The REM de l'Est Report is here, and it's not good.

If you were hoping the ARTM had good reasons to tunnel the entire eastern REM, I think they do not.

If you are wondering what’s up with all the posts this week, Montreal released a new report about a project I care deeply about and I’ve wanted to make sure I cover it in detail, expect a return to the usual content on Friday.

I think it might also be good to read my piece on the issues around the Mount Royal tunnel being used on the first REM (REM A from here on) and the article I published last night on a vision for a suburban REM network.

After a few days of writing on this, I am exhausted – so it will likely be a while before you hear anything else about Montreal from me in videos or articles.

---

Image may be NSFW.
Clik here to view.

Today is the day that the ARTM releases their report on the plan for the fully underground $36 billion dollar eastern REM project. To the ARTM’s great credit (I am going to be quite critical of their report) they did have me at the technical brief, so as much as the report annoys me – kudos to them for being open, it’s something that deserves respect. I was able to ask a question and a follow up and I will present the question and answers below.

The report is linked here.

Before I dive into my more in depth analysis, I want to address a few points which have come up around the project and the report release:

1. The cost of the project includes financing costs and contingencies etc. – this is well and good but, the cost isn’t 10% too high, it’s an order of magnitude too high (the Canada Line all in was about 10% as expensive 15 years ago) – this is the same issue that comes up every time with the inflated costs paid for transit in places like North America.

2. The report states the reduction of cost from going above ground is only 6% – if this feels wrong that’s because it probably is, my cursory reading of the report (my French is just ok) suggests the proposed elevated structures are needlessly large and expensive (they could still look nice while being much smaller and less pricey) which makes elevating seem artificially less likely to reduce costs – which makes sense because if you’ve been following this project the political consensus in city hall is that elevated is unacceptable.

3. There is talk of an expensive upgrading of the Green line to handle new passenger from the REM – upgrading the Green line should essentially be seen as an extra cost of this project (I am not sure if its included in the $36 billion but, it’s likely a rounding error anyways). To be sure, upgrading the Green line is sensible – but, only if the Green line itself has the ridership to justify it. The opportunity cost of delivering capacity through an expansion of the Green line rather than through a downtown REM leg is that you don’t provide that faster service and more coverage that the downtown branch provides – not great.

With that out of the way let’s dive into the report.

The report goes into a few main subjects, modifying the existing project to extend it, alignments, ridership and more. But the thing I want to focus on is the surprising amount of energy given over to rationalizing (if it can be called that) burying the whole thing underground – often through industrial and suburban areas.

I think the best way to digest this report is by way of a sniff test on that section – in my opinion it frames the whole report as unserious, hyperbolic, and just perhaps not very well developed (and this is after what I believe is a 6 month delay on it’s release).

I want to preface this by something which most people in the transit and planning industry and advocacy place know but, which might not be obvious. Reports are often not impartial (not talking about any particular report here but, rather in general) – that’s to be expected, public agencies are made of real people who have political masters and other pre existing biases. The point to make is that sometimes studies are produced either intentionally or unintentionally in a way that thrusts the discussion towards or away from a certain solution. You can give a correct “answer” to the problem of creating a new transit project that is not the optimal answer. It might meet the building code and put stations in the right location but, for example design those stations in such a way to suggest they are prohibitively expensive. The problem with this is that it’s impossible to make a true apples to apples comparison when one option has been made to look artificially weak (sometimes called sandbagging) – and this is why reports should always be read with a critical eye. You might suggest simply mandating that reports are “fair” but, herein lies the issue – these processes are often for better or for worse black boxes, so such mandates are essentially impossible to perfectly implement.

The Issues of Elevation

As I mentioned before the report spends a lot of time discussing concrete elevated transit structures or guideways, which are commonplace in cities around the world from Paris, to Hong Kong, to Vancouver and even in Montreal with REM A. The language used to describe such structures in Montreal media and by some politicians often pertains to their “social acceptability” and the report repeatedly refers to example structures as “industrial” in nature.

Image may be NSFW.
Clik here to view.

The reality is that not all guideways are created equally, the pillars and the deck structures can be varied widely in height, width and design. They can also be designed to harmoniously integrated with things like sound barriers, stations, street finishings and electrical systems.

Of course, different elements of the line – such as stations, also have an impact on the design of the elevated structure. Stations can be designed be taller in dense places to reduce their horizontal width – forcing the tracks and guideway higher, or they can be lower slung – context matters a lot here. In dense areas it is often preferred to put the stations over the street because there may not be space on either side but, that forces the station to have an extra floor or level where passengers can move between multiple platforms (side platforms are typically preferred on modern elevated lines so that the guideway need not “split” at stations) you can see this at a station like Brentwood in Vancouver.

Image may be NSFW.
Clik here to view.

This design works but, it increases price and visual impact – so most elevated stations in Vancouver do not use this layout. Instead they look something like this.

Image may be NSFW.
Clik here to view.

This station has a circulation area which plays the role of the mezzanine at ground level significantly cutting the height and complexity of the structure. Such a design really only works on the side of the road, but given the width of suburban streets and the plentiful open spaces (think parking lots which many SkyTrain stations have taken over) that isn’t a major issue. But, despite the wide roads and plentiful open parcels in Montreal east the report does not suggest this more cost effective design.

Image may be NSFW.
Clik here to view.

This is quite striking because you can see there was clearly an opportunity to select a less imposing or costly station design compared to the one pictured above, which is directly adjacent to several good parcels of land (strip mall parking) for a station. The precedent here is Burquitlam station in Vancouver which was built on a strip mall parking lot (the mall is now being redeveloped).

