On March 29, 2021, the Toronto Region Board of Trade released a proposal Getting on the Right Track – Connecting Communities With Regional Rail as the second of a planned series that will eventually include:
- Erasing the Invisible Line: Integrating the Toronto Region’s Transit Networks
- Solving the Last Mile (Summer 2021)
- Building Infrastructure (Fall 2021)
I reviewed the first report in The Siren Song of Regional Fare Integration and will not duplicate my comments on the Board of Trade’s fare proposals beyond the level needed to explain how the scheme in Getting on the Right Track dovetails with this.
From the title of the article, one can easily guess that I was not entranced with the Board’s proposal, and I should make clear why right at the outset.
First: Although the plan includes a very robust regional network with frequent service on all GO corridors, there is too much talk of how everything will work when it is finished, and not enough about how we actually get from here to there.
Second: As with so much regional planning that comes out of Metrolinx, there is no discussion of last mile costs and service, nor of the burden local municipalities would face in providing them. Yes, a “last mile” report is in the offing, but this could range anywhere from massive increases in publicly funded local transit to an embrace of ride sharing services. The report contains not even a hint of how the vastly improved service will get riders to and from its stations.
Third: The focus is very strongly on Toronto (the 416) where there is an established transit system that can provide frequent service at connection points, but less on how this would scale outward into the 905 and beyond.
Fourth: The Invisible Line report and its fare-by-distance proposal is assumed as a pre-requisite even though there is no agreement that this is how fares should and will be calculated. In particular, its gerrymandering of fare zone boundaries and the tariff has not been subject to critical review outside of venues such as this blog.
Many proposals in Getting on the Right Track are good and provide a level of background we have not seen from the nominal regional planner Metrolinx, an agency that prefers to save proposals for Ministerial photo ops and routinely hides details under confidentiality provisions.
To give Metrolinx their due, a key shortcoming in the Board of Trade’s report is that it does not clarify which parts of its proposal are works already in the Metrolinx pipeline, and which are net additions to the scheme. Indeed, maps purporting to show regional networks and travel times do not even acknowledge rapid transit lines planned and under construction that will open within the timeframe of the Board’s proposal.
An untutored reader might think that almost nothing is underway, that the Board has returned from the mountaintop with the one true word on regional transit.
Finally, and particularly toward the end of the report, elements creep in which feel like pet rail projects with only minimal evaluation. They are included either because the Board sought to curry favour with politicians in the affected areas, or because someone had too many crayons to play with. I leave it to the dedicated reader to peruse those parts of the report.
There is a sense throughout that what might have been a reasonable proposal for Metrolinx to aim higher in its plans evolved into a design exercise that substitutes detail and volume for practicality.
“Organization Before Technology Before Concrete”
On page 20, the Board makes a key observation, if only by implication, about how transit is planned in the GTHA by citing a practice elsewhere:
The German-speaking world has propounded the planning and engineering doctrine of “organization before technology before concrete.” The highest priority is to resolve issues of organization, which includes factors like fare and service integration between agencies. Then, technology, such as better signalling systems and rolling stock, should be improved. The last priority is the building of new infrastructure, like additional tracks and grade separations on corridors. This prioritization provides the most economically efficient means of improving service and capacity on a network.Getting on the Right Track, p. 20
This is one of several cases where there is an implicit, if not explicit observation that the way “we” do business is out of step with good practices elsewhere, or even just common sense here. However, the Board has violated its own principle by driving through an entire network design exercise without clearly figuring out goals, not to mention the basic question of how much we might be prepared to spend on this transit network.
“We” is a tricky term here because there are three levels of government each of which prefers to fund only certain types of service and infrastructure, and each has significant blind spots in the financing and funding of public transit.
In this article, I will not attempt an exhaustive review because even my readers have limits to their patience. Moreover, there are points where one must peer very deeply into the crystal ball, make too many assumptions about actual future circumstances. If our current situation teaches anything, it is that the future will change.
This section is buried down on page 21, and yet it is absolutely key to the entire discussion. It is so important that I will include its text here.
Drawing from international best practices, it is possible to demonstrate five guiding principles that form part of successful implementations of regional rail. Based on these principles, it is possible to design a network and operations plan for the Toronto Region.
