Q2Q, again.

This Post is actually an extended response to the comment by Ken, a Quayside resident and community builder, to my previous post about the Q2Q bridge. I thought his comments raised enough issues that I couldn’t do it justice just replying in a comment field!

Thanks Ken,

I will try to address your questions, but recognize that much of what you talk about occurred before my time on Council (so I was not involved in the discussions) and I respect that you have a much more intimate knowledge of the conversation on the Quayside over the last decade than I do.

The project has indeed gone through various iterations in its history, and the initial plans ( here is a link to a report from the time) were to reach 22m of clearance to develop a fixed link that would get adequate clearance that we would not need Navigable Waters permission (read- not specifically need Marine Carriers permission) which required essentially the same height as the Queensborough Bridge. Conceptual drawings were developed based on the site conditions and some baseline engineering, and very preliminary cost estimates prepared. That concept was indeed reviewed by the Port (at that time, the Vancouver Fraser Port Authority) and note they even at the time preferred an upstream (east of the train bridge) location (see page 12 of that report I just linked to). Note also: that report suggests elevators at each end to improve accessibility. This is the concept that first went to public consultation, and concerns were heard about the need for long ramps that would have nonetheless been very steep, the overall height, the fate of the Submarine Park, etc.

The only alternative to all of that height was a swing/bascule bridge. To explore this option, the City asked some engineers to sketch and (very preliminarily) price some alternative concepts, including a bascule and a sidewalk attached to the rail bridge. The City again took these preliminary concepts to public consultation, and the bascule design clearly came up as the preferred approach, even recognizing it was potentially more expensive.

Now that a preferred concept was (hopefully) found, and the Q2Q crossing once again received endorsement from the new Council, it was time to actually pay a little more money to engineers to further develop the preferred concept to a level of detail that would allow screening for Port review. Not enough development for a full review, mind you (that will likely take several hundred thousand more dollars in engineering and environmental consultant fees and will no doubt also result in adjustments of the concept), but enough that it is worth the Port’s time to look at our concept and provide a detailed regulatory screening and provide us a pathway to approval.

That is pretty much where we are right now, and for the third time, this concept is coming to the public for review. The only thing I can guarantee you at this point is that if (and it is still an “if”, despite general Council and public support) this project is completed, it will not look exactly like the drawings you see on the page today. There is much engineering to do, environmental review to perform, and more public discussion to be had. Satisfying the Port’s environmental review will be months once we get to that point, and we can guarantee it will require some design adjustments.

There are also other adjustments I think we need to see based on public feedback this time around. Although I have held my cards close to my chest because I don’t want to prejudice the public consultation, I will admit up front that there are two things in particular I cannot tolerate in the plans as presented at the open house: the 8% ramps simply do not meet modern standards of accessibility; and the closing of the bridge at night is not an acceptable way to treat a piece of public active transportation infrastructure. I’m prepared to accept that we cannot have the Copenhagen-style transportation amenity I would prefer, but I am still hopeful we can find a compromise that provides an accessible, reliable, and attractive transportation connection. We are not there yet. (And please remember, I am only one member of a Council of seven, and I cannot speak for them).

To answer what seems to be your main concern, I don’t know when the Marine Carriers were first consulted on this project, but the Port (who provides the Marine Carriers their authority) were clearly involved from day 1. They preferred an upstream location (now prefer a downstream one) and created the 22m by 100m “window” that led to the original 22m-high bridge concept, and have now led to evaluation of several swing/bascule concepts. Clearly, the City and our engineers have been searching for a creative solution to make what the politicians and public want mesh with the rather strict requirements of those who regulate the river and transportation. But serving those two/three masters is why the City is taking this iterative, slow approach, and why “plans that keep changing” are a sign of progress, not failure.

One thing to think about is that every step of this process costs more than the previous step, and moving backwards costs most of all. As engineering analysis and design gets more detailed, it gets more expensive, so we don’t want to do the detailed work twice. We could have asked for a ready-to-build concept a decade ago, and done enough detailed design that we just needed to pull the trigger and we could have it built within a year, and then taken it to public consultation. But if things are found that don’t work (i.e. the initial 22m height), we have spent a lot on a concept we now need to spend more on to change. Instead, we do feasibility studies, take it to stakeholders, the public, the regulators, and are given feedback. We then develop the concept to get more engineering done, and again have a look at the result and either move forward or change track depending on feedback.

