From October 20 to 30, Metrolinx will host a series of public meetings for those interested in commenting on the draft Regional Transportation Plan. Four background papers are now available on the RTP page providing additional information about various aspects of the plan:
- Climate Change and Energy Conservation
- Transit Technologies
- Modelling Methodology and Results
- Mobility Hubs
This article deals primarily with the modelling of ridership, likely the most important of the four backgrounders because it shows how the proposed network is hoped to behave and the impact it will have on travel in the GTAH.
There are many caveats in this process set out in some detail in the report, and I won’t replicate them here beyond the standard warning that any model is only as good as the data it is fed, that the likelihood that the real world will match the modelled one falls off as we move into the future, and the basic fact that models cannot project the effects of changes beyond the range of known circumstances. We know what changes are expected in population, jobs and housing fairly well for a 5 to 10 year horizon, but the 25 year view is hazy. We know how people react to comparatively small changes in the relative cost of travel, but we don’t know what happens when changes are large and sudden. Metrolinx is quite open about these problems, as anyone publishing modelled data should be, and it is important that we view the projected network behaviour in this light.
From my own point of view, there is a much more profound problem. The only data shown in the backgrounder are for a completed network of routes in 2031. If anything is certain about Metrolinx, it is that the proposed network will not all be built, and will not be built when or where today’s draft plan suggests. The draft RTP itself acknowledges that this is a conceptual plan and is subject to change.
Moreover, the primary discussion today is about what will be done with the $11.6-billion of MoveOntario money we hope to see from Queen’s Park, plus, if the gods smile on us, an additional $6-million from Ottawa. That’s less than half of the total cost of the RTP, and does not cover any costs for local transit systems such as ongoing maintenance, fleet expansion and service improvements.
Whatever is built, the work will happen in stages. From a simple marketing viewpoint, if nothing else, it would be useful to know what the situation will be in five, ten, fifteen years as new routes come onstream. Will the public (and the politicians who depend on that public for support of large public works like the RTP) see significant change in a meaningful timeframe? What will we have to show for all the money we will spend?
Metrolinx already has a short list of projects to start immediately, in effect a seven-year plan, plus a fifteen-year plan in the RTP. However, the backgrounder only shows the situation on a 25-year horizon.
How much of the demand projected so far in the future comes from existing or soon-to-be trips, and how much depends on building or rebuilding that is decades off? If we only look at the 25-year timeframe, we will see projections skewed by populations and jobs that won’t exist for at least a decade.
How will travel patterns behave when only a third or a half of the network is in place? Will we have temporary crowding problems caused by an inapt (or inept) sequencing of projects?
The second major flaw lies in the presentation of the projected data. The backgrounder gives us figures for the AM peak and the peak point on each line, but does not show the volumes on links within the network or even identify where the peak points are (although in many cases they are easy to guess). Evaluation of proposed routes is difficult if we don’t know the contribution each link makes to the network as a whole, or the impact addition of a new route has on links in the existing system. Decisions about project staging, how much of a new route should be built when, depend on knowing the time periods when each stage would be most effective for the region as a whole.
The full backgrounder is a large file (about 10MB) and it is not the most printer-friendly document (a common problem with many Metrolinx publications). For readers’ convenience, I have reproduced the appendices with the detailed numbers as much smaller files.
Appendix B: System Performance
Information in the System Performance table appears in summary form in the Draft RTP itself, but it is convenient to see it consolidated in one place. There are three sets of numbers shown:
- Actual 2006 data
- Modelled “business as usual” 25-year projections (only existing, committed projects are completed)
- Modelled RTP network 25-year projections
Travel and transit use do increase even with the “BAU” case showing the effect of population and economic growth even without substantial additions to the transit network, but this is overshadowed by the parallel increases in auto use. The modal split for transit is essentially unchanged even with an increase in transit riding of 50%. By contrast, the RTP projection shows transit riding going up by over 100%, and the modal split moving from 16.5% to 26.2% for the AM peak.
The underlying problem, of course, is that auto trips still go up (by about 25%) after 25 years of RTP construction, although not as much as in the BAU scenario (by 50%). Either way, auto use will continue to be a major part of travel within the GTAH. On the environmental front, various emission factors will improve with the RTP relative to the BAU model, but it is important to remember that some improvements are due primarily to assumed technology change, not the transit network. This reflects the fact that if transit is only carrying one quarter of the trips, the other three quarters still contribute emissions based on then-current auto technology, and the absolute number of these trips will be higher than today.