Two-way, All-day Service
The majority of trips in any region – even work trips – do not involve the downtown core and do not take place at rush hour. A service plan that provides service all day, every day is essential if a regional rail system is to become a core part of the regional transit network.
High Frequency (turn up and go)
Research by Transport for London indicates that riders on routes with a frequency of 12 minutes or less will not need to consult a schedule and can instead simply “turn up and go.”
This level of service has been demonstrated to drive major increases in ridership. Frequency is even more important when making connections because wait times can multiply when a trip involves several connecting segments, and a missed connection could result in an unacceptable delay.
Seamless Integration with Local Transit
On a busy commuter rail service like GO Transit, park-and-ride lots fill up early in the morning. That makes them effectively useless for mid-day travellers. For two-way, all-day service, there needs to be another way to access the station. Transit-oriented development can play a role – and provides a major opportunity for recovery of regional rail investment – but as the TTC subway demonstrates, the most effective way to deliver large numbers of riders is by seamlessly integrating rail with local bus and streetcar services. That means fully integrated fares – a transfer is an inconvenience, so you should not have to pay more for it. It also means having bus routes designed to connect with stations, additional rail stations to connect with busy surface corridors, and schedules with timed transfers where necessary. The objective is to create the equivalent of a subway backbone for the whole region, serving local trips as much as long-haul. By being a backbone of a broader transit network, regional rail does not just serve residents of neighbourhoods adjacent to stations – it serves everyone in the region.
Focus on Equity
Planning should intend to prioritize improved access to employment opportunities and services for equity-seeking communities. This means reducing travel times, locating additional stations where they would serve communities like the City of Toronto’s Neighbourhood Improvement Areas, and ensuring that fares are not prohibitively expensive. Transit must function as an integrated network, particularly for those who rely on it for all their trips so it is imperative that no transit mode be deemed “premium.”
Integration with Regional Planning
With its region-wide extent and high level of service, regional rail should become a centrepiece of regional planning. In Copenhagen, for example, all substantial office developments must be located within walking distance of a rail station. This would not be possible today in Toronto, given the limited size of the existing rapid transit network, but it could be possible with regional rail. Greenfield suburban developments could be designed around rail stations, creating “15-minute communities” oriented to walking and cycling, rather than following the traditional auto-oriented pattern centred on concession road blocks. Regional rail is the most feasible path to a truly transit-oriented region.
These are key principles not just for a regional rail network, but for transit in general. They run counter to so much of what would-be transit riders are fobbed off with.
“High frequency” really does mean frequent service, not a train now and then when it is convenient to run one. This requires a commitment to both capital and operating costs for the rail network.
“Seamless integration” means an end to assuming that parking will solve all access problems, and that the rail system’s revenue stream is sacrosanct. The concept of a “premium” service as a justification to charge higher fares on part of the network simply does not work if the rail lines are the key, backbone component of a whole. This is an example of how looking at only one aspect – fare revenue – distracts from the larger picture of the potential contribution and value of the rail network for mobility.
As for regional rail and planning, this is a fascinating position for the Board of Trade because it implies that we would dictate where development could and could not occur. Will we also consider network effects of overbuilding at selected “hot” development nodes, and the implications for road congestion and pollution of allowing growth away from transit stations?
The Trillium Network
Yes, it’s a branding exercise, and the Board makes no bones about this. It has a nice sound, and it uses the provincial flower. The name and logo might even survive a change of government. There is a spiffy map.
The key point in this design is that services are through-routed at Union Station and arranged in a manner to avoid conflict between four main corridors: Lake Shore, Kitchener-Don-Richmond Hill, Barrie-Don and Milton-Lincolnville. This is not new, and Metrolinx has talked about the need to reorganize its service in a similar way as part of its expansion program.
Services have route numbers all starting with “T” although the nomenclature could be confusing if a scheduler decides that trains will operate between some other pair of endpoints. The combined service through Union Station is impressive with the intent of a massive increase both in GO’s capacity and its usefulness as a regional and local carrier.
Electric Multiple-Unit Trains
The Board emphasizes that the scale of service proposed cannot be achieved without a wholesale conversion of GO Transit from locomotive-hauled to electric multiple-unit (EMU) trains. What’s that when it’s at home? The term EMU refers to train where every car is powered rather than pulled by a locomotive (electric or diesel) at one or both ends of the train. The technology is identical to a subway car with adjustments for a different operating environment and top speed.
