The Great Canadian Transit Thread

[Doady]

Originally Posted by isaidso

It's only good if we compare our cities to the bottom of the barrel, the United States. 10-20% public transit share isn't anything to brag about. It's good that we're not last but it makes far more sense to compare ourselves to the best. That would be European/Japanese cities and in north America, Mexican cities.

Canadians need to move beyond their perennial US tunnel vision.

Transit mode shares in Canadian cities is on par with similar-sized cities in the UK and France:

http://forum.skyscraperpage.com/showthread.php?t=236523

Lack of transit isn't the problem in Canada, it's the lack of walking and cycling that's the problem.

[MalcolmTucker]

So, we have a spectrum:

Own Work.



We can conclude that with computer control, the elimination of level crossings, and some other measures (speed restrictions in residential neighbourhoods), Edmonton's capital line should be able to get a 5 KPH boost. Ottawa with more aggressive timing would probably be able to converge with Montreal. Bloor Danforth with automatic control should be able to increase its average speed by a lot. Calgary with the downtown tunnel will get closer to the trend, but never reach it (a segment will within our lifetimes remain above ground downtown on one line).

[ssiguy]

Thanks for the stats guys, really very interesting.


These stats are great for comparisons but like all things statistical, the devil is in the details. Going from the 2 main cities I have lived in all my life, Vancouver and London Ont, it is hard to compare the 2. For instance Translink serves all of the metro area and so is an excellent reflection of how well the city is doing on the transit file. London on the other hand is having itès per-capita ridership levels reflected by 20% lower than they actually are. This is because the stat is for a London metro of 474k but London transit only serves the city itself with 400k because as soon as you leave the city limits, it is basically small town or rural so there is no transit service.

[Metro-One]

Originally Posted by Aylmer

Regarding the "surface vs grade-separated" debate, I've been thinking about this in the context of Ottawa's urban main streets. The way I see it, there are benefits to both approaches:

- Grade-separated (GS)

Pro: Faster vehicle speeds (because fewer stations, exclusive ROW)
Con: Longer walk to station (because of fewer stations), longer pedestrian access times to platforms (because not at street level)

- At-grade (G)

Pro: Shorter pedestrian access times (because at street level and more stations)
Con: Slower vehicle speeds (because more stations, non-exclusive ROW)

This means that GS systems are more useful when you're going long distances because you spend a greater percentage of your time in the transit vehicle as opposed to walking there. For example, you have to go 20km and you either have a 20km/h tram 5 minutes from your door/your destination or a 40km/h subway 10 minutes from your origin/destination. The added 10 minutes of walking is more than offset by the 30 minute reduction of in-vehicle time.

G systems are better for short distances because a bigger share of your time is likely to be spent walking to and from stations. In the same situation, travelling 2 km by subway would total 26 minutes but the tram would get you there in only 16 minutes.


I'm assuming that surface LRT (like Edmonton's Valley Line) runs about 20km/h with stop placement every 500m and that a subway/skytrain runs at 40km/h with a stop every 1000m and require 1.5 minutes to get from the station entrance to the platform (ascending 15m by escalator is about 40 seconds, plus walking on the concourse). I'm also assuming the average walk applies on both ends of the transit trip. With these parameters, surface transit is faster for trips under 8.5km, and GS systems faster above that.

For those interested, here's my formula: (X km/20kph*60min)+10min walk=(X km/40kph*60min)+20min walk+3min access, where X is the point at which GS is faster than G.


There are a few interesting implications as well as some limitations to this model.

For implications, this is an indication that for urban transit corridors where people will generally be walking/cycling to stations, a subway might actually result in longer trips than a surface alternative. I'll soon experience this phenomenon; Currently, I'll often take Ottawa's Transitway BRT for 1-2 stops downtown just because I can easily hop onto a bus from the curb (and because I'm lazy). Once LRT opens, the trains will be faster, but once I factor in the time it takes to get down to the platform and back up again, it'll likely be faster for me to walk those same distances.
So for urban corridors, good surface transit will result in the fastest trips even if the vehicles are slower. Ottawa's Bank/Rideau, Montreal's St-Laurent, and Toronto's King are good examples of this; most people on this corridor walk from their house to the street and have their destination within walking distance too.

However, there are limitations. If this is a transit line which mostly relies on transfers as opposed to walk-in traffic, the difference is largely evened out since there's no difference in access time to the line on one end of the trip. In that case, surface transit is only faster for trips under 5km (excluding the time on the transit line to/from which passengers transfer). The YUS Line is an example of this; although it has walk-ons, I believe that most of the passengers transfer onto the line and then walk to their downtown destination.

And of course, if most passengers transfer on both ends of their trip on a line, GS is always faster. Montreal's Yellow Line is an example of this; the lion's share of passengers transfer by bus on one end and transfer to the metro on the other.

