because they believe it is the best for Germany or because they think they will never get the Germans to agree with the true high speed of the type you suggest. I don’t have the answer to that question. Maybe Alon who lives in Germany has an idea.
I did not invoke the concept of fairness when discussing HS2. I said that HS2 Ltd’s argument that the Chilternites had to accept the line for the good of the country was bound to fail. I said repeatedly that a HSR project needs to promise tangible benefits to a large enough segment of population in the regions it traverses. I am not accusing you of any nazi sympathy. I said that everybody, including the nazis claim to act in a spirit of fairness. The US is currently going through this in the debate about DEI. Is promoting some groups on the basis of fairness or reparation an act of fairness or the opposite?
Invoking fairness during high speed rail planning is equally risky. CAHSR could have built an efficient alternative to flying between LA and SF. Given the distance that meant Tejon Pass, I-5 corridor. Instead we got Palmdale and the Central Valley towns on the basis that these were disadvantaged communities who needed a boost. As you know, if HSR can link two metropoles in 3h30 or less, it will win the battle against aviation. CA HSR gave up on this sure win to pursue the losing strategy of selling short distance HSR tickets when a car is generally required before and after the train ride.
There isn’t any debate. It’s old rich straight white guys wanting for things to go back when everybody else let them get away with being assholes. It isn’t going back. That’s too too bad.
I think the obsession with saving a handful of minutes to get journey times under 3 hours or under 4 hours is misguided. Yes sure faster is better and sure ‘under 3 hours’ makes a better advertisement.
But ultimately people are prepared to go by train when it is reasonable and there is a fair amount of evidence to back this up. Brightline is getting very respectable numbers between Miami and Orlando with a 3.5 hour service, and London Edinburgh is now looking pretty decent with a ~4h15 service.
Do not forget that as well as flying being more tempting for longer trips they also pretty much always have bad timetables and/or ridiculously high fares.
It’s not an obsession, it’s well researched and documented that at the 3-hour marked rail demand share is highly elastic with respect to journey time.
https://hotrails.net/2024/04/rails-market-share-against-air-a-global-review/
The focus ought to be – ‘all else being equal, how can rail be as good a product as it pragmatically can be’, not ‘let’s hope other things work in rail’s favour so we can afford to be a bit mediocre’.
@Weifeng, that link is interesting, but as well as getting faster the city pairs under discussion also saw stronger service.
Where you have a city pair such as London-Edinburgh with strong service it attracts above average ridership – especially as the London-Edinburgh flights are also getting people from the Thames Valley on them where the train time is more like 6 hours not 4 and involves a change via London.
And Hakata-Tokyo where the service is decent enough also has extremely high fares – $320 return – vs 1/3 to 1/2 of that on a flight.
You don’t quite get the ‘all else being equal’ do you?
Yeah but Weifeng, all else never is equal.
There aren’t many trips over 4 hours with reasonable service frequency, speed and fares and where they do have those they appear to have decent ridership.
And the fact that there isn’t a fast service on the Sanyo Shinkansen that at least runs non-stop from Kokura to Shin Kobe is pretty bad. You would save real time with that.
Also an earlier departure than 6am/later arrival than midnight at Hakata would help too – as would through service from Tokyo to Kagoshima.
Just under X hours is often important. Because odds are good much of the local transport options are on a clock face schedule and so if you can ensure the long distance train is at the station in time to catch a local bus/tram when you get there you can get where you want to be. It is almost certain that someone on a long distance train doesn’t want to go to the train station, and odds are high they don’t want to be some someplace within walking distance of that station. They want to be someplace within the city.
It isn’t always possible to pull this off (cities are not regularly spaced), but it should be a consideration anytime you can. A small improvement in some area can allow this and in turn make the system more usable. And HSR needs to be about getting humans around not trains.
Odds are good that you have no clue how cities work.
I would never think to check the schedule for Philadelphia’s Market-Frankford line for a connection. Because it will be there “soon” however you want to define “soon”, most of the day. I might want to check if SEPTA still cross honors Amtrak tickets for travel to Suburban because I could get to Suburban on a …. suburban …. train. Or the trolleys. Or the buses. Or since it’s only 15 blocks to Suburban and today’s weather is very good, walk.
Or Boston’s Red Line at South Station or the Orange Line at Back Bay. … Ya made me look. If I wanted to take Amtrak from New Haven Connecticut to Portland Maine I could get off the Northeast Corridor train train in Back Bay and take a one seat ride on the Orange Line to North Station to catch a Downeaster. The Orange line is running every seven minutes this time of a weekday afternoon. Does Amtrak attempt to meet the Orange line or the Red line?