With each car powered, the driving force is distributed to all (or most) axles throughout the train rather than concentrated at the locomotive, and this permits faster acceleration, a key part of frequent, high-performance service. Metrolinx’ own electrification study went down that path too, but current plans focus on locomotive-hauled trains. This preserves the existing fleet of bi-level coaches and allows electrification by the simple move of swapping a diesel loco for an electric. The Board argues that this approach will constrain future service growth by preventing a move to very frequent, fast service that EMUs can bring.
The report includes a substantial discussion on issues such as train size, capacity, platform height, signalling and motive power as examples of changes that make a unified whole of the new network. Much of this is familiar to anyone with knowledge of railway operations and certainly to Metrolinx.
An important change would be to move to single level cars with multiple doors and level platforms to speed loading and unloading just as the subway does today. There is an obvious tradeoff on train capacity, and the current GO model is based on the need to run few trains within the limits of existing rail operations and signalling.
EMUs and the stations they serve would operate much more like subway trains with short station stops, speedy arrivals and departures unlike the lumbering 12-car GO trains. Each train would have a lower capacity, but there would be more of them. The Board does not sort out the net change in capacity versus anticipated demand.
Counterintuitively, using cars with less capacity can increase overall line capacity if it means passengers can board and alight more quickly, and therefore trains won’t need to dwell so long at stations. A GO bilevel car has about 80% more capacity than a Toronto Rocket subway car, which is significant. But if a subway-style train can run every 2.5 minutes while the GO train can only run every ten minutes because of long station dwells, the subway realizes a 300% increase in train throughput. Even taking into account the smaller number of people who can fit on a subway car, that’s still more than twice as much overall line capacity.Getting on the Right Track, p. 33
A typical European single-level EMU, like the new Class 490 trains of the Hamburg S-Bahn, would have a little more than half as many seats as a GO train of the same length (though the difference including standing riders is smaller). That means that doubling frequency would exceed existing capacity – something that is easily achievable with the far higher performance, and faster loading and unloading, of single-level EMUs.Getting on the Right Track, p. 33-34
The problem here is that the Board has latched onto a specific technology rather than addressing the basic “what do we want to do” question. If one presumes the decision to create, in effect, a surface subway network on the existing rail system, this certainly explains what is needed, but there is a cart-before-horse feeling to the proposal.
We have been here before. Remember the surface-subway that was to be SmartTrack? That scheme is mentioned from time to time, possibly to assure Mayor Tory that somebody actually remembers it exists, but SmartTrack is a plan like a tree in autumn whose leaves lie more on the ground than on the branches.
I must come back to that map above: it shows service frequencies, but not capacities, nor does it show how these would differ from what Metrolinx provides today or plans to provide within the next decade, let alone how this might relate to future demand.
Without question, the frequent off-peak service proposed can carry a lot of riders, but whether they will show up is quite another matter. There is no point in talking about all those new riders at noon, evenings or weekends if one does not also talk about peak demand. This may be less of an issue with future commuting patterns, but we are not yet sure, and latent demand could very well backfill work-from-home travel reductions.
Getting Autos Off The Road
Freight and the needs of the trucking industry are often mentioned in transportation reports, but we have been waiting for a “freight strategy” for almost as long as Metrolinx has existed.
[…] trucks find it increasingly difficult to get to their destinations because the area’s highways are at a crawl for much of the day. Regional rail can get thousands of cars off the Highway 401, freeing up space for the trucks that keep our economy moving and help bring our exports to market.Getting on the Right Track, p. 9
A common thread in many transit proposals, and a key part of the Metrolinx Benefits Analysis process, is the premise that getting people on transit will reduce congestion. This is a bogus claim for several reasons including:
- There is a substantial latent demand for road space. If this were not so, we would not live in a region where highway widening projects are part of daily commuting life. A basic rule of traffic planning is that demand will grow to fill available road space.
- The percentage change in demand represented by any individual network change is too small to have a noticeable network effect, and even this benefit must work against population-related demand growth. As I noted recently, the Yonge North Subway Extension will make only a tiny dent in non-transit market share on the Yonge corridor. Autos will continue to handle over 80 per cent of commuter traffic because the transit system does not serve these trips.
