Are the Chch Cycle Design Guidelines up to scratch?

A little bit of controversy has been brewing this week with the release of a blogpost by well-known Dutch-based UK cycling advocate David Hembrow, which has taken aim at the City Council’s Cycle Design Guidelines.  Hembrow has produced a fairly critical commentary on these Guidelines and clearly feels that many of the things presented are flawed or downright unsafe. Is this criticism warranted?

{Disclaimer: I had a little bit to do with the development of the Guidelines, in being on a stakeholder group for it and supplying some of my photos}

Hembrow’s blog, “A View from the Cycle Path”, provides a fascinating and well-documented insight into how The Netherlands have provided so well for cycling over the past few decades (he also runs study tours for those who’d like to see first-hand). On occasion he also takes a swipe at “less-than-optimal” examples of cycle planning and design elsewhere, particularly in his UK homeland (which lately has been making a lot of song and dance about supposedly “going Dutch” for cycling).

Before responding to Hembrow’s concerns, I think it is important to appreciate what the Guidelines are and aren’t:

1. This is a conceptual design guideline, not detailed design. Hence many of the images are what I’d consider “artist’s impressions” – for better or worse, they present a vision but without all the little nuances. There are very few dimensions given in the Guidelines – this was deliberate to leave them off the drawings so that someone wouldn’t just take a design concept from here and try and insert it somewhere without consideration for context. Page 8 states “The guidelines are not technical engineering standards,… the guidelines will be used to inform an addendum to technical engineering standards in the City Council’s Infrastructure Design Standards”. The current task at Council is now to convert these generic images into actual detailed designs for the first cycleways being planned, and a lot of consideration is being given to provide a very good level of service for people using them.
2. This is not a Dutch cycle design guideline. It is a guideline for a city on the far side of the world that has an existing road/cycle infrastructure (and an existing road user culture) and a desire to step this up. This is likely to be based on a number of useful techniques that we have seen from all around the world. While the Netherlands provides many exemplary examples of how to provide for cycling, I have certainly seen other nice examples from elsewhere (some of them documented here on this site) that I believe can also work effectively here (and in many cases attract new people to biking as well). This will particularly be important in situations where we don’t have the immediate means to physically provide what might be the “gold” standard ($69 million will still only get you so much) – so what’s the next best option for now?

OK, so what concerns does Hembrow have about the Chch Guidelines?

• The first concern was about the portrayal of a “Dutch intersection” on p.32-33 (some of you may recall we even mocked up a layout of one during the Open Streets event). Hembrow states that “The Christchurch design misses many key details. The geometry of it is also completely wrong.” While I grant that the “artists impression” layout is a bit simplistic (yes, they should have shown a rider stopping in advance by the island), I think it mostly captures the intent, which is to provide more protection for through and turning riders (it could have made a better effort to indicate that left-turning riders can go at any time, though). Hembrow is also concerned that a right-turning cyclist won’t be able to see their signals because they will be behind them. But standard NZ practice is always to have secondary signals on the other side of the road too (and I also wouldn’t mind trialling some of those smaller eye-level cycle signals here too). A related concern seems to be that it is largely based on a very popular intersection layout created by Mark Wagenbuur doing the internet rounds, that was his take on converting a standard US (and probably NZ) layout to a more Dutch-style one, but wasn’t meant to purport to be a “standard” (or even currently much-used) Dutch design. OK, I get the point being made, but the accompanying illustration showing an actual Dutch intersection (on the right) in comparison to Mark’s hypothetical layout doesn’t seem exceptionally dissimilar to me in concept…

• I suspect that the labelling of the previous layout as a “Dutch intersection” in the Guidelines didn’t help; Hembrow may have felt that we thought this was a standard intersection layout in The Netherlands. But I think it was just a handy shortcut way to refer to it – it came from Holland; therefore it was a “Dutch intersection”. Evidently a more common approach in The Netherlands these days is to provide a “simultaneous green” intersection, or what has been described in the Guidelines (p.34-35) as a “cycle Barnes Dance” (i.e. similar to the popular all-crossing intersection setup for pedestrians). Here Hembrow raises concerns that the Guidelines “have unhelpfully provided extra paint on the ground seemingly to guide cyclists to ride across diagonally rather than using the curved paths that are normally taken“. Actually the white markings shown are the standard NZ Barnes Dance markings for pedestrians – presumably this intersection also features a separate phase for them. I imagine that the average cyclist during their phase will go wherever they feel is reasonably direct and avoids collisions. Hembrow also misunderstands the text which says that “ideally” there would be “a separate cycle signal” – this doesn’t mean that a separate phase for the cyclists is optional (that clearly defeats the point of having a Barnes Dance); it means that the physical signals for cyclists may not necessarily be on the same poles as the motor traffic ones.

