Monday, 6 October 2008


York Road is a disused station on London Underground's Piccadilly Line. It opened in 1906 but never saw a high volume of traffic. Eventually Sunday services were withdrawn and it closed completely in 1932. The station catchment suffered due to being on the edge of a large area of goods depots and marshalling yards behind King's Cross station with no habitation and little commuter traffic. This despite being about 1km from the adjacent stations on the line (King's Cross and Caledonian Road) and on the edge of central London.

To the east of the station is the run-down residential neighbourhood of West Barnsbury, in the Borough of Islington. This neighbourhood is an excellent example of an area that fails to thrive precisely because of poor transport links. It would benefit greatly in prestige and development by having a local tube station.

York Road station on the Piccadilly Line. Closed since 1932, current evidence suggests it would now be very useful.

The case for reopening the station has not previously been strong. That is until the King's Cross Railway Lands Development scheme was announced (now called King's Cross Central). This scheme involves the mixed-use development of 65 acres of land to the north of King's Cross station - precisely the area of depots that have been the vacuum in York Road's catchment until now. This should bring thousands of new jobs and residents to the area. Preliminary work has begun.

A 1930s map of the area, showing the location of Maiden Lane station (1) and York Road (2)

The reopening of York Road and also the construction of a new York Way station on the North London Line (London Oveground) close to the site of the former Maiden Lane station a little way to the north, have already been suggested by concerned parties (see map) though both Islington and Camden councils would seem to be in favour of the latter option.

As is clear from the map, the York Road station is ideally placed for the northernmost part of the King's Cross central development (bound by Wharf Road). Pedestrian distances from the nodal centre of the site are far shorter than walking to King's Cross (250m as opposed to 808m). A viability study performed by London Underground found the station would attract about 3.2m passenger journeys/year.

However on a cost:benefit analysis the idea was rejected for two reasons:

1. Inconvenience of longer journey times for the 38.4m through journeys/year

2. Station refurbishment would be too expensive

Let us investigate the evidence for this:

1. The extra station would add around 30 seconds to through journeys in standing time and deceleration/acceleration. Multiplied by 38.4m this is a total time loss of 320,000 hours/year.

Now, analysing the mean walking distances of passengers on the northern sector of Kings Cross central we see that the absence of York Road means average additional walking distances of 558m to reach the next nearest station (King's Cross St. Pancras) instead (808m-250m). At a walking speed of 4.5 km/h this is a 7.4 minute walk. Let us take this as the average additional distance for all passengers on our notional York Road station. Extrapolated over 3.2m passengers this represents a loss of 395,000 hours/year - 75,000 hours more than the delay to through passengers.

Furthermore if we compare the station usage of adjacent stations:

Kings X/St Pancras 66.36m/year on 6 lines
York Road (projected) 3.2m/year on 1 line
Caledonian Road (actual) 5.33m/year on 1 line
Holloway Road (actual) 7.49m/year on 1 line
Arsenal 2.73m/year on 1 line
Finsbury Park 24.8m/year on 2 lines

* Tube entry and exit statistics 2007

what is clear is that York Road, although not on the scale of King's Cross or Finsbury Park would have a passenger traffic comparable to the next three stations on the line, and indeed greater than Arsenal station. In fact it would have passenger figures higher than 17 other stations that uniquely serve the Piccadilly Line. By London Underground's logic shouldn't they be all closed down?

2. York Road is a station already constructed. Lift shafts, ticket office etc. exist already. The station would need major refurbishment and new platforms, but minor structural works. London Underground has the money to rebuild Shepherd's Bush (£65m) and Camden Town (£135m), compared to which the refit of York Road would be pocket change (LUL estimates for reopening South Kentish Town are £12m). King's Cross Central is a multi-billion pound development and the developers ought to be able to contribute. Reopening York Road would be in their interests. Long-term, running costs of the station would be no more than any other small Piccadilly line station though York Road would generate more revenue than many.

The reasoning against reopening the station appears to be very weak.

The advantages would be:
  • Convenience to King's Cross Central development
  • Relief of congestion at King's Cross
  • Revitalisation of a forgotten corner of Islington, very close to central London.
The alternative

The alternative proposal - to construct a York Way North London Line Overground Station near the site of the former station called Maiden Lane, has the backing of Islington and Camden councils. Being around 350m metres apart there would be no practical interchange component if the two stations were open. Whilst on its own Maiden Lane would undoubtedly be advantageous, it suffers compared to York Road on six counts:

1. It is further from the King's Cross Central development
2. It is further from the unserved catchment of West Barnsbury
3. It is located in an industrial area with low passenger demand
4. Services on the North London Line are oriented East-West and are not convenient for services to or via central London
5. Lower train frequencies

The other alternative

There is a further, more costly option. That is to leave York Road abandoned, and York Way/Maiden Lane forgotten, and instead create a new interchange station between the NLL and Piccadilly Lines. This would be somewhere towards the back of Bunning Way, alongside the Eurostar Tracks out of St Pancras. As it currently stands station access to this site would be poor and would require major works. Further, the station would impact the viability of Caledonian Road & Barnsbury Station, 400m along the line to the east. The interchange component of such a station would be extremely useful however.

Friday, 3 October 2008

TRANSPORT SOLUTIONS FOR CAMDEN TOWN - The good, the bad and the costly

Camden's market area in north London is immensely popular. At weekends upwards of 100,000 shoppers and tourists congregate on its 3 major market sites to search for clothes, jewellery and brik-a-brak. The majority of them arrive at Camden Town station.

The station is important also as a major interchange between the two branches of the Northern Line (Edgware-Kennington and High Barnet-Morden). Despite mixed train routing on the two branches passengers frequently have to change at Camden to reach their destination. This leads to significantly more congestion across its four platforms than would otherwise be the case. Added to this, Camden Town itself is a well-populated area close to central London.