- The 401 corridor, and specifically the area round Pearson Airport, a major, national hub, is not paralleled by major transit routes to which trips could be diverted. GO Transit and the subway system already brought in pre-pandemic times the vast majority of travellers to Toronto’s core from outside of the city. This is not the case for most other employment nodes in Toronto and beyond.
- GO depends on a high ratio of parking to riders with about two parking spaces for every three riders. The ratio is lower if one counts GO trips, but each rider makes two trips per day. Most of them would not do so without a parking space. Those “last mile” trips take up road space and drive the need for ever more parking if GO is to gain more riders. This is not a sustainable business model.
A similar argument applies to claimed pollution reduction with trips diverted to transit. Latent demand will offset much of the saving, and most trips will continue to be by car. A shift to electric cars will change this, but that too will be gradual, and does not eliminate the need for commuters to have their own vehicles and the congestion they bring.
Union Station Rail Corridor
A fundamental problem for the entire GO rail network is the layout and usage of tracks and platforms at Union Station. The Board proposes a complete reorganization of this and a GO route structure that would be optimized to the new structure. This is not news, and Metrolinx itself has proposed something similar.
The changes needed and proposed are complex, and to avoid bogging down readers here, I have placed some of this discussion in an appendix.
In brief, three problems with current operations and facilities limit the station’s throughput of trains:
- The track layout and platform assignments require services to cross each other. These create conflicts limiting service frequency.
- Service levels on each corridor vary with the intent of keeping costs at levels appropriate for demand. This makes through-routing difficult and requires trips to terminate at Union. In turn that drives up dwell times on platforms.
- Platforms are capacity constrained both by vertical circulation structures (stairs, escalators, elevators) and by the design of some platforms. Those originally intended for baggage and freight are narrower than those intended for passengers.
The Board proposes new track and platform layouts and the addition of three satellite stations converting the rail corridor into something more like a subway line than a commuter rail system.
The Board proposes that some tracks through Union Station be eliminated and replaced with consolidated platforms. This is similar to Metrolinx’ own view, but with some significant differences:
- The Board proposes that some tracks might be relocated although this would shift them off of the support columns beneath them. This would require structural changes so that columns continued to bear the shifted load.
- Space east and west of Union now occupied by storage yards would repurposed for satellite stations whose spacing is much more like a subway line than commuter rail.
The satellite stations only work with EMU trains where the extra stops would not represent a major delay, although stops there would eat up some of the savings from lower dwell times at Union. The intent is to offload Union so that passengers bound to/from the shoulders of the core would not have to use Union.
A major shortfall of this is that the concentration of destinations (mainly jobs) is much lower at the satellite locations and nearby areas that would be reached by walking or by a short local transit trip. Just because there are four stations in place of one does not produce a massive reduction in demand for Union and its surroundings. Fare integration with the TTC would strengthen demand at Union with its direct link to the subway.
A vital part of the design for the area east and west of Union Station is the replacement of existing yards used to store trains between the peak periods with stations at Spadina, Sherbourne and Cherry. The intent is to spread demand away from Union Station, but this assumes that there is a substantial number of trips that are bound for these locations. Moreover, much of the conversion to EMU operation and replacement of the yards is required before these stations could be built.
This has the air of an attempt to make GO regional rail a substitute for local transit to the waterfront and shoehorn into its mandate a task better left to local transit service. There is already the planned Ontario Line on the north side of the corridor, and the waterfront east extension from Union Station.
There is no discussion of the staging of conversion between the current and future arrangements.
A truly breathtaking example of transit planning hyperbole appears in a footnote explaining the claimed capacity of the new configuration. This completely ignores the many changes that would be needed throughout the network to achieve anywhere near the claimed capacity, notably a massive increase in local transit feeder services.
Yes, the Union track bottleneck could be eliminated, but that is only one part of a much larger picture. Moreover, “solving” this problem only provides better regional network service to the extent that travel is on existing corridors and local transit supports this.
Based on a service pattern of 12 trains per hour on Kitchener, Lakeshore, Stouffville, and Milton corridors and 6 trains per hour on other corridors, using Paris RER MI09 rolling stock of the same length as existing GO trains. At 12 trains per hour, the trains would have a capacity of nearly 47,000 passengers per hour per direction. At the intensity of operations and station design typical of European or Japanese operations, ultimate capacity on the ten-track downtown corridor would increase to as much as one million people, per hour, per direction, compared with GO Transit’s pre-COVID total rail ridership of 215,000 per day.Getting on the Right Track, p. 10, emphasis added.
