Updated July 6, 2016 at 5:10 pm: “Yonge Subway Now” has updated their website to remove the double counting of capacity improvements, and to clarify that their claims about subway capacity apply to the peak point south of Bloor Station. The revised text is included in the main article.
Although in theory there will remain 4% of free capacity on Yonge south of Bloor in 2031, this is hardly the sort of margin we should be planning for. The Relief Line’s demand projections show that it has a major effect if it runs north to Sheppard, and it will have to be in place sooner rather than later to avoid deadlock on the Yonge line.
A related problem is the question of station capacity to handle passengers with trains arriving about 30% more frequently than they do today.
York Region has wanted a subway to Richmond Hill for years, and there is even a completed Environmental Assessment and its Addendum for this project.
Today, July 5, 2016, a new website extolling the virtues of this project went live. It contains the usual things one would expect about the growing need for transportation and how a subway will improve the region’s future. Unfortunately, it also contains a misrepresentation of available and future subway capacity.
But what about overcrowding you say?
- Metrolinx’s Yonge Relief Network Study analyzed options for crowding relief to the existing Yonge Subway line by examining new local and regional travel opportunities and improving mobility across the GTHA. Key findings include:
- Significant relief to the Yonge Subway line will be achieved through already committed transit improvements, including:
- TTC’s automatic train controls [adds 29% capacity];
- New subway signals [adds 10% capacity];
- New six-car subway trains [adds 10% capacity];
- Toronto-York Spadina Subway Extension [adds 5% capacity]; and
- Regional Express Rail/SmartTrack/DRL will add even more capacity.
- With the above capacity improvements in place the Yonge Subway line will be running under capacity when it opens and beyond 2031.
- The Yonge North Subway Extension only adds 9% demand at peak period.
Updated July 7: The text above was the original version. The page now reads:
But what about overcrowding you say?
- Metrolinx’s Yonge Relief Network Study analyzed options for crowding relief to the existing Yonge Subway line by examining new local and regional travel opportunities and improving mobility across the GTHA. Key findings include:
- Significant relief to the Yonge Subway line will be achieved through already committed transit improvements, including:
- TTC’s automatic train controls [adds 29% capacity];
- Toronto-York Spadina Subway Extension [diverts 1,300 riders to free up 5% capacity]; and
- Regional Express Rail diverts 4,200 riders to free up 15% capacity.
- With the above transit improvements in place the Yonge Subway line will be running under capacity when the extension opens in 2031.
- The Yonge North Subway Extension has a projected ridership of 14,000 to 22,000, but is only expected to add 2,400 demand during the AM peak hour, at the peak point south of Bloor.
Let’s start off with the increased capacity for the Yonge Subway. The Metrolinx report cited here says (p 15) that the existing capacity is 28,000 passengers per hour per direction, and that by 2021 this will rise to 36,000. That’s roughly a 29% increase, and is possible because of the new signal system which includes automatic train control. This will allow trains to run closer together, roughly every 110 seconds in place of the current 140 seconds.
Capacity of the new subway cars is already included in the 28k value as these trains have been exclusively providing service on the Yonge line for a few years. They no longer represent a marginal improvement that is still available. The design load for service planning (average loads over an hour, not peak loads on a train or car) for the new trains is 1,100 passengers. If trains run every 140 seconds, that is equivalent to 25.7 per hour or a capacity of about 28k/hour. Moving to a 110 second headway gives 32.7 trains/hour or a capacity of 36k/hour.
Traffic diverted to the TYSSE (Toronto York Spadina Subway Extension) at 5% of current capacity represents 1,400 per hour. This is in line with the value shown in the Metrolinx study (see chart below).
GO/RER has only a modest effect on the Yonge corridor because the Richmond Hill line is not part of the RER network, and other routes paralleling Yonge (the Barrie and Stouffville corridors) are too far away to have a meaningful impact. There is also the issue of the fare differential between GO/RER and the TTC which could discourage some riders from travelling on GO.