These three factors make it one of the busiest stations on the network and mean that at some periods on weekends access to the 101 year-old station is limited to set-down only, so as to avoid dangerously congesting the platforms. i.e. during these periods no-one may enter the station. This is obviously inconvenient for shoppers wishing to return home as well as locals wishing to make regular journeys. Through passengers, transfer passengers and those arriving are not disrupted.

Camden's agglomeration of street and alternative markets attract enormous crowds, especially at weekends.

London Underground project a 20% increase over the next 20 years. Furthermore, in the near future the line will be defnitively split into two routes, further increasing the demand for interchange at the station. London Underground quite rightly wishes to remedy the situation so that Camden Town can meet the demands placed on it. They hired Ove Arup to design an entirely new Camden Town station, from the bottom up.

At an estimated cost of £135m (these things always overrun), the design requires the demolition of a whole city block, including certain local landmarks of sentimental significance. Among them, the listed Electric Ballroom, one of London's oldest nightclubs; Buck Street market, one of the smaller markets in Camden; and a church. The resulting plan would incorporate new accommodations for the Electric Ballroom and the church as well as commercial office space and would take seven years to construct but significant opposition from residents and Camden Borough Council has led to over 16 versions of these plans being rejected over the last few years. Currently plans are at an impasse, with no adequate solution to satisfy both London Underground's needs below ground and local interests above ground. Allegedly, however, in 2007 the government declared that the Electric Ballroom is a dilapidated building, which could swing the tide in the favour of LU's ambitions.

The proposed £135m overhaul of Camden Town would look like this. Demolition of a whole city block has led to enormous local opposition.

A cursory glance at the plans reveals many of the hallmarks of modern LU station planning.  I hesitate to mention words like waste and vanity, in what is merely a rebuild. Including commercial development is undoubtedly part of the business model and it is not clear what long-term income TfL might extract from its tenants. Whilst Ove Arup are an extremely well thought of company, their inclusion may be what the bosses of Madrid Metro warn precisely against in metro construction. Prestige architecture is a luxury that the taxpayer cannot afford and requires cost-cutting in other more essential areas, such as reaching neighbourhoods where there is no service at all. Lottie gets new ballet shoes while Tom and Charlie go barefoot. At £135m it's an expensive enlargement.

There are currently three other stations in the vicinity of Camden's markets other than Camden Town: Chalk Farm, on the Northern Line Edgware branch, Kentish Town West (KTW) and Camden Road, both on the Overground North London Line. (see map)

For the sake of simplified analysis let us take the junction of Camden High Street and Castlehaven Road as the nodal centre of the principal market area of Camden Town. The station resources relative to this point are as follows:
  1. Camden Town 350m (all destinations on Northern Line, central and south London)
  2. Chalk Farm 600m (Edgware branch destinations only)
  3. Camden Road 600m (East London destinations only)
  4. KTW 715m. (West London destinations only)
Factoring in the route connections of each station tilts the bias towards Camden Town for passenger uptake. Furthermore, train frequencies on 3. and 4. are typically only 4tph, as opposed to Northern Line frequencies around 20tph/branch. N.b. Northern Line frequencies are projected to increase to over 30tph once the line is split in two.). Chalk Farm can serve central and south London but requires counter-intuitive strategy on the part of the passenger as it is:

a) almost twice as far to walk as Camden Town
b) in the opposite direction to central/south London
c) has half as many train services as Camden Town
d) may require a change at Camden Town anyway

Even without detailed statistical data it can be seen that stations 2, 3 and 4 cannot significantly reduce passenger demand for Camden Town station. The same conclusion has been reached by London Underground and Camden Borough Council. Doing nothing is not an option. But with the redevelopment plans on the buffers, are there any other alternatives? Here I shall briefly put forward some new scenarios.

Option 1 - Rebuild Camden Town station in a different place.

  • The existing station could operate mostly unimpeded whilst works progress.
  • Sensitive conservation sites and landmarks need not be disturbed as in the current plans, or at worse, less sensitive ones demolished.
  • New ticket halls and escalators can be excavated directly under the street.
  • Would be cheaper than the current scheme
  • All the capacity and facility improvements of the Ove Arup scheme would be achieved, albeit in a slightly different place
  • Some more serious excavation would be required to bore the four tunnels to platform gauge on the existing route
  • Tube traffic disruption during tunnel boring
  • Serious traffic disruption on Camden High Street
  • The new station would necessarily be marginally further away from Camden markets
  • Would probably still take 3-4 years
Guesstimated congestion relief at Camden Town - 0% (but all the capacity to contain it)

Option 2 - Build a new Overground station next to Camden Market

Two railway lines pass very close to Camden Market. One is the North London Line, with stations already at Camden Road and Kentish Town West. The other is a link between the NLL and the Watford-Euston Line. This line is not currently an Overground route though maps of the network's future do display it offering through services between the two lines. Station construction on this stretch would be relatively straightforward and inexpensive as there is plenty of elbow room around the site. The main problem with this idea would be the need to rethink Overground routings and the issue of train frequency (which currently stands at 4tph on the NLL).

To be viable the station would have to offer through services between Willesden Green (or even Watford) and Stratford thus allowing useful interchange at Dalston (ELL), Highbury and Islington (Victoria) and Queen's Park (Bakerloo), as well as the termini; and increased train frequencies to say 8tph. It would replace the function of Camden Road and KTW relative to this site.

  • Relatively cheap, quick and easy construction (18 months?)
  • On the site of the markets themselves
  • No local heritage damaged
  • If option 3 were to be pursued in tandem this could become an important Underground/Overground interchange
  • Would require adaptation of LO's long-term strategy to be viable.
  • London Overground does not yet have the image of "convenience" that would attract passengers in the same way that an Underground station would. That may change.
  • Interchange on LO is limited and the east-west axis of the line would not necessarily be conducive for journeys to or through central London.
N.B. There is also space to locate a station directly on the NLL, thus not impacting Overground routings. Once again  train frequency would have to improve and the station itself would be marginally further from the markets. In effect, this would be a replacement for Camden Road station which would likely become unviable.

