Royson James has a pair of columns in the Toronto Star discussing the perennial LRT vs subway transit debates.
City needs a transit lesson (Nov. 17, 2010)
Commuters won’t fill LRTs, much less subways (Nov. 19, 2010)
James sets out the pros and cons without becoming mired in either side’s arguments. As with any overview, there are points for or against either technology that are not made with the vigour that advocates would prefer. The important issue, however, is not to choose one technology to the exclusion of the other, but to look at the appropriate one for each implementation.
One critical issue — regardless of which side one might be on — is the matter of land use and how the evolution of Toronto will affect demand on routes and the overall network. There are two fundamentally different views of of future development — the Official Plan’s “Avenues” with major streets lined by mid-rise buildings and shops giving an active pedestrian environment at ground level, or the more traditional “tower in a park” design that has shaped much of Toronto’s growth since the 1960s. A third variant has appeared over the past decade — both tall and dense, as exemplified by the railway lands, parts of Liberty Village and most recently the Queen West Triangle (Queen & Dovercourt).
Each of these produces transit demands which vary both due to the built form and to the neighbourhood in which development occurs. A building located in an existing walkable neighbourhood with shops and transit will have very different transportation demands than the same building located on a suburban arterial where the nearest shop is the mall a short drive or a lonely, windy walk away.
The perennial myth about subways is that their high capacity will be consumed by redevelopment around stations. This is utter hogwash. The Yonge line is full well north of Eglinton not with Willowdale condo dwellers, but with traffic fed in on surface routes. Developments along the line add to the demand, but the subway exists to serve a much wider catchment area. Similarly, the BD subway depends on feeder services to many stations, and the decades-long absence of nearby development did not prevent the buildup of demand eastbound from Etobicoke or westbound from Scarborough.
LRT lies somewhere in between by serving both busy “local” corridors and, in some cases, acting almost like a subway in speed, if not capacity. We must remember that the SRT would have been an LRT line (and to Malvern too, decades ago) but for Queen’s Park’s intervention with the ICTS technology. Regardless of technology, it is a medium capacity line whose principal function is to feed the BD subway at Kennedy and, much more recently, to serve the high-density residential development at Scarborough Town Centre, developments that did not occur until decades after the SRT opened.
In many ways, LRT has always been a misunderstood, orphan technology in Toronto. Some within the TTC have never accepted the retention of streetcars, much less the creation of an LRT alternative to full-blown subway construction. At a time when LRT was coming back into favour around the world, Toronto pursued ICTS and lost the chance to show what real LRT could do. At more than double the cost of the LRT proposal, ICTS “proved” that there was no cheap way to implement transit lines, and system expansion stalled. The TTC did nothing to advance the LRT alternative.
Spadina, Harbourfront and St. Clair are really not LRT, but rather upgraded streetcar lines. That statement brings me to a common question: what’s the difference between streetcars, LRT and “Heavy Rapid Transit” (or HRT)? Everyone knows what subways, streetcars and buses are, but things get mushy in the space between them.
The boundary between HRT and LRT is fairly straightforward: if the technology cannot run at grade in medians or crossing streets and walkways, then it’s HRT regardless of what vehicle actually operates on the structure. There can be “light” railways such as the SRT, or full-blown subways, but in either case the lines are confined to an exclusive right-of-way. This imposes costs and complexities wherever they are built.
The boundary between LRT and streetcar is not as clear-cut. How exclusive is the right-of-way? How much mixed-traffic operation does a route have? How aggressive is the traffic signal priority? Do passengers board through all doors? How far apart are the stations? How long are the vehicles or trains? All of these issues and more produce a range of answers, and there is no magic point at which a light blinks on “LRT”. That’s the strength of the technology — LRT does not have to be the same thing all the time on every metre of a route or a network. The challenge is to strike a balance between the “light” and “rapid” parts of the name — exclusivity and speed versus the footprint a line can have in a street and neighbourhood.
The term “LRT” has been oversold in Toronto. We have never seen something in the style of other Canadian LRT implementations in Edmonton or Calgary. We lost that chance when the Scarborough LRT became the “RT”. It’s still dubious whether we will see that route incorporated into an LRT network, or swallowed by a subway extension.
Toronto’s “LRT” routes run through downtown areas with frequent cross-streets where traffic signals grudgingly give priority to transit (but just as often serve to delay it). They have slow on-board fare collection with high-floor cars and low-floor platforms. They suffer a planning context where transit must fight to be acknowledged.
There is only so much road space and money to go around. Subways make for flashy announcements and lots of work for the construction industry, but endless waits by riders whose trips are not served by the most recent subway extension. LRT lines (and busways while we’re on the subject) take space that would otherwise be used by motorists. On some arterials, this space is available, but on many it is not (even VIVA’s BRT network is constrained in places by a narrow right-of-way).
LRT advocates have an uphill battle because Toronto’s version of this technology pleases few. St. Clair was a disaster for “LRT” (and for transit in general) — there were too many design tradeoffs and construction was appallingly mismanaged. Operations have improved over “the old days”, but still depend on keen route supervisors who actually manage the service rather than letting cars roam back and forth in packs taking generous layovers at terminals. We may be rid of traffic congestion, but not the infamous TTC culture.
The political climate may shift back to one where we make announcements to appear to be “doing something”, even if that won’t bear fruit for a decade or more. Such plans will serve only small parts of the GTA when finished (if ever), we will have yet another “lost generation” of transit investment. Decisions about how to build, where to build, what to build are difficult and need more than an endless supply of magic markers, maps and press kits.
We have seen how a proposed LRT network suffered from funding cutbacks. Major new revenue streams (tolls, regional taxes) cannot be implemented in the current political climate without a huge fight and an expenditure of political capital nobody seems willing to make today.
“The Big Move” could turn out to be little more than a modest expansion of GO Transit, busways, and a few rail lines of indeterminate technology within Toronto. That’s not a network, and certainly not a recipe for convincing people that transit can offer an alternative to driving. The challenge is to find a plan, a network, a quality of transit service that people are willing to pay for, however the money is raised.
LRT has a role as do full-blown subways and busways with each fitting into the mix under the right circumstances. Advocates would do well to focus on the strength of each technology rather than trying to justify a full network of one option. The goal is to improve and expand transit, not to prove that my subway is better than your streetcar.
Footnote:
Within James’ second article, the TTC is quoted as saying that ridership on the King car is 1,800 per hour. It’s worth noting that the AM peak service is 30 cars/hour of which 7 trips are served by ALRVs. The TTC’s service design capacity is 74 for CLRVs and 108 for ALRVs, and this gives a total for the route of about 2,450. Crush capacity is higher. A common complaint from riders is that they cannot get on, and this suggests that the demand cited by the TTC is rather lower than the actual level.
Steve Munro 
I’ve been going through Friday’s (November 19) Star trying to find Royson’s column, thinking I’d missed it.
It’s actually in the Saturday (November 20) Star. (Yeah, I could read it on-line … but if I’ve got the paper sitting in front of me anyways …)
Steve: The dates refer to the posting date online.
LikeLike
And I would argue that Transit City, as it is currently designed, is not LRT. It’s an upgraded streetcar plan. The only true LRT line in Toronto will be the converted SRT (if that ever happens).
LRT proponents are idealists and purists — and there’s nothing wrong with that. But, we’re not talking about LRT, we’re talking about the TTC’s idea of LRT. Those are two different things.
LikeLike
I heard that Eglinton was supposed to get a BRT line, and then would be converted to subway later on, if the ridership justified. Did politicians or TTC truly believe that a subway along Eglinton would eventually be justified, or was the subway promise just political appeasement?
And I’ve noticed that the Network 2011 plan called for exclusively subway expansion (correct me if I’m wrong about this). Was it politics that no other technology (like LRT) was considered, or was there another reason?
Also, what does “subways are justified” really mean? If the subway on Yonge Street is justified, does that mean it requires no additional subsidy, and the operating costs is recovered by ridership?