Coming back to the absence of published model data, the other huge omission is, of course, the behaviour of the road network. We know that there will be more trips, but where will they be? What will be their origins and destinations? Is there anything we can do either in land use planning, transit network design or a combination of factors, to offset the growth in auto travel that is otherwise predicted by the model?
To a limited extent, this is addressed by sensitivity analyses (pages 19-21 of the backgrounder), although the detailed network behaviour under each scenario is the sort of thing we really need to see.
- Road pricing. If a 20-cent/km fee is assumed on the major highways, transit ridership and modal split are projected to increase three percent with a 6 percent drop in auto travel measured in vehicle-km. This suggests that motorists are comparatively impervious to tolls, at least at this level, likely because the convenience or necessity of driving takes precedence.
- General auto cost increases. If auto costs go up by 400 percent, rather than 200 percent as assumed in the base model, the impact on transit riding is 9 percent, with an 8 percent drop in vehicle-km. Note that we are likely seeing the effect of car pooling here, and again this points out that price alone will not change travel modes, at least within the constraints of the model here.
- Intensified land use. This model assumes the same population and job growth but concentrated in “urban growth centres” and existing built-up areas especially centres at Mississauga, North York, Scarborough and Markham. In this scenario, transit riding goes up 5 percent, the modal split goes up 4 percent and vehicle-km go down 2 percent. This is a particularly interesting result because it implies that major incursion in the land-use planning process will not, by itself, make a huge change in travel behaviour. To some extent, this could be due to dilution of the totals by the existing trips.
- Uncontrolled sprawl. If sprawl is permitted, transit loses a small amount of riding while auto use goes up a bit. The reason the numbers are so small is that although sprawl gobbles up a lot of land, it does not accommodate many people, and the trips contributed by the sprawling form are a small proportion of the total. However, as we have already seen in the 905, this built form creates large areas where there is little hope of transit making any impact except for park-and-ride services oriented to major job nodes.
- Changes in road capacity. These two models were described in White Paper 2, and the analysis concluded that road expansion, although helpful in some ways to address congestion, would also contribute to increased non-transit travel over longer distances and would be counter-productive to the goals of the RTP and the Ontario government. This is a difficult issue because the same road expansion that might be advocated on economic grounds, especially for freight, also attracts people away from transit and diverts capital and political resources from transit expansion.
The analyses of alternate network behaviours already exist, and the details should be published so that everyone can better understand the implications of various scenarios.
Appendix C: Individual Project Performance
When we turn to projections for individual routes, things start to get quite interesting. As I said before, one big problem is that this is an end-state projection for year 2031, and we have no indication of how each component of the planned network will behave as lines are added through the 25-year buildout.
Express Rail and Commuter Rail
Express rail projects do extremely well, and that is no surprise at all. The provision of extremely frequent, fast, all-day service on what are now GO Transit corridors transforms the commuter rail network into a true regional rapid transit network. We will take it on faith that it is physically possible to build these corridors, and assume that the railways are always amenable to suitably large infusions of cash.
Peak point peak hour riding on the five express lines is modelled at levels comparable to the existing subway system:
- Lakeshore East (23K), Lakeshore West (26.3K)
- Mississauga (Milton line) (17K)
- Brampton (19.8K)
- Richmond Hill (18.1K)
Those are huge demands, and they pose a major concern for actual accommodation at Union Station, likely the peak point for all of these routes. The total of the demands shown above is 104,200 per hour. By contrast, Union Railway Station handles about 155,000 passengers per day on GO Transit, and current planning assumes that this will double over the next 20 years. The demand projected by Metrolinx is considerably above that target.
Over and above the express rail system, we have the remaining commuter rail network with good peak hour demands notably on new routes to Bolton (5,700) and Seaton (6,300). The commuter lines peak point demands (again likely at Union) would add a further 40,000 passengers or so during the peak hour.
The combined effect shows up in the Mobility Hubs backgrounder where the combined boardings and alightings at Union Station hub (which would include the subway/LRT station) are projected to be 387,000 during the three-hour AM peak period. To put this into context, current usage of Bloor-Yonge Station is around 200,000 per day.
Even assuming that all of the trains implied by these services will fit through the rail network around Union Station, the handling of such a vast number of pedestrians will be extremely challenging.
