The question of population density has come up in comment threads here in relation to various competing transit proposals.
As part of a planning course at Ryerson, Anthony J. Smith reviewed the SmartTrack and Downtown Relief Lines together with detailed data on population density, income and other measures. His paper Toronto Transit Choices: Evaluation of the Downtown Relief Line + SmartTrack Options including maps is available online at the Healthy City Maps website.
Steve Munro 
How do all the “downtown elites” manage to afford all those lattes they sip on such little income?
LikeLike
On a more serious note: this is an interesting paper, that applies some real demographic data to identify which of the two alternatives (DRL vs ST) should have the more immediate priority. One factor mentioned earlier in the paper, but absent from the analysis in section six is the ability of the two alternatives to move people. Even though the DRL will have fewer residents in its walkable catchment area, it will be able to move more than five times as many people. I realise the full capacity is projected out to 2031, so at least in the early phases, compared to ST may not have such a stark advantage. However the potential gains in ST over the decade after it is built will be limited by realistic speeds and headways that can be achieved with the type of equipment that is likely to be used on the line. There has been loads of discussion on this issue in the comments on other posts, so everyone knows what I mean.
I think if I had been the one writing the conclusion of the paper, I would have said that ST (with careful thought to route alignment) can be a quickly implemented solution to some of the current crowding on the system, but there should be no delay in getting construction underway on the DRL, because it will be the solution needed to growth beyond the capacity of ST over the next decade and a half.
LikeLike
It’s interesting, but this is really a high level analysis, not detailed … I’d like to see the same thing done with 800m from specific stop locations (I imagine that smart track would do even better, because there are more stops!)
However, it’s also clear that the DRL really serves a different purpose than the SmartTrack — in that it’s diverting people already on the system through a shorter/faster route to their destination (DRL), vs trying to be the only route that people use (ST). I think we will find when the analysis is done that ST fails to really divert many people from Bloor (given that the three stops south of Bloor on ST will likely only divert streetcar riders, and the connection with Main and stops further east are already low overall demand to the core (relative to Pape)).
The city will need to make the call as to whether the situation at Yonge/Bloor is as dire as everyone is saying, and Vaughan will need to make some noise about extending Yonge for the DRL to happen.
I think in a perfect world we do both, but without heavy Federal funding for a DRL (something which I would suggest Tory suggest as part of a Harper legacy, similar to the Confederation Bridge) they both won’t happen together.
Steve: The level of analysis is constrained by the size of census tracts which will not necessarily get you down to the fine level needed. Also, one has to factor in transportation survey information about the likely destinations of people living near stations, and of those who will access the network via connecting routes. It is not a simple modelling problem, and modelling efforts in past decades have been constrained by computing power.
Probably the biggest issue with transit modelling is that a lot of the “models” have been based on auto traffic and with fairly large cells in the network. This works for drivers as long as competing paths do not lie in the same cell, but for transit it is more difficult because there are factors at a fine-grained level that affect route choice. (And this presumes that factors such as land use and population demographics are valid within the model, not goosed for a future condition that is, shall we say, unlikely to occur.)
Finally, there has to be an understanding by the planners/modellers of the validity of what they study. If the organizational attitude is that the DRL is a waste of time (the prevailing wisdom for much of the period I have been active in transit politics), then networks will be crafted to downplay the DRL, and explorations of how it might work will be minimized. Similarly, it was quite common for TTC studies to ignore the possibility that a future GO network would be a vast improvement over the then-current version with the inevitable result that lots of subway construction was “justified” by future demand. That’s outright bias in the model, not to mention professional incompetence.
I remember asking in years past why more network configurations were not studied, and the response was “it costs too much”. Yes, network decisions to spend billions were constrained because it “cost too much” to model all of the possibilities.
LikeLike
One of the challenges we have to get past as a Toronto/GTA/GTHA is the idea that we do not have enough money to plan more than one section of one line at a time.
It is strange that while people are dreaming of networks and politicians are drawing lines, planners are focusing on small sections of lines.
The end result is too much focus on the type of mass transit technology which ends up blurring with public confusion about the functions of different type of mass transit. We end up with subways that should be LRT Lines, subway extensions built to move commuters downtown (instead of providing better, cheaper GO service), proposed LRT lines that might run better as buses etc etc.
If Toronto/GTA/GTHA is going to look beyond this very narrow thinking and see transit networks offering level of service rather than individual lines offering a level of technology, we will have to start by paying the money needed to fund real studies rather than dropping money in the hands of political supporters and asking them to apply it where they see fit.