Steve: Amazing, isn’t it, how an all-subway plan like Network 2011 gets away with ignoring LRT while Transit City is hammered for ignoring subways. If all one looks at is subways, then there are places where even the subway jocks cannot justify their infrastructure, and that’s why the west end of Eglinton was going to be BRT (not to mention the property available from the unbuilt Richview Expressway).
Subways are “justified” primarily by demand, real or concocted. It’s so funny that we often hear about development along a line “justifying” its construction at the same time as the zone of influence, for ridership estimates, is often taken well beyond the immediate walking distance of subway stations. Sheppard’s projected demand was inflated through this technique, and by leaving out any competing services such as GO to bleed off regional demand.
The “investment” in a subway is not just the construction, but also the cost of infrastructure operation and maintenance, and the much more frequent service (capacity that would never be provided if it were a surface route).
Of course there are a few examples where we built (or are building) subways to make some developer’s land rather more valuable than it would be if served only by a bus every ten minutes.
LikeLike
My own thinking is that since there’s no well-known example of light rail close at hand – sure, Buffalo has a light rail line, but how many people know that? – the debate tends to be framed as “streetcars, but awesomer,” and whatever preconceived notions that exist about streetcars get welded on whether or not they’re appropriate. I’ve only had direct experience with light rail in Los Angeles; there, there’s no valid streetcar comparison. The way I see it, streetcars are as close to buses as light rail.
Steve: But even “streetcars” can carry more people than buses if the route requires it.
LikeLike
And I would argue that Transit City, as it is currently designed, is LRT. It’s not just an upgraded streetcar plan.
Seems rather ignorant to claim that 60-metre long LRT trains running in grade-separated subway tunnel under Eglinton isn’t LRT.
Steve: As I said in my post, LRT isn’t one specific implementation. An SRT type right-of-way with complete traffic segregation is about as “heavy” as it gets without actually using subway cars. The only thing “light” about it is the length of the trains, and therefore of the stations. If the SRT is the only true LRT, then neither Edmonton nor Calgary’s systems qualify because both have points where they cross or run in streets.
The TTC doesn’t help the debate by using “LRT” to talk about streetcar lines like St. Clair, while at the same time insisting that if a Scarborough “LRT” train ever ventured off a protected right-of-way, the quality of service would collapse. I won’t say anything about what it does on a regular basis already, including situations where crew changes don’t work out quite as planned.
LikeLike
Being a newcomer to transit discussions, I had difficulty getting my head around esoteric terms such as LRT, SRT, HRT, none of which make much logical sense since nobody seems to agree on a definition. Surely it’s possible to discuss transportation options without the associated bafflegab. The Toronto public wants fast comfortable and clean public transportation. They don’t care what form it takes or what it’s called.
What is clear also is that Toronto is in the public transit dark ages when compared to every mid-sized city in Europe. While we debate the number of angels on the head of a pin, the Europeans have gone ahead and built magnificent transportation infrastructure after receiving and implementing advice from transportation professionals – not politicians. In fact politicians are largely responsible for the abysmal quality of the GTA’s transportation network.
Steve: I agree that the terminology wars have not helped, but when we are trying to advocate for something that does not exist here, we have to give it a name. This inevitably leads to explaining why “LRT” (or whatever) isn’t the same as streetcars, or the SRT, or whatever.
As for politicians and “professionals”, the “pros” are just as responsible for the mess we are in thanks to their subway fetish and anti-streetcar bias which for years prevented the TTC (and others) from looking at any alternatives.
LikeLike
nfitz said …
“Seems rather ignorant to claim that 60-metre long LRT trains running in grade-separated subway tunnel under Eglinton isn’t LRT.”
That is NOT LRT. That is a SUBWAY.
Steve: But when it emerges to run down the middle of Eglinton, it’s an LRT. The essential difference is that a train of TR’s is never going to do that.
LikeLike
Before adding to this debate, I’d be really curious to know something….that I think Steve might know, or certainly could find out…
What was the per hour ridership of the Yonge Subway when launched, Bloor, University, and Spadina? (I could ask Sheppard, but its early days yet)
I’m curious (without having a pre-defined position) whether the thresholds suggested today would (for HRT) have caused the non-construction of current lines. and whether those same lines meet or exceed such standards now?
How long is it reasonable to wait for a line to meet ridership minimums/likely long-term usage? Surely any new line should be over-built for its day one use, you don’t want to be expanding on day 2, or year 2. But….at the same time, you don’t want to build something that is cavernously overbuilt, and will be under-utilized for 30 years plus.
***
I’d also be really curious to compare per hour ridership figures for a several transit systems (HRT/LRT lines) around the world. Let’s look at Paris, London, but also L.A., Chicago and N.Y.
As Steve well knows, as do regular readers here, I fall into the middle on the subway/lrt question. Mostly feeling there should be more subway than is currently proposed, but more LRT than most subway proponents would like in the balance.
But I’d really like more solid comparisons as I weigh the options/choices.
Steve: I don’t have the original ridership figures, but what I can talk about is the surface system that existed immediately before each subway line opened.
The Yonge corridor operated extremely frequent streetcar service (I think most people have seen photos of two-car Peter Witt trains nose to tail), and I have seen figures quoted of over 10,000 passengers per hour for this line. However, there was also frequent service on Bay and on Church (not to mention other north-south streets) that fed passengers into downtown by a variety of routes. The day the Yonge line opened, there was a pre-existing demand well above 15,000.
The University subway did not fare as well because it really didn’t duplicate a heavy existing travel corridor. Opening this line was used by the TTC as an excuse to take streetcars off of Bay Street, but even they acknowledeged that this didn’t work very well. The real change over time on Bay came from the shift of all traffic into the core onto the subway leaving the Bay bus mainly serving the government offices between Bloor and College. For many years, the University subway closed after 9:30 pm.
On Bloor-Danforth, the situation was a bit different from Yonge because, in a way, this can be thought of as two routes — Bloor to the west and Danforth to the east — with heavy transfer traffic to the Yonge subway (which by 1966 had consolidated loads previously handled by those north-south routes I mentioned above). There was a one minute headway of two-car PCC trains, or 120 cars/hour/direction. Using the TTC’s current service standard, that’s a design capacity just under 9,000 per hour. Like the Yonge line, the Bloor car also had some parallel services notably on Dupont and Harbord. Service on all of the east-west streetcar lines was much, much better than it is today. Bus routes fed the streetcar lines at many points, and there were routes into downtown that did not involve taking the Bloor streetcar. The original subway line ran from Keele to Woodbine, with extensions into the fast-growing suburbs soon to follow.
The Yonge extensions north to York Mills and, a few years later, to Finch, continued expansion in an existing heavy transit corridor (extremely frequent local and express service on Yonge north of Eglinton vanished leaving only the local Yonge bus running now and then).
The Spadina alignment was, of course, motivated by a desire to sanitize the expressway proposal, even though arguments were made for routes on Bathurst or Dufferin. Initial riding was under 10,000 per hour and the TTC received a special subsidy for several years. Riding has grown over time mainly due to surface feeders, and the fact that half of the service ends at St. Clair West in the AM peak tells you the real story for the suburban part of the route. The TTC has planned to extend the short turn to the pocket track at Glencairn, but this has yet to be implemented. Once the line reaches York U, the short turn will probably move further north and I expect to see political pressure to take full service at least to York U. Who will pay for this is anyone’s guess.
There is no hard rule for the point where HRT becomes the only option but the choice is driven by two basic factors. There is a limit to the frequency of service one can operate on a surface route depending on the alignment (middle of the street, frequency of grade crossings, available room for station platforms, length of trains) and on the circulation of pedestrian traffic at stations. For example, if the Bloor streetcar had not had a closed transfer platform to the Yonge subway, the transfer traffic walking between these lines would have dwarfed anything we see on streets today. Also, one must consider reasonable growth projections for a corridor and whether this would be handled with more surface service, additional capacity elsewhere in the network, or if the corridor really has potential for subway demand in the medium to long term.