So to sum it up, surface transit is faster under:

Walk-transit-walk: <8.5km
Transit-transit-walk: <5km
Transit-transit-transit: [GS always faster]

So instead of debating which one is always better in all cases, I think that it's important to look at the context and to choose the best mode accordingly. In certain contexts, a faster vehicle may end up in higher trip times for most passengers. And in others, surface options, although they may be attractive from an urban design standpoint, they might not actually be the best option in terms of getting people where they need to go.

Interestingly thought out and there are times where at grade is preferable, but I find those cases for most projects in Canada to be few and far between since we are still building back bone systems and not relief / local complimentary lines.

For me on most routes it makes far more sense to have grade separation for the mid to long commutes while still maintaining a bus service for short commutes. At grade LRT often attempts to combine the two but more often than not ends up with the worst of both worlds rather than the best.

Also your time estimates don’t account for train frequencies (how long people wait at stations for the next train). Also at grade LRT (especially road grade) is far more susceptible to delays.

Then there are the capacity and safety issues.

And for some projects, such as the Langley extension, even station spacing barely favors at grade (the proposed LRT was to have 9 stations over 16 km, the Skytrain will only have one less at 8).

[nname]

Originally Posted by Metro-One

Interestingly thought out and there are times where at grade is preferable, but I find those cases for most projects in Canada to be few and far between since we are still building back bone systems and not relief / local complimentary lines.

For me on most routes it makes far more sense to have grade separation for the mid to long commutes while still maintaining a bus service for short commutes. At grade LRT often attempts to combine the two but more often than not ends up with the worst of both worlds rather than the best.

Also your time estimates don’t account for train frequencies (how long people wait at stations for the next train). Also at grade LRT (especially road grade) is far more susceptible to delays.

Then there are the capacity and safety issues.

Another thing is the consistency of grade-separated (automated) system.

For instance, when I worked at Metrotown almost 10 years ago, I leave office at exactly 6:06 everyday. I timed myself to reach the SkyTrain platform by 6:10:30 and there will always be a train waiting. The train will arrive at Braid at exactly 6:30:15, and it gives me 45 seconds to transfer to my bus at 6:31. Did that every weekday for 6 months, and only missed the connection 4 times out of more than 100 attempts.

Would anyone dare to try that on a LRT system?

Also, LRT station may have shorter (by distance) access, but may not necessary be shorter by time. If the station is located in the middle of a street, the wait at the light to cross the street will be much more unpredictable (and potentially longer) than just going up and down the escalator.

Originally Posted by Metro-One

And for some projects, such as the Langley extension, even station spacing barely favors at grade (the proposed LRT was to have 9 stations over 16 km, the Skytrain will only have one less at 8).

The proposed LRT have 9 stations because they need to build a new one for King George station. SkyTrain had 8 because they don't count existing station

[milomilo]

Originally Posted by Metro-One

Interestingly thought out and there are times where at grade is preferable, but I find those cases for most projects in Canada to be few and far between since we are still building back bone systems and not relief / local complimentary lines.

For me on most routes it makes far more sense to have grade separation for the mid to long commutes while still maintaining a bus service for short commutes. At grade LRT often attempts to combine the two but more often than not ends up with the worst of both worlds rather than the best.

Also your time estimates don’t account for train frequencies (how long people wait at stations for the next train). Also at grade LRT (especially road grade) is far more susceptible to delays.

Then there are the capacity and safety issues.

And for some projects, such as the Langley extension, even station spacing barely favors at grade (the proposed LRT was to have 9 stations over 16 km, the Skytrain will only have one less at 8).

Yeah, in an ideal world where money was no object cities would have complimentary systems with streetcars providing short distance transit with metros and commuter rail on top etc. But since a Canadian city is likely to build only a line every decade or so, we can't afford to waste money on the poorer options. The LRT lines we build here are rarely/never of the type of line that provides that short distance service, but city council types often see this sort of line elsewhere and want it in their city. And we can see from the US that streetcar systems aren't very successful in NA cities. It would be interesting to see an example of a new Canadian streetcar, but that isn't likely because we know they aren't cost effective.

[Reecemartin]

Originally Posted by Metro-One

So now these LRTs are tram ways? That sounds like a pretty poor choice for a back bone system.

I am not fully against LRTs with at grade segments, for example Vancouver’s proposed Arbutus LRT, but that will be a more local relief line, not the primary line for the general area.

I agree that reliable permanent transit does want to make people want to live downtown / near stations, but in an urban setting far more people will ditch their cars / choose to do so if that system can compete with driving speeds and have high frequencies.