SmartTrack was originally claimed to be capable of subway-like service down to a 5 minute headway (12 trains/hour) that would serve Unionville and Milliken stations. However, we now know that “SmartTrack” will really be just a few more GO trains (part of the already planned RER improvement) stopping at a few more stations within Toronto, not the “subway like” operation some in York Region might have expected.
The Metrolinx study includes a chart showing the interaction of demand and capacity changes to 2031.
The current 2015 demand is shown as higher than the actual capacity (31.2k vs 28.0k) based on the level of overcrowding now experienced on the line. The light blue dotted line shows the capacity before the new signal system is activated, and the solid blue line shows the added capacity. Even this will not be sufficient to handle the projected growth to 2031 absent other changes.
The TYSSE and other changes are expected to shift 1,300 per hour from the Yonge line, and a further 4,200 would be attracted by GO/RER. This mostly, but not completely, offsets the anticipated growth so that by 2031 the “base case” demand is 32.3k, slightly higher than the demand today, but in less crowded conditions thanks to more trains/hour.
The Yonge North extension adds only 2,400 peak hour passengers and brings the line up to 96% capacity. Note that this is the peak hour average, and there will be some overcrowding due to variations over the hour.
This leaves no room for growth, but it also shows the paltry additional demand expected on a very expensive subway extension. Indeed, this makes the Scarborough extension to STC positively shine by comparison with 7,300 peak hour riders. The projected demand on the Richmond Hill line appears to be lower than the existing ridership of the SRT!
But things are really not that bad.
Those 2,400 are net new riders attracted by the subway in place of existing bus service. Total ridership will be a combination of then-current bus passengers feeding into Finch Station plus the 2,400 new riders.
Metrolinx shows that the “long” version of the Relief Line to Sheppard produces a sizeable reduction in projected demand on both the Yonge line and for the Bloor-Yonge transfer movements.
If Metrolinx, Toronto and York Region are really serious about providing capacity for future extension and ridership growth on the Yonge Subway, then construction of a Relief Line is absolutely essential despite its cost.
Meanwhile, York Region should update its website to provide accurate claims about future changes to subway capacity. Blatant inaccuracy such as we see here are the marks of hucksterism designed to sell a project, not a professional representation of what is actually needed.
Update: The new version of the website addresses these issues, but I must wonder why the incorrect information appeared there in the first place.
Steve Munro 
Is 110s time separation actually realistic? I can run subway trains as close together as I like, but if dwell time at any station is more than (at most) 80s then we’re just parking subway trains in the tunnels. And at (c)rush hour, a minute to clear the doors at Yonge+Bloor isn’t unrealistic these days…
Just to throw numbers at the wall, it takes about 18s for a subway train to accelerate at 1m/s^2 clear of the platform, and around 2s after the doors finally close to get confirmation and start moving. Assuming the next train to be _directly_ behind the departing train (0m trailing distance) with an identical acceleration curve, it still needs at least an additional 7s to finish pulling into the station (12s braking time from 1/2 distance). I’m reasonably sure the train separation would have to be non-zero, so that would add time. And then it takes at least 1s for the doors to open.
So, here in the real world, are they actually going to be able to run the subway at 110s rush-hour headways?
Steve: The TTC is well aware of this problem, and it is among the reasons they now shoot for 110 seconds at best rather than even lower claims. Dwell times at major stations are also related to crowding, and attempting to operate at a similar level to what we have today simply won’t allow trains to get in and out of stations quickly. Then there is the matter of terminal operations which area affected by several factors notably the length of trains and crossover geometry.
A big problem the TTC got themselves into years ago was that they needed funding for a new signal system, but it would never be approved just as “state of good repair”. Therefore, it was sold as a stepping stone to a huge increase in capacity. Oversold, more like.
LikeLike
An article in the Star last month said that the TTC “expects to increase capacity to over 32,000 riders an hour by implementing an automatic train control system,” which is somewhat lower than the 36,000 peak rider number given by Metrolinx. Would you know the reason for this discrepancy?
