Stretching the Line: Why We Do (and Don’t) Extend Tube Lines
With Crossrail under construction and extensions to both the Bakerloo and Northern Line being mooted, it seems a pertinent time to provide a brief reminder into some of the whys and wherefores for those less familiar with subsurface engineering and London’s underground history.
One lesson quickly learnt by the early entrepreneurs who built early tube lines (and by this, as for the duration of the article, we mean the deep level lines) was that the longer the line and the bigger the network, the more profitable it was. To some extent this may seem obvious – a tube line between only two stations is of limited use (although exceptionally the Waterloo and City line manages to perform this role).
As usage tends to tail off at the extremities, it made sense to have the ends only being a small portion of the line. It also made sense to maximise use of resources. Trains sitting in terminal platforms were not in revenue earning service and a lot of the infrastructure – such as power supply – had large initial costs but the add-on cost for these items when extending the line was not that great.
Extending a tube line had a potential second benefit. If it could surface where there was an existing branch line that served the suburbs, then it could take over that branch. The users would have a fast frequent, electric service to the centre of town as an alternative to the slow, dirty and probably infrequent steam train that took them only as far as the mainline terminus. Furthermore, instead of the cold air and the London smog they would be warm in the bowels of the earth.
The Northern line to High Barnet and the Central line up to Epping and Ongar were extended in this way. Others were extended on the surface but took over the overground section of existing sub-surface lines (e.g. Piccadilly line to Hounslow), shared an already suburban electrified line (Bakerloo to Watford Junction) or built tracks parallel to an existing railway (Central line to West Ruislip and even the Jubilee line extension to Stratford) or simply built their own line mainly on the surface with minimal diameter tunnels as necessary (Northern line to Edgware, Piccadilly line to Cockfosters).
So if extending a tube line is such a good idea why is it not done for all lines?
Ultimately extending a tube line is very expensive. Modern day underground stations are even more expensive and tube track in bored tunnels is incredibly expensive to maintain. Nowadays this becomes even more difficult to justify as current rules mean an extension to an existing line must be built with an emergency walkway – adding to the cost of tunnelling but providing no offsetting revenue benefit.
As a result, there has to be somewhere worthwhile to extend it to that has sufficient traffic demand and sometimes the problem is simply that all the capacity has already been used up. This is why it is unlikely that the Victoria line will ever be extended.
Given the efficiency of modern tunnel boring machines and the fact that the size (bore) of the tunnel is no longer a critical fact in the cost, the pendulum is swinging in favour of building totally new lines capable of taking full size trains rather than tube extensions. And once you have started boring, you might as well keep going so megaprojects such as Crossrail are now seen as the cost-effective way to spend money.
So what about taking over existing lines – assuming there are any suitable ones still around? Well for a start a lot of the advantages have gone. We don’t have steam trains so the image of the sliding-door electric underground train doesn’t offer any appeal over mainline suburban stock. The smog is gone and the cheering thought of a warm underground has been replaced by the dread of being trapped in a sweltering Underground train that isn’t going anywhere. Indeed it might be a good idea to look back at the some schemes that were implemented and see if they really were fit for the 21st century.
Some above-ground extensions really do not seem to be really ideal for the modern world:
- Northern line to High Barnet. Dreadfully overcrowded and served by six-car Northern line trains. A tragic waste of capacity on a line built to mainline gauge size with platforms that could be extended, but this would be pointless as long as tube trains run on it.
- Central line to Epping. Demand at Epping, which serves a large catchment area, means that these trains get quickly packed. A tube service means there is no easy opportunity to miss out calling at some stations so journeys are unnecessarily long. Again the line was built to mainline gauge but is now constrained by tube line dimensions.
- Piccadilly line to Hounslow. This subsequently got extended to Heathrow using small bore tunnels. In hindsight using small bore tunnels seems to have been a really short-sighted decision. With a bit of switching of branches between the District and Piccadilly we could have had high-capacity S7 stock running to Heathrow.