Looking at the Transit City routes, none of them has the existing density of bus service to warrant subway construction, and that’s why they are proposed as LRT (among other reasons). The one exception is the south end of the Don Mills line below Eglinton, but other factors come into play here. Fitting a surface LRT through East York will be extremely difficult, although the TTC never tires of trying to find a way to do this. In the process they damage the credibility of the LRT proposal. This part of the route will almost certainly require its own new infrastructure including a crossing of the Don Valley and a tunnel from there south to Danforth.
Coupling this with a Downtown Relief Line would make sense operationally (it would eliminate transfer traffic between the Don Mills LRT and DRL). The question remains of whether this combined route should be built as a high platform subway or as part of the LRT network. Demand projections for a Queen subway to Don Mills on roughly this alignment made in the 60s suggested that the demand south of Danforth could strain an LRT subway, but this debate is moot if the DRL never emerges to run as a surface LRT anywhere.
I will leave international operations for now because this requires both research and context. For example, many systems’ trains are shorter and run with smaller cars than Toronto’s, and simply quoting headways does not directly compare capacities. Some subways have multiple routes sharing the same infrastructure, while others (notably New York) have parallel subways serving the same corridor. Lines with full automation and stations designed to handle very large passenger flows cannot be directly compared to older systems. Some cities have extensive networks built at a time when subway construction was relatively cheap and easy, and if one were starting from scratch today, a very different network might be the result. And, finally, many cities have extensive commuter rail networks that supplement the “subway” system and avoid loading regional travellers onto the local network.
LikeLike
Chicago’s Brown Line has a few at grade crossings, yet is considered “heavy rapid transit”. Edmonton’s rail line too is grade separated or runs in its own ROW with at grade crossings, yet is considered “light rapid transit”. Both move a comparable amount of people, at about 100,000 and 75,000 respectively.
If you look at Stockholm’s rolling stock on Bombardier’s website, both their light and heavy trains are the same width. The only real difference (power supply aside) is that their metro trains have an extra articulation.
I think as we progress into the future, the differences between heavy and light rail will continue to be blurred. Eventually it will simply be “rail rapid transit” for any system which has spaced out stops and operates separate from traffic (grade separated or right of way) – with possibly a small segment of mixed traffic operation. And “rail local transit” which have closer stops and operate in mixed traffic or a right of way with limited to no signal priority.
LikeLike
The TTC doesn’t help the debate by grossly mismanaging the current streetcars.
It’s very amusing to constantly watch in Nextbus seeing two streetcars leave the terminus close together, while everyone can see there is a 20-minute gap until the next one. Common sense would dictate that the second car wait 10 minutes …
It’s stuff like this that make people think we need subways instead of streetcars. We don’t let subway trains leave the terminus in the same haphazard way that streetcars come and go.
Steve: The TTC’s inability or unwillingness to manage its surface routes will become even more evident once the bus system is visible in 2011. The ongoing BS about how only streetcar routes are screwed up will be revealed for the sad excuse that it is. Even routes with entirely private right-of-way like St. Clair and Spadina are regularly screwed up thanks to poor or completely absent supervision. Operators get away with taking long layovers and running in packs, and the riders suffer.
LikeLike
Actually, LRT was always meant to be in between streetcars and Heavy, the big advantage of LRT was/is, you don’t have to take property to build it, sure you can have a gold handled thing that, like the SRT is HRT in all but name but reality is, LRT was a name the was used to bring back streetcars in the US in the mid 70’s without conjuring up pictures of old Brill semi-convertibles grinding up and down streets. In Melbourne, the tramways took over two Heavy electric lines that were not carrying the patronage to justify heavy rail retention, the HRT ran into the main station (Flinders st) which is on the southern edge of downtown, when the lines were converted to LRT, they were tied into the rest of the system close to the city and run through the city using lines that formerly terminated in the city. One of the LRT lines (the 96 St. Kilda) was also extended at the outer end to join another streetcar line into the heart of that suburb. These lines were an instant success, the 96 has become the heaviest line on the system of some 33 routes and rates 5 section Combino low floor cars. The big advantage over the HRT was the way it now goes through the heart of town rather than the edge. The other shorter line (the 109), like the 96 has been a great success.
In Europe heavy subways have become a bit passe and most subways now are built for sections of streetcar/LRT operation. Indeed, subway construction itself has dropped off since the halcyon days of the 60’s and 70’s as it has been found that, rather than removing transit from the street, it created a need for a bus service along that street for short distance riders who did not want to go down into the subways to go a couple of blocks, a perfect example of the this is the massive success of the San Francisco F line, designed as a tourist line, it has become a great success for commuters who choose the streetcar for short trips down Market st instead of going underground where they have a choice of the Muni Metro or BART. In Europe, it was long ago decided that public transit and people owned the streets, not the tin god motor car.
A complete change in thinking is required to get the selfishness out of people, why should we pamper some moron in his Lexus who gets pissed off at having to wait for 30 seconds or a minute whilst a streetcar loads and then holds up the streetcar at the next intersection doing a left turn? The whole culture needs to take a good long look at itself and to coin a phrase I loved in Canada, it needs to “suck it up buttercup”!
In this day and age, cities just cannot afford full blown subways, LRT can and will do the job if it’s properly implemented, just like streetcars are far superior to buses especially when they get serious about priority.
Greg
LikeLike
When can politicians stop bickering about what kind of transit to build and start building something? BRT, LRT, and subway are all fairly reasonable choices for many of the corridors we are discussing (e.g. Eglinton and Sheppard), the choice mostly depends on budget and ridership projections (which can vary by orders of magnitude depending on projections of development along the line/population growth in Toronto/the state of the economy/oil prices etc.)
LikeLike
” … St. Clair and Spadina are regularly screwed up thanks to poor or completely absent supervision.”
What happens when the Sheppard LRT is running. I have visions of three LRT’s going down Sheppard bumper to bumper.
In May I was in Istanbul. They have a Tram line (T1 Line), or it’s labelled that way, but it looks like a very large LRT (Bombardier). Most of the line is in a dedicated right of way but in the Historic district (Sultanhamet) the line shares the road with taxis and buses. The so called tram winds up through narrow streets, with hills and curves. It’s a big machine operating in a confined space and the operators are not driving in first gear. Sidewalks in some areas are only 2’6″ wide, with the rails 2’0″ from the curb. Sometimes you have to step off the sidewalk onto and walk over the rail to get by people. You tend to be looking over your shoulder to see if the next tram is coming around the corner.
Within a very short while you notice that the trams are not clumped together but actually run at equally spaced intervals, more or less.
Steve: A great deal of irregular service on the TTC comes from decades of an attitude that nothing can be done, coupled with a lack of headway discipline. You don’t need many operators on a route playing games (dawdling, taking long breaks, running immediately behind the preceding car/bus) for the public to feel the results in poor service. Little is done to correct this.
LikeLike
I think that part of the public confusion and furor that has frequently been associated with Transit City, while many factors were/are at play, has in part been attributed to the lousy political spin and deceptive marketing of what LRT actually is, including what the abbreviation even stands for, which can make some suspect of the whole thing because the simple reality is they aren’t being given straight answers. Multiple parties are at fault for exacerbating this, but the general public, caught in the middle, is bombarded with claims they can’t make heads or tails of.
If we’re going to talk about nomenclature and terminology, we shouldn’t let political spin and the like muddy the waters.