Steve: The TTC has claimed various numbers from time to time. I used the value in the Metrolinx report because that’s the one cited by York Region. It would not surprise me if the TTC were now using a lower number in recognition that 36k may be a challenge to achieve. They had claimed 38k/hour and 105 second headways in a 2012 study.
I understand that Metrolinx does not intend to electrify the Richmond Hill Line within the foreseeable future. So, does the estimated diversion of 4,200 peak riders all depend on electrification of the Richmond Hill line? Or is some service improvement to be achieved before electrification?
Steve: It is unclear just how RER will achieve this diversion or where it will occur. One possibility, based on some of the demand numbers we have seen for SmartTrack, is that most of the diversion will occur from trips that now travel to the core via the BD subway and Bloor-Yonge. If so, this does nothing for the capacity problem north of Bloor Station.
It is worth noting that Metrolinx’ own projections shown in the chart in the article for the configuration RER+ network give almost no effect on subway demands, even though in their projection of cumulative demand additions and relief they show 4,400.
I understand that peak hour demand to justify a subway should be 15,000. Would you know how well the YNSE meets this measure? Will it be better than for the TYSSE?
Thanks.
Steve: The 2013 Benefits Case Analysis for the Richmond Hill extension (figure 4.1 on page 35) projects a peak hour load of just over 10k leaving Richmond Hill Station with a cumulative load by Steeles Station of over 20k. The projected demand at Bloor is 44k, well in excess of the line’s service design capacity. The report is clear that crowding is expected, but even this is based on a presumed 38k/hour design.
The obvious point is that we should not be designing a subway to be overloaded the day it opens, and certainly should not base crowding projections on a level of service that cannot be achieved.
LikeLike
Thanks for this further post Steve, and to the commenters.
One option for real relief theoretically is with the Richmond Hill GO line to the east of Yonge, but I’m lazy or don’t know where to look to see if/how it is being upgraded to provide that faster sub-regional travel that is perhaps a key to shifting demand. It was part of a Smart Ride LRT idea of a decade-ish ago for more frequent transit, and there’s a spur line/shortcut now a bike trail that would be a good idea to add back in to the mix for faster trips in.
Sure, it goes through more moneyed areas that are well-represented by Ms. Wynne, Ms. Robinson and a few other good politicians, but isn’t this mobility crisis so severe we need to have ALL options considered? Could this be part of a relief scenario far far sooner than some other multi-billion project?
Steve: The issue is not that it goes through “moneyed areas”, but that it is comparatively expensive to upgrade given problems with flooding and the winding nature of the route. Also, GO does not own the tracks north of the York Subdivision, and RER is intended to be (mainly) on GO-owned trackage. Metrolinx does have a scheme to raise the trackbed northward from the Union rail corridor, but it’s not a trivial job.
An obvious question is whether they have been downplaying this line in their plans on the assumption that other relief valves including the DRL will avoid the need for it. In conversations a year or more ago with Metrolinx folks, I learned that their concern was further to the east because the centre of development has shifted from Yonge Street over to Markham. SmartTrack’s Unionville service was supposed to somehow address this although the last mile problem is severe with the development a few kilometres west of the rail corridor clustered at the 404.
LikeLike
Richmond Hill subway extension and Scarborough subway extension are two sides of the same coin, they are both needed. On the other hand, Downtown Relief Line subway and SmartTrack are two sides of the same coin i.e. either one will suffice and given that SmartTrack costs less than one tenth of the Downtown Relief Line subway, the choice is obvious.
Steve: Actually SmartTrack costs considerably more than 1/10th of the DRL, and it provides substantially less relief to the Yonge line. The choice is far from obvious.
LikeLike
If the discussion over the practicality of providing subway service to Richmond Hill relies on the DRL first being extended to Sheppard, does this introduce the option of extending the DRL up to Richmond Hill, instead of extending the Yonge line to Richmond Hill?
One thing for sure, extending the DRL to Richmond Hill, instead of the Yonge line, will prevent the Richmone Hill subway riders further overloading the Yonge line.