- Queens Park to Watford Junction. Watford Junction in retrospect was really too far out for a tube line and the Bakerloo line nowadays does not go beyond Harrow and Wealdstone. Now that TfL have control of the surface line from Euston to Watford Junction their enthusiasm for extending the Bakerloo line to Watford Junction seems to have vanished.
One must not be over-harsh in criticism. Most of these extensions were probably the right decision at the time and even viewing these 20th century extensions with 21st century glasses one can see that they could be regarded as a good, cost-effective solution. It is only now with 21st century traffic levels and expectations that we are really starting to see them as slightly inadequate.
One learns from what happened in the past. One also learns from what didn’t happen in the past. So it pays to look at a couple of schemes that didn’t make it.
It the 1960’s there were parliamentary powers obtained to extend from Aldwych to Waterloo. At first sight this would appear to be quite a neat little proposal. It would be short extension, built under the Thames (which despite all the bridges is still a hindrance to free movement) connecting to a major mainline rail terminal. The problem, apart from lack of money at the time, was that the resulting branch line was just too short to capture much traffic. In any case the bus over this short distance is still a viable alternative. If such a proposal couldn’t produce a financial case then, it is unlikely to produce a financial case now. Not only would there be the cost of the tunnel with walkway, the entire cost of reinstating Holborn-Aldwych would have to be factored in.
The most famous of all is the 1950’s extension of the Bakerloo line to Camberwell that just didn’t happen. This in itself was a revamp of an earlier scheme that didn’t happen. The scheme is well documented in the book “The Bakerloo Line” by Mike Horne. Mike is an ex- LU manager who researches his books and articles to a standard that we can only dream of at LR. He generally goes back to original sources that are either publicly available or in his own vast private collection.
The critical point that Mike makes in relation to the proposed extension to Camberwell was that its primary purpose was not to provide a better service to the inhabitants of Camberwell – strange as that may seem. The problem with the Bakerloo at the time was that it had two branches north of Baker Street which led to an extremely overcrowded central section. The limit to capacity was the ability to reverse trains at Elephant and Castle and by building a three platform terminus further down the line one could increase capacity on the whole line.
Camberwell has always been recognised as somewhere not really adequately served by public transport. Since 1916, when Camberwell Station closed, it has not had any rail service. It is probably one of those places in which one could justify an underground station, if one happened to be building a line through it (something Bermondsey benefited from on the Jubilee line) but it would be very difficult to make a case of building a line specifically to serve it.
Currently there are very few realistic options to extend existing Underground lines. With the proposed separation of the two main Northern line routes through central London there has been an opportunity to consider an extension south of Kennington from what is generally known as the Charing Cross branch. There is a developer-backed proposal to extend to Battersea which we have covered before (and no doubt will again).
The only other realistic possibility is the “big one” – the southern extension into southeast London of the Bakerloo line. It has been felt for many years, perhaps even a century, that the Bakerloo line really terminates too close to central London. It is also true that there is a serious disparity of northbound and southbound flow on this line with southbound trains in the morning being packed out but northbound trains being relatively quiet. So the capacity is already there. You just need something to tap into it and take advantage of it.
It seems an obvious line to extend into an area not served by the tube and with modern tunnelling techniques the harsher geology south of the river is less of an issue. Camberwell and possibly Denmark Hill are the most obvious destinations but they would have to be justified on its current merits and not a nostalgic desire to see a 1950’s scheme finally come to fruition. Peckham would be another alternative. The relative locations of Peckham and Camberwell mean that all serious proposals have included, at most, only one of these locations as otherwise choosing a suitable line of route would be challenging. Better to serve one of them properly rather than both of them in a half-baked manner.
The alternative is to be much bolder and take over a branch line out to the suburbs. None are conveniently located but the Hayes branch stands out as the most realistic candidate. Bromley North has also been suggested but this would almost certainly involve yet more tunnelling. Again, we have looked at suggested Bakerloo Extensions before (here also), and as recent activity surrounding Hayes amplifies, we no doubt will do again…