View previous topic | View next topic

Building bigger cities

Page 7 of 9
Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8, 9  Next

PDR
1321296.  Wed May 08, 2019 10:21 am Reply with quote

They were dealt with by having a shorter timescales. They didn't have to build a whole underground system first and then start on the buildings,. Being based on roads and cars they could build and occupy buildings progressively, which you couldn't do if the whole concept depends on this extensive infrustructure being in place to support people living there.

Unless I've misunderstood (which is more than possible) your proposal is to build a complete new city based on an extensive 3-dimensional underground transport system.

To minimise the disruption and expense of building that extensive underground system you propose to build that first, for the whole city, and have it fully operational before you start constructing surface buildings because it would not be practicable to live/work in this city without that transport system.

Therefore the capital for the transport system would have to be serviced for the time taken to complete it PLUS the time required to build the actual buildings. The transport system would only start to get any revenue to service the capital after people actually move in. Its revenue would only start to cover its *operating* costs once a significant proportion of the buildings had been completed and people/businesses moved into them, and only after THAT could it even begin to service the capital.

Milton Keynes was a completely different case. In Milton keynes each building could be occupied as soon as it is completed. The road trasnport system could support the occupants with little further work (even when it's a building site it would still work) and the cost/effort required to bring it up to the full operating condition would in any event have been much smaller. The major capital to be serviced is that used to construct the buildings, and they either got sold or started earning revenue pretty well as soon as they were finished. They also benefitted from the way that a road transport network can be built incrementally as the city grows, where your proposed externsive underground system needs to be fully completed at the start. You can't build a few hundred yards of tunnel covering the first few buildings and have them working as a transport system.

That's why these cases are not comparable.

PDR

 
cnb
1321300.  Wed May 08, 2019 10:41 am Reply with quote

PDR wrote:
They were dealt with by having a shorter timescales. They didn't have to build a whole underground system first and then start on the buildings,. Being based on roads and cars they could build and occupy buildings progressively, which you couldn't do if the whole concept depends on this extensive infrustructure being in place to support people living there.

PDR


There's also the fact that it wasn't built on an empty piece of land. There were 40,000 people living there (and therefore paying taxes to the development corporation from day 1) in existing villages.

Milton Keynes wasn't a particularly ambitious project in scale. They had 30 years, and only built about 1500 homes per year. That's about the current construction rate in Leeds.

 
PDR
1321303.  Wed May 08, 2019 10:48 am Reply with quote

Indeed. It's just not a good reference case to the capital question.

PDR

 
GuyBarry
1321306.  Wed May 08, 2019 11:23 am Reply with quote

dr.bob wrote:

I think the main attraction is the ability to create an extensive 3D transport network several levels deep without having a major impact on the people living above ground.


Why does it have to be several levels deep, though? It's expensive to bore lines deep underground. The London Underground "sub-surface" network (Metropolitan, District, Circle and Hammersmith & City) was built by digging a trench and building a roof over it. Subsequent lines were built in deep-bored tunnels because of the disruption caused by the construction of the sub-surface network, but that wouldn't be an issue if you were building the city from scratch.

(Incidentally, the difference between the height of the trains used on the two types of lines is quite noticeable - see photo here.)

Quote:
Although, if we implement cnb's suggestion of a grid system, then we'd achieve the suggestion you made earlier of only ever having a maximum of two lines at each junction, and these could be made to pass over/under each other to increase throughput.


Indeed.

Quote:
This is partially motivated by reducing people's carbon footprint, but it also gives an opportunity to create a space that people would actually enjoy living in. I'm interested to know what that would be like.


I quite enjoy living where I am at the moment, actually! I can walk to pretty much everywhere I want to go (though I do catch buses a couple of times a week). If I had to travel everywhere by Underground train my carbon footprint would probably go up.

Quote:
GuyBarry wrote:

Usually in situations like that there are signs advising passengers "If no through train is shown then take the first train and change at XYZ". Not a significant problem in my experience.


I'm not sure what you mean by "situations like that".