Guesstimated congestion relief at Camden Town - 0-15% (dependent on routings and frequencies)

Option 3 - Refit Camden Town to a basic level and provide extra capacity in a new station nearby.

A "Camden Market" underground station halfway up Chalk Farm Road (see map), close to the nodal centre of the markets would attract significant shopper traffic away from Camden Town (and also Chalk Farm). Being at only 150m from the nodal point, the station would undoubtedly attract the majority of passengers from south, central London and the Edgware branch wishing to visit the market. Camden Town would be left to cope with predominantly High Barnet traffic, regular local catchment and transfer passengers - well within its capacity.

How big would Camden Market station have to be? Well, on a rough assumption, around 50-75% of all market traffic would alight here. Once the Northern Line's running is definitively split passengers on the High Barnet-Morden branch would still use Camden Town as changing to the Edgware branch for one station would be inefficient. However since most journeys will not originate on the Northern Line, many passengers would be expected to reroute their journeys to interchange onto the Edgware-Kennington line at Euston, Tottenham Court Road and Embankment, etc. Add to this the residential catchment but remember it would not have the interchange problems that Camden Town does. It would likely have to be no significantly bigger than any of the other Edgware branch stations e.g. Chalk Farm, although weekend demand spikes would be very marked, so wider platforms would be an asset.

As to construction sites, this proposal is in the vicinity of Camden's stable market. This large site still contains semi-permanent structures and could feasibly be excavated for construction of an Underground station, without major planning issues. Some local protest would undoubtedly be stirred because of disruption during works, but upon completion the market could return to normality, barring one or two well-located station entrances.

Try to see this as an enlargement of Camden Town with an extra pair of platforms up the road.

  • Minimal disruption to Camden Town station during works and all buildings remain.
  • Could be done relatively quickly (2-3 years, as opposed to seven at Camden Town)
  • Free up capacity at Camden Town by removing a large proportion of market traffic
  • In effect, an enlargement of Camden Town with extra platforms
  • Improved general catchment on the line hence increased fare revenues
  • As a single line station it would be far cheaper than rebuilding Camden Town interchange
  • More convenient access for market shoppers
  • Easy planning due to semi-permanent structures in the vicinity
  • Slight increase in journey times on the Edgware branch
  • Extra operating costs of a new station
  • Some disruption to traffic on Chalk Farm Road and possibly demolitions during works.
  • Disruption to stable market could stir some protest
  • Not useful to High Barnet branch passengers
Guesstimated congestion relief at Camden Town - 20-50%

Option 4 - Reopen South Kentish Town Station

South Kentish Town station lies about 600m north of Camden Town up the Kentish Town road. It was one of the original stations on High Barnet branch of the Northern Line. It closed during a strike in 1924 but was never reopened as passenger demand was never very high. The station structures remain in place and the station building is now a commercial outlet. The walking distance from the nodal point is 570m. It would therefore be the second closest station to the market area, but still generally less convenient than Camden Town (depending on which side of the markets you are on).

It could be expected to divert a major chunk of northbound High Barnet branch passengers even if the doors are open at Camden Town. The natural incentive of escaping the crowded area around Camden Town might contribute to this organically. As the sole improvement in the area it would be a great convenience to many more passengers (south and northbound) whilst Camden Town remains set-down only, although obviously not ideal. As in the case of Chalk Farm, the use of the station for south and central London would mostly be counter-intuitive for reasons listed earlier. However, the pedestrian route from Camden Market doglegs and there may be ways to improve passenger access with a more direct right of way, possibly incorporating a subway, and any feasibility study should investigate this.

The great attraction of this scheme is that in Underground terms it would be extremely good value for money. No excavation is required, just a refurbishment. The short timescales involved would make it a "quick fix" solution, if not a comprehensive one.

Leslie Green's original South Kentish Town station. Being less than 600m walk from Camden's principal markets makes it a viable alternative to Camden Town, and extremely good value for money.

  • Could be done very quickly (18 months?)
  • Extra capacity and catchment on the line, hence more fare revenues
  • Likely to naturally divert most of High Barnet passenger traffic from Camden Town
  • Highly cost effective as the station structure is already in place. It would require minimal structural works and refitting - a small fraction of Camden Town redevelopment (£135m)
  • No demolition/controversy required and Camden Town's buildings undisturbed.
  • With adequate passenger encouragement could remove traffic for central/south London routes away from Camden Town also
  • Slight increase in journey times on the High Barnet branch
  • Extra operating costs of a new station
  • Some disruption to traffic on Kentish Town Road during works.
  • Not useful to Edgware branch passengers, nor central/south London passengers without adequate encouragement to walk slightly further
Guesstimated congestion relief at Camden Town - 2-10%

Option 5 - Option 3 + Option 4

Creating a new "Camden Market" station on Chalk Farm Road and reopening the disused South Kentish Town Station in combination would add significant station capacity to both the Edgware and High Barnet branches in the Camden market vicinity. This would negate the need for expansion works at Camden Town. A simple refit would suffice. Additionally passenger convenience is increased for the local catchment area by providing journeys closer to origin and destination.

Returning to our nodal point analysis, the station resource list for the Camden Market area in this scenario would then look like this:
  1. Camden Market 150m (Northern Line Edgware Branch, central and south London)
  2. Camden Town 350m (all destinations on Northern Line, central and south London)
  3. South Kentish Town 570m (improved access notwithstanding, Northern Line High Barnet branch)
  4. Camden Road 600m (East London destinations only)
  5. KTW 715m. (West London destinations only)
A reduction in pedestrian congestion is also likely in the crowded streets as pedestrian journeys would be cut. Most significantly of all, the two station projects in combination would likely cost far less than the £135m Camden Town projection, could be completed within three years and would not cause the upheaval amongst local residents that demolition of such local icons as the Electric Ballroom can cause.

