A Look at Vancouver’s Downtown Bike Lanes

→ Photo from StreetsBlog.

Yesterday, staff at the City of Vancouver released a report recommending that the bike lanes on Hornby Street, Dunsmuir Street, and Dunsmuir Viaduct should be made permanent fixtures in Vancouver’s transportation infrastructure.

The recommendation is based on a few factors, not least of which is the increase in ridership on these (and Burrard) bike lanes since they were installed and the increase in safety along the bike routes.

The report goes before council later this month, and I’m hopeful that they will adopt staff’s recommendations.

Let’s take a closer look at the report — embark with me, if you will, on a wonderful journey of statistics and graphs! (I promise it will be better than it sounds.)

The staff’s report includes these nuggets of stats, telling us that more and more people are riding their bikes on the separated lanes downtown:

	April 2010 – March 2011	April 2011 – March 2012	Year-over-Year Growth
Burrard Bridge	997,000	1,045,000	+5%
Dunsmuir Viaduct	346,000	411,000	+19%
Combined	1,343,000	1,456,000	+8%

There are no stats for Hornby and Dunsmuir Streets because we don’t have two years of data for those lanes yet, but we can still get a picture of how they are doing. As a reward for taking the first step on the wonderful journey, we find that Dunsmuir Street has seen a large year-over-year increase in ridership from 2010 (orange line) to 2011 (green line). In fact, 2011 was better than 2010 for every calendar month.

Figure 1. Click on the graph for a larger version.

So far, so good: ridership has increased.

Using data downloaded from the City of Vancouver’s web site, we can take an even closer look at the stats and get an even fuller picture. If you want to play around with the graphs on your own, feel free to do so at this link.

Temperature

As we would expect, there are more riders in the summer than there are in the winter. This is not news, but it’s nice to have our intuition confirmed by the statistics:

Figure 2. Click on the graph for an interactive version.

We also expect that there are more riders when it’s warmer, and this is also confirmed by the statistics:

Figure 3. Click on the graph for an interactive version.

So, we would expect more cyclists on a warmer year, and fewer on a colder year. Or, put another way, if 2011 was colder than 2010 but we saw the same number of riders, this would mean that things have improved. Or, if 2011 was colder than 2010 but in fact we saw more riders, then we know things are really improving.

In fact, this is what happened. The period of April 2010 to March 2011 was colder than April 2011 to March 2012, by about half a degree on average, and ridership still went up. Half a degree doesn’t seem like much, but statistically it makes a difference of a couple hundred riders on Burrard Bridge on any given day. We can imagine that ridership would have gone up even more if 2011 had been the same temperature as 2010.

Precipitation

Unlike temperature, precipitation isn’t really linked to ridership. The only exception is that there is a definite bump on days when there is no precipitation at all. This is due to the presence of what some might term “fair-weather cyclists”, or, as an old coach of mine might have put it, “people who are made of sugar.” You know who you are.

Here’s a graph to show this phenomenon.

Figure 5. Click on the graph for an interactive version.

Warning: this next paragraph is quite nerdy. Feel free to skip it since it mainly serves just to increase my nerd cred. (You may now be thinking, as an old coach of mine might have put it, “quit yer yapping and hit somebody.”)

One thing we can say is that the presence of precipitation affects the relationship between ridership and temperature. We can see from this next graph that the coefficient of determination (R²) between ridership and temperature decreases as precipitation increases. In other words, the more it rains, the less temperature matters to people deciding whether to cycle.

Figure 6. Click on the graph for an interactive version.

In summary, it doesn’t really matter how much it’s raining, just whether it’s raining at all.

The Hornby Effect

Last month, researchers at UBC published their BikeScore, modeling the bikeability of several cities. BikeScore considered four main factors:

1. The presence of bike lanes
2. Topography
3. Destinations and road connectivity
4. Bike commuting mode share

The most interesting factor, when it comes to the Hornby lane, is road connectivity. Separated bike lanes on their own are useful, but when they are connected into a network they are much better. One of the big arguments in favour of the Hornby lane was that it would click into place between Dunsmuir and Burrard and link those two together. That seems to have worked out, and I’m calling it the Hornby Effect.

With the Hornby lane, commuters from the east and south can use safe, accessible bike lanes to get that much closer to their jobs downtown. It’s likely not a coincidence