Subway Lines
Life on the subway network, by contrast, is rather quiet. The projected peak demand on Bloor-Danforth drops to 16,000 thanks to riding diverted to the Downtown Relief line. The Yonge-University line sits at 25,100, a lower value than today’s peak, again likely due to transfer trips from BD diverting onto the DRL.
The Sheppard subway (note that in this model it is still a subway even though some versions of the RTP treat it as a generic rapid transit line with the possibility of conversion to LRT), has a paltry 6,100 riders at peak (roughly equivalent to a 5 minute headway of 4-car trains). The Spadina extension gets 7,000 at the peak point, and the Yonge extension to Langstaff gets 8,800. Meanwhile, the DRL has a projected demand of 17,500 at the peak point.
This is a particularly important example of the need for details about intermediate network configurations. What do the BD and YUS loads look like if we build the subway extensions but do not build the DRL, or at least delay it for a decade or more beyond completion of the northerly extensions? Where do the projected DRL riders come from, and would this line benefit from a staged implementation?
Even more to the point, are the projected flows credible given the major implications for the network’s behaviour? We need to understand these factors as part of the overall planning for the RTP network and capital plan.
Other Rapid Transit
This category is a catch-all for BRT, LRT and other possible technologies such as Skytrain. Some of the 905 network elements such as the Highway 407 Transitway, VIVA Yonge and the Hurontario line are above 5,000 per peak hour taking them into LRT territory. Within Toronto, Don Mills, Eglinton, Finch West and Waterfront West have good peak point demands but all within the capability of LRT especially considering that the peak point for Eglinton will be in a tunnel. Even Jane does well in this model, by contrast to the TTC’s estimates.
All of this has to be read with suspicion. One thing that experience has shown is that demand models are good at coarse-grained trip assignments to networks, but that the more finely woven a network becomes, the greater the possibility the model will mis-assign trips because the relative differences between route choices are smaller. On top of this we have the 25-year modelling timeframe.
Again, we really need to see detailed data, and we need this for various intermediate dates and network configurations.
Average Trip Lengths
A fascinating number, not listed in the backgrounder, but easily derived from it, is the average trip length on each route. This gives some indication of whether the route is heavily used over its length, serves long-haul regional trips or primarily short-haul local trips.
If we divide the peak hour passenger-km by the peak hour boardings, we get the km per boarding or the average length of a trip taken by a passenger on that route. Note that the total trip will be longer as it may include other routes (local transit services or RTP lines) and walking/driving at either end. This number only tells us how far the average peak hour rider travels on a specific route in the network.
- Lakeshore West (34.6 km)
- Lakeshore East (30.2 km)
- Mississauga (19.7 km)
- Brampton (25.0 km)
- Richmond Hill (22.2 km)
These are fairly long trips and there is a good chance that most of them are bound for Union Station.
Also worth comparing is the length of a line versus the average trip. For example, Brampton is 32.3 km long and the average trip is 25.0 km indicating that most trips are over the full length of the route. By contrast, Lakeshore West is almost twice as long as its average trip length.
For the regional rail network, we find similar results with Barrie topping the charts at 75.7 km/trip on a line that is 98.6 km long. The Seaton route is intriguing because its average trip length is 22.6 km on a 39.8 km route indicating that many riders originate well west of the terminus, likely in Scarborough in the same catchment area as the extended Scarborough RT. It is noteworthy that the Metrolinx projected demand on the SRT is considerably lower than the TTC’s, and this is likely due to diversion of trips onto the commuter rail network.
On the subways (BD and existing YUS), the average trip is between 5 and 6 km (each route counts separately), and this is no surprise. The average trip on the Yonge and Spadina extensions is also in the 5-6 km range indicating that most trips will travel the full length of the route extension probably as commuters, not as local traffic. On the DRL, the average trip is 4.6 km, only 30% of the route’s length. This is no surprise considering that it is actually two routes with an east and west leg.
On Eglinton, the average trip length is 6.75 km, a clear indication that demand in this corridor is overwhelmingly local, although obviously if the route were designed for regional service with fewer stops, it would attract more long-haul trips. That tradeoff will, I suspect, continue to be debated for a year or more through the detailed analysis of Eglinton and other high-priority projects.
On Finch West, the average trip length is 8.77 km. This is higher than on Eglinton, but still only slightly more than from Yonge to Jane Street. This line also acts as a feeder to both north-south subways.
The real surprise (well, not a surprise for anyone who has been watching the issue) comes on Waterfront West where the 14,300 peak hour riders travel only 34,700 km or 2.4km each. In effect, all of the traffic on this line originates close to downtown, at least in the modelled view of the world. The projected peak demand is 5,200 per hour, impressive for what will be almost a streetcar line, and vastly more than the existing loop at Union can handle. This whole project needs careful review.