Guesstimated congestion relief at Camden Town - 22-60%


Taking 30 random stations on the Underground or Overground and planning the most convenient route to visit Camden Markets shows the following changes in station usage. Preferred station currently is listed first, preferred station with the 3 notional stations (Camden Market Underground CMU, Camden Market Overground CMO, South Kentish Town SKT) available is listed second.

ORIGIN Actual Notional

Embankment Camden Town CMU
Roding Valley Camden Road CMO
Upminster Bridge Camden Town CMU
Tooting Bec Camden Town CMU
Ladbroke Grove Camden Town Camden Town
Highgate Camden Town SKT
Surrey Quays Camden Road CMO
Cal. Rd & Barnsbury Camden Road CMO
Arsenal Camden Town Camden Town
Baker St. Camden Town Camden Town
Acton Central Kentish Town West CMO
Hendon Central Chalk Farm CMU
Bank Camden Town Camden Town
Bond St. Camden Town CMU
South Harrow Camden Town Camden Town
Tottenham Hale Camden Road CMO
Fulham Broadway Camden Town CMU
London Bridge Camden Town Camden Town
Canary Wharf Camden Town CMU
Victoria Camden Town CMU
Brixton Camden Town CMU
Shadwell Camden Road CMO
K'ton (Olympia) Kentish Town West CMO
Osterley Camden Town CMU
Hillingdon Camden Town Camden Town
Holborn Camden Town CMU
West Ruisip Camden Town CMU
Clapham Junction Kentish Town West CMO
Lambeth North Camden Town CMU
South Kensington Camden Town CMU
Euston Camden Town CMU
Finsbury Park Camden Town CMO
Covent Garden Camden Town CMU
Southfields Camden Town CMU
Edgware Chalk Farm CMU

In this sample 21 of the journeys (70%) currently terminate at Camden Town. 7 of the Camden Town journeys are unaffected (33%), 12 change to CMU (57%), 1 changes to CMO (5%) and 1 changes to SKT (5%). i.e. the new stations divert 67% of journeys away from Camden Town.

5 journeys (17%) terminate at Camden Road. Of these, all 5 (100%) would prefer CMO.

2 journeys (7%) terminates at Kentish Town West and would also prefer CMO (100%).

2 journeys (7%) terminates at Chalk Farm and would prefer CMU (100%).


Camden Town receives the bulk of traffic for Camden Markets. Camden Market Underground station would have the potential to remove around 50% of Market passenger traffic from Camden Town. What proportion of Camden Town's total traffic that would be (including interchange passengers, and normal catchment) is not clear but it is undoubtedly very significant. It would also take traffic away from Chalk Farm, although the volume is less significant there.

South Kentish Town would displace far fewer passengers, but, again a 5-10% reduction may be highly useful considering the modest cost of refurbishing that station.

The case for an Overground station at Camden Market looks weak as it does not displace many journeys from Camden Town in this sampling (just the one from Finsbury Park). It is undoubtedly preferable to Camden Road, or Kentish Town West, however, for Overground journeys.


Barring laws, bills, acts of parliament, public enquiries, planning permission and of course, the money to do it this might be a good solution to remove 20,000+ passenger journeys to and from Camden Town per weekend, and much cheaper than the reconstruction of Camden Town (£135m):

1. Reopen South Kentish Town for some very cost effective immediate light relief. estimate 18 months, £12m
2. Construct Camden Market Underground Station for a very significant diversion in passengers from Camden Town. estimate 2 years, £30m
3. If necessary refurbish Camden Town with some platform widening works, estimate 2 years, £20m

Thursday, 2 October 2008

ON AIR-RIGHTS, FOOTPRINTS AND DENSIFICATION, or How we're running out of space and shouldn't be wasting it

We're running out of space. In 2000 the government told us the UK would be needing 3.8m new homes by 2016 and may have to start tearing up greenbelt policy to meet demand. Our cities simply don't have room to grow any more. Even if every brownfield site were built on, we'd still need heaps of greenfields to meet demand for new single-occupant properties. (Great article about this here.)

Well, that sounds kind of serious, doesn't it? We don't want to ruin our countryside. It's a small, overcrowded country we live in and we want to keep our green and pleasant land, thank you very much. Sounds like something should be done about this. Something that will allow us to keep our gentle suburbs, keep our countryside and provide all these much-needed new homes. "Densification" (to quote Richard Rogers) was the answer. i.e. to pile in higher-densities of occupancy into our towns and cities and in so doing create more viable, efficient and pleasant urban environments. Kind of like Paris...

Abbesses Metro station in Paris. A classic piece of Hector Guimard art nouveau. Like almost all Paris stations, there is no surface building. It's infrastructure that gets out of your way, and feels kinda sexy at the same time...

Paris, for example, does everything London does in half the space, with the same number of people (to quote Ken Livingstone). Which is a good thing in terms of urban viability and countryside protection. Problem's always been, us Brits just love to stretch out. We love our suburbs, our low-rise and our open skies. We're going to need some tough love from our politicians, quangos and planners if we're going to get in line.

The kind of people that politicise, quang and plan our public transport, for example. They ought to know that one of the unique marvels of an underground rail station is that it needn't occupy any surface space at all. All you need is a hole in the ground with a wrought iron awning (a la Hector Guimard), or perhaps an elegant little building, tucked neatly under a bigger one, a la Leslie Green or Harry Bell Measures. They knew all about "air-rights" 100 years ago! The station buildings were constructed with flat roofs and reinforced steel frames with the deliberate aim of selling off that valuable skyward square footage.

Leslie Green's classic Oxford Circus station: classy burnt russett tiles - check!; reinforced steel structure - check!; sufficient contextuality affording a neat contribution to the density of the metropolis and urban viability - check!!!

So what's the current deal? Take a look at London's newer stations on the Jubilee Line. Southwark follows this same principle, but has so far not attracted an air-rights tennant. Points for trying. Further east though, Bermondsey, Canada Water and North Greenwich all have big monumental surface buildings with no prospect of air-rights development. You kind of start to forgive. They seem to look at you, wink and whisper "it were a bit shit round 'ere before we come along like". You nod back and force a smile. Another 150 single mums out in the farmyard...