The median of the 407 is very wide. Is it available to run a rapid transit line, like Allen Road, down the middle of the 407? Would this be a good alignment for crossing from Don Mills to Yonge? Mike Harris sold the 407 for a mess of porridge. Would the private owners of the 407 be able demand an exorbitant amount to allow the median to be used?
If it is available, would running a rapid transit line down the 407’s median be useful, some decades from now?
Steve: If we’re talking a new line to Richmond Hill, there is a perfectly good GO corridor that should be used up first. As for LRT line in York Region, you might want to talk to their transit planners for whom Viva is, in theory, a precursor to LRT service.
LikeLike
I’m not going to argue for YNSE priority over DRL, because DRL is clearly needed first.
But I think it’s not really fair to say that YNSE is worse than SSE because it will only carry 2,400 passengers and SSE 7300. That’s apples and oranges. 2,400 is the NEW ridership for YNSE, 20,000 is the peak ridership. SEE is 4,500 and 7,300 respectively.
Steve: Yes, I make that clear in the article because York uses the 2,400 figure to imply that they really are not adding much to demand that is already there. Total ridership shows up in a chart in the BCA, and it is over 20k at Finch Station.
YNSE is a better candidate for subway than SSE because of peak ridership, not new riders. YNSE is intended to reduce the bus traffic on Yonge St., which is already at or beyond capacity north of Finch.
We need to get DRL moving for many reasons, and one of them is that there isn’t really more capacity for buses on Yonge St. past 2031. Now, we could build LRT, on Yonge St. to relieve bus congestion instead, but no matter the mode, it’s yet another project that needs to get moving.
FWIW, I don’t think LRT is a good choice south of Steeles because the LRT will also reach capacity soon after construction.
Steve: I agree that a Steeles subway extension is necessary and should have been underway for some time now independently of whatever we might do further north.
LikeLike
For years, I have been advocating extending the subway only to Steeles (primarily for an alternate turn-back location along with Finch, and partly to remove the huge number of bus routes on Yonge between Steeles and Finch), but have any rapid transit north of there provided by LRT.
The first reason is that very few people live where a short walk to the subway with a one-seat ride to a location south of Steeles is a benefit. Everyone will have to take a bus, or drive, to the subway. Building LRT north from Steeles, with an underground across-the-platform transfer or a St. George type transfer to the subway will be far more convenient to far more users because, for the same money, more people will be a short walk from the LRT that gets them to that easy transfer. With a subway to Highway 7, those people will have to take a bus or drive to the subway, then endure a vertical transfer of two, if not three, levels down, using escalators that the TTC does a bang-up job of keeping in top-notch working order all the time, to get on a subway to continue their trip.
For the same money as building the Steeles-to-7 part of the subway, LRT can be built 6 km further north all the way to Elgin Mills Road, plus a nearly 10 km Highway 7 line from Dufferin to Woodbine. This includes a short underground portion at Steeles, and a tunnel on Yonge for about 1 km from just south of Major Mac to just north of Crosby, as there is no room on the street (think of the centre part of the Crosstown line). Adjust this estimate for today’s costs, and you are looking at rapid transit reaching at least 12 km FURTHER than the subway. That’s a lot of people closer to a rapid connection.
The other reason for this is that what is built north of Steeles should be built by York Region. For York Region, that means being able to extend it when York wants to and can find the funding, not waiting for Toronto’s priorities (even if they are valid, like the building of the Relief Line IS). For Toronto, that means not getting stuck with the operating costs, like the TYSSE.
Finally, VIVA Phase 3 is supposed to be the upgrading of the busways to LRT. With the early construction on Yonge from Highway 7 north now underway, there would be a cost savings to just do this as LRT now and skip from Phase 1 to Phase 3.
Unfortunately, York Region Council is fixated on subway the way Springfield was fixated on the monorail. We will probably be stuck with the Jones Express to go with the Sorbara Subway.