When Toronto placed an order for LRVs, these were “Light Rail Vehicles.” Their counterpart for subways are HRVs, although this term is rarely – if ever – used, and of course are “Heavy Rail Vehicles.” APTA annual reports have, at the back, table data on the amount of infrastructure by type, including “HR” and “LR,” which are “Heavy Rail” and “Light Rail,” respectively. LRT is “Light Rail Transit,” and HRT is “Heavy Rail Transit.” Nobody buys “Light Rapid Vehicles” or “Heavy Rapid Vehicles” and so there is no such thing as “Light Rapid Transit” or “Heavy Rapid Transit.” Only in “BRT” and “SRT” does the “R” stand for “Rapid” (and the latter is only a line name, which was also twisted into a brand; the technology for the SRT is ICTS, which falls neither under HRT nor LRT as it uses so-called “4th rail technology” (see also: Linear Induction Motor)) The “RT Map” of the TTC is simply anything that is fully 100% grade-separated (almost all of which is the subway) and has a frequency of 6 minutes or less at all times except between 1:00am (depending on direction, can be after 2:00am) and 6am. Of note, GO trains counts as Heavy Rail Transit, too, even though they have at-grade crossings.
It should also be accepted that there are good and bad implementations of any mode, and just because an implementation is poor does not change the mode. Case in point, the Sheppard Subway is an extremely lousy implementation of HRT, but it’s still HRT. The same is true for Spadina, and St. Clair, and Queens Quay as LRT lines. When there’s low-floor vehicles with all-door loading, plus timed transfers and signal priority operating on these lines, they become good LRT instead of poor LRT. They’re still technically LRT today, regardless of the quality of their implementation, which is obviously in need of meeting a higher standard.
The stop spacing, I argue, has zero impact on whether or not something is classified as LRT, as stop spacing should be dictated by the needs of the built form, which will vary across the line. This is true for Sheppard, where some stops, like those around and including Warden, are around 300m apart, and for Spadina, where some stops, like Willcocks-College or Spadina Station-Sussex, are around 300m apart as well. Both are dictated by geographic conditions along the line.
LikeLike
Steve said (in response to Roy Murray): “As for politicians and “professionals”, the “pros” are just as responsible for the mess we are in thanks to their subway fetish and anti-streetcar bias which for years prevented the TTC (and others) from looking at any alternatives.”
The problem is, they’re dealing with the wrong pros. Try hiring some companies who have actually designed and built European systems (many in confined spaces) instead of the usual suspects available in North America, who, in my opinion, exhibit the same smug attitude as some TTC boffins and minions (notice I’m not painting everyone with the same brush – there are folks on staff who actually give a damn).
Those who want Toronto criss-crossed with massive tunnels and frequent service do not understand the density of population that needs to be in place, or moving in now, to support it. New York, London, Paris, etc have core and inner-suburban density to allow such extravagance (and most of those lines were built many years ago with far less bucks than it takes now). We do not have the density, nor will we, not with the automobile ruling travel.
Some of the problems can be laid on the long-dead. The designers of the city in the 1800’s had no vision as to what could make a great city and thus we are stuck with four lane roads (if that), or in inner suburbia, 6-lane, that cannot be expanded without uprooting vast numbers of those they need to serve in the first place. We’re stuck with that, so transit planners need to deal with it with the most suitable methods.
Replacing inner city streetcars with buses won’t do it, certainly not economically (Mr Ford, kindly note the poor use of taxpayers dollars here, something you and we are against). Think of all those extra bus drivers tailgating each other in an effort (perhaps the wrong word) to move our carcasses hither and yon. What is needed is some plain common sense, a commodity sadly lacking in many of those we elect, or hire.
Find the right experts, listen to them, and cut the crap. We’re all tired of it.
LikeLike
Steve said: “The boundary between LRT and streetcar is not as clear-cut. ”
One option is to draw that boundary based on speed. If the design speed of a line is significantly higher than the typical speed of local non-express routes in the area, then call it LRT; otherwise, a streetcar line. With that definition in place, Sheppard East (at 23 kph) and Eglinton (at 25 – 30 kph) will be LRT, whereas Spadina and St Clair are streetcar lines in dedicated lanes.
Such definition might not be technically precise (as Karl Junkin pointed out, even vehicles for mixed-traffic downtown streetcar lines are now called LRV), but IMO it would be clear and convenient for many participants of the debate who are not transit engineers.
Steve: The problem remains that an LRT line, because it has many different possible implementations, may be “LRT” in one part of a route and “streetcar” in another. On bus routes, there are huge variations in speed thanks to differences in demand and in street layouts. Even the Queen car can, if it tries, make very good time from Long Branch to Roncesvalles, but it’s not LRT except east of Humber.
LikeLike
James … I have those numbers.
Bloor-Danforth was about 95,000 passengers per day with the streetcar, and about 130,000 per day after the subway opened (Keele to Woodbine). Ridership on University was zero before the subway opened (there was no surface route on it), and about 25,000 per day between ’63 and ’66. University spiked with the Y (ridership jumped to 75,000 per day during the trial) and then dropped down to 55,000 per day after the Y was abandoned.
Streetcars could have kept on running on Bloor. It wasn’t that Bloor needed a subway — it was that an east-west subway was a given and it had to go somewhere. The University line was a ghost town outside of rush hours in those days and well into the 70s — it had a very eerie feel to it.
LikeLike
Steve said: “A great deal of irregular service on the TTC comes from decades of an attitude that nothing can be done, coupled with a lack of headway discipline.”
I’d say compounding those problems with the streetcar network is that the infrastructure has been constructed with those same attitudes which makes it more difficult to fix. For example, there really should be a passing track at every subway station loop. This way, when a group of streetcars arrives, the second one would unload and head out empty while the first one would be loading. You might not break up the group, but you would ensure a more comfortable ride for the riders getting on after the station. At the same time, if an operator decides to take a break, they would not end up stopping all streetcars behind them.
Steve: This runs into severe space problems. It is common to see both tracks full and cars queuing on the street at Broadview Station during peak periods. The geometry at Bathurst Station would make a passing very difficult to install. St. Clair Station could accommodate one, but the running time on that line is so padded at times that I’m not sure being able to pass is the problem.
LikeLike
Steve wrote:
“The important issue, however, is not to choose one technology to the exclusion of the other, but to look at the appropriate one for each implementation.”
Kevin’s comment:
Very true. And true not only for LRT vs. subway, but buses and heavy rail (GO trains) as well. What makes a technology appropriate is the demand for transportation throughout the network and, in the case of heavy rail, the availability of a railway corridor. So, without further ado, I put together a little chart with all the transportation methods and their range of hourly service.
Bus: 0-1,000
LRT: 1,000-6,000
Subway: 4,400-11,100
GO Train: Up to 23,300
Since I know people will love to nit-pick at the numbers, I’ve included my calculations below.
Steve jumps in: The calculations below focus on seated loads, but for capacity planning, some standees must be included. The proportion of standees is higher on routes with comparatively short trips, and lower on routes (like GO) where people ride a fair distance and pay a premium fare.
Buses are appropriate technology in residential areas, industrial areas with a surge at shift change times, and for express “rocket” services such as the ones currently run by the TTC.
Once hourly demand gets above 1,000 per hour it is time to introduce LRT service. All major arteries in Toronto have a current peak passenger throughput well above this level, so LRT is the appropriate service to have on all of Toronto’s major arterial roads.
Similarly, once the projected demand from the route and all its feeder lines gets close to 6,000 per hour, it is time for subway service.
Finally, once the demand goes over 11,000 per hour on a subway line, it is time to look at relief subway lines or parallel heavy rail lines. An example of this is the Yonge line which, at peak hours, delivers spectacularly poor service with people crammed in so tightly that passengers are frequently left standing on the platform because they simply cannot force themselves onto the train.
This “Take The Car” type service is undoubtedly discouraging many passengers, leading to substantial pent-up demand on the Yonge line. I also strongly suspect that there is substantial pent-up demand on Eglinton due to heavy congestion in its central part. This leads me to predict that, when the Eglinton LRT is opened with its subway-like service in the central area, it will be crammed to capacity shortly after opening.
Details of calculations:
Bus calculations use an Orion VII bus with 44 seats. Maximum use headway is 2.5 minutes or 24 per hour. Capacity is therefore 44 X 24 or 1,056 per hour, which I round down to 1,000.