I mean situations where it's quicker to get to one's destination by changing trains than by waiting for a direct service.

Quote:
I'm talking about arriving at a station and, due to the number of possible destinations, people have to wait for the right train. Someone may be able to "take the first train and change at XYZ", but choose to wait because they see that a direct train leaves in 25 minutes and that's less hassle.


Well they might, but if advised to change trains to get there quicker I imagine that most passengers would do so, unless they had difficulty changing trains for some reason (such as disability). There are a number of stations on the Underground (e.g. Baker Street) that have a through service at peak times, but where you have to change at off-peak times. It's a perfectly workable arrangement.

Quote:
If people are being encouraged to wait on the platform rather than just take the first train, then the numbers of people on the platform would necessarily increase.


But that's the opposite of what I said. You can encourage people to take the first train whenever possible, but give them the choice of waiting for a through service if it's more convenient.

 
suze
1321311.  Wed May 08, 2019 12:03 pm Reply with quote

dr.bob wrote:
Presumably there aren't half a million train changers every day. If you average out the figures above, you get a quite modest 80-odd thousand per day, so the train changers must be quite concentrated. There can't be too many days with half a million train changers.


Damn and blast. At this point I have to hold my hands up to a schoolgirl error. I saw that same figure of 29,604,407 ... and promptly divided it by 52 to get a figure per day ...

So half a million a week, yes. A day, no.


cnb wrote:
That's why I think you need to get away from the idea of a centre. Building a city that's equal density all across its area means that you can expand it piecemeal without overloading a core.


You can, but you end up with Los Angeles. Is there much appetite for a medium-density urban sprawl the size of Belgium?

dr.bob mentions Milton Keynes as perhaps the closest we have to such a city in Britain - although I would mention in passing that Glasgow is grid-based too. He is quite right to say that providing effective public transportation in MK has proved challenging. It's better now than it used to be, I am reliably informed, but there are two downsides.

Guy hints at one. Walking anywhere in MK is tedious, and people don't do it a great deal. What's more, the buses stick mostly to the grid roads, so it can be quite a long walk from home to the bus stop. My reliable informant tells me that buses do penetrate the estates more now than they ever used, but there are still large parts of town where they don't.

The other is the government's insistence that buses have to make a profit. It's very difficult for them to do so in a lower-density city, especially if the places where people want to go - the office blocks, the supermarkets, the hospital, and so on - are scattered around the place rather than concentrated in a traditional centre.

 
cnb
1321322.  Wed May 08, 2019 2:30 pm Reply with quote

suze wrote:
You can, but you end up with Los Angeles. Is there much appetite for a medium-density urban sprawl the size of Belgium?


Consistent density doesn't necessarily mean low density. I'm proposing something almost ten times as dense as Los Angeles, which should provide plenty of custom for a frequent and reliable public transport system, and remove the need for many of the worst things about LA: private cars, the amount of space devoted to parking, and the lack of public open space.

LA has, and this surprised me, the highest average population density of any US urban area. It's still barely higher than Milton Keynes.

Quote:
Guy hints at one. Walking anywhere in MK is tedious, and people don't do it a great deal. What's more, the buses stick mostly to the grid roads, so it can be quite a long walk from home to the bus stop. My reliable informant tells me that buses do penetrate the estates more now than they ever used, but there are still large parts of town where they don't.


The original intention in Milton Keynes was that the buses would run only along the grid roads, to make them faster. The grid squares average 1 sq km, so the average walk from the original bus stops would only be 350m or so. It's only because buses have become, largely, the transport mode of the elderly, that the distance has become an issue. Commuters would rather walk 5 minutes to a bus stop with a 20 minute journey time, than have a service with a 40 minute journey time outside their door.

Quote:

The other is the government's insistence that buses have to make a profit. It's very difficult for them to do so in a lower-density city, especially if the places where people want to go - the office blocks, the supermarkets, the hospital, and so on - are scattered around the place rather than concentrated in a traditional centre.