• Hembrow also has concerns with other treatments proposed that we have discussed here before, including hook turns (or “two-stage turns), advanced stop boxes, and sharrows. I think this misunderstands their likely applications in a Christchurch context. We already have an existing cycling infrastructure based on traditional “on-road” designs that is not going to be replaced overnight. Hence we need ways to pragmatically improve the existing cycling networks in the interim, while acknowledging that some of them are more about improving the lot of the existing riders than attracting many new riders. Something like hook turns for example are a relatively simple way to make right turns a much less stressful manoeuvre where there are no other dedicated cycle turning facilities. I know that Hembrow regularly points out that cycling numbers in Denmark are in slow decline, but it is significant that the Danish make very common use of these and still achieve far higher cycling numbers than us. Hembrow also takes issue with a “pointlessly badly designed” cycle bypass shown on p.38, actually the one on Buckleys Rd that seems to work just fine; his example of “the same concept done properly“ seems to be a completely different context (i.e. side-road crossing a two-way off-road cycleway).

• Hembrow feels that there’s “far too much emphasis on on-road cycle-lanes”. However this reflects the distinction between the “Major Cycleways” (which will be the focus of the immediate works programme and emphasise physical separation or calmed streets) and the “Local Cycleways” that will fill in the rest of the network at least for now. That is why the Local Cycleways section of the Guidelines still includes quite a lot of on-road treatments (which as we have demonstrated elsewhere, can provide significant safety benefits in their own right). Personally, I would like to think that most of the Local Cycleway routes will eventually look and feel very similar to the Major Cycleway designs, but that won’t happen overnight. The intersection layout on p.61 comes in for a particular grilling by Hembrow as “especially horrifying” and “a recipe for disaster”. I think this is a bit alarmist (and in some cases quite wrong – there are always traffic signals visible in front of cyclists); even the text says that cyclists could be given an even higher priority in this scenario by having a cycle-only phase or by delaying left-turning vehicles. And let’s be clear – this is not an intersection layout for a Major Cycleway.

• Hembrow feels that the Guidelines suffer from a lack of actual experience with the Dutch designs in practice; he even offers his famous Dutch cycling study tours as one solution. I must certainly confess to have only spent a few days in The Netherlands myself (and that was mostly in The Hague), so I would certainly welcome the chance to take a closer look some time. But of course, Christchurch has been receiving Dutch input into their cycleway developments courtesy of our recent visitors Mark and Leo (and Leo will be back soon to assist further). The Dutch duo recently provided some very useful formal feedback to Council on their initial impressions about the city and its cycleway planning process; I note that they had very few concerns with the existing Guidelines other than to point out that “the devil is in the detail” – yep, that’s why they’re only conceptual…

• One issue that clearly annoyed Hembrow was the fact that one of the figures in Guidelines (p.34) had come from his own website unacknowledged. As a fellow private blogger, I can appreciate this concern; I try to acknowledge sources where I can (and have seen my own photos all over the place…). Certainly a belated apology and acknowledgement is due by CCC for this oversight.

I’ve overlooked responding to a few other little bits & pieces but, in summary, my personal feeling is that the Guidelines are not as fatally flawed as Hembrow makes out; in many places it is a case of differing expectations about what the document is for and how it will be used (plus a few technical misunderstandings). Some of it is for the new Major Cycleways and some of it is to enhance existing cycleways, and in neither case will designs be photocopied straight off this document. That’s not to say that there aren’t improvements that could be made to the details presented, and Hembrow makes some valid points (esp. about some of the dimensions/speeds quoted – I actually wish that no numerical values had been included, as they are very context-specific). No doubt we will get a few things not quite right, even with the benefit of international experience. As the installation and monitoring of various new cycleways in Christchurch gets into full stride, it is likely that a “version 2” of the Guidelines would be a useful update (notwithstanding that the addendum to the Council’s technical engineering standards will provide more of the detailed design guidance). But ultimately, the success of the future cycleway plans in Christchurch will depend a lot more on the collective skills (and continuous learning) of the design teams making them happen than what appears in a glossy public document.

What do you think of the Council’s Cycle Design Guidelines?