Station with attitude: Canada Water. It knows it's the most important building around here...

It gets worse though. Out west and as I write this, Harry Bell Measures' dear old Shepherd's Bush station is being replaced with an even bigger monster. The new Central Line station sits like some genetically-modified conservatory that couldn't make it round Holland Park Roundabout on its way to the Chelsea Flower Show. Gone is the human scale of the hole in the pavement as is the pragmatic capitalisation of what is prime central London real estate. It has what planners call a large "footprint". You can't miss it and it ain't gonna get out of your way. Its' a huge glass atrium, has no internal access to the new Overground station next door, is costing around £65m and is totally...empty! "Nul points" Shepherd's Bush.

Artist's impression of the new Shepherd's Bush Central Line station. As you can see, it's space well-used. It contains, erm... nothing.

It looks unlikely that London will ever discover the charms of Hector Guimard's discreet human-scale chic but the contextual pragmatism of the Central and Northern line pioneers may have a future. The proposed Camden Town redevelopment features a steel and glass office building incorporated into the original design. But at a price. So far more than 16 plans have been rejected for the £135m (!), seven year project, mainly on the grounds of conservation as several Camden landmarks would be demolished.

Camden Town's latest proposal (rejected): points for using the space, shame about the landmarks (and the price!). You can't have it all.

What seems clear is that if we are serious about combating the problem of suburban sprawl and "densifying" our inner cities our planners need to reconsider their priorities. Whilst schemes like Crossrail are beneficial in some ways (although vastly expensive), who will want to stay in inner London if it's quicker to commute from the home counties? This is going to require some hefty transport improvements for inner (and especially south) London. Whilst perhaps insignificant in the larger scheme of things, wasted space above underground stations demonstrates how seriously the issue is being taken.

Assuming it can't keep the Electric Ballroom et al intact, the Camden scheme is right about one thing - incorporating its new tenants from the start. That way the space does get used, and presumably London Underground will make a buck or two as the landlords. That's the lesson from Southwark and others. In 108 years no-one ever did build on top of Harry Bell Measures' dear old Shepherd's Bush.


It goes like this:

The Piccadilly Line was overcrowded. The Victoria Line was built to relieve it.
The Central Line and Paddington are overcrowded. Crossrail 1 will be built to relieve them.
Victoria and Kings Cross are overcrowded. Crossrail 2 will be built to relieve them.

In areas that have no lines or stations there are no overcrowded lines or stations. Therefore we do not need to build new lines to relieve the overcrowding there...

The result of this logic is "concentration":- of resources, investment and opportunities and, naturally, further overcrowding! Central London stations become more overcrowded, overconnected, overblown, as more lines join them up.

Reaching underserved neighbourhoods first has the more pleasing effect of "distribution".

Wednesday, 1 October 2008


In the past decade Madrid has astonished the world with its explosion of Metro construction. It has amazed not only for the scale of its ambitions, but with timeframes and budgets that would make other cities blush. With two major project phases (1999-2003, 2003-2007) Madrid Metro has cast itself as one of the world's most extensive systems.

Phase One included the new Line 12, or "Metrosur" which runs in a loop around some of Madrid's southern suburbs, and a Line 10 extension to reach it from the city. The Second Phase was even more ambitious, including 10 Metro extensions and three new light rail/tram lines (see map below).

Scale map of the Madrid Metro network. Thicker lines denote new construction 2003-2007 (47.4 km of metro lines with 42 new stations plus 28 km of light rail lines with 39 new stations). Metrosur is the green loop line at bottom left.

This was achieved under the guidance of Metro President Manuel Melis, quoted from this article.

Talking of the Metrosur project Mr Manuel Melis said construction had gone according to plan, though he admitted that the timescale could have been even shorter. He said: "We completed everything on time and within budget In fact, we could have finished six months earlier because we were too conservative in our planning. Tunnel construction went faster than we expected."

The entire cost of the 1999-2003 metro development programme amounted to [euro]3.16 billion. These projects, incorporating Metrosur and the Line 10 extension, included planning, civil works, electrical and mechanical installations, interchanges, maintenance facilities, and rolling stock at an average cost of [euro]42 million/km. Previous projects have been undertaken successfully on the same basis, as Melis explained. He said on that occasion: "I believe that rail transport projects are simple engineering projects, easy to design and build, and, with the appropriate staff and management techniques, they can easily be completed on time and within budget. I refer particularly to those Madrid Metro projects where completion dates have not only been met, but have been beaten by several years in comparison with similar projects elsewhere."

"Civil works amounted to about 70% of the total cost of our programmes. The most important part of this cost is the tunnel part. Transport infrastructure projects can be divided easily into manageable parts. Each section of the project can be designed simultaneously and all contracts can be awarded simultaneously, so that any manageable contract worth up to [euro]150 million, for example, can be completed within three years. Even enormous tunnelling projects such as the Channel Tunnel have been excavated in this timescale. Therefore, provided that funds are available, any lineal project such as a metro can be designed and built in 40 to 45 months, as we have demonstrated.

"Station architecture is an important factor, too. It should never be handed over to world-renowned architects. A transport project is a serious engineering work that should not be confused with a museum or an emblematic building for a city. Several million passengers/day may move through metro stations, so their design must take into account this fact by giving easy access from the street to the trains, via wide escalators and corridors and shallow station platforms.

A typical new Metro - Metro Ligero (light rail) interchange in Madrid's northern suburbs. Madrid punches above its weight with down to earth "does what it says on the tin" architecture. (c) Luke Peters 2006

Metrosur (Line 12) is 41 km long and has 28 stations. It took four years to complete from authorisation to completion at a cost of 42m Euro/km, and is almost entirely underground. Compare this to the Jubilee Line Extension, the last London Underground Project of any note:

16km long with 11 stations, about 80% underground and 20% overground on existing rights of way; took 9 years to build from authorisation to completion at a whoppingly over-budget £3.5bn, or £218m/km.

