The UP Express is a rare recent Toronto transit success, but it’s not headed in the right direction.
Today’s article is split in two — the first half details the history of UP. Read on for my discussion of how I think the service should evolve.
When the Union Pearson (UP) Express first opened in Toronto, coverage was positive — nice trains, fast & frequent service, onboard WiFi, and attractive stations — it had everything. And then there was the famously very low ridership, in large part because of the astronomical fares.
Fortunately, Metrolinx was ordered by the government to drop the fares, and the line’s ridership (and notably also it’s financial performance) improved significantly.
But since then, it seems that service that moves more than 10,000 riders every day has gotten little attention in the media besides claims that the service had primarily become one for “commuters”. My anecdotal experience riding the service and data I have seen previously doesn’t actually suggest that most riders are commuters, but if they are… so what? It’s extremely weird to “purity test” a transit service to make sure it’s serving a particular group of people — more people on transit is an undeniable win and it seems obvious that many commuters would rather use an all-day, every 15 minute train service than the much less frequent GO service that stops at two of the UP’s four stations. It’s not at all uncommon for transit projects in cities around the world to be successful for unexpected travel markets — it should simply lead to a recalibration of models and expectations.
Even calling the UP a true “airport express” isn’t necessarily all that accurate, given that it makes a number of intermediate stops (with at least one more on the way), something which most airport express services do not (Arlanda Express, CDG Express, Heathrow Express, KLIA Express, and most service on Narita Express do not make intermediate stops). It’s more like an early taste of what rapid regional rail to Pearson could be.
Now, astute readers will note that ridership of over 10,000 a day is not particularly good for a rail service in Toronto, and this is true: the Sheppard subway, frequently criticized for being underused moves roughly five times that. But it is worth considering that for one, the UP runs only about a third as frequently as the Sheppard line, and that each UP train can move less people; the UP is also providing far more passenger kilometres of service, meaning if it’s shifting people from cars, it’s having a bigger impact. Ultimately, given the UP is quite a bit newer than Line 4, I consider it being in the same ballpark of capacity utilization as being quite respectable.
The ridership trajectory for UP is also very good. For one, Pearson airport has not fully recovered from the pandemic passenger wise, but when Metrolinx was running the UP at its reduced 30-minute service level late in the pandemic, trains were packed. My experience riding in the last few months has also been a lot of very busy trains and lots of traffic at the airport and Union station terminals. It’s also important to remember that given population growth in the GTHA, not to mention the growth of low-cost air carriers, total passenger numbers are going to continue to increase dramatically in coming years, which should translate into additional ridership growth (especially with more price-sensitive low-cost carrier passengers).
And then there are the other stations on the line, I detailed in a recent post that the long promised indoor connection between Bloor station (UP, and GO) and Dundas West station (TTC Line 2) will probably have a positive impact on ridership because of just how much better making the connection will get.
On top of that, there’s also Mount Dennis (the terminus of Line 5 Eglinton), which will make the UP a really great option for getting to midtown and other parts of the city that are not downtown or downtown adjacent.
Unfortunately, despite the positive headwinds, the service has had a fair number of issues in the last few years — or at the very least has felt a bit neglected.
For one, the trains have continued to be pretty problem plagued (they have had engine, door, and brake issues over the years) which was part of why service coming back to every 15 minutes was delayed. The wayfinding systems (station announcements, names of stations on the screens in the train) on the train have also not been working the last several times I’ve taken it — this really sucks because the wayfinding was actually quite good when the service initially opened. I’ll also note that track conditions have also been poor on sections of the line just south of the crossing of Highway 401 — I wouldn’t normally bring something like this up, but when your out-of-town visitor looks afraid the train might derail and the door for the bathroom is opening and slamming shut, it’s not good!
On top of all of this, the future of “UP” as a distinct service is unclear.
Several years ago an article came out discussing Metrolinx planning for “GO trains [servicing] Pearson from stations on the Lakeshore East and Stouffville lines”, which seems like a bad idea. Even if it is possible for bilevels and locomotives to navigate the guideway into Pearson (the UP DMU cars and bilevels are similar lengths), they would surely have to do so even more slowly than the current sets (and the speed with the current sets is not great!), and it would be a logistical headache — both because you wouldn’t be able to run a long consist (the platform will currently only fit three cars) and because the platform height is different, requiring significant construction at Pearson. Now, to be clear, there are certainly good elements to this plan — through-running airport trains to more destinations would be very useful, and moving the airport trains to “main” Union Station platforms while wasting the very nice existing platform should allow for more frequency.
But, given I rather like the UP service and I want it to succeed, I decided to sketch out my own plan for how the future of the service should look.
A reasonable reader might ask: “what is the role of UP in the future with the planned extension of the Eglinton Crosstown to Pearson Airport?”
I think this is a really interesting question, and fortunately I actually think the services are very complementary. The first phase of the Crosstown to open will already provide useful connections to UP from midtown so there is synergy there, but more importantly I don’t see the western extension really “competing”. For one, unlike say the Piccadilly line to Heathrow, Chicago’s Blue line, or Line 2 in Shanghai — Line 5 in Toronto does not connect the airport and the city centre. For two, Line 5 will just be a whole lot slower, while you could get from Union to Pearson using Line 1 and Line 5 once the extension is complete, that journey would likely take close to an hour, compared to UP with a sub-30 minute journey.