The Board perpetuates the myth that SmartTrack still exists despite the fact that it has dwindled to a handful of stations, some of which are not on the SmartTrack corridor. The emperor’s new clothes are very, very tattered.
The City of Toronto’s SmartTrack plan builds on these efforts with its aim of fully integrating part of the Stouffville, Lakeshore, and Kitchener corridors into the local TTC transit with frequent, rapid transit service. This means that riders would pay the same fare as the TTC, including free transfers from bus, subway, and streetcars to the new SmartTrack trains. It also means redesigning bus routes where necessary to bring passengers to the stations, as is the case on the subway, and it means additional stations to facilitate those connections and to serve key urban nodes. SmartTrack is an important step toward realizing regional rail, and this report is intended to build on the city’s and province’s plans, highlighting globally proven practices to aid in their implementation. It calls for implementing and going beyond two-way, all-day GO service to true electrified regional rail.Getting on the Right Track, p. 11
There is only one problem: SmartTrack barely exists any more as stations fall off of the map. The planned service level (to the degree one can get a straight answer on this out of Metrolinx) is far below the “surface subway” analogy used when then-candidate John Tory announced it as a centrepiece of his campaign.
We need to stop using that brand name and recognize that the real issue is network-wide improvements to GO Transit both inside the 416 and beyond the city limits. The Board of Trade at least moves the regional rail discussion beyond SmartTrack’s limited scope.
Who Owns What and Limits on Service Growth
The GO network has expanded over the years as the province of Ontario bought up portions of the CNR and CPR networks that served industrial areas that have disappeared or are now served primarily by trucks rather than trains. Portions still in the railways’ hands are parts of their national freight networks and would be given up only with great difficulty. (That also applies to trackage that is not part of the GO network, notably the CPR line running from West Toronto across midtown and out to northeastern Scarborough.)
GO service takes second place to freight on railway-owned lines, and any service improvement requires negotiation both for track time and in some cases additional track capacity add at GO’s cost.
The Board has strongly supported the work of the Connect the Corridor organization, which has long been advocating for the infrastructure upgrades needed to enable two-way, all-day service on the Kitchener corridor without impeding CN freight traffic.Getting on the Right Track, p. 12
The Board is rather cavalier in thinking that these problems can be easily fixed and implies that simply putting enough money on the table will soothe the railways’ concerns. However, they need their lines, and would face huge costs and challenges to duplicate some of them elsewhere.
A February 2017 article on Urban Toronto’s site delves into the freight railways in more detail, and I recommend it to readers to fill out their knowledge of the rail network beyond GO Transit. Note that it is four years old, and plans for a new CN/CP shared corridor west of Toronto appear to be stalled.
Network Reach and Fares
The Board of Trade, while talking about better use of the rail corridors, makes a key point about how the GO network should be truly part of a regional service: fares must be fully integrated with and comparable to local transit.
[…] the Lakeshore corridor between Burlington and Oshawa, the Kitchener corridor from Union to Bramalea, and the entire Stouffville and Barrie corridors are already owned by Metrolinx and have limited freight traffic. Fundamentally, the limited service on these lines is a choice. These routes have as much potential as any subway, but subway-level demand will only materialize if their service levels and fares are made comparable to local transit.
Hundreds of train cars and invaluable trackage sit idle for all but a few hours of the day, five days per week, while carrying loads artificially depressed by fare policy.Getting on the Right Track, p. 12
The Board goes on to points rail corridors that might be added if only they could be pried loose from the owners (mainly CP):
- the CPR line through midtown and northeast to Scarborough;
- the link from Leaside to Union via the Don Branch which Metrolinx now owns, but plans to use as train storage;
- The CPR line to Bolton;
- CNR track on the eastern waterfront;
- a connection from the CPR to the GO Richmond Hill corridor that has not existed for years.
This really has the feeling of throwing every piece of track one can find onto the map, and creating the impression that a regional network could solve many problems. This is reminiscent of SmartTrack which was touted as being the one service that would cure all of Toronto’s transit ills.
Notable by their absence on this and other maps are many routes including Eglinton Crosstown, the Eglinton West LRT to Renforth Gateway and the airport, the Eglinton East LRT to UTSC and Malvern, the Scarborough and Richmond Hill subway extensions and the Ontario Line all of which are in various stages of planning and construction. This grossly misrepresents the gaps in the existing network or the ways in which that growing network can integrate with the proposed regional system.