I'll repeat that because it's an important figure and won't even bother converting it to Euros. Jubilee Line Extension: £218m per kilometre. Metrosur: 42m Euros per kilometre. That's roughly 6 times the price/km. i.e. if the JLE were being built in Spain they could have built 6 of them for the same money!

Canary Wharf station on the Jubilee Line. Designed by world-renowned architect Sir Norman Foster, the station has been likened to a cathedral. It's an impressive achievement, but for the same money five or six simpler stations could have been built.

Furthermore, construction of the Victoria Line took 10 years. The Chelsea-Hackney Line was thought up in 1901 (!) and has still not come to fruition (although it arguably shouldn't, in its current form). The East London Line Extension was put forward in the 1980s and phase one completion is due in 2010. Relatively simple extensions to the Bakerloo, Victoria and Northern Lines are operating on a, shall we say, "open" timescale.

Could it be London has a project management problem, rooted in a flawed philosophy? After all, why should projects take so long and go so much over-budget when cities like Madrid demonstrate time and again that with the right approach, large infrastructure can be built far more easily and cheaply than we in London perceive?

Consider these two contrasting philosphies for a moment:

Philosophy A:

1. Metro Lines are expensive, difficult projects of questionable benefit, that take a long time to design and construct (high investment, low return). Therefore:

2. Metro Lines are rare. Therefore:

3. We must spend a lot of money and time ensuring we do not waste this rare opportunity, or we will not undertake it. Therefore:

4. Metro Lines are expensive, difficult projects of questionable benefit, that take a long time to design and construct (high investment, low return).

Philosophy B:

1. Metro Lines are affordable, straightforward and worthwhile projects that take a short time to design and construct (low investment, high return). Therefore:

2. Metro Lines are common. Therefore:

3. We will not waste time and money worrying about this common opportunity. Let's get to it. Therefore:

4. Metro Lines are affordable, straightforward and worthwhile projects that take a short time to design and construct (low investment, high return).

We invented the underground railway but somewhere along the line we fell from the faith. Somehow it belongs to an age of Edwardian pioneers and top hats. Aside from prepubescent networks in Glasgow, Liverpool and Newcastle, the rest of the UK never bothered with the idea at all. It was always far too much trouble. And yet Sr. Melis and his colleagues in Madrid have no psychological obstacles to Metro construction at all. They just get on and perform, what could seem in some circles, miracles:

"This latest project demonstrates again the importance of our philosophy. Many cities around the world desperately need new metro lines, but they cannot afford to build them at an estimated [euro]150 million to [euro]200 million/km, nor be forced to wait a dozen years for the lines to become reality. These estimates of costs and time are simply wrong. In Madrid, with all humbleness, we have now proved it on more than one occasion.

And it does take on something of a religious aspect when you consider that Seville, Barcelona, Bilbao, Valencia, and Malaga are all doing the same, within an economy far weaker than the UK's.

London has one or two lines in the distant pipeline, the £16bn Crossrail  line and the aforementioned Chelsea-Hackney Line. Relatively simple extensions to the Bakerloo and Victoria Lines are indefinitely mothballed. Outside London there are no new metro lines planned at all. Even very large cities such as Manchester and Birmingham have opted for surface light rail schemes, though the potential for full metro exists.

Clearly one of the mistakes made on the Jubilee Line Extension was to employ just those "world-renowned architects" Sr. Melis said we should avoid. Over-ambitious, monumental stations overran the budget for no gain to the passenger other than a brief "wow" on first visiting the station, followed by another "wow" upon seeing the ticket price. Britain has probably learned its lesson on this front, but is still waiting for its own evangelical Sr Melis to spread the gospel. Until then, all its pagan superstitions about Underground construction are likely to be self-fulfilling.

By the way an average journey in Madrid costs around 70 euro cents (about 50p). Sobering...


View Larger Map

The Aldwych spur from Holborn was originally conceived to be extended south of the river. As with so many schemes it never came to fruition and the service to Aldwych operated as a single track shuttle until closure in the 1990s (despite having twin tunnels). Some modification of Holborn would be necessary and this could be run as a spur of the Piccadilly line with trains direct from Cockfosters or be a self-contained line terminating at Holborn.

Aside from connecting Peckham to the network this would usefully serve the major development area around Elephant & Castle, serving not just Elephant & Castle but also Walworth Road, some way north of the proposed Thameslink station. A new station on the South Bank opens up access to major leisure facilities there.

Estimated cost to Madrid Metro specs 273m Euros; to JLE specs £1.4bn


In the beginning there was the Railway network. Then someone dug a tunnel and created the Underground network (though a lot of it runs overground). Then, somewhere between the two a light rail network was conceived for the Docklands. Not long after a Tramlink network for Croydon, and an Overground network, which was pinched off the Railway network and the Underground network (and some of it runs underground). Soon to be added is a Crossrail network, which is a bit like a Thameslink network, only it isn't, and maybe a cross-river transit or two, which is sort of a tram.

Isn't it all getting a teensy bit over-complicated?

Some off-the-cuff suggestions:

Couldn't DLR and London Overground be integrated? Train profile and power supply differences notwithstanding, they kind of do the same thing.

Couldn't Thameslink be integrated into the new Crossrail scheme. Again, it's kind of the same concept.

Couldn't London Overground absorb some more of the inner south London rail network to start providing a more rational "metro" service.

Three networks is about all I can get my head around...


Now, I have a problem with any "Metro" system that names its lines "East London Railway" or "West London Line" etc. Why? To me it conjures up fat controllers and steam engines, complicated timetables and no Sunday service.

Now, that's fine for a system with no pretensions beyond being a suburban rail service. But TFL clearly do have big ambitions for the Underground's little sister. It's on the Tube map, its being extended, its being homogenised, it's getting new trains and an orange roundel of its own. It almost looks like an S-Bahn.