LRT calculations use the Bombardier Flexity car with 84 seats. For the minimum service, I assume single cars operating with a five minute headway. This gives 84 X 12 or 1,008 per hour, which I round down to 1,000.
For maximum LRT service, I assume a three-car train (as planned for Eglinton) with a 2.5 minute headway. This gives 84 X 3 X 24 or 6,048 per hour, which I round down to 6,000.
Subway calculations use a six-car train of Bombardier “Toronto Rocket” cars with 62 seats each. For minimum subway service, I assume a five minute headway, which gives 62 X 6 X 12 or 4,464 which I round down to 4,400. For maximum service, I assume a two minute headway, giving 30 trains/hour, which gives 62 X 6 X 30 or 11,160 which I round down to 11,100.
GO calculations assume Bombardier bi-level cars with 162 seats per car. For maximum service, I assume 12-car trains running at a headway of five minutes, which gives 162 X 12 X 12 or 23,328 which I round down to 23,300.
Justifications:
1. Some people may object that subways can be run with four-car trains, as Sheppard is currently doing. However, this puts the subway into the LRT capacity range. It is far cheaper to build LRT, so four-car subway trains are deemed infeasible and excluded from consideration.
2. Some people may question the minimum standard of five minute headways. However, this is the TTCs current minimum for rapid transit. People have got used to it for a very good reason: longer than five minutes is such a long wait as to make it difficult to justify the capital expense of rail transportation.
3. Some people may wish to throw the crush load into consideration. Crush load capacity does, of course, exit and will be used in an emergency, such as a revolution in Saudi Arabia causing gasoline prices to skyrocket. However, on a regular day-to-day basis, having to stand is such a crappy level of “Take The Car” service that it must be excluded from consideration.
4. Some may ask why 90 second headways are not used for TR subway cars with ATC. For reasons previously discussed here, I do not believe that 90 second headways are possible.
Steve: I really cannot concur with the conclusions above because they are based on cutoff points that are generally too low. For example, the design capacity of a bus for peak periods is about 50, and this means that 1,000 passengers/hour translates to a peak point service of 20 vehicles/hour, or a 3-minute headway. This is quite feasible to operate with buses, and indeed several routes get down to 2-minutes comfortable. If a street can also support express buses and locals, the combined service can get even lower as on Finch East.
Streetcars on a 2-minute headway have a capacity of 2,220/hour (CLRVs) or 3,240/hour (ALRVs). Obviously an LRT implementation is going to provide for trains, not single cars, and the capacity scales accordingly (although within the constraints of stop length and passenger congestion at major stations).
The design capacity of a subway train is about 1,000 (although this has not been adjusted for the roomier TR trains), and at 30 trains/hour (a level that can be achieved in theory if terminal design does not constrain headways) that’s 30k/hour peak.
Note that in all cases, actual peak demands at higher rates can be achieved, but they cannot likely be sustained due to passenger congestion and the rise in dwell times at stations.
GO trains cannot operate on a 5-minute headway without very substantial changes (electrification, resignalling and huge capacity upgrades that may not, in fact, be practical at Union). Twelve trains/hour/corridor is a fantasy on the current or any likely future network.
You are miles off base on much of your calculation because you are using off peak design capacities and headways for some situations (subway) and beyond peak values for commuter rail. This type of analysis is not helpful in the overall discussion.
LikeLike
Let me add one more twist to whether or not the Eglinton LRT’s underground component can be considered a subway or not. I for one, have to consider it that but What I’d call the entire line to be in general would be a hybrid. That’s a term that I can remember seeing back before anyone ever used the term LRT. Hybrid was a term used to describe what is now the Green Line in Boston as well as all those tram/subway lines and systems in Europe.
LikeLike
I agree: According to 2006 TTS data, Spadina was still only at 12,000-and-change, and then jumps to 20,000 when it becomes the University Line at St. George from B-D transfer traffic. Spadina, for 32 years of growth and counting, has certainly left something to be desired on multiple fronts.
I don’t expect the AM short-turn service to go north of Finch West, since that’s where the pocket track has been relocated to in the tunnel design. This would be the northern-most pocket before Vaughan Centre, and my understanding is that it is not practical from an engineering perspective to put the pocket anywhere else, which politics can’t change regardless of York U’s influence. Of course, it’s not a long walk for able-bodied students to York U from Finch West Station’s north exit.
LikeLike
One thing TTC does well, or has done well so far, is the subway. The TTC subway is one of the better subways around the world, a bit boring maybe but operates very efficient, lots of room, lots of doors for entry/exit and straight platforms at perfect height mean quick stops. Most stations are covered and finally subways are not subject to surface route delays. So the public sees this, and the the public compares streetcars with LRT’s of course.
Of course the public sees subways as a good solution. Who knows what this streetcar like LRT is that is being pushed on us is, I don’t even really know the TTC version is either. Build us a good example of an LRT. That is hard to do, but build one LRT and make it work. Then see if the public likes it………..
I know it is late and we are running out of time, but that is the proper way to do it.
LikeLike
David Gunn’s pragmatic approach: increase the number of buses on a route until they no longer can meet the demand, then put in light rail until it can’t the demand, then subways.
Steve: TTC management’s approach is to keep extending the running time and the headway until all the cars laying over at the terminal will no longer fit in the loop.
LikeLike
I think that if LRT advocates would cut their losses and just accept the fact that Eglinton and Sheppard ought to be built as full-scale subway lines, the public would be more willing to entertain the case benefits of building LRT elsewhere in the city. Line-haul commuter corridors such as those would attract far higher ridership levels with subways; because factors of trip speed, reliability, frequency, available seating, preboard fare collection, and enclosed shelter from the elements makes subways far more palatable. Always have been and always will be. Most stations could just have basic stair and ramp portals to the surface (a la NYC’s system) instead of extravagant grand entranceways that tend to spike up construction costs. I also do not think either Eglinton or Sheppard at major intersections has maximized their on-site growth and development potential. Walk-in traffic at every station (roughly spaced 1 kilometre apart from each other) would number into the thousands easily (Chester with no surface connections does 7000 passengers per day). Feeder bus connections carrying tens of thousands daily from northern Etobicoke, Mississauga and Brampton would also more than justify building a subway to the airport.
People wouldn’t mind taking a short bus trip to/from wherever the subways end so much as having to endure an arduously long voyage on the LRT to YUS with the requisite stalling/bunching, collisions, malfunctioning transit signals, human error, fare disputes, etc. And seeing as future fare hikes and prolonged construction periods will inevitably occur no matter which mode Toronto ends up pursuing, maybe we should think practically about the usefulness of a mode rather than introducing new technology for the explicit purpose of showcasing what that mode’s capable of doing. It was that line of thinking that stuck us with ICTS for 25 years now instead of a one-seat subway ride to Scarborough Ctr.
Steve: While I can see the point of your argument, it’s off base on a few things. First, you ascribe to both subways and LRT characteristics that don’t necessarily apply in a Toronto context. Entrances to stations on Eglinton (and on the existing Sheppard line) cannot be simple ramps or short stairs because these lines are not close to the surface. That’s a function of Toronto geography (not flat) and the fact that there are many utilities in the way of a near surface alignment. New York had the advantage of building much of its subway a long time ago, in a location generally without steep changes in surface levels and in a political climate where the locals just put up with cut-and-cover construction. Call it nimbyism, but there is a very strong resistance to this in Toronto.
Fare collection on Transit City routes (and on the streetcar system too) will not require people to queue and pass by the operator, but instead will have all-door loading. Your characterization of travel as an arduously long voyage (plus the shopping list of problems) overstates the situation for LRT, and by contrast delays are not confined to surface routes as anyone who uses the subway knows. A basic point about the network in general is that at some point, subway construction is simply not affordable (unless we only want a small number of lines) and there must be a transition to a surface mode.