I will leave it to readers to mull over other numbers in the Metrolinx backgrounder.
Summary
The demand projections are a vital part of the evaluation and political support needed for the new regional plan, and bluntly put, they should have been published a long time ago. Many of the debates about various aspects of the plan could have taken place in a more informed setting had this information, as well as model runs for alternative proposals, been on the table in the public arena through the Green, White and Draft RTP processes.
The demand figures we do have are interesting, but they fall far short of the overview we need to properly evaluate the draft DTP specifically because they omit:
- models of intermediate stages in the network rollout
- models of alternate networks and rollout strategies
- link volumes to indicate which parts of the network are the most productive and have the greatest effect on other parts of the system
I could ask for a lot more, but those are the basics. Indeed, this is the sort of thing we used to expect from so-called Environmental Assessments, and information which any proper business case analysis would demand in evaluating the benefit of various options. I can’t help thinking that a lot of Metrolinx’ work has consisted of drawing many lines on maps and hoping that the numbers came out right. That is, in fact, how a lot of planning happens, but it helps to know what influenced the plan’s evolution. The lack of visible progress from concept to plan and the comparative secrecy with which Metrolinx has worked make the whole exercise more suspect than it might have been otherwise.
We are facing a very large public expenditure for a transportation network that will make things better than they might have been for some, and much better than they are today for the region as a whole, but in many places we are only standing still, just keeping up with population growth.
In the next installment, I will turn to the other backgrounders to the Draft RTP.
Steve Munro 
Steve
Are you aware of any similar studies that have tackled these issues in a better way than Metrolinx has?
Also, given the capacity issues noted above, should we start thinking about relocating low capacity routes out of Union station? Maybe we’d get more bang for the the buck if VIA Rail services were routed into a low capacity Dupont or Summerhill station while those slots and platforms were given over to 12-car electric GO services?
This might be a more successful endeavour than turning a freight line through tony areas of this city into a commuter line, especially considering the flat crossing between Richmond Hill GO and the CP Midtown.
Steve: Other studies even in Toronto have included link volumes so that the performance of route segments is evident, as well as general traffic flows to show origins and destinations. Metrolinx is unusual in that it is a large, regional study, but that does not absolve them of doing a good job. Some of this material must already exist (it is implied by references in their published documents), but they’re doing a poor job of explaining how we get from “here to there”, or why we would even want to take their particular proposal.
By the way link and flow volumes go back a century or more, and there are lovely hand-drawn charts of early 20th century travel patterns for Toronto. This is not a new technique.
I am trying to encourage Metrolinx to be more open with their data and actively pursue public examination of alternatives. What has actually happened to date has been an almost obsessive secrecy, even from planning professionals within the very municipalities who make up the GTAH. This is not healthy, and gives the impression that Metrolinx has something to hide.
As for the CPR line, there are a few points. Diverting VIA and maybe a few of the GO services to North Toronto Station (or a new station on the Spadina subway) won’t deal with the lion’s share of the problem, the huge new demands flowing into Union. Also, there’s a good argument for keeping VIA and GO services together to allow for GO to act as a VIA feeder.
The Richmond Hill line does not make a “flat” crossing with the CPR. The CNR runs on the floor of the Don Valley, while the CPR crosses above it (and the river) on a trestle. No chance for an interchange here. The CPR crosses the tracks that are used by the Brampton/Georgetown service at grade at West Toronto, although a grade separation is planned for that location. It is not, however, a great place for a station because the subway is inconveniently less than 1 km south of the crossing.
LikeLike
Steve,
If Union Station can’t be expanded to accomodate the staggering volume they’re proposing, What do you propose be done? Is there some other feasable way to get all this service to the downtown core? I can’t imagine any option that wouldn’t be either awkward (terminating GO trains at TTC stations) or unecomonical/undesirable (tear apart downtown to replace Union).
I’ve also wondered myself whether it’s wise to have so many important lines connected to one point. It seems to me that if something drastic were to happen; natural disaster, terrorism, whatever, at Union station the whole GO system stops dead.
Thanks for the analysis,
Ian
Steve: I’m not entirely sure, to tell you the truth. Part of my problem lies with the absence of detailed data from the model that would indicated how this huge volume builds up in time (over the years as lines are added) and space (where they are actually going). The plan seems to do a very good job of supporting further intensification of the core area with commuting traffic from a considerable distance, and I am not sure that’s exactly the plan Metrolinx or its board had in mind.