If this approach to LRT is so ideal then why does Vancouver’s lowest capacity line, the Canada Line, have a higher daily ridership than the entire Portland LRT Max system? (which has nearly 100km of rail vs. the Canada Line’s 19km...)

Edmonton isn’t building some complimentary system to an existing metro network, this is the heart of their system, and to handicap it like this is a major mistake.

There is a lot of truth to the saying, you get what you pay for.

Is grade separation more expensive? Yes. Is it more versatile and likely to generate higher ridership levels? Yes.

So what is your benchmark population for grade separation? If I recall you are in support of the Broadway Subway? Why doesn’t Edmonton deserve similar?

This will probably be the last I comment on the Edmonton LRT because I am honestly not that invested in their projects (unlike Vancouver where I will fight to the end for grade separation) and I also don’t want to get into a circular argument with you. We may just have to agree to disagree.

PS, besides large trucks in need of a tune up (or full replacement) I actually haven’t really noticed car exhaust to the level of actually bothering me in years along a major downtown street

I'm not sure why the argument against LRT must be made via the US, Calgary has much higher ridership as was mentioned.

Originally Posted by milomilo

Don't tell me to do my research when you're purposely ignoring the point of my posts. Yes, a low floor vehicle could be automated, but you'd have to grade separate the whole system, and if you've done that what benefit would low floor vehicles give? Absolutely nothing. That's what Ottawa has done, built a system that gives zero benefit to low floors yet used them anyway, and also has the stupidity of wasting money on unnecessary operators.

Calgary is slightly different, we are building a line where the claimed benefits of using that technology have been touted, but will have almost no value as the line is almost entirely grade separated and in it's own ROW. The platforms being a few inches lower does nothing. And it will only, maybe, be a small future benefit if they run a future phase down the middle of a street. A scenario everyone hopes they don't do.

You keep talking about Ottawa "as if" is will have ATC but it does have ATC, and as was mentioned there is a good reason why they went with low floor vehicles . . .

Originally Posted by Metro-One

So tonight Surrey officially cancelled the LRT plan. All preparation work is being suspended as of tomorrow and Surrey will begin talking with Translink for Skytrain.

Every community this project directly affects is in support of this change (Surrey of course and both Langley City and Langley Township).

Fantastic day. Now, all we need is for the mayors’ council’s approval. Given the optics above, not allowing this to pass would though a major wrench into regional cooperation and planning.

This is not fantastic news, it sounds just like the TYSSE in Toronto (and this comes from someone that grew up in Langley where the extension is potentially going). Once again we are choosing to serve commuters in car-dependent suburbs rather than serving the high-density core areas that are much closer to being walkable.

Virtually nobody in Langley takes the bus, all the dense areas of housing are served every 30 minutes at best currently. Choosing to build to Langley over the core of Surrey makes no sense. I have serious doubts that you will see any significant off peak ridership.

[Reecemartin]

Another thing in the favor of the OLRT is the very nice lines of faregates, for some reason even some of the busiest skytrain stations only have 4-5.

[MalcolmTucker]

Originally Posted by Reecemartin

This is not fantastic news, it sounds just like the TYSSE in Toronto (and this comes from someone that grew up in Langley where the extension is potentially going). Once again we are choosing to serve commuters in car-dependent suburbs rather than serving the high-density core areas that are much closer to being walkable.

Virtually nobody in Langley takes the bus, all the dense areas of housing are served every 30 minutes at best currently. Choosing to build to Langley over the core of Surrey makes no sense. I have serious doubts that you will see any significant off peak ridership.

Important to remember that the plan is SkyTrain with BRT along the potential, now rejected, LRT corridors, with service just as good, and in just as exclusive of lanes.

The Surrey LRT was also going to be on the slow side, 21.4 kph average speed. It only saved a travel time of 1 minute on average compared to the existing service. At least the SkyTrain + BRT has the potential to actually improve service.

[milomilo]

Originally Posted by Reecemartin

You keep talking about Ottawa "as if" is will have ATC but it does have ATC, and as was mentioned there is a good reason why they went with low floor vehicles . . .

If you're going to reply to my posts please take the time to read them properly. You are making the deliberate choice to not understand my points.

The only real benefit of low floor is the small reduction in platform height allowing better street integration. As I said, since the Ottawa line is fully segregated allowing automation, the low floors offer no operational benefit, at the cost of poorer train interiors. If it truly was cheaper (I find this dubious since the whole thing was rebuilt, but I won't argue that), then I concede that is a fair reason.

[nname]

Originally Posted by Reecemartin

Virtually nobody in Langley takes the bus, all the dense areas of housing are served every 30 minutes at best currently. Choosing to build to Langley over the core of Surrey makes no sense. I have serious doubts that you will see any significant off peak ridership.