This also applies to service to key neighbourhoods that are more transit dependent. The GO rail network is not the only game in town, and the Board would do well to remember that little system called the TTC. An example of a trip improved by the Board’s network ignores the effect of the future Scarborough Subway, or the fact that service in the Stouffville corridor will exist in Metrolinx and City plans. The stumbling blocks are fare integration and frequent “subway-like” service.
[…] instead of needing to ride the bus all the way from Malvern to the overcrowded Yonge subway to get downtown, Scarborough residents could take a short bus ride to the Stouffville corridor, which would save them as much as 80 minutes on a round-trip downtown.Getting on the Right Track, p. 13
The problem of transit access is critical in low income neighbourhoods far from good transit, and regional rail can help change this. However, it would be part of a network of services, not the only one, and the Board would do well to remember this. They fall into a trap commonly seen at Metrolinx where their own proposal solves everything, while projects and services on other agencies’ maps count for little.
Currently, the score for transit service and accessibility in high-income areas of the city is almost four times higher than low-income ones. The radically improved access afforded by regional rail would be life-changing for residents of the region who currently struggle to get to their jobs and to other services.Getting on the Right Track, p. 13
There is a huge problem in the GTA that high intensity transit concentrates on getting people downtown while leaving those travelling elsewhere to their own devices.
While commuter service currently effectively serves commuters to downtown Toronto, the area represents about 16% of the region’s nearly 3.5 million jobs.
Regional rail is not just about regional trips. With additional stations, frequent trains, and connections to local transit, it can function as a kind of local subway within cities across the region. By knitting together the region’s patchwork of local transit systems, regional rail can quickly and cheaply improve their efficiency and increase their ridership.Getting on the Right Track, p. 14
That sounds wonderful, but the fact is that the rail corridors are radial and they cannot serve non-core oriented trips without a big help from local transit. There has to be a local service to connect with. That “patchwork” is a direct result of the lack of provincial funding support, and of municipalities running only enough service to get by depending on transit’s budget priority. Even in Toronto, transit does not get all the support it deserves.
An important way to view transit service is to map how far people can travel in a given time on various network configurations. The maps below show how the reach of transit travel is improved by the addition of frequent fast service on the rail corridors. What is missing from these maps is the contribution of new rapid transit corridors such as Eglinton Crosstown.
In a vignette describing the effect of the new network, a rider is described as travelling from Scarlett Road & Eglinton Avenue to Bramalea on 4 local buses, taking 90 minutes, and costing two fares. Many things would improve this journey including the Crosstown extension westward, improved bidirectional service on the Kitchener GO corridor and fare integration, but some connections between routes will still be needed. The big changes would be in reduced fares and shorter wait times on all legs of the journey.
It’s A Network
We take until page 18 to reach a key observation: designing at the individual project and component level will always miss the mark.
Each aspect of the system fundamentally affects every other aspect, so nothing can be designed in isolation. The performance characteristics (particularly braking and acceleration) and capacity of trains (rolling stock) determines how much track capacity will be required, how much electrical infrastructure will be needed, the weights for which infrastructure like bridges and overpasses must be designed, and how stations should be designed. Conversely, station design determines the choice of rolling stock – if the main station has small platforms and narrow access points, smaller but more frequent trains are likely a better approach than large trains. These are only a small number of ways that different aspects of the infrastructure affect each other, necessitating integrated planning.Getting on the Right Track, p. 18
A huge gap in Metrolinx planning is that they review projects in isolation from each other so that their combined effects, benefits or shortcomings are seen individually, not in a network context. “Investment” in new transit (to use the current market-oriented lingo) produces many benefits but these are not always captured. Conversely, projects may be evaluated on a small scale to avoid sticker shock, a problem we have seen in TTC planning and at Metrolinx going back to the original “Big Move” plan that was trimmed to fit within a politically acceptable budget.
In some cases, the apparent cost of a project such as the original Relief Line sandbagged the project for decades because this was not offset by the cost of attempting to handle demand growth without relief from a new line.
The Board’s proposal takes a network view, at least to the point it considers multiple lines, technologies and infrastructure requirements rather than the simplistic approach of, say, the SmartTrack plan.