There is always the possibility that the model is wrong, but if there is additional employment in the core that is not balanced by housing relatively nearby that is affordable and attractive to the people who will work there, then we will have a problem with capacity. It also begs the question of the degree to which the model reflects and simply builds on existing commuting patterns rather than redistributing trips to adjust for new population centres.
It may well be that the growth in the 905 so overwhelms population growth in the 416 that long-haul commuting is inevitable. This has implications for much of the land use planning throughout the region.
Meanwhile, we need to take very seriously the question of the rail corridor and Union Station’s role and capacity. Quite clearly, operational feasibility (track time, platform use, pedestrian flows) have not formed any part of Metrolinx’ planning efforts.
LikeLike
Steve: No chance for an interchange here.
Except this is exactly where an investment really should be made for some new infrastructure, to run the Richmond Hill Line down the much straighter CPR Don Branch and allow Richmond Hill and Midtown corridors to connect at Don Mills, connecting to both Don Mills and Eglinton LRTs.
Steve: There’s a small problem. The connection track between the Richmond Hill line and the CPR was ripped up several years ago and it is now part of the city’s nature trail / bike path network. There is no way to get from the Don Branch onto the CNR.
LikeLike
Another interesting comparison for the GO projections is between AM peak hour boardings and AM peak hour peak point riders. In essence, this is telling us what percentage of riders on each line are bound for Union Station.
For most of the GO routes remaining as “regional rail”, somewhere between 80% and 90% of all riders get off at Union Station, even with service expanded to two-way travel at all times. (On the Milton line, the peak point ridership is greater than the boardings.) There may be some reverse peak trips, but the vast majority of travel will continue to be into downtown Toronto.
For the “regional express” routes, there is a greater percentage of riders that are not bound for Union Station. This is especially true for Lakeshore West, where less than half of all boardings (43%) are bound for Union. More than half are either getting off at earlier stops or are actually traveling westbound. This suggests a very significant shift in travel patterns in this corridor, hypothesizing that riders will begin to find GO to be a much more attractive travel mode to suburban centres than they do today (or, as an intermediate link between other modes as one might travel bus 1 > subway > bus 2 today).
LikeLike
Steve – sorry about the brain fade there. I meant the GO Bradford, not Richmond Hill.
As for the CPR / Richmond Hill connection’s fate – this is why I am so sceptical about the West Toronto Railpath. Once the rails are gone and the bike wheels replace them, it’s hard to reverse.
LikeLike
If, in fact, much of the Lakeshore West traffic is not Union bound, then couldn’t some trains there be short-turned to Kipling Station? Then again, why is it that when we regard commuter rail, we only look at existing routes? Since we’re dreaming big with this plan anyways, why not look at a new tunnel and downtown station, with construction possibly coinciding with the downtown relief line?
LikeLike
Interestingly, Table C of “Modelling Methodology and Results” contains two entries for the Richmond Hill GO service: one in Express section, other in Regional Rail. The annual ridership is projected at 51.3 million for Express version, but only 1.2 million for Regional version. I wonder how is that possible.
Steve: It also shows the peak demand on the regional rail as a paltry 900. I suspect that they forgot to take the “regional” line out of the model. It exists in the 15-year plan, but is replaced by express rail in the 25-year plan. This sort of artifact shows clearly that Metrolinx has far more data than they are sharing with us, and that some of what they have published needs cleaning up.
A related problem is the high projected volumes on the express rail lines. These could only be achieved with very frequent service, and yet Metrolinx does not talk about the lines as having a train every 5 minutes (the only way they can possibly achieve a ridership of 20K/hour or more). This tells me that the model is out of sync with what Metrolinx actually proposes to build and therefore the results from the model are suspect.
LikeLike
Steve, I was talking about new infrastructure, because of the reality of the ripped up portion (that portion also doesn’t allow a connection at Don Mills along the CPR). I’m envisioning a new connection directly north of Wynford, between the CPR and Lawrence arcing over the DVP. It’s not cheap, this is certainly a more expensive junction/wye than typical, but the opportunity is certainly there, increases the transit network’s competitiveness, and should be taken advantage of.