You talk about stations one km apart, and Chester would not even exist by that standard. Station spacing affects access time to transit for those unlucky enough to not live right next door, and long walks are at least as annoying as standing on a surface platform waiting for a vehicle. Chester is in the middle of a residential neighbourhood well served by closely-spaced stations. The same does not apply to Yonge north of Eglinton, nor to Sheppard.
As I said before, I understand the basis of your argument, but we disagree on the point at which the subway network should end and an LRT network should begin.
LikeLike
Thanks for the article explaining the differences of transit stations for different areas. I am on other transit sites as well and I was ripped apart on one site (for some reason) for endorsing the work being done on Victoria Park Station (which I use daily and am very grateful for this work in progress). The person told me that transit stations should be small and part of the streetscape like they are downtown!!! I think that is fine for downtown, but out of the high density downtown area we don’t get nearly as much walk-in traffic as the downtown stations do and that a lot of the outer stations get a high percentage of their commuters from the feeder bus routes that serve each station.
I am from Montreal, and have lived most of my adult years in Vancouver and this design of having buses enter right into stations is a design feature that this transit user greatly appreciates. I know Vancouver has a moderate climate but it was very very uncomfortable waiting for my connecting bus ‘outdoors’ compared to waiting in a station like I do at a lot of stations here in TO. I hope that the TTC and the city of Toronto continues with this philosophy of investing a lot in our stations, it isn’t frivolous, as I was told, I feel it adds to our comfort and will increase transit usage because of this added feature to Toronto’s transit system making commuting via public transit that much more comfortable.
LikeLike
I read both of these articles. It only confirms my view that in the “subway vs LRT” debate, most know or think they know what a subway is but have little idea of what LRTs are and tend to buy in the idea that they are “just” streetcars. Moreover I suspect the reason subways have been favoured over LRTs is that the former never impact on road capacity while the latter often do, thus with subways we can “have our cake and eat it too”, that is provide mass transit to those that need it while leaving roads free for those that prefer to drive (this certainly seems to be the unspoken theme for those arguing against LRTs and even streetcar lines).
Having lived in San Francisco at the time they replaced their aging PCCs with LRVs, I saw an example of a hybrid system, where in the downtown the LRVs ran as 3 or 5-car trains underground at widely spaced stations but would uncouple and run along the surface in mixed traffic or in a protected ROW depending on the location. One had to be careful which car one boarded when in subway-mode lest they wind up on the wrong route. The one curious feature of these cars is that stairs for the centre doors could be raised for subway mode and lowered when travelling on the street (they were not low-floor). Since our LRVs will be low-floor such would not be required where they run underground.
What’s interesting is that in his second article, Royson quotes the TTC’s estimated cost of converting the exiting Sheppard “stubway” to LRT, which I’d often contemplated on the few occasions I’ve used the line. Has the TTC actually contemplated converting the Sheppard “stubway” to LRT? It would be really nice not to have to transfer just to continue west or east past Don Mills. Obviously the expense is huge and given the amount of retro-fitting that would be required.
Phil
Steve: The Sheppard conversion proposal was raised a few years ago by Richard Soberman, but it has never been embraced by the TTC.
LikeLike
Steve wrote:
“The calculations below focus on seated loads, but for capacity planning, some standees must be included.”
Kevin’s question:
Why? I am quite curious to know what percentage of standees you are willing to plan for and from where you derive that number.
Planning for standing seems to me to fall into the “planning for failure” category. I don’t have any survey data (anyone know of any that exists?), but my experience is that the #2 turn-off that drives people away from the TTC amidst jokes that it stands for “Take The Car” is excess crowding and having to stand during trips. #1 is, of course, late or non-existent service.
For many people having to stand is real torment. They may not be officially disabled, just elderly or living with lesser injuries, but they just can’t stand standing (pun definitely intended!). And it is not just standing, but being thrown around with every lurch and jolt of the vehicle. Having to stand is a very, very unpleasant experience and a major reason why many people refuse to take the TTC.
Crush load does exist, and on the Yonge line during peak hours many people are physically unable to get on the train in spite of much pushing and shoving. This makes for a spectacularly unpleasant, crappy experience. It also means that there is zero excess capacity available for any emergency. And that people who really are disabled are prevented from boarding. At least we don’t have Japanese style “pushers” shoving people in!
Steve: You need to do some reading. The TTC has service standards which list the standee load used for design purposes when planning service levels. They don’t always achieve this, especially on streetcar routes, but that’s what they aim for. Although a CLRV can hold a crush load of 100, the target for service design is 74. Given that the car has 46 seats, this means that 28 would be standing. Some cars will carry more, some less.
If we operated service designed to give everyone a seat, even for comparatively short rides such as on Spadina, we would have to increase service by at least 50%. Frankly, I can think of better uses for the money (capital and operating) that would entail. You may not like planning for standing, but just try to convince tax-weary citizens that we should shell out for a vast increase in system capacity when what they really want is better reliability.
Of course there are crush loads on Yonge, but I am not advocating that level of crowding. The problem is that we don’t have enough service in the Yonge corridor generally (both on GO and in parallel subway service as has been discussed at great length here). However, you cannot extrapolate from that to a demand for a seat for every rider on every route.
Steve wrote:
“… the design capacity of a bus for peak periods is about 50.”
Kevin’s comment:
The 40′ Orion VII bus has seats for 44 people. Source: Orion Bus
Steve: You really need to check against the layout of buses as operated by the TTC once in a while. Low floor buses on the TTC have a seated capacity ranging from 35 to 38. The page you link says “up to 44”, but achieving this requires narrower aisles that make moving through the vehicles more difficult.
Steve wrote:
“Streetcars on a 2-minute headway have a capacity of 2,220/hour (CLRVs) or 3,240/hour (ALRVs).”
Kevin’s comment:
My calculations were based upon the new Bombardier Transit City vehicle. It has 84 seats and is currently planned to run in three-car trains on Eglinton. I do not have any information about planned headways (is there any?) but assumed a 2.5 minute headway or 24 per hour. This gives an hourly seated capacity of 84 X 3 X 24 or 6,048 which I round down to 6,000 per hour.
Note that this means that Royson James is quite wrong. Peak hour ridership of 5,400 people will nicely fill the LRT. Normal statistical fluctuations in loading will probably push some trains over their seated capacity.
It is my opinion that 5,400 by the year 2031 is far, far too low a forecast. I believe that the phenomenon of induced demand will apply to Eglinton and there is a lot of pent-up demand that is currently being suppressed by severe congestion on Eglinton. We will see.
Steve: I wasn’t talking about new LRVs, but about the existing fleet. Also, Eglinton will not likely run with three car trains initially. The TTC will not design service based on all seated loads, and you will never get a Commission or City Council to approve funding to operate service this way. You are creating a false argument by lowballing capacity numbers to suit your own purpose, namely arguing that we need a subway. If I were to take the same approach, we would need at least two, probably two more subway lines into downtown to meet your standard of service comfort. It ain’t going to happen.
Steve wrote:
“GO trains cannot operate on a 5-minute headway without very substantial changes (electrification, resignalling…”
Kevin’s comment:
True. The best GO is currently doing is 10 minute headways. I note that there are six trains currently leaving Oakville to Union on the Lakeshore line from 0655 to 0745 on 10 minute headways. I was writing about future planning in which things like electrification, resignalling etc. can be taken into account. There are places in the world that do operate such trains on 5 minute headways. We can too.
The important thing is to ensure that the appropriate technology is used to meet existing and future transportation demand to provide a pleasant, reliable ride. Keeping in mind the capacity and limitations of each form of service will enable us to do just that.
Steve: What you cannot do is to operate Lakeshore east and west, plus Georgetown, plus Milton, plus Barrie all of 5 minute headways, even though some of the dimmer lights at Metrolinx once thought this possible. However, there is an upper bound on the pedestrian traffic Union Station can handle between the platforms, mezzanine and surrounding streets. There is a capacity constraint at John Street for trains thanks to our having given up land so that little tower could be built relatively close to Front Street.