Cheers, Moaz
LikeLike
I still believe it would be worthwhile for the province to request postal code (side and portion of street) and hours of work data for all employees from each employer across the region. This highly resolved information would allow a detailed travel time and likely ridership for alternate routes, including various stop locations and attached bus routes. The models could then be refined, and should be a low cost enough exercise to permit comparing mixing and matching many routes in potential networks, including auto usage. Such data would also help us understand the impact of improved transit priority on overall commutes and how changing and tweaking run times and sharing that info with the public should affect ridership on and between routes.
LikeLike
A more realistic comparison would have assumed that just the eastern alignment of the DRL was to be built, (at least at one time). This would have a major impact cost & timing among other factors.
The maps are well done. For transit purposes, a map with jobs, or ideally population & jobs is optimum given the disproportionate impact of jobs, particularly in dense walkable areas, on ridership.
While not part of this study, it is important to keep in mind frequency & ridership limitations. Smart Track uses existing infrastructure & stations that are in danger of being maxed out by the time this is realistically built, unless the Province reneges. ST adds substantial new capacity (tracks & stations) at the outer end, for example from Mt. Dennis to ACC. By contrast, the DRL adds substantial capacity right into the financial core.
Looking at demographics & walkability along the corridors, rather than at likely stations, risks producing major distortions. GO, UPX & (planned) Smart Track, express past many of the high density areas where adding a station would be difficult. These express lines generate much of their ridership from outer stations.
Unlike subways, which travel underground and can place their stations in the middle of dense walkable areas, rail corridors tend to be more isolated and less walkable, with conditions generally deteriorating as one heads out. Stations are often centred on huge parking lots, and frequently accessible on foot from only one side of the tracks, eg. Unionville, (south of) Steeles.
While predicted capital costs are included, operating costs are also key. What operating subsidies would be needed to run ST at TTC fares, GO fares? What are the ridership implications of each? The report’s original author says his version of ST would be quickly profitable, but as these studies are private, and for a different line, comparisons are left to rely on assumptions & figures (eg. 7 years) that might prove false.
LikeLike
I agree with Roger. From a planning point of view, it is far easier to predict success in ridership if I know that transit takes me to where I want to go….
Considering the how the TTC is based on a surface feeder route, upgrading RT corridors based on connecting transit to destinations (jobs, post-secondary schools, airports, malls, tourist attractions) vs. origins (residential), would probably do us quite well.
LikeLike
I agree, I think that a western side of DRL is not required until condition similar to what exist today on Yonge start to appear on Spadina. A line in UPX (which would make sense), would fill this function anyway, and more critically now provide real linkage for the northwest of the city, the CrossTown LRT, Finch West LRT and beyond into Brampton and for the Hurontario-Maine LRT.
Yes, unless Smarttrack manages to get TC exclusion and is placed in a tunnel across the core, it will suffer the spacing from the limits that RER would. Union Station would be pressed to handle an increase to 6 trains each side in this corridor, let alone a truly game changing 8-10. If capacity in Stouffville is capped at say 11,500, and the same is true from Kitchener side (as opposed to adding this to UPX) it will not represent a truly game changing transit solution. This means as you say, changes the connection to the core for the outer end, but will not leave capacity (or run frequently enough to attract) for ridership on the inner end. If the eastern leg of the DRL had been timed to meet the Crosstown, shortly after it opened these issues would be seen in a completely different light. The corridors would be seen as offering capacity to the outer ends, and service would be designed accordingly. Regionally in the long term I believe this to be a preferable outcome.
LikeLike
I suppose there are still those who pine away for a full subway under Eglinton. I, for one, fully understand why it’s not justified, at least for now. I believe that the one thing that’s always fuelled any sentiment for a full subway there is the fact that the line has to be tunneled for much of it’s length anyhow. I’ve always tried to be as pragmatic as humanly possible about it and hope I don’t come across as too terribly biased on this issue. I’m 100 per cent for what’s being built there now but at the same time feel that looking ahead to future possibilities is absolutely the right thing to do.
Steve: The key words are “much of the length”. As originally proposed, only 1/3 would be underground with the sections between the airport and Weston Road, and most of the section from Brentcliffe to Kingston Road (and then north ward) on the surface. The amount we lost to political interference and shortsighted cutbacks does not reflect well on the evolution of Toronto’s transit “planning”.
LikeLike
Interesting how Anthony uses projections from the TTC which uses projections from the Provinces Places to Grow. Doesn’t anyone ever check to see if these projections are realistic? The Places to Grow forecast are already so far of on employment projections as to be laughable. The 2011 employment projections are off by nearly 200,000. Yet there is a continued reliance on them.