LikeLike
As the Metrolinx plan shows that Union will be a very busy station it seems ‘somewhat unfortunate’ that GO is in the process of selling off a section of the tracks just east of Yonge Street. This is the area, which used to have sidings on it, running just south of the Esplanade. It is unused now but it seems rather risky to put up an office building there (5-7 The Esplanade). The proposed building will also contain the parking for the L Tower – or whatever its new name will be now the “L” part has vanished!
LikeLike
One point that I have thought about for a while is about the Downtown Relief Line. Toronto didn’t think big enough when building the Yonge and B-D subway line in regards to express tracks. That mistake can be more then solved if the DRL is built with express tracks. With the line going on Queen, the one the city wanted for so long, having express tracks would alleviate everything! It would ease the B-D in the central section and it would do the same for the busy Queen – Bloor section of the Yonge subway line.
Express means that it would make the route much faster with stops that are spaced out more like the Sheppard line as opposed to the B-D line. Then, the local tracks would act like the B-D line serving all major cross streets.
For example, Express would probably go…West to East
…Dufferin, Bathurst, Spadina, University, Yonge, Jarvis, Parliament…
The local tracks would have intermediate stops, with finally a stop called “CITY HALL” (for Bay st) ….something that many other cities already have!
It would be costly but it would definitely be worth it and would be as smart as the engineer with foresight that decided to build the Prince Edward Viaduct with a lower bridge for subways!
Steve: The projected demand on the DRL is nowhere near what is needed to support express tracks, and the route itself is so short that they won’t make much difference. Express services only work if there is a reasonable balance between local and long-haul demand, and if there is enough demand for both of them to justify frequent services. Without frequent service, the time added by waiting for a train will offset some if not all of the saving from the express run.
Remember that the Yonge line, when it opened, was only about 8km long and the idea that someday it would extend from Richmond Hill to Vaughan Centre was unthinkable. In 1954, good-size chunks of what we now call the 416 was still farmland (this was also roughly the era when Toronto went from 6 digit to 7 digit phone numbers, and call from Scarborough to Etobicoke was “long distance”).
The “foresight” of building the second deck on the Prince Edward Viaduct was intended to allow construction of a subway line into downtown (something like the Green Line in Boston), but that project never got started. It wasn’t intended to sit unused for over half a century.
LikeLike
Thanks for all this; comments too.
The future is now, and to harp on a theme (still, ad nausem at times for me as well), the use of Front St. as a strong corridor for transit really should be pushed towards the top of the agendas to protect the transit potential of the two routes past Bathurst, where Front now ends.
A surface DRL using the Weston rail corridor is quicker cheap capacity, and yes it can and should go further northwest, but with enough stops to ensure that it becomes a reasonable route. But rather than complicating life too much at the railyards, and to bring extra connectivity at Spadina, it should rise up to street level, eastbound using the remnant strip of the Lands and Garden trust south of Front St. that’s now a rail spurline. Using Spadina as a transfer spot eases pressure on the subways.
The other major Front St. transitway routing is to keep going west past Bathurst albeit with a bridge (not a tunnel) over these Weston railtracks to start a more sensible WWLRT than currently is rumbling forward with a deadline this week of resisting the piecemealing chunk to link the Harbourfront line to Dufferin, and continuing to miss providing transit for Ontario Place [for] 35 years.
The WWLRT EA actually says a third study is needed of a more direct line into the core instead of the proposal now going ahead as this isn’t “cost effective”, and we should tack on another c. $150 to $200M for rebuilding the Union Station loop for further intensity at Union Station.
This EA also forecast this direct line in would be about a 22-minute ride vs. a 30-ish from Etobicoke.
I’m delighted that you write “This whole project needs careful review” – it’s a looming disa$ter though we need to spend c$400Mish, maybe more. But not $700M plus for inferior transit, and hopefully officialdumb, blindp or not, will wake up to the utility of squeezing the multi-millions a little bit harder to provide both resiliency and effective transit.
GO does need a back up, we need to provide good direct competition with the mobile furnaces on the G/Lakeshore ahead of tolling it, and it’s kinda scary just how much time and effort needs to go into trying to raise transit options against hundreds of million$, with well-paid status quo stupidity enforced and with little personal resource.
Grrumble, but the future is now, and we must restore Front St. transit for the greenhouse century.
Steve: I am not sure I agree with everything you have written, Hamish, but do concur with the general thrust. If GO Lakeshore West does receive the huge increase in capacity Metrolinx proposes, it’s hard to see that a local line on Front or anywhere else will make much difference in that corridor.