That depends on how big each of the distributed centres is. I'm proposing that each one serves a population of 50,000 or so, which is enough for each centre to support a couple of supermarkets, a small hospital, a full range of schools etc. It should be plenty of people to support a high frequency (every 5 minutes or so) rail service on a grid system making inter-centre transport easy when needed.
[/quote]

 
suze
1321337.  Wed May 08, 2019 6:25 pm Reply with quote

cnb wrote:
LA has, and this surprised me, the highest average population density of any US urban area. It's still barely higher than Milton Keynes.


Sorry, but can I challenge that? If it's correct I'm not just surprised but absolutely staggered. If you had asked me to name the most densely populated US city, I would have had no hesitation in saying New York City.

If it really is LA, there's probably some peculiarity of municipal boundaries involved, and the limits of New York City stretch half way across the Atlantic or something.

cnb wrote:
Commuters would rather walk 5 minutes to a bus stop with a 20 minute journey time, than have a service with a 40 minute journey time outside their door.


I'm sure that's right, and you won't often hear anyone who lives a 5 minute walk from a railway station complaining about how far away it is. Even so, it really does appear to be considered "too far" to be expected to walk to a bus stop.

Whether that is because, as you suggest, bus users tend towards the elderly and the infirm, or whether there is some other psychology involved, I don't really know. Have buses become a Giffen good, and there is an element of "distress" attached to using them at all?

cnb wrote:
I'm proposing that each one serves a population of 50,000 or so, which is enough for each centre to support a couple of supermarkets, a small hospital, a full range of schools etc.


A couple of corner shops and a GP's office, perhaps. But I live in a town of 35,000 which doesn't have a proper supermarket, and Jeremy Hunt tried to close the hospital in a town as big as Huddersfield (170,000). (That plan was abandoned, but Mr Hunt spent two years supporting it before changing his mind.)

How confident are you that the present government wouldn't just build the houses and sod any proper infrastructure?

cnb wrote:
It should be plenty of people to support a high frequency (every 5 minutes or so) rail service on a grid system making inter-centre transport easy when needed.


Would rail be the best transport option here? Can I take advantage of the opportunity to mention them twice in one day, and suggest trolleybuses? If your city had 64 cells, you'd only need eight routes if they operated on a high enough frequency. The eight routes are the red and blue lines on the rather crude drawing below. Free transfers (common in North America, rare in Britain), of course.

 
barbados
1321345.  Thu May 09, 2019 1:11 am Reply with quote

suze wrote:
cnb wrote:
LA has, and this surprised me, the highest average population density of any US urban area. It's still barely higher than Milton Keynes.


Sorry, but can I challenge that? If it's correct I'm not just surprised but absolutely staggered. If you had asked me to name the most densely populated US city, I would have had no hesitation in saying New York City.

If it really is LA, there's probably some peculiarity of municipal boundaries involved, and the limits of New York City stretch half way across the Atlantic or something.

There is apparently a recent report from the census bureau that is suggesting that NYC is actually sitting at 5th currently, and I think the reason most people consider it to be the most densely populated is kind of touched upon by your suggestion of a peculiarity - when people see New York, they see Manhatten, and forget about Staten Island and the other boroughs.
More people per sqmi live in Manhatten than in Brooklyn and Bronx (the nerxt two in the list).

 
cnb
1321355.  Thu May 09, 2019 3:22 am Reply with quote

suze wrote:
cnb wrote:
LA has, and this surprised me, the highest average population density of any US urban area. It's still barely higher than Milton Keynes.


Sorry, but can I challenge that? If it's correct I'm not just surprised but absolutely staggered. If you had asked me to name the most densely populated US city, I would have had no hesitation in saying New York City.

If it really is LA, there's probably some peculiarity of municipal boundaries involved, and the limits of New York City stretch half way across the Atlantic or something.


The Census Bureau press release is here.

The map of the Los Angeles - Long Beach - Anaheim Urban Area is below (the heavy black outline) and it doesn't follow administrative boundaries.