There are other places where comparisons with Vancouver make sense, sometimes it’s hard to judge whether something is plausible – and if you find another city with the same problem and a solution you can have a fairly confident answer.

Image may be NSFW.
Clik here to view.

Transmission lines pictured above in the lower part of the picture are also presented as a reason why Montreal could not build the suburban part of the REM de l’Est elevated, however one of Vancouver’s current projects – the fully elevated suburban Surrey Langley Skytrain is dealing with the exact same thing. What was the cost of mitigating these power lines? $2 billion? $5 billion? Nope, $138 million baked into a larger package of “advanced works”.

Image may be NSFW.
Clik here to view.

I am skeptical that a cost on the order of ~$200 million is a showstopper for a project estimated to cost $36 billion, and indeed the cost savings from that “small” 6% cost reduction by building elevated mentioned earlier? It would be over $2 billion dollars. That’s almost the entire cost of the Vancouver Canada Line – the entire line.

The scale presented throughout the report is also clearly meant to tilt things against elevated rail, take a look at this render of what I guideway would supposedly look like in Montreal, compared to one from Vancouver. Note the size of cars relative to the guideway.

Image may be NSFW.
Clik here to view.

The cherry on top of it all, and the moment when I felt the most frustrated reading the report? This quote (translated): “The metropolitan highway is an example of the impact of an aerial structure on a urban environment of a size comparable to that of the residential sector of Pointe-aux- Aspens, creating an urban border effect. Even if the aerial structure proposed for the PSE is higher than the highway metropolitan area, somewhat reducing its visual impact, its location along the rue Sherbrooke Est risks accentuating the effects of the urban border, especially in commercial and residential sectors.”

Comparing an elevated highway to elevated transit is so completely disingenuous it makes my blood boil, at a minimum an elevated highway is going to be several times taller, wider, louder, and more environmentally impactful. Here’s a final comparison:

Image may be NSFW.
Clik here to view.

Technical Brief

Fortunately, I was given the opportunity to ask some questions at the technical brief, here are my questions, responses and some post-hoc comments. I have edited the responses for brevity, to the team’s credit they did bring up Copenhagen, a good example of an elevated rail system – but, I do not recall where it fit in

Question from Reece: “I appreciate you included examples from Paris, but the structure of Line 6 is over 100 years old – why not include a modern structure such as those created for the modern grand Paris express project? At the same time why are these prices so much higher than the grand Paris express, a project which is over 80% tunneled?”

Response: “We didn’t find solutions from elsewhere that would be suitable for the Montreal reality. On cost, the cost of an overhead structure is more or less the cost of an underground structure – stations are the differentiator – an underground station is about $100 million more”.

I can’t comment too much about the Montreal reality besides to say that Paris is also a highly regarded city in a developed country which manages to build for a fraction of the price. I am skeptical that the cost of tunneling is equivalent to elevated structures, though it is in theory possible with an extremely low cost tunnel (doesn’t appear we are getting one in Montreal) and a very high cost guideway.

Follow Up from Reece: “What is the cost for an elevated station? And why is the guideway so tall, transmission lines and rail lines do not seem to have been a major barrier for SkyTrain in Vancouver – it crosses many?”

Response: “The cost of an elevated station is roughly $40 million. As for a project on Vancouver or elsewhere, I was not working on them, I cannot explain why they found solutions for this, in our environment when you add environmental problems crossing railways and power lines, this was just too much to justify to go forward, and this is why we suggested an underground project.”

First, the cost provided at least for an elevated station sounds about in line with past experience with individual stations built in Vancouver, so that checks out. On the second part I am less impressed – surely when building an automated elevated metro in Canada you should be familiar with the solutions used to address common problems like hydro and rail lines in other Canadian cities?

A few more questions were asked, one mentioned the timeline of the project – and officials suggested that if things moved well the project could be done by around 2036 which is 13 years from now. That’s a long time especially because construction time (which is liable to delay) was quoted at 4-5 years.

There was also a general question about why the lines were no longer elevated, and some of the reasons given were gas and oil lines, hydro transmission infrastructure, a river crossing, and Sherbrooke street being too narrow. Some of these like the river crossing may be reasonable but, as far as I know Sherbrooke is a wide suburban thoroughfare – perfectly appropriate for elevated rail, utilities are also not a problem unique to Montreal and so I am skeptical of that justification.

Stepping Back

So stepping back from all of this, with some questionable planning decisions and a high price this project is clearly not going anywhere. In the report it is suggested that stations added deep into the suburbs would add ridership but, the cost of those stations per rider generated (~3000 riders, $140,000,000 cost) is nearly $50,000 per rider. Given transportation alternatives that exist this is an astronomical price, in Vancouver stations often have a ratio close to $10,000 per rider.

Ultimately, with all of this in mind, I cannot see how intended or not this report does not kill this version of the project. Academics and business leaders in the media have said “now that we know that this is the automated metro option we need to know what could be built for the same cost for other options like trams.” And the fundamental logic is sound but, the comparison won’t be because this is NOT the automated metro option, this is a facade of an option. It is not something we will or would build, and it shows almost no attempt to balance cost with other considerations.

Fundamentally even if you don’t like automated metro as an option, this is bad – since this report does not present a realistic comparison to benchmark anything else against we are not really in any better a position than we were several years ago with the original CDPQ plan. And I think that’s the important comparison that has to be made here – the CDPQ plan is not perfect – but, it could have conceivably been built, this plan is also not perfect but, it’s not even worth comparing against because with it’s cost it will never be built.

This makes me sad and leaves me questioning the value of this entire exercise. What’s clear is that we are left with no clear workable automated metro option to benchmark against, and that will either mean needing more studies, or a bad outcome based on needlessly flawed information.