There is an implicit criticism of existing plans, but with no specifics, in the following statement:
Designing infrastructure without a detailed projected timetable that incorporates a full and integrated understanding of all aspects of the planned operation is a recipe for overbuilding – potentially wasting billions of
dollars on unnecessary infrastructure.
These kinds of unnecessary expenditures can be avoided when a comprehensive infrastructure plan, guided by a detailed operations plan, is created before infrastructure is built.Getting on the Right Track, p. 19
By implication, we are now planning or building some “unnecessary” infrastructure, and by implication the Board’s plan avoids this pitfall. They owe us specific examples to illustrate what they might change and how the network would benefit.
Earlier I quoted a key passage about the order in which aspects of network planning should occur, according to the Board. Here is longer version of this quote:
The German-speaking world has propounded the planning and engineering doctrine of “organization before technology before concrete.” The highest priority is to resolve issues of organization, which includes factors like fare and service integration between agencies. Then, technology, such as better signalling systems and rolling stock, should be improved. The last priority is the building of new infrastructure […]
The second step is to invest in technology where appropriate—using advanced signalling systems and higher-performance rolling stock to squeeze more capacity out of existing infrastructure. Then, only when necessary, new physical infrastructure should be built.
Developing a comprehensive timetable for a system as large and complex as Trillium may be challenging. That is why it may make sense to build one corridor at a time, enabling challenges to be uncovered and lowering overall project risk.Getting on the Right Track, p. 20
Here, the Board stepped on a land mine. After all of the fine words about a massive network transformation, we get a proposal for staged implementation, but without explaining how various network configurations might co-exist or the transitions between them. Moreover, there is no discussion of how service build-up might fit together with infrastructure needs.
We have a truly marvellous plan, but like so much transit in the GTA it would be achieved in dribs and drabs, and probably never at its full extent. The unasked (and hence unanswered) question is what intermediate stages might look like, especially if they turned out to be end states for decades to come.
What Is This Mythical Local Transit?
Through close integration with local transit, Trillium will radically increase transit accessibility in the Toronto Region, bringing people closer together.
This is just one of many explicit and implicit references to the local transit services across the GTA, but there is no discussion at all of the changes in scope and funding required to make the regional rail network work. Just as the subway depends on surface routes as feeder/distributors as well as walk-in trade, a regional network must have good local transit serving its catchment area. Adding stations to existing routes, as the Board proposes, will reduce access distances, but will not eliminate them nor the need for something other than a 15-minute walk.
The Board proposes a study of what factors might draw people to transit to which I respond you don’t know already, but you are redesigning the entire network?
The report talks about the importance of reliable, frequent transit in the context of the rail network, but this applies equally if not moreso to local transit service.
People who have a choice rarely use transit systems where they need to check a timetable. Research has shown that the frequency at which riders can simply turn-up-and-go is about 12 minutes. 10-minute frequency provides the opportunity for a “clockface schedule,” where trains depart at the same time every hour. Australian scholar Paul Mees has referred to a “network effect” to explain that when all routes are sufficiently frequent, it becomes possible to conveniently make transfers between routes. This has a compounding positive effect on the utility of a transit system. Instead of only being able to reliably reach destinations on the line that happens to run past the rider’s origin, a network that facilitates connections enables the rider to go from anywhere to anywhere in the region.Getting on the Right Track, p. 25
That is precisely the argument behind so much transit advocacy work: a line on a map is not enough if the bus shows up infrequently and irregularly. A network does not have “reach” if waiting for service consumes the lion’s share of journey times and compounds because of dubious connections between services.
The Board notes that GO stations are getting expensive. Part of this is due to the high cost of parking structures that should, in theory, not apply to a truly “transit oriented” area.
While some new GO Transit station projects are very impressive facilities, their high cost is not always necessary – especially for stations that are anticipated to have relatively modest ridership. The five proposed City of Toronto stations on the GO network are planned to have an average cost of $239 million each, excluding additional amenities like pedestrian improvements requested by the City, while the new Bloomington GO station cost $82 million. In Montreal, a recently completed commuter rail station cost $14.2 million,17 while in Italy, typical regional rail stations cost between $3.8 and $9.2 million. A group of seventeen new regional rail stations in Germany averaged $3.61 million apiece.Getting on the Right Track, p. 44
The Board trots out Transit Oriented Development, another planning fantasy that sees more use in studies and reports than in actual practice. Development may occur around transit, but because of market forces, not because planners put it there. Also, the idea that “there’s gold in them thar hills” is a common misconception when one compares the revenue available from developments and the high cost of stations in an urban area. Yes, stations are easier and cheaper to build for a surface rail line but property will be expensive and stations should be integrated with development, no simply be a little building next door.