Baltimore sucks, I’d take a cab. How often is the Red line running in Washington D.C.? Does the schedule for the whole Northeast Corridor get fucked so the long distance trains meet NJTransit’s River line in Trenton or do adults look at these things called schedules or timetables and see that the River line runs every half hour and tolerate whatever the wait turns out to be? And which one meets the River Line. Acela or the Regional? Northbound or southbound.
Does Amtrak schedule itself to meet the “City Subway” at Penn Station Newark? Or one of the many many many bus lines. Perhaps PATH. How often is PATH running this time of day?
Or perhaps the obscure little village’s Toonervillle trolley, like the Toonerville trolley itself, arranges the schedule to meet every train.
Not all the world is the US where costs are so high we can only afford to run service in cities so dense that they can run all day show up and go frequency transit. Most of the world runs their transit to less dense areas where they have to use clock face scheduling to afford to run service and make it reasonable to use. Switzerland is famous for this. In fact every US city I know of (though I’ll admit to not having done a full study) has the majority of its transit not show up an go frequency, and for as much as we (rightly) talk about how bad transit is, most seem to at least get clock face schedules and try to do that.
If you need to get a cab when you get to the destination that is a sign you should skip the train and drive your own car. It is probably cheaper overall, not much slower (door to door), and you have your own car when you get there.
I encourage and advocate for show up and go frequency wherever possible. However the reality is large numbers of people live in those less dense areas that cannot support such transit at a reasonable price. Thus long distance travel should attempt to arrive in sync with the local transit options at the station. (Amtrak’s inability to keep a timetable is a problem they need to fix)
The idea of running trains at the same time each hour is good, i.e. so if there is an hourly service the train to London leaves at :14 each hour.
If you have a single service and a branch line then sure a takt makes sense. But often it is probably better just to run a stronger branch line service really.
At Oxenholme for example you have trains to and from London, Birmingham, Manchester, Liverpool, Glasgow and Edinburgh arriving and departing between :10 and :40. When should the service to Windermere leave and when should it depart?
Really you need to run it every 30 minutes or more often and then it doesn’t matter as much. It kinda goes back to a point I think you have made before, either run a frequent service or close the line.
And yes the bus should be timed to meet the train at Windermere, but LOL it doesn’t – in part because the train tries to meet the departing services in the morning and the arriving services in the afternoon so it isn’t consistent.
You don’t know how it works in less dense areas either. They aren’t stupid. They will schedule the bus to meet the train. With the few people on it who can’t drive for some reason. Because it’s a less dense area where there is plenty of free parking and there isn’t any congestion.
I know how it works in Switzerland where they have sped up trains as needed so the long distance train is in the city when it needs to be to meet up with local transit.
If you are only talking about the US where costs are so high that we can’t afford to build HSR from DC to Boston (via NYC) despite very high population then we can ignore this. (not to mention Amtrak’s famous inability to keep a schedule anyway) However in places where construction costs are reasonable you can afford to make things more complex and quickly it becomes impossible to schedule local transit to meet the trains.
Trains being key – The trains should be running both ways, not just into the city in the morning and out in the afternoon. Since city stations are not spaced according to ideal geometry that means one train will have to travel farther than the other, so if you want to meet both trains one of the two needs to run faster to arrive on time. If you are building in the real world you care about 2d factors, which means you have both north-south trains and east-west trains – and often more directions than that because cities are not in ideal geometric locations, and with reasonable construction costs you will build lines from the big cities to surrounding smaller cities (which might or might not continue on to other big cities).
Sure there are a lot of people in the core of major cities who have frequent local transit. If you only care about serving those people well you can ignore the issues that the many many people who live in less dense areas have and just run whatever service. The lower your costs are (both construction and operating) the more such people you can serve so I don’t want to discourage this too much, but you should really acknowledge those people who live in less dense areas.
Note that the subject here is Germany. People there do not live at Asia density so it becomes more important to serve people who just don’t live dense enough to support show up and go frequency.
Yet you keep changing what you want to imagine.
Contort the trains schedules so the trains going in either direction are at Fumbuck Central at the same time what happens 5 minutes down the line. Or 17 minutes down the line. Which is 11 minutes up the line and 18 at the second station. …
The trip from beyond “meh, four hours is good enough” to just past it, is then five and half hour instead of four and half and they fly. In this particular case Copenhagen-Frankfurt or Hamburg-Zurich. Looking at things from Times Square, Alon thinks four hours between Washington and Boston is good enough. It isn’t because there are places other than Washington and Boston.