LikeLike
DRL I think it would be a mistake to build through low density areas along Pape when higher density areas are available further east along Main or Victoria Park and another benefit of using the Main alignment is the interchange with GO Trains. Coxwell is also a possibility which would serve the hospital. Having said that certain personal interests are heavily vested in advocating a subway through low density Pape and so if a subway were to be built through low density Pape, then we can save up to 3 billion dollars by building it via cut and cover. The density is simply not there around Pape to warrant a bored tunnel and perhaps the high density Downtown portion of the DRL may be bored tunnels.
Steve: I have said this before, but it bears repeating. A Pape or Donlands alignment sets up for a connection to Thorncliffe Park, something that routes further east would miss. There is a huge amount of land there ripe for intensification. Main and Vic Park are too far east to get to Don Mills and Eglinton.
LikeLike
Whose personal interests? Does Steve have an undisclosed interest in a billion or two dollars worth of real estate around and between Pape and Donlands stations?
Not every differing opinion is the result of corruption. I’ll even grant that many Scarborough subway boosters’ opinions reflect their sincere belief that the subway is better for Scarborough, not that they must live near a proposed station location and will benefit personally. I still think they are wrong, for all the reasons discussed previously, but there is no need to invoke ulterior motives.
Steve: There are many comments (mostly deleted) accusing me of having my hand very deeply in the cookie jar, and they are complete crap. I actually prefer the Donlands alignment because it makes a connection to Greenwood Yard much simpler (from the south, not from the north) and avoids a Sheppard-Yonge style construction project at Pape and Danforth.
The other major chunk of land that is on “Pape” is Gerrard Square, and it is no secret that the owners would love to redevelop this. However, this site will be on any alignment that follows the rail corridor, including SmartTrack, although fitting a surface rail station there might be a challenge.
LikeLike
After reading what you say about the Donlands alignment, I’ve become 100 per cent for that one. As far as Gerrard Square is concern, I would hope very much that the challenge of putting a station there would be too horrifically insurmountable challenge.
LikeLike
The only problem with Donlands as a high traffic transfer station is the current location of the station box. The last thing we need is a design that repeats the errors of Bloor-Yonge.
At any rate people can keep repeating the myth that the Pape/Donlands/Thorncliffe corridor is low density but they’re just ignorant.
LikeLike
There is also the question of course of the possibility of this being extended further in the future beyond even what was in the initial plan. I would hope that it will end up being even longer than the version that was shortened by the province. It should also be part of a network that hopefully one day will encompass at least the equivalent of all the original LRT lines. If LRT were to be on Finch instead of Sheppard (or as well) for instance, another tunnel to get through the narrow area close to Yonge might make sense. I have no problem imagining a much higher density along considerable portions of the major roads in Toronto, especially if there is rapid transit near. Further if Toronto is smart about managing increases in capacity so it appears where it is really required, the city could add a million residents and reduce problem congestion.
Steve’s point above about where a DRL needs to run also relates to what can connect there from the east and the north. Also the logical connections for whatever it is that runs from there. I personally like the idea of a Don Mills LRT being a direct connection to a Sheppard one where it meets the Sheppard subway, so I could opt to transfer either way if I was core bound.
Steve how does Victoria Park compare to Don Mills as a possible candidate for a future LRT? What is along Victoria Park that makes a logical connection? Is not the existing service along Vic Park better than that offered for Flemingdon & Thorncliffe Parks also is the area along the Don Mills route not also present a high level of density? Also look at the area around Cosburn, while the entire area may not be that dense, there are 40 or so apartment buildings within 400 or so meters of that intersection. I will grant you there is smaller number of larger buildings at Main and Danforth (although I suspect the resident count is likely lower), but there is already at station there, whereas there is not at Pape and Cosburn. However the purpose of this line is not local service but connectivity for the network.
Steve: AM peak service on 25 Don Mills is 3’10”. This is supplemented by 100 Flemingdon Park at 3’15” and 81 Thorncliffe Park at 4’40” for a combined headway west of Thorncliffe Park of about 72 seconds. The 24 Victoria Park runs every 3’32”. 87 Cosburn runs every 4’20”.
I think Don Mills is the stronger candidate for an LRT line because it is an easy extension of the corridor served by a DRL to Eglinton. A Victoria Park line would simply be another feeder to the Danforth subway. Main and Danforth has nothing to do with either corridor.
LikeLike
The other thing about the Don Mills/Pape alignment, other than actual density (please see maps for highest density area) is the income levels appear to be lower along much of this length, which would also explain why, despite the fact the bus ride to subway from this area is painful bus in this area is so well used.