I find it hilarious that by your measure, GO trains have a higher capacity than subways. The reason, of course, is that GO coaches are laid out for maximum seating area and limited circulation space, while subway cars are the exact opposite. The new TRs actually reduce the number of seats/car compared to the T1s. If we were to follow your argument, we should reconfigure subway cars to ensure seating space for all. This would be challenging.
Please note that I do not intend to continue this conversation further.
LikeLike
One argument for the Sheppard “stubway” is that it can be used as a park and ride for commuters coming from the north and east ends into the city. I know how such setups are looked down upon on this blog, but the 401/404/DVP interchange is arguably the worst in the GTA, if not the country! 3-2 lanes from the 401 eastbound, 3 lanes from the 401 westbound, and 5-4 lanes from the 404 southbound merging into THREE lanes! Let’s not forget that this is the only thoroughfare connecting the busiest urban center in Canada to its northern and eastern suburbs – not to mention eastern Canada. Even at the most remote off-peak periods that interchange can bottleneck, during the rush hour it is hell on earth. If the Sheppard subway was never constructed, this interchange could potentially be one of the worst on the continent today!
With that said, there are certainly several design flaws. Why didn’t they build it as a branch line to Yonge? Or why don’t we today have every other train on the line skip Sheppard-Yonge and branch on to the YUS? Parking could have been better implemented too: before Metropass parking was discontinued, you had to go to Leslie station to park for free – reducing its efficiency as a commuter line. With the 404 and its HOV lanes, why not run express buses out of Don Mills to reach the further out suburbs of Aurora, Newmarket, and northern commuter towns instead of having them crawl up Yonge St? Or hell, why not make extending it to Downsview as a priority to improve commute times to York University – even interline it with the subway extension to remove a transfer for students commuting in from Scarborough and points further east.
Steve: Two points here. First, the junction at Yonge is complex enough without making the Sheppard line a branch of the YUS. Also, as anyone who compares ridership on the lines will know, there are far more riders coming south from Finch, and splitting the service at Sheppard wouldn’t serve them at all well.
Second, on the question of park and ride: My position is that with respect to subway lines, the capacity of a parking lot is trivial compared with the capacity of the subway and the bus network feeding into it. Disproportionate resources, including land, are consumed catering to this market. With respect to commuter rail, there is no question that the only way to get to many stations is by car. However, Metrolinx/GO foresees a day at most a decade away when there will be all-day bidirectional service on most if not all routes. For inbound travellers, we know that many lots are already full early in the day, and there is a point beyond which building parking structures is not a viable option. For outbound travellers (or inbound ones whose destination is not served by a subway station), getting to a location surrounded by cars but with little local transit is utterly useless. GO Transit needs to recognize that its service model will change, and Metrolinx must start thinking of bus feeders (much like the TTC), not just parking lots.
LikeLike
Steve wrote:
“The TTC has service standards which list the standee load used for design purposes when planning service levels…”
Kevin’s comment:
I have been utterly unable to get a convincing answer as to where those numbers come from. My suspicion is that they are merely an ex post facto justification of the existing system. Which, of course, makes them useless for future planning – unless you want the future to look just like today!
Steve wrote that I was:
“…lowballing capacity numbers to suit your own purpose, namely arguing that we need a subway”
No, actually. Cost and capacity numbers actually show that the biggest “bang for the buck” comes with more LRT. “Plain vanilla” LRT running single cars on a five minute headway with no tunnelling and no St. Clair style “street enhancements” is less than a tenth the cost per km as a subway. And capacity can be easily increased by running cars more frequently or hooking them up as multi-car trains. LRT gives roughly half the capacity of subways for a lot less than half the dollar cost.
Although Royson James criticised the slower average speed of LRT, that has to be balanced against the time taken to descend into the subway station. Some stations (eg Bayview) are so far underground that it takes 5-10 minutes simply to get from the surface to the subway platform.
Although each technology has its place, it is my personal opinion, based upon costs vs. benefits, that Toronto needs far more additional LRT than it does more subways.
Steve: I agree with your position on the need for more LRT, if not the way you arrive at the conclusion.
LikeLike
Phil Piltch said: “Has the TTC actually contemplated converting the Sheppard “stubway” to LRT?”
Michael’s response: not TTC, but Metrolinx has performed a “Finch-Sheppard corridor” study, and considered the aforementioned conversion amongst other options. They estimated the cost to be $670 million.
Royson James gave a higher estimate, $800 million, in his article. Perhaps he scaled the Metrolinx number to “current dollars”.
Even if it is $670 million, that’s more than 50% of the whole Sheppard East LRT cost, without adding a single new LRT or subway stop. I don’t think this is an acceptable cost of eliminating one transfer.
Moreover, if the Sheppard line is ever expanded west of Yonge, about 1.5 km between the end of existing tunnel and the Don West bridge will have to be underground regardless of technology, as the street there is too narrow for surface LRT.
LikeLike
Kevin Love said: “I have been utterly unable to get a convincing answer as to where those numbers come from. My suspicion is that they are merely an ex post facto justification of the existing system. Which, of course, makes them useless for future planning – unless you want the future to look just like today!”
Michael’s response: You are right that TTC’s load standards are largerly “an ex post facto justification of the existing system”. However, your proposed standards that give everyone a seat require a huge increase in the service level, accompanied by only a modest increase in revenues (some more choice riders on the system thanks to more comfort).
Given that TTC struggles every year to get an operating subsidy just to maintain the existing service level while keeping the fare increases to a minimum, how can it fund a 2-fold or 3-fold increase in service level? To my knowledge, no major city in the world manages to operate its public transit system with every rider getting a seat during peak hours.
Steve: The standards were the outgrowth of work done by then Mayor Sewell and me along with TTC planning staff decades ago. The idea was to have a yardstick (Canada had not yet gone metric) against which to measure overcrowding. During the budget problems of recent years, the TTC was regularly publishing lists of all of the service improvements they could not operate because they didn’t have enough operators or buses (I reprinted that info on this site).
You have to remember that ridership fell to about 360-million in the mid-90s, and there were widespread service cuts. When riding started to recover, the TTC didn’t have the resources to react to it.
Part of the Ridership Growth Strategy (something I also had a hand in) was to revised the standards a few years ago to reduce the acceptable load in peak periods. There are still standees, but now service is supposed to be improved before they’re down to only room on the roof. Off peak standards are already a seated load, but this is violated in many places as any regular rider knows. We can’t afford to run more service, don’t ye know.
So, please, enough of this ex post facto crap. Many transit advocates have been working for decades to have more comfortable service recognized and formalized in TTC planning standards and policies. If nothing else, the incoming Council will have to explicitly change these standards if they think that current service is excessive, and be counted for worsening service on the TTC. If there is an ex post facto justification of anything, it will be the penny pinching of Council.
LikeLike
Steve:
I don’t commute regularly on the 504 anymore, but based on my occasional use I still think that the information below is true. You mention that the total capacity of the 504 in morning rush hour is greater than the reported loads. Coupled with the fact that people have trouble getting on, it appears that the actual loads are greater than reported. However, I fear that the old villain, bunching, has lead to an enormous waste of capacity. At Shaw, the service eastbound is not every minute or so, but rather a group of five or even six cars every 10 to 15 minutes. The 4th 5th and (if applicable) 6th cars are all below capacity. I have often waited for an empty car and it never became even moderately full before people started to get off after Spadina.
On a related issue, why do people always insist on getting on the first car when there is another empty or at least emptier one right behind. I always wait and invariably find that I arrive at my destination only seconds after the passengers in the car in front.
Clarification: My poor wording suggests that the later cars in the queue never get full. That is not true. What I was trying to say was that, at the extreme, some occasional cars never get full.