LikeLike
I think Richmond HIll can still be improved with (more) one-way service and by adding stations on John St and Steeles Ave. The subway could be extended to Steeles (or maybe even Clarke) but beyond that would be wasteful. The Richmond Hill GO line runs pretty close to Yonge that any new ridership can and should be absorbed by improvements in that line.
LikeLike
Thanks again Steve, and commenters.
With the flooding, a recent history of the Don River by Ms. Jennifer Bonnell on p. 193 “after heavy rains stormwater makes up 70% of the river’s flow.” So – what about driveway and asphalt disconnection, which are part of a larger subsidy to cars/automobility. If we were to put on a drainage connection fee and an asphalt tax, could we cut out half of the storm surge, and raise some money?? Sure, this would be fewer millions for the construction industries etc., or a switch in some work to smaller scale depavings and remedial work, and yes, roads do need drainage, and I wouldn’t want to bike on a swale/permeable gutter, but rather than spending a further larger sum on flood ‘protection’, why hasn’t Moronto worked on this pavement issue?
It’s probably part of the caronic denial we have, and taking heat from votorists isn’t fun.
With the indirect nature of the route, if it could be alleviated a bit by reclaiming the shortcut of what is now a Rail Trail N of Eglinton, I’ll get friction from some cyclists, but go ahead.
And there are some other options too, if we got to truly thinking about it. And yes, that includes having a reversible rush-hour transitway on the Don Valley Parkway. I think there’s a batch of space if we rebuild the centre and reduce lane widths and reduce speeds to c. 60kms to enable fast travel, and hopefully having signals to ensure one-way travel remains safe is not too much a new thing for our transit systems to do.
Another option, written about in today’s Star, is to have a development freeze – not just on Yonge but also in the King/Queen/Etobicoke west waterfront corridor too, where it’s a feeble to grossly inadequate response to very clear overload and transport demand.
LikeLike
Thanks for your attention – we know the details are important in transit planning so we’ve taken a look at our website, made a few updates to the already-committed transit improvements and also responded to a few of your tweets on the topic.
One key point that we would like to clarify is the ridership on the YSE north of Finch:
You noted that
The 2,400 peak hour passengers and 96% capacity are for the increased persons/hour southbound from Bloor (specifically at the peak point), during AM peak hour. A parallel comparison to the STC ridership would be the YSE peak hour ridership north of Finch, which is 14,000 to 22,000, according to a joint TTC/YRRTC study (p3 in document).
Steve: You are missing the following two paragraphs added shortly after the first version of the article went up.
The way your site described them was unclear, but the updated version is a big improvement. The same 2,400 riders show up in the Metrolinx summary of capacity and demand changes which I included in the article.
All that said, the increase in riders for the Scarborough line will be greater than for the RH line in part because Scarborough is capacity constrained today.
LikeLike
I’d be interested in seeing a study comparing the cost and travel time for LRT and Subway from Steeles to Richmond Hill Centre (RHC). Including the option of burying the LRT through any portion of Thornhill where it can’t fit easily. I fear that without selective tunneling, the travel time for LRT would be little improved from today’s VIVA buses.
I have to side with Metrolinx that travel time matters a lot in the case of the YNSE. When 90% of riders are not new, one of the key purposes of switching to higher order transit is not just capacity, but also decreasing travel time. There are a lot of people on long commutes that funnel down Yonge St. Once LRT is added (north from Steeles or Hwy 7), they’re going to have 2 transfers rather than one (bus to LRT, then LRT to subway). Of course, it would be nice if we could divert more to RH GO than is possible at present too.
On a slight aside, I wonder if York will ever get serious about LRT in the medium term. Currently they are plucking all the low-hanging fruit, such as ‘curb-side BRT’ and building centre-ROW BRT through Richmond Hill, but not doing any improvements through the denser heritage district. The cost of LRT, including tunneling under that section of RH, is going to be a lot more expensive than VIVAs BRTs.
What I’m saying is that York needs to get serious about what it wants on Yonge St. 20 and 30 years from now. So far it’s only planning out 10 years, and some of its intentions are woefully inadequate.