Flemingdon Park 15,800
Don Mills 38,000
Thorncliffe 7,100
Cosburn 8,200
69,100.00
Vic Park has a ridership of about 23,200.
Also substantial development is possible right at Eglinton and Don Mills, which kind of makes it a natural.
LikeLike
Just in case anyone thinks actual data is important in transit planning, Royson James makes a (depressing) case for the Scarborough RT.
Nostalgia, maybe? The line is around 30 years old and that’s approximately the time when fashions return are revived.
Cheers, Moaz
Steve: Royson cites, selectively, TTC reports about the RT technology. Notably, he ignores (as did the TTC) the fact that keeping the RT with new cars was only cost effective in an environment without a surrounding LRT network and, more importantly, without extending the line to Malvern.
He also talks about how the RT cars outlived the H6 subway cars which were their contemporaries to buttress the claim that the RT equipment is great stuff. In fact, the H6 cars were the lemons of the “H” series fleet and had many problems after their delivery including a need for a complete replacement of the original trucks and numerous issues with their control systems. The TTC retired them early because (a) they had low reliability and (b) there was an incentive to keep the TR production line running at Thunder Bay.
As for the Neptis/Schabas report, I have already spent four long articles eviscerating it and will not comment further. The Star thinks it is the greatest thing ever. They are wrong.
LikeLike
@Steve & Moaz, correct me if this is a misperception but does not the goofy small tunnel size on the RT make adopting the Vancouver type cars impossible and thus forces massive reconstruction of one kind or another? To make this work as the best tech would it not require an extensive network where it was the right capacity and where the in tunnel or above ground tech could reasonably be applied?
So if Toronto had a large un-served area with density like much of downtown or where we could collect enough traffic and could dedicate a row it might make sense, but running in median anywhere or as a short stand alone line does ICTS make sense? If we could use it in say Stouffville and UPX it might, but is this possible?
Steve: Yes, the tunnel at Ellesmere is too small for the Mark II (Vancouver) cars and would have to be rebuilt. Also, I believe that the curves at Kennedy Station are also too tight. Rebuilding the line for Mark IIs would not involve as much work as for LRT, but it’s not just a quick changeover.
Bombardier has been salivating at the thought of “extending” the SRT to the airport for years, going back into the days of Transit City, but considering the mess that company is in, now is hardly the time to jump to yet another technology choice. Also, there would be problems with an ICTS line and railway operations co-habiting in the Stouffville corridor just as there would be with LRT. It would have to be elevated because some parts of the right-of-way are quite tight.
Can we please stop re-re-re-re-designing the system to make yet more work for the engineers and get on with actually building something?
LikeLike
Saddest part of the story is that the media are stating that transit planning must be transparent and evidence based … Then someone who makes a case for that goes ahead and manipulates the way the evidence is portrayed to the public to suit their own perceptions.
Here’s another piece of the planning story … taking you back to Hamilton, 1981.
Proof that the more things change the more they stay the same.
Cheers, Moaz
LikeLike
I am in complete agreement, as long as what we are going to build makes sense. There was a plan for Scarborough that was doable on the current budget for subway that would have permitted extensive coverage and was using the same technology as a plan for much of the city.
The place in my mind to look at changing plan is where it will not provide enough service, and that is usually where plans have not been completely fleshed out. The answer in my mind remains complete the planning for Transit City and build it at greatest possible speed. Increase service on all GO corridors and implement RER on the corridors where it will have the largest impact (where large increases in service are required and not possible without it). Design where there are still clear holes (capacity on and parallel to Yonge). Come up with some ideas to back fill where there is an unmet need that arises due to additional issues, but keep them in reserve.
Toronto’s current problems stem from building too heavy, and waiting too long to move. I have to admit Steve, however, that one of the reasons that transit has caught my attention, is it occupying the same political space that health care – emergency rooms did 15-20 years ago. It is being driven by politics and a misperception of where the real issues are and what the real solutions are on the part of the voters. Metrolinx could probably do a fine job if it had the money and was not the most popular political football in the province.
LikeLike
Steve, I couldn’t honestly agree with you more about stopping all this constant redesigning the system over and over.
LikeLike
OK, to me those maps point towards a Dundas St. Subway. Look at it with an open mind. Assume RER-like service on the GO tracks, and LRT-style upgrades to all the streetcars.
Steve: There is a limit to what RER can do for local travel, and the ability to upgrade the streetcar lines to LRT is constrained by the street geometry and land use.
Dundas would not make a good subway location because it does not serve the heart of either the commercial or academic communities downtown.
LikeLike