As I wrote in my comments, the average trip length on the WWLRT shows that most of the riding originates in the dense, downtown part of the line. I am sure that the model has placed the longer-haul traffic, even from Etobicoke, on what is in effect a Lakeshore rapid transit line that the WWLRT can’t possibly compete with.
What I think is badly needed is to decouple planning for whatever will serve the western waterfront from serving the new population between Liberty Village and downtown. In that context, Front Street does have a role even though life gets tricky around Union Station. It would likely have to be combined with an LRT branch up the Weston corridor sharing a common path, somewhere.
Note to other writers: Please do not try to turn this thread into an excruciating technical discussion about alignments and technology choices for a DRL. The DRL is only in the 25-year plan, and the real question, given the projected demand, is whether it should be moved forward even though that is contrary to the political climate.
The issue here is to look at the Metrolinx plan in the context of the demand projections, not to debate how many angels can dance in a low floor streetcar.
LikeLike
Regarding Union Station, going out on a bit of a limb here, but with studies about to start for the Gardiner to be dismantled east of Jarvis, if we include in that study a relocation of the onramp for westbound Gardiner access to abutt the eastbound Gardiner offramp/to-be-new-terminus at Jarvis, wouldn’t it be conceivable to have a second Union Station starting at Yonge St. on the south side of the rail corridor and extending east from Yonge?
I was just studying a satelite image of the area on Google and imagining the amount of space the Gardiner would free up. If Lakeshore’s redesign is carefully thought out and integrated with some other initiatives for transit and taxi connections running through the station above it, as well as the East Bayfront plan, this “Union Station East” could be something seriously worth looking at for the challenges we’re facing for the current station.
Some of the current platforms already extend across Yonge St. beyond the train shed, although for GO Train users these areas are currently off limits. A re-design would allow these areas to be extremely useful in connecting the two stations together and allow more integrated pedestrian flows between services and allow them to work together without actually leaving either of the stations.
There is a challenge in getting through service to proceed west as there are only two tracks currently available (excluding CN’s high track) to allow trains to head west from this new station without passing through platform areas of the current station. There’s potentially room for an additional track between tracks 13 and 14(which has no platform access). If we do that, and restrict CN’s use of the high track to off-peak usage, it could still work.
LikeLike
4-track was briefly considered for Yonge in the late 40s (with all 4 tracks on one level), but it was quickly ruled out as the street simply wasn’t wide enough to support it.
LikeLike
What I’ve been wondering is why the legends on Metrolinx’s planning maps and their project lists don’t show any differention as to which lines under “other rapid transit” are BRT, LRT or AGT. Is this some kind of subterfuge on the part of Metrolinx to keep people in the dark and push the wrong modal choice onto them?
Steve: Yes. This is a compromise presentation to push the discussion about mode choice off into the “business case analyses”. Having said that, the rationale for an AGT line on both Eglinton and even the SRT has collapsed now that we see the demand projections, corrected to reflect a local, not a regional service in that corridor. Meanwhile, potential demand for the SRT is drained off by new GO services in its catchment area.
Metrolinx made a big mistake by talking technology choices too early in the process and giving away their prejudices, possibly even their leanings for potential PPP arrangements. That shouldn’t have happened if they were following the process they claimed to have.
LikeLike
Hi Steve,
I know you don’t want to get bogged down in DRL routing questions but can you clarify some things for me:
1) Why is the DRL often mentioned to be at Pape instead of Greenwood, where there is the subway yard leading to the rail line? If Greenwood is an option, and the DRL was moved up, would this have any affect on the routing of the Don Mills LRT?
Steve: The Pape option is a side-effect of this whole process being a descendent of the “Don Valley Corridor Transportation Study” which was originally very, very bus-oriented. The Don Mills bus goes to Pape Station, and so this became one of the optional routes for the BRT. The others are Broadview and Castle Frank Stations.
Subway alignments south of Danforth are not at issue for a BRT line and, frankly, any BRT regardless of the route would have big problems once it got into East York due to road constraints both north and south of Danforth.
There are several challenges for any subway alignment including the route north from Danforth across the Don Valley to Thorncliffe/Flemingdon Park, the connection with the Danforth subway (especially if any trips would be interlined to run from Kennedy direct to downtown), carhouse capacity, and the alignment south and west to downtown (via the CN Kingston subdivision or straight south and then west).
Based on the demands for the line they’re suggesting, would you suggest a phased approach, i.e. before completing extension of Yonge Line to Richmond Hill, at least the eastern part of the DRL (from Pape/Greenwood to downtown) should be completed first?
