Fantasy Map for Toronto Transit

By the Numbers: The Case for the Downtown Ottawa Transit Tunnel

It always amazes me how people can form such intense opinions about something they know so little about. After reading numerous discussions on forums like the Ottawa Citizen’s editorials, it never ceases to amaze me how people always think they know better than the people who have dedicated their lives to understanding their field, and knowing what they’re talking about.

Truthiness is all to prevalent in our society. The idea of “my opinion is just as valid as anybody else’s”, even when in reality your opinion is based on little to know actual evidence or research. Because I feel like it’s the right answer, experts be damned. I know better.

What is even scarier is that when given the information that directly proves why their “great idea” won’t work, they simply dismiss it and go on believing what they’ve always believed. It is one thing to be ignorant of the facts, but it is another to be presented with the facts, and not only choose to continue to be ignorant of them, but to refute them entirely.

The fact is that the City of Ottawa requires a downtown transit tunnel in order to maintain effective service, not only at present service levels, but also at predicted service levels in 2031. I know, a lot of people can say “those are just predictions, they may not happen” or “you can manipulate the numbers to say whatever you want to say”. This is true. However, even when the ridership forecasts were being done for the 1^st incarnation of the LRT plan, possibly one of the most deeply flawed transit plans ever conceived, the numbers clearly leaned that a tunnel would “eventually be needed”. This “eventual” they spoke of was 2031. Even for plan that didn’t call for a tunnel, the numbers still said it should have, it was the politics that said otherwise. As a result, the tunnel was excluded.

What that plan neglected to mention was that even at current service levels, the proposed at-grade LRT through downtown would have been treading water, at best. The increase in capacity would have been not nearly enough to justify the $881 million price tag, especially considering that the ridership numbers warranted a substantial portion of the system being replaced less than 20 years after opening.

Without getting too technical, let me break the numbers down for you. The “comfortable” capacity of the a single LRV (Light Rail Vehicle) that the City will be purchasing is 150 people (practical load). The “crush load” (the point where you physically cannot cram another person onto it) is 200 people. The current ridership is 10,500 passengers per hour per direction in peak hour (pphpd). The accepted forecast for 2031 is 20,000 pphpd. The generally accepted minimum headway in a tunnel is about 90 seconds using ATC (Automatic Train Control), however this number can be brought down slightly if signalling is done properly. To run smooth at-grade service, minimum headway is about 2 minutes, with anything less leading to the possibility of bunching along the line, particularly at red lights.

The numbers break down like this; at-grade can safely run no more than a tripled vehicle, due to stopping times, etc. This effectively limits the capacity of a single train at 450 people, or 600 crush load. None of the proposed Transit City routes (Toronto’s LRT Plan) propose running 4-car LRT trains, except for along the central portion of Eglinton, which will be an LRT tunnel. All the rest of the Transit City lines propose using paired vehicles. The only comparative example of a city that uses 3-car LRT through its core is Calgary, and the system operates through a transit mall, not in a dedicated lane like was proposed in Ottawa. As a result, I have used 2-car LRT estimates for the analysis, as that is what was initially proposed for Ottawa (and what is used on most other in-median LRT systems around the world). This limits the capacity of a single trainset to 300 people for practial load, and 400 people for crush load.

In order to meet current demand, the required headway would be 1 min 43 sec for practical load, and 2 min 10 sec for crush load. Even with the current ridership, it would be a challenge to deliver smooth LRT service, given that the trains would need to be run at headways around 2 minutes.

When the numbers are crunched for the predicted demand of 20,000 pphpd, at-grade becomes even less attractive, if even possible. Assuming a practical load of 300 people, in order to meet demand the headway would need to be 54 sec, or 1 min 12 sec for crush load. This headway is nearly impossible to maintain with the at-grade scenario that was proposed in the 1^st LRT plan. The headway for practical is half of what can be reasonably accommodated without running the risk of bunching.

However, if the line were to be grade-separated, the length of each trainset is not limited to what is safe by on-street standards. The stations on the system are being built to accommodate 6-car trainsets, although 4-car trainsets are being proposed on launch.

Keeping in mind that minimum effective headway in a tunnel is around 90 seconds, the current ridership combined with a capacity of 600 passengers per trainset (150 x 4 LRVs) is 3 min 34 sec at practical, and 4 min 32 sec at crush. This is roughly the same headway that the Sheppard Subway in Toronto (the only subway line in Toronto to use 4-car trainsets) operates at during peak periods.

When the projected ridership is examined, it would result in headways of 1 min 48 sec for practical load, and 2 min 24 sec for crush load. At this point, 6-car trainsets may be warranted, increasing headways to 2 min 42 sec for practical, and 3 min 42 sec for crush.

So, as one can clearly see, the numbers simply don’t work for an at-grade system through downtown, at least not in the configuration that was proposed. The math to get to these numbers was not difficult, a 5^th grader can do it (ridership per hour / per trainset capacity = number of trainsets required per hour).

So when you see somebody make the claim “the tunnel isn’t needed”, ask them how they arrived at that conclusion, and if they have run the numbers for themselves. Chances are their conclusion is based on either truthiness or a decided bias against transit, as opposed to simple arithmetic.