Except for one thing. The East London Railway doesn't sound like it's going to get me anywhere fast. Looking on the map, all the lines are a similar double-stripe orange, and it's not clear where services are running from and to.

What would convince me? Well, a branding and diagrammatic scheme in line with London Underground. Now, I don't much care if it's a double line to distinguish it from Underground lines. That's fine. Or if they use pastel colours or dashed lines. Don't care at all. What is important is that the colour should be distinct and that the lines should be distinctly numbered.

I'm far more inclined to hop on Overground Line 3 Westbound, than attempt to make the o8.53 stopping service to Gospel Oak on the "Gospel Oak to Barking traction Railway (steam trains to all parts of the metropolis)", especially if it had a neat light blue, or lilac colour. Then I could swiftly change onto Line 1 Eastbound (the green one), and continue my journey to Highbury & Islington confidently knowing I won't be waiting more than 8 minutes on the platform. I need something to sweep away the images of rattling points and clouds of steam.

My general rule of thumb; how easily could you explain it to a foreigner? Make any sense to anyone?


Now, the Crossrail concept is discussed here. Unlike Crossrail 2 however, this one looks like it might get built. On balance this would be a good thing, but for the ever-pervasive problem of value for money. Crossrail 1 is going to be immensely expensive (current estimate £16bn), and yet, as with Crossrail 2, what it brings to the table in terms of new services is not what it ought to be.

Unlike Crossrail 2, this project does plug into a major commuter route out to Maidenhead (which could conceivably be extended to Reading, (high-speed train competition notwithstanding), but in the east falls short at Shenfield. (Why isn't it reaching Southend?).

The London Transport Users Committee had this to say on the matter:

LTUC note that the draft train service specification for west of Paddington has few of the features of a metro service and some stations (e.g. Hanwell) with only 2 tph would have a service that is completely inadequate. Burnham – current off-peak journey times to and from Paddington are 25 and 29 minutes respectively. Crossrail will increase the journey time to 38 minutes. Taplow – improved service off peak from one train per hour to two per hour. Current off peak journey time to and from Paddington between 29 and 35 minutes. Crossrail will increase the journey time to 43 minutes.

Crossrails south-eastern branch would offer extra capacity for growth of rail traffic in Kent, but does little to improve rail facilities in Thurrock or South Essex, in that there would be no station east of Custom House and that the regenerated Stratford area would not be directly served. Although some may argue for a service to cover this area, LTUC would not be in support for a third eastern branch of Crossrail. It may be that the existing c2c route via Tilbury to Fenchurch Street can be developed to meet the rail needs of Thames Gateway North, but we are concerned that little work appears to have been done to verify this and that no section of the rail industry seems to be taking full responsibility for this issue. LTUC also believe a station at Silvertown should be built from the outset, instead of passive provision. This would enable a short pedestrian or bus link to London City Airport.

Through central London Crossrail 1 is little more than a big fat Central line, linking the same old stations again, which are likely to collapse soon under their own gargantuan complexity, whilst peripheral neighbourhoods languish. Overall number of new stations for London? Technically 2 - Woolwich and Isle of Dogs, but both are a stone's throw from existing stations. We could say that no new neighbourhoods are being served therefore. NONE!

Does this represent value for money? (They say £16bn at the moment but it'll be 6 times what anyone tells you!) Not really, not whilst Chelsea, Battersea, Camberwell, Walworth, Peckham, Hackney etc. etc. don't have a single tube station amongst them. Add to this projected services that are slower and not significantly more frequent than current timetables.

New extensions to the Northern Line, Bakerloo line, and Waterloo and City could be achieved for 1bn Euros (if Madrid Metro were in charge, at 2003 prices and with Spanish base costs). Even if you conservatively multiply that figure by four, to 4bn Euros, it's less than a quarter of Crossrail 1's projected cost. That, including 21 new station sites in south London, as opposed to the TWO that Crossrail 1 offers. You do the math...


The Chelesea-Hackney line was conceived decades ago as a way of linking two inner London areas (Chelsea and Hackney) into the tube network. Despite relatively high population densities and their central locations, they had been overlooked ever since the Underground started its life over a century ago.

The safeguarded route (see image) shows the line passing through central London then onto two existing lines - the District Line to Wimbledon and the Central Line to Epping (both on the LU network).

Although initially conceived as a LUL line like any other, the idea was taken over by the same group that conceived Crossrail 1 and the idea took root that this line too should be built to Crossrail standards.

For those unaware, the Crossrail is in concept an express, long-range, heavy metro line that bores through the centre in tunnels and extends far out to suburbs and peripheral towns on the surface. It is closely based on the RER system in Paris. An "underground" line, by contrast, would ordinarily have no interurban pretenses, lighter infrastructure and greater station densities, without express facilities. The Crossrail/RER advantage lies in the relief of congestion it would provide at rail termini, as it distributes commuter traffic closer to their final destination, whereas normally they would be funneled onto and off the underground network at very high density. It also provides the potential for much quicker intra-urban journeys.

The line would serve existing national rail lines in Hackney and add one new underground station in Chelsea, providing benefits to those areas.

Currently it is caught between two concepts and can't decide which one it is. In its current form it has the profile and expense of a heavy Crossrail line and the lower station densities, but the range of an LU line (Epping to Wimbledon). Until extended routes are published onto say, SWT rails, or East Anglia trains out to Chelmsford.

The main flaw in the concept is that it provides a grand total of one new station to London. ONE! One station for Chelsea, which urgently requires two or three. Its central route doubles up capacity between Sloane Square and Victoria (is this really necessary?) and merely joins the dots through central London's same old stations which are set to become major, passenger-unfriendly labyrinths. In this profile the inclusion of both Piccadilly Circus and Tottenham Court Road stations seems ridiculous. The Crossrail stations are enormous, with platforms over 250m long.