This is a responsible way to plan, design, and pay for a public amenity. It is an iterative process, because as a government, we need to do our best to meet the needs of residents, of taxpayers who are footing the bill, of the regulations at 4 levels of government that have a thousand ways to limit our excesses, and of people who may be impacted by every decision we make.

If a government claims to do three years of stakeholder and public engagement, detailed engineering analysis and business case development, then turn around and deliver to you the exact same proposal they managed to render in a 3D model three years ago when the analysis started, then you know their consultation was bunk.

And I guarantee you, for every person who complains “this project has changed since the public consultation”, there are two who will say “public consultation never changes anything, they are going to ram their idea through regardless of what we say”. Actually, the same person will often say both, completely unaware of the irony. And that is why I appreciate your honest comments Ken, it sounds to me like you are trying to understand, not just complaining. So please provide your comments to the Engineering department and to Mayor and Council, and you will be heard!

7 comments on “Q2Q, again.

  1. Thanks for the detailed posts on this subject.

    FYI, the link to the Record article is broken, as is the link to your previous post at the top.

    When the 22 m height was deemed too tall, was it recognized that the swing/bascule alternative would have to be “open” by default (if it did not have a large clearance)? To me it seems like we are possibly going in circles… 22 m is too high, but now the alternative is 14.5 m, which is still quite high and problematic… and now it has to open as well!

  2. My new idea and vote goes toward a ski lift type apparatus! It doesn’t even need to be a gondola style! 😉

  3. Patrick, thanks for sharing all these details and making the process more transparent to the residents! Are there any assurances that the Port won’t come up with yet another change in requirements? IMO this is a perfect example of life giving us lemons, so it may help us all to admit that yes, we are living on the working river, and yes, for that reason the Q2Q bridge will have to mirror that fact by having some features that don’t necessarily mirror the idyllic, classical mini-bridges seen in the “little venice” gardens on the Quay.

    Also, if there is a requirement for a 20m high deck to appease the Port Metro Vancouver jerks, I say let’s not mess with a lower deck height and opening/closing. Why not have fun with the 20m height requirement? Let’s put out a concept design competition with a bounty for the top submission(s) as chosen by New West’s engineering department and a panel of residents from affected buildings on either side? Who knows, someone might actually come up with a funky monorail/girder simicircular crossing with a circulating gondola on it that looks like it’s from the Jetsons, and it will prove workable?

    All the best, and keep the communication going!

  4. Stevenson: orders of magnitude more expensive. The crossing is too narrow to do a tunnel for the depth required. The approaches on each side would be very large.

    The problem is that there are two ways to do a tunnel. The massey bridge sunken structure route, and drilling under the river route. Drilling requires more depth, lots of room on either side to insert and remove the boring machine, and the grade it can do is limited. So extremely expensive.

    The sunken structure route is a non starter due to port requirements. It would take up too much height, this reducing the size of ship that could transit over.

    So bridge it is.

    My solution is simple, elegant, and only a liiiiitle bit dangerous. Giant slingshot on each side. Just fling the people across.

  5. Thanks for the informative update, Pat, and describing the iterative design process I work in every day.

    Architecture, and most any “project”, is delivered from from the general to the specific: from early schematic design concepts sussing out big issues, ideas, constraints and opportunities; then design development for more detailed study when it looks like the scheme is “road worthy”; and finally working dwgs. that fully describe the “thing” for tender, permits and construction.

    At each stage issues arise: some that reinforce and advance the scheme and others that force a rethink and revisions. Indeed these gremlins can come out of nowhere (city depts!). The port reversal on where they want the ped. bridge is a good one.

    This messy creative milieu can be challenging to clients or the community when certainty and details are not readily apparent in the early stages. As I say let’s not worry about the door knobs when we’re not even sure there’s going to be a door. A strong initial concept can weather design changes and still be a great work. In fact it is better for it.

    Stay the course on this one and the community will be there. The concept is sound: we want a bridge, it will respect its context and it will look great. Some clever design team will respond with a cool design that touches lightly on the earth (make sure they have an artist on the team). And yes someone will try to fish from it.

    The RFP must challenge the designers to not produce something that is a banal engineering solution (no offense p.engs. out there!) just to please everyone and keep costs down. The Eiffel Tower was considered beyond ugly and a blight when first presented. And please, no coloured LED lighting!

  6. It may be more expensive, but as a fascinating example of Victorian ingenuity and a point of interest for the city why not build a transporter bridge? That allows any desired amount of clearance for shipping and spans exceeding 100m were fairly common place at the beginning of the 20th century, all with no need for any gradient whatsoever.

Leave a Reply