I think the reason is that LA, or at least the parts I've been to - has denser suburbs than other US cities. Most cities slowly reduce in density until you reach single storey houses on two acre lots. LA is constrained by the mountains and ocean, so in general the lot size is far smaller even at the extremes of the sprawl. The comparable map for New York - Newark extends a very long way into very low density suburban New Jersey.

 
barbados
1321364.  Thu May 09, 2019 4:34 am Reply with quote

Thanks for adding the link cnb, every source I could find led to a 404 page on the census site, and while some of the sources for the link may well have been user fed data (Wikipedia and the like) There were many from more reliable sources and enough to suggest the target had been moved rather than “made up”

 
cnb
1321368.  Thu May 09, 2019 4:56 am Reply with quote

suze wrote:
cnb wrote:
I'm proposing that each one serves a population of 50,000 or so, which is enough for each centre to support a couple of supermarkets, a small hospital, a full range of schools etc.


A couple of corner shops and a GP's office, perhaps. But I live in a town of 35,000 which doesn't have a proper supermarket, and Jeremy Hunt tried to close the hospital in a town as big as Huddersfield (170,000). (That plan was abandoned, but Mr Hunt spent two years supporting it before changing his mind.)

How confident are you that the present government wouldn't just build the houses and sod any proper infrastructure?


Not meaning to be rude about it, but I'd argue that where you live isn't really a proper town in its own right. It's part of a larger urban area which has roughly the number of supermarkets any other conurbation of the same size has. I've ignored 'convenience store'-type supermarkets for this comparison.

The whole Medway urban area (population ~250,000) has (as far as I can tell from Google maps) 8 large supermarkets (over 2000sqm) and 10 smaller ones (and at least one planned).
In Reading, where I live, the whole urban area (rather than just Reading borough) has a similar population, and has 8 large supermarkets and 15 smaller ones (of which 3 have opened in the last year).
My experience of other European cities suggests that these numbers are not unique to the UK. As an example I've spent a few weeks in Braga, Portugal recently, and that city of 200,000 or so people has 5 large supermarkets and roughly 14 smaller ones, with a couple more planned.

On that basis I think it's reasonable to assume that a densely-populated area of 50,000 inhabitants could support at least one large supermarket and two or three smaller ones.

In my grid plan, everyone bar those in the outside rows of the grid would also be within fairly easy reach of 8 neighbouring modules, and therefore have a choice of probably a dozen major supermarkets within 3km.

50,000 people isn't enough to support a full-service general hospital - you'd probably want one of those for every 9-module block in the grid, but that would still put every resident within 4km. 50,000 is enough to cost-effectively provide a limited range of outpatient services within each module though.

The question about the current government isn't really relevant. This proposal would only work with a stable multi-decade commitment which couldn't happen under the British political system. It's irrelevant to the UK anyway, as we don't have the population growth to need a new multi-million person city.

 
dr.bob
1321373.  Thu May 09, 2019 5:25 am Reply with quote

GuyBarry wrote:
I quite enjoy living where I am at the moment, actually! I can walk to pretty much everywhere I want to go (though I do catch buses a couple of times a week). If I had to travel everywhere by Underground train my carbon footprint would probably go up.


Why would you have to travel everywhere by Underground train? I've already mentioned upthread that the best kind of developments would be a mixture of commercial, residential, and retail. cnb has mentioned his idea for a city with no defined centre where services are spread out more evenly. Surely you would still be able to walk pretty much everywhere you want to go.

GuyBarry wrote:
I mean situations where it's quicker to get to one's destination by changing trains than by waiting for a direct service.


Hang on a minute! This whole subthread only came about because you mentioned in post 1321097 that allowing trains to switch lines would be a big advantage because that would allow for "removing any need for passengers to change trains at all. Think of all those escalators on the Underground - how much energy would be saved if they didn't exist?"

Now you're saying that, given a situation where train can change lines and people can take a direct service, they'd still choose to take a changing service (using those escalators) if it was quicker.

I'm really not sure what point you're trying to make.

GuyBarry wrote:
Quote:
If people are being encouraged to wait on the platform rather than just take the first train, then the numbers of people on the platform would necessarily increase.