The cost of a scheme like this is immense, much more than a normal tube line, since stations have to be much bigger. Don't believe what the experts project. You usually have to multiply this figure by 6! Immense cost, for the luxury of having an alternative route between Wimbledon and Epping, and one new station in Chelsea! Does it make sense?

What's the solution? If this scheme is to work, firstly it has to decide what it wants to be: either a scaled-down light, LU line, between Wimbledon and Epping, but providing sensible new services to London i.e. some route alterations and more stations! ;


An express heavy rail line linking onto existing surface commuter routes that is going to usefully free up capacity on lines into Liverpool Street and Victoria/Waterloo, perhaps between Guildford (over SWT routes) and Chelmsford, by extending the existing Ongar route. (?)

Tuesday, 30 September 2008


View Larger Map

The problems with the envisaged Crossrail 2 "Chelsea-Hackney" scheme are discussed in another post This is a radically different proposal which would see Stevenage or even Letchworth joined to the Medway towns. This would use existing rights of way into London as far as Moorgate and London Bridge, requiring a new linking tunnel from London Bridge to Moorgate via a new Cannon Street station in tunnel. This would leave Cannon Street mainline station as obsolete so it would probably have to close (freeing up some pretty valuable real estate!). A new tunnel section under London Bridge would emerge onto new stations envisaged through Bermondsey, one of the most depressed areas of inner London.

Sunday, 28 September 2008


Hands up if you'd like to see some of these stations on the Tube Map. Some may have been closed for reasons of underuse, expense or fire, and not reopened. Some would be entirely new.

The principle: London has several inner city areas with poor transport links and socioeconomic problems, all within a stone's throw of the city centre. New lines aside, to encourage regeneration of these neghbourhoods and revitalisation of residential communities, the following stations could be built relatively inexpensively and be of immense benefit to inner London communities.
Furthermore, we are all aware that we face a housing shortage, and need to increase our urban densities if we are to safeguard our countryside for future generations. The logical target of such densification (to coin Richard Rogers) would logically be the fringes of central London. Naturally we're facing a chicken or egg situation. Development won't come without transportation and investment tends to go to successful places first... This obstacle notwithstanding, more people need more station capacity and I believe the best way to provide it is with more stations, not necessarily enlarging existing ones, thereby increasing the convenience of the passenger at the same time.

N.B. I personally don't believe you need to employ world-class architects and use gilted escalators such as happened on the JLE (OK I exaggerate about the gilted escalators, but only just). New stations need only be functional and need not be prohibitively expensive. Of course, it's always a lot easier and cheaper to build them during construction of the line, but hey, it's too late and London does need to find solutions.

1. Wood Lane (Almost Complete) - included for reference.

2. Lords. Not the old Metropolitan Station but on the Jubilee line on the large gap between Baker Street and St John's Wood.

3. York Road. Reopen this station! The King's Cross Railway Lands development is next door, a rundown backwater of Barnsbury behind it. It's a long old way up to Cally Road!

4. Mount Pleasant. A long old trundle on the Circle Line too between King's X and Farringdon. This is a densely populated area.

5. Barnsbury. Although the concept of the Victoria was to be effectively an Express Line between existing stations, it may be time for the line to mature and reach more of the neighbourhoods it passes under, this being one of them.

6. Pentonville Road. An entirely new station for the Northern Line City branch.

7. City Road. An old station which could be reopened in an area which needs some vitality.

8. Shoreditch High Street is about to have a brand spanking new station on the East London Line, passing right over the central line. It's a huge distance between Liverpool St and Bethnal Green and this is a well-populated central neighbourhood. Dosn't it make sense to have an interchange here?

9. Tower Bridge. The JLE was all about quality, not quantity, but once again it bypasses busy neighbourhoods entirely.

10. South Kentish Town was closed decades ago and is now a Cash Converters. We all know the problems of overcrowding at Camden Town. I propose this as a better solution to a massive overhaul of Camden Town station.

Northern Line extension southward 1 (proposed)

One possible route south from Kennington. N.B. In the near future the Northern line will likely be split in 2. This extension would apply to the Charing X branch, the City branch continuing to Morden as usual. Most likely this line will be coloured lime green on the map, and either retain the Northern Line name or something else (west end line?)

Total cost, to Madrid Metro specs = 315m Euros; to JLE specs = £1.64bn

View Larger Map

Bakerloo Line Southward Extension 1 (proposed)

One possible route southwards for the Bakerloo Line. This plugs a gap in services around Old Kent Road, Walworth and Peckham, some of the most deprived and worst-connected inner-city neighbourhoods in London. The idea has been mooted for some time, but the political will does not yet exist to take on the relatively modest scheme, whilst massive schemes such as Crossrail have been prioritised.

Bakerloo services to New Cross means London Overgound services need no longer stop there and the clumsy spur can be ripped up. The station will allow interchange onto the tube for mainline commuters from Kent.

There exists the potential to takeover National Rail tracks beyond Greenwich, which are at surface or in cutting, on to Catford, Woolwich, or even as far as Dartford.  

Total cost (E&C - Greenwich); To Madrid Metro specs = 294m Euros; to JLE specs = £1.52bn

Waterloo & City Line extension (proposed)

A proposed extension to the Waterloo and City line to Battersea and Putney.
At present the W&C Line is merely a fancy "travelator" for Waterloo commuters to and from the City. The Line could be vastly more useful to Londoners, bringing tube service to Battersea and Wandsworth, including developments at Battersea power station, and interchange on the District, Victoria, Bakerloo, Northern, Jubilee, Central, DLR, Circle and Mainline services. Platforms and stations at Waterloo and Bank would have to be enlarged (they are currently very short).
However there exists the possibility of keeping it as a short-train light rail service with on-street running from Nine Elms through to Putney.

Total cost: to Madrid Metro specs = 504m Euros; to JLE specs = £2.6bn

NOTE: This is approximately half of the proposed new City Line, which would include a southern routing like this and a northern extension.

View Larger Map