No one in NE Coquitlam took bus too, before TL ramping up service in the mid-2000s.

The route with best service were every 30min peak, and hourly off-peak. Only 3 routes had service on Sunday.

And the worse is pretty much comparable to present-day 748/749.
https://imgur.com/qOzH4wg

[CityTech]

Originally Posted by MalcolmTucker

So, we have a spectrum:

Own Work.



We can conclude that with computer control, the elimination of level crossings, and some other measures (speed restrictions in residential neighbourhoods), Edmonton's capital line should be able to get a 5 KPH boost. Ottawa with more aggressive timing would probably be able to converge with Montreal. Bloor Danforth with automatic control should be able to increase its average speed by a lot. Calgary with the downtown tunnel will get closer to the trend, but never reach it (a segment will within our lifetimes remain above ground downtown on one line).

It should be noted that initially, the Confederation Line will only cover the central area of the city where stop spacing will be higher. It's average speed, and average stop spacing, will both increase when Phase 2 opens. With Phase 2, the city has said it expects travel time from Moodie to Trim (about 38km and 27 stations) to be about 50 minutes, so average speed will increase to the mid 40s and stop spacing to about 1.4km, similar to the SkyTrain on that chart.

[MalcolmTucker]

Originally Posted by CityTech

It should be noted that initially, the Confederation Line will only cover the central area of the city where stop spacing will be higher. It's average speed, and average stop spacing, will both increase when Phase 2 opens. With Phase 2, the city has said it expects travel time from Moodie to Trim (about 38km and 27 stations) to be about 50 minutes, so average speed will increase to the mid 40s and stop spacing to about 1.4km, similar to the SkyTrain on that chart.

Pretty close the trend line! I'm surprised it is that accurate. Probably the only reason it is slower is at least in my experience the speed of doors opening and closing on LRTs increases dwell time. Also without intrusion detection, the trains even under ATC will have to slow down earlier.

[White Pine]

Originally Posted by MalcolmTucker

So, we have a spectrum:

Own Work.



We can conclude that with computer control, the elimination of level crossings, and some other measures (speed restrictions in residential neighbourhoods), Edmonton's capital line should be able to get a 5 KPH boost. Ottawa with more aggressive timing would probably be able to converge with Montreal. Bloor Danforth with automatic control should be able to increase its average speed by a lot. Calgary with the downtown tunnel will get closer to the trend, but never reach it (a segment will within our lifetimes remain above ground downtown on one line).

Hey, just curious, do you know the stats for the Yonge/University line? I suspect the Yonge section is quite fast, although it might be brought down with the more curvy University section, plus the new addition stops are fairly close together in some parts. Does it perform better or worse than B/D?

[MalcolmTucker]

Originally Posted by White Pine

Hey, just curious, do you know the stats for the Yonge/University line? I suspect the Yonge section is quite fast, although it might be brought down with the more curvy University section, plus the new addition stops are fairly close together in some parts. Does it perform better or worse than B/D?

At 1.02 km between stations, it should be a bit worse. 79 minutes to complete the U. 29.47 kph, probably perfectly on the trend line.

[Innsertnamehere]

The Yonge line is really slow around the U downtown with the dense stop spacing. It flys north of Eglinton though.

[McC]

Originally Posted by MalcolmTucker

So, we have a spectrum:
).

Cool work, thanks for sharing. Does your Translink data include West Coast Express? The stop spacing seems high to me.

[Chadillaccc]

Originally Posted by McC

Cool work, thanks for sharing. Does your Translink data include West Coast Express? The stop spacing seems high to me.

The Skytrain network is 79.6 km long and has 53 stations.

79.6/53 = 1.502 km distance between stations

[osmo]

Originally Posted by milomilo

If you're going to reply to my posts please take the time to read them properly. You are making the deliberate choice to not understand my points.

The only real benefit of low floor is the small reduction in platform height allowing better street integration. As I said, since the Ottawa line is fully segregated allowing automation, the low floors offer no operational benefit, at the cost of poorer train interiors. If it truly was cheaper (I find this dubious since the whole thing was rebuilt, but I won't argue that), then I concede that is a fair reason.

Low floor vehicles are the reality of Ontario accessibility laws and not much else. Eventually, all transit in Ontario will have to be accessible (TTC is slowly putting elevators in evrey station in the shower way system for example).

[corynv]

Originally Posted by osmo

Low floor vehicles are the reality of Ontario accessibility laws and not much else. Eventually, all transit in Ontario will have to be accessible (TTC is slowly putting elevators in evrey station in the shower way system for example).

Low floor is only a requirement when there's street level boarding. As long as a train system has level boarding at a station it doesn't matter if it's a high floor (subway/metro) or a low floor (LRT/Streetcar etc) train.