Steve: One of the saddest things to watch is the Nextbus display where it is easy to see cars leaving the terminal in bunches. TTC planning staff make the point, when they get a chance, that overloading a route by running to little service actually wastes resources by driving up crowding, increasing stop service times and annoying passengers. However, when they report counts that suggest the service is underused, they destroy their own argument. Of course if operations would space out the service, loads might be better distributed and “surplus” capacity better used.
LikeLike
“One of the saddest things to watch is the Nextbus display where it is easy to see cars leaving the terminal in bunches.”
Oh wow, I had no idea this information was so readily available online. Has anyone ever gone to the trouble of recording a whole day’s worth of Nextbus imagery? I’d be very interested in seeing what a typical day on a given line looks like in fast forward.
I imagine you could make some very enlightening visualizations out of all that data.
Steve: George Bell has a live view of TTC vehicle locations (data available through the City’s Open Data initiative), as well as an historical view of data from selective extracts which I obtained for route analyses. Although the Nextbus site does not show the Dufferin and Bathurst routes, data for them is available, and George displays this on his maps.
LikeLike
With respect to the Soberman plan to retrofit Sheppard to LRT, there is an alternative I’ve considered but never heard discussed. Instead of converting the subway to LRT, could the subway be extended along Sheppard at-grade (median running with simple raised platforms a la Calgary or Buffalo). I would imagine that retrofitting a small collection of subway cars to use pantograph while running at-grade would be more cost-effective than retrofitting the tunnel for LRVs.
Is this feasible? Is there precedent for third-rail HRVs running at grade in this fashion?
Steve: It’s not just a question of retrofitting pantographs. A cutover system would be required so that when the pans are the active source for power, the third rail shoes and related exposed gear (the fuse on the shoe beam and the connection point for a “bug” for carhouse operations) would have to be disconnected so that this was not “live” during street operations. There may also not be either structure support in the car roof or a location for a high-voltage feeder to run from the pan down to the car’s electronics.
Subway cars are almost two feet wider that streetcars/LRVs in Toronto, and this would mean a four foot wider right-of-way (even more at pocket tracks). This would have to be designed in from the beginning, and as the Agincourt grade separation is already under way, we have just about missed the boat for a subway-width right-of-way.
There would be an issue with fleet size and service standards. Would we operate a five or six minute headway of four car trains to the outer end of Sheppard East until 2 am?
LikeLike
“Operators get away with taking long layovers and running in packs, and the riders suffer.”
I think I read earlier on this blog is that the reason why the streetcars are not run on headway is because drivers want to exit their shift on a particular route, while not incurring significant amounts of overtime.
If this is indeed true, then this, combined with Steve’s comment above means that the Union is still calling the shots on management on streetcar routes. The way I see it, the move to headway-based running is being blocked because no one on the ATU 113 wants to be in a situation in which they end their shift at the opposite end of the city where they would like to be. The problems with abhorrent line management does not solely lie with Management alone, the Union needs to step up to the plate with this and work with management on this issue. So if it means that an Operator ends his shift on Queen and Roncesvalles where he really wants to be at Queen and Woodbine, so be it.
Obviously, I may be wrong, but I think this issue is more than simply management. It almost looks like Management wants to, but can’t.
As for the soon-to-be TTC Chair Karen Stintz, how much history does she have with the Transit system anyway? One hopes she can approach TTC issues with an open mind like Rob Ford, but listening to the grapevine, she is more stubborn and inflexible than Ford is.
Steve: Your characterization of this as a labour issue shows your predilection for blaming the unions for the ills of the world. The problems lie on both sides, but particularly with TTC management who cannot get beyond decades of claiming that nothing can be done to improve operations in mixed traffic.
As for Karen Stintz, she gave some indication of movement on Transit City’s Eglinton line late in the campaign. It will be interesting to see whether she agrees with Ford’s idea of scrapping it completely and leaving her constituents to deal with all those buses.
LikeLike
Michael Greason: “On a related issue, why do people always insist on getting on the first car when there is another empty or at least emptier one right behind. I always wait and invariably find that I arrive at my destination only seconds after the passengers in the car in front.”
Possibly they’ve had my experience — that when the cars are bunched like that, the following cars get short-turned.
LikeLike
I think we’ll have to agree to disagree on park ‘n rides for the time being, as it would take this thread off track, but I think you understood my Sheppard branch strategy.
When I said every other train, I meant every other train westbound to southbound. So if you are at Bayview station, one train would go to York Mills while the other would go to Sheppard-Yonge. This could also help southbound crowding, as some trains would not be picking up passengers at Finch.
Since the Yonge line operates far more frequently than the Sheppard line, maybe every fourth or fifth train would end up skipping Sheppard-Yonge and go eastbound on the Sheppard line.
Steve: Every fourth train on the current headway would be one every 10 minutes. Passengers are not going to wait around that long for a through train to show up. Even if the TTC gets the YUS headway down to about 1’50”, this would still mean a gap of over 7 minutes between through trains. I won’t mention the problems of sending a six car YUS train onto a line with four car station platforms.
In any event, there is no provision for west to south movements at Sheppard and Yonge, and this discussion is moot on that count.
LikeLike
Hi Steve and Michael:-
“On a related issue, why do people always insist on getting on the first car when there is another empty or at least emptier one right behind. I always wait and invariably find that I arrive at my destination only seconds after the passengers in the car in front.”
I get on the first car because the second one will frequently get short turned and I won’t get to my destination, even though when getting on the car, it was rolled up for my desired stop. I’ll never forget the last time I did that, it was a cold night and I had the time to get home so I thought, sure I’ll wait for the next Carlton car and get a seat home. It was only a minute behind and it clearly stated that it was destined for Main Stn. Well wouldn’t you know, at Ashdale Avenue the motorman called out that the car was short turning to Queen and to catch the car behind. The car ahead didn’t wait for us even though we were right up his back side. Alas, there was no car behind for more than a half hour. So giving up, I ended taking a diseasel on Coxwell and the subway to Main where there was still no car and a large crowd waiting for it. Sooo, from now on I take the first car. This by the way wasn’t the only time, but the most memorable and the last.
Steve: There is an ongoing problem with through cars not waiting for short turns behind them to exchange passengers. Some operators do it, but many do not and at off hours this can cause huge delays and annoyance to riders. Yet another piece of “customer service” the TTC has to work on.
Dennis
LikeLike
Steve comments, “There is a capacity constraint at John Street for trains thanks to our having given up land so that little tower could be built relatively close to Front Street.”
Steve, is this a serious criticism, or something that you like to mention (I think you’ve said something similar in another topic). Given that construction started in early 1973, I wonder if anyone really had honest reason to worry about pinching down the rail corridor. Those were early days for GO Transit, and Metro Centre proposals were pretty current which envisaged the demolition of Union Station. And the “rail lands” were really the rail lands back then, hence “CN” tower.
Were there actually concerns about rail corridor capacity when the CN Tower was planned? As you have pointed out in different contexts, it’s very hard to plan for demand forty years in the future — which is about the position we’re in now.
Steve: The question of rail corridor capacity was an issue back in the early 70s, but nobody wanted to get in the way of redevelopment of the railways lands. Indeed, we are lucky to have as much of a “corridor” as we do. Passenger service at Union was not exactly breaking records for demand and even the station itself was threatened with demolition. However, the suburbs were already growing and the limitations on expressway capacity into the core were obvious. GO, after all, was created as an alternative to expansion of the Gardiner/QEW corridor.
LikeLike
Matt G: Until quite recently, the western lines of the Oslo metro used overhead wire, while operating through services with the central and eastern third-rail network. (Two routes are still partly closed for the upgrades.) Exactly as you anticipate, a train arriving at the transition station would simply raise its pantograph, deactivate the third-rail shoe, and then proceed.
Now, though these lines had numerous level crossings, they weren’t actually running on the street itself, and I’m not sure that huge TTC subway trains in the centre of Sheppard Avenue would be desirable. But it is indeed technically possible – in Europe it would probably be called a ‘premetro’.
LikeLike