The way I see things, the purpose of the two rail links breaks down like this:
Essentially, long term, I see UP as providing the regional connecting service, while Line 5 provides a service that connects you to destinations mostly within the city of Toronto. Of course there will be some overlap, but the division fits quite nicely.
My Plan for UP
Service
The first thing that I think needs to be improved with UP is the service level. Right now service is every 15 minutes all-day, which for one doesn’t align with the higher demand during peak periods and for two is less frequent than the upcoming REM link to Montreal Trudeau airport (every 10 minutes in the peak) or the Canada Line to Vancouver International Airport (every 6 minutes in the peak). It just seems nonsensical that UP has a lower peak service level than Vancouver or Montreal’s services despite serving a bigger city and a much bigger airport. There’s no reason UP should not run every 10 minutes at peak even with Line 5 extending to the airport, and even more frequently if demand is sufficiently high. You could even have some services run “super express” direct to Union.
The other big thing that needs to be fixed with the service is the span. Right now the UP starts operating around 5 AM and ends at 11 PM, which is just not good enough. Clearly service should run at least a few hours later, as late flights (or delayed flights) certainly can arrive well after 11 PM, and that’s before you consider potentially slow baggage unloading or customs screening. There’s also the fact that a service that starts at 5 AM or 6 AM on weekends is probably a bit too late if you have a very early flight, at least if you are a cautious traveller.
Once you realize that you probably want the last train to be sometime after 1 AM, and that the first trains should already be running again by 5 AM if not sooner, I think there’s a case to be made that you may as well run an hourly service overnight (which could probably be done with a single train) and fill the small gap. Airports are 24-hour activity centres (particularly big ones like Pearson) and the corridors UP runs over certainly have the track capacity to manage one train per hour overnight (a nice side benefit of having platforms on more than just two of the tracks).
Trains
Of course, these improvements are not possible with the current unreliable trains, which we do not have enough of to run 15-minute service today with all full-length three-car trains. The perfect solution to this problem in my opinion is to expand the fleet, but doing so with DMUs that were produced by a company that has left the North American market does not make sense.
My preferred option is a new fleet of high-floor trains, and given we have some lead time, I think it would be wise to have them be electric (EMUs to be precise). I’d also suggest shorter cars more in line with the length seen on the TTC subway, which might be able to navigate the curves into Pearson more quickly, and could have a slightly higher-capacity layout and a few more doors (this would require swapping the screen doors at Union at Pearson, but that should be trivial). If you want a specific train option, we could probably grab something designed for the UK like Bangkok did, or something like a high-floor FLIRT from Stadler. And this actually works pretty well — the UP Express line was designed to be electrification ready (you can see pole bases on the guideway to the airport), and this was promised by past governments, so making good on it is the right thing to do.
You could also go for low-floor EMUs, and these would have the advantage of easily being able to interoperate with GO infrastructure, enabling an easier extension of service east of Union Station and a “starter” fleet of EMUs for GO to get accustomed to. However, the issue with this idea is it requires serious construction at Pearson that likely couldn’t be completed without a service shutdown, which makes me lean in the high-floor direction.
I don’t think bilevels to the existing Pearson station make sense — you don’t really get more capacity with that option, and since your train would need to be short, you compromise capacity on lines east of Union just as much as you would with EMUs. If we really want full GO trains to serve Pearson, they should be full GO trains on a new through-running route (which we should be planning and protecting for) that continues onto the Kitchener Line. Modifying the current infrastructure with all of its problems seems like way more trouble than its worth.
With EMUs, you could reduce journey times and increase frequencies (I think you could probably do every 10 minute service or maybe even 7.5 minute service without relocating the platform at Union) to an extent that between the UP and Line 5 you would likely have enough capacity for at least a couple of decades.
Stations
Now, the final thing to consider changing are the stations. As I have already discussed, Mount Dennis will come online in the next few years, and I’m personally of the belief that it’s worth trading Weston station (which should have frequent GO service by the time any of this is done) for a new station at Woodbine. Not only would that new station serve the major TOD and entertainment destination that is forming there, but it would also connect naturally to an extended TTC line 6, which would give all three intermediate UP stations connections to east-west running urban rail services — which feels nice.
Of course, I do like the idea of extending services beyond Union, and despite what you might think, this isn’t really an issue with high-floor trains. Given UP is an express service, it shouldn’t be stopping at every station and so you simply decide what stations it will stop at and build small high platforms like those seen at Weston and Bloor today. Given this constraint, the obvious station to extend airport service to is East Harbour, especially if it ends up being the “Canary Wharf of Toronto”. Beyond that, I don’t feel strongly that airport service needs to be extended along the Lakeshore East or Stouffville lines (you could probably make a case for Kennedy or Guildwood), so that is something I’d probably limit to a plan that uses low-floor EMUs.
What you get:
So in summary, what this plan looks like is this: We decide to upgrade the UP service, we tender contracts to electrify the corridor (I am presuming we already have an electrified route to one of GO’s yards, which I hope is a given), and we order a new larger fleet of high-floor electric trains. All of this could easily be completed within five years (we probably should do it by ~2030 lest capacity becomes a real issue) — at which point UP can run far more frequently (and possibly faster), 24/7/365, and all the way to East Harbour — with 3x+ the capacity it has today. It would provide a first foray into proper electric regional rail with modern trains for the region, and it would finally pull the UP Express out of the hangar and prepare it for liftoff.
Leave a Reply