Again the Board cites local transit feeders as a benefit for development at stations, but neglects to mention the scale of change needed.
While most riders on a Trillium Regional Rail Network will likely reach their stations by connecting bus routes, as on the suburban stretches of the TTC subway, transit-oriented development will still be an excellent means of augmenting ridership at a station while raising revenue.
While all development around the network’s stations will be desirable, employment will be of particular benefit to ridership. Firstly, it is a way to generate bidirectional traffic on lines, improving infrastructure utilization. Secondly, office employment has higher density than residential, as office buildings contain more people per square foot, and their occupants are more likely to be making commuter journeys. GO Transit currently has acres of parking around most of its stations. By shifting access to the station primarily to surface transit through integrated fares and increased frequency, as well as improving pedestrian and cycling access, it will be possible to monetize much of that land through high density development. Local transit feeding regional rail trains can facilitate “missing middle” development throughout the region.Getting on the Right Track, p. 45
The Board foresees a day when local-to-regional connections are simple because of station placement, but this ignores the fact that rail corridors are where they are for historical reasons and they are not going to move. The paragraph below ends with the observation that on-board displays would tell riders about connecting services at stations. If those services operated to the same frequent service standard, on board displays for connections would not really be necessary.
To be able to serve as the backbone of the regional transit network, it is essential that regional rail be fully integrated with local transit. Bus routes must be designed, and regional rail stations situated, to facilitate intermodal transfers. For example, instead of locating stations behind acres of parking lots, additional stations could beGetting on the Right Track, p. 47
designed with direct pedestrian access to and from the sidewalks of major arterial roads, minimizing the need for buses to divert from their route or for passengers to walk long distances to make transfers. More frequent, urban-style stop spacing – potentially with stations spaced as close together as subway stations when the circumstances justify it – is possible with faster-accelerating modern electric multiple-unit trains. This will also minimize the need for buses to divert from their most direct routes. Stations should also be designed to integrate seamlessly with other mobility modes, such as through the inclusion of bike storage, scooters, and facilities for ride-hailing pickup and drop-off. Integration with local transit will allow people to take a short bus trip to the station and then the train for the rest of the journey, rather than needing to use a slow local bus for the entirety of long journeys. Trains can provide realtime information about connecting routes at upcoming stations.
Union Station Rail Corridor Capacity
Here is a diagram of the current track layout at Union, and this diagram does not clearly show all of the links where tracks cross on the diagonal.
Metrolinx is proud of this complex track layout (the diagram below does not clearly show all of the links) although it is relic of intercity train operations unsuited to high capacity regional rail services.
The proposed new layout and track allocation is shown below. Note that the colour scheme in the key is out of sync with the chart, and the route identities (“T1” etc) should not be confused with track and platform numbering.
There is a very strong argument that Union could handle more trains/hour than it does today. The charts below show how many trains pass through the station eastbound and compares this with major stations in Paris and Berlin. (Note: Lakeshore West numbers are only for trips ending at Union. Through trips are counted with Lakeshore East.)
The high throughput at the European stations is a direct result of each line operating independently of others without conflicts combined with signal and train control systems that support very frequent service. Achieving this on GO requires a complete change in the train fleet, signalling and Union Station’s layout. This is not an overnight change and it affects the entire network.
The Board fudges the question of how Union Station might be reconfigured, and frankly I have to ask why they didn’t simply go to Metrolinx who have already pondered this question.
These changes pose some challenges, and it is unfortunate that they were not implemented at the same time as the broader reconstruction of the station. Still, they are absolutely essential if the goals of regional rail are to be achieved. It should be possible to relocate tracks where needed. Most of the existing tracks are situated directly above support columns. If it is necessary to move the tracks, it should be possible to support them with a beam running between the two adjacent columns. If shifting the location of tracks should prove to be entirely impossible, it would still be feasible to simply remove some tracks and build expanded platforms overtop.Getting on the Right Track, p. 42