But that's the opposite of what I said. You can encourage people to take the first train whenever possible, but give them the choice of waiting for a through service if it's more convenient.


I was arguing for a service where all the trains go the same way, so there's no advantage to waiting for the next train (unless the current one is full). You're arguing for a system where the trains have a choice of destination.

The point I was trying to make is that, compared to the system I was proposing, your system would result in more people waiting on the platform rather than just taking the first train. In that sense, they are being "encouraged" to wait on the platform.

 
cnb
1321377.  Thu May 09, 2019 5:38 am Reply with quote

suze wrote:
cnb wrote:
It should be plenty of people to support a high frequency (every 5 minutes or so) rail service on a grid system making inter-centre transport easy when needed.


Would rail be the best transport option here? Can I take advantage of the opportunity to mention them twice in one day, and suggest trolleybuses? If your city had 64 cells, you'd only need eight routes if they operated on a high enough frequency. The eight routes are the red and blue lines on the rather crude drawing below. Free transfers (common in North America, rare in Britain), of course.



I think trolleybuses are probably a dead concept. Battery-powered buses today can run most urban routes all day on a single charge without any of the infrastructure costs or inflexibility of trolleybuses.

A small number of electric minibuses would probably be needed in my proposed city to accommodate those who can't get to the nearest station by themselves, but for the core grid transport network I don't see any better way than trains.

Trains are pretty easy to automate, especially on a new-build straight line route, and they are very safe when you have things like platform-edge doors. They are more reliable than road vehicles, easily made wheelchair accessible, and can be varied in length to suit demand without reducing frequency. But the major reason for using trains over buses is capacity.

Let's assume that 20% of the population go to work or school outside their own grid module. That's 10,000 people per peak period on 4 routes - a standard bus in each direction every 3 minutes. That's just about possible. However, as soon as some of those people travel more than one module away from home, they block space on the vehicle that would be needed by people getting on at the next stop. If we assume that the average journey is 6 modules long, we need 6 times the capacity, or a bendy bus every 45 seconds. That's not a level of service that small vehicles can reliably provide. A 5-car train every 5 minutes would achieve the same capacity.

I took free transfers between lines as essential, as almost every journey would require a change.

 
dr.bob
1321378.  Thu May 09, 2019 5:42 am Reply with quote

PDR wrote:
Indeed. It's just not a good reference case to the capital question.


I disagree. You make a fair point that a city needs to start servicing its capital investment as soon as possible. This is another strong argument in favour of the "modular" approach to city building that cnb outlined in post 1321274.

Indeed, in the early days of the building programme, there would probably be no need for an underground network at all. I imagine all transport needs could be met by a good network of buses or, if suze was in charge, trolleybuses, in the early stages.

Once the city has grown to sufficient size to require a more efficient mass transit system, only then would the capital need to be raised to implement this. Once again, cnb's suggestion of a grid system seems eminently sensible. It would do away with any need for some kind of underground Clapham Junction, thereby making the implementation process easier.

The question then becomes how best to allow for a later implementation of this mass transit system. If Guy was in charge, it would be easy since all the rails would be overground with junctions passing over and under each other using raised tracks or cuttings. If suze was implementing her trolleybuses, then it would be a simple expansion of the existing system (if that was practical).

If we wanted to make the whole thing underground, maybe it would be possible to build the grid of tunnels at the initial stages of construction (an increase in initial costs, but with potentially much fewer problems later on), and only build all the rails and platforms once the city is big enough to sustain such a large mass transit system.

 
barbados
1321379.  Thu May 09, 2019 5:45 am Reply with quote

Quote:
I was arguing for a service where all the trains go the same way, so there's no advantage to waiting for the next train (unless the current one is full). You're arguing for a system where the trains have a choice of destination.

Am I reading this wrong?
The trains all go in the same direction anyway so you are arguing against a system that currently exists, looking to replace it with the same system.

 

Page 7 of 9
Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8, 9  Next

All times are GMT - 5 Hours


Display posts from previous:   

Search Search Forums

Powered by phpBB © 2001, 2002 phpBB Group