LikeLike
According to their website, “the Yonge North Subway Extension is the top transit priority of York Region.” Mississauga is getting their Hurontario LRT as well as the Transitway. There are lines being built in Hamilton and Ottawa. I think that York has every right to not only advocate to build a Richmond Hill subway, but to do everything in their power to secure the funding and build it. If they can adequately justify its existence, regardless of capacity issues on YUS, they absolutely should build a subway from Steeles to Richmond Hill Centre.
But, if they want their subway to run from Steeles to Finch, and connect to the Toronto owned and operated YUS line, well then they can wait until Toronto is ready. And we’re not right now.
Steve: Maybe we could build a moving walkway from Finch to Steeles?
LikeLike
As someone in Richmond Hill (near Yonge and Major Mackenzie), I could easily see taking a bus to the Maple GO station to get downtown. In fact, I am looking forward to it! MM runs surprisingly smoothly, even in rush hour congestion (certainly better than Yonge does).
Side topic, but could we not have the Sheppard line hook south to become part of the DRL? Obviously it cannot turn directly south, but perhaps extend it a bit east and have it U back to Don Mills? Not only would it do SOMETHING with the line, but it would give legitimate reason to the LRT instead of a transfer to continue in the same direction.
Steve: Better to run the Relief Line as a separate service. Building that “U” would be messy considering what it has to go under.
LikeLike
Not exactly. With the lower cost per kilometre of LRT, an LRT line can reach further (6-14 km further, depending on how much of it is tunnelled), placing its access withing walking distance to a whole lot more people than the subway will.
Add to this a transfer at Steeles between LRT and subway that is a minimal disruption (St. George style, or even across-the-platform), and taking away the bus-to-subway transfer that will involve a vertical transition between two or three different levels gives people an easier overall commute. Even for those who will need to transfer bus-to-LRT, I argue that the two transfers will be easier and less time-consuming that the one transfer from bus-to-subway on a good day when all the escalators are working properly.
LikeLike
To be perfectly honest ATC in itself will not allow 110 second running. It will require the short turning of every other train before the end of the line or a turn back system beyond the terminal station. The existing terminal stations will not handle 110 sec. service. Before the collision on the Spadina line when trains could key by and close up to the train ahead the Yonge subway would regularly have on train entering a station before the one in front had left.
ATC with moving block system of train controls allows the trains to operate closer together but if that is the only change then they will not get a 29% increase it capacity increase. This is another case of the TTC overselling one benefit to get something they want. Problem is it can come back to bite them in the ass.
Steve: With the Vaughan extension, they will be able to run a split service on the west side and overcome terminal limitations. On the east side, the projected demands north of Finch are substantial and would make for capacity problems to Richmond Hill. Only if the RH terminus were designed for tight headways are they going to get down to 110 seconds.
I think that an essential trial once they have ATC working on the Spadina line will be to run a live attempt at a 110 second headway. It doesn’t have to be over the entire line, just with enough trains stacked up south of Downsview to keep the pipeline “fed”. Then we could see what they will actually achieve under ideal circumstances.
LikeLike
The question of course becomes very quickly, would you be able to sustain this with real loadings that likely everywhere south of Eglinton, and especially at Bloor, which will soon create far from optimal conditions even at 33 trains per hour the platforms will be overloaded soon enough after the capacity to do it is there, assuming that the TTC can actually get there.
LikeLike
PS. There is also the simple issue of course, of why correct the issues at Yonge/Bloor, when spending resources to address this means that the entire system becomes increasingly vulnerable to a single point of failure. Would not a single blown transformer, in effect shut down that station, and in effect cripple the function of system for a massive portion of its riders, without any viable work around. Would it not make more sense to expend those resources on alternate paths? Does it make sense to extend the loading of Yonge to say 40k per hour, or does it make more sense to try and redirect some of that load elsewhere?
Steve: “Sense” has nothing to do with elections and voters who are convinced that they “deserve” a subway.
LikeLike