Steve: This is one of those questions where my real answer is that I don’t accept the premise. The Metrolinx projections show that a frequent commuter rail service to Richmond Hill will shave off the demand in that corridor, and yet we plan to invest hundreds of millions in additional YUS capacity to accommodate a peak demand that could be diverted elsewhere. Metrolinx has lumped the Richmond Hill regional express line into the 25-year (not the 15-year) plan, and I think this is a mistake.
The TTC has, I fear, underestimated the complexity of ramping up to handle projected demands in YUS, and overestimated the ability of the line to actually handle all those passengers. If there are ways to avoid hitting that peak in the first place, we need to look at them.
Alas, the YUS upgrade has been sold as a pre-requisite to the Richmond Hill subway. In effect, the very necessary signal upgrade to replace the antique 50’s-era signals has been lumped in with the extension as a way to pry money out of Queen’s Park and Ottawa to pay for it.
Another complex question is the matter of whether a DRL should run from Downtown to Eglinton with a major interchange at Don Mills & Eglinton where construction of an underground complex will be comparatively easy linking the DRL to the two proposed LRT lines.
Finally, what other possibilities have been studied to reduce demand on the Yonge Line if the extension to Richmond Hill goes ahead? i.e. Has the TTC studied the possibilities of more express bus services from the eastern part of Toronto to downtown, at least as an intermediate measure?
Steve: The number of bus trips needed to make a serious dent in YUS demand is huge. Let’s be generous and assume an average load of 60 per bus. Headways below 2 minutes are problematic unless there is a dedicated right-of-way and a large plaza-type station to accommodate the bus and pedestrian traffic downtown. Imagine if all the service at Finch Station were looping via Richmond and Adelaide and trying to pick up and drop off at on-street stops.
The current subway cars and signal/track layouts can handle a headway of about 130 seconds at best before serious congestion arises at terminals and major interchanges. Signal changes coupled with split terminal operations (say half of the service turns back at Downsview and at Finch) are hoped to get this down to about 110. That’s a capacity improvement of 130/110 about 18%.
The new cars are expected to have about 9% more capacity due to the gangways, and a further 8-10% could be achieved with a seventh 50-foot section added to the trains.
Putting all of this in context, a one-minute headway of buses would give us about 3,600 per hour, leaving aside the question of how we would actually fit all those buses on downtown streets. Assuming the traffic on these buses is all diverted from the Yonge subway, that would shave something in the range of 10% off of the demand. A similar diversion might be achieved with four additional trains per hour on GO, although the passengers using each service would be different.
2) The demand on Union Station is huge as you mentioned. I know we’ve already discussed the CP midtown route utilizing North Toronto station. I’m wondering though if residents there though would accept it becoming a regional transit hub, as intensification demand for the area would surely rise. I think, if an alternative to Union Station is considered, a more politically feasible site would probably be chosen.
Thanks!
Steve: There really are few operationally feasible sites, never mind the politics. One question about a service on the North Toronto sub is the location of a subway interchange. Summerhill has an existing station (masquerading as a liquor store), but it is also the worst possible place to add load to the Yonge line. Dupont Station may offer an alternative, but would also require a major new railway station and a new connection into the north end of the subway station.
The politics of the situation really turn on whether freight traffic on the CPR can be diverted elsewhere such as a shared cross-GTA corridor with the CN (there has been talk of this, but nothing definite). The complaints about trains on the CPR through the Annex and Rosedale arise as much from concern about hazardous cargos than noise issues.
In all, it’s a very complex set of issues, and we have not been well-served by the current disjointed studies (including Metrolinx itself) to understand the implications of various alternatives. Too many existing schemes are assumed to be untouchable, and we are burdened with poor decisions of decades past.
LikeLike
Hi Steve:-
Just for info sake. The Dupont station already has the tunnel connection to the south side of the CPR’s right-of-way. It is presently a long narrow store room accesed by one of the ubiquitous, annonymous grey doors one sees everywhere in stations.
I truly believe that one does not need a ‘major’ station, as you are suggesting, for this site as exampled by GO’s Lakeshore startup in the 60s. Aluminum and glass bus shelters and similarly appointed ticket kiosks were the norm and served well enough to help GO become successful and a model for many of the newer commuter lines in North America. Once demand dictates and budgetting can be found, then improve the new GO station as the Lakeshore line is now seeing along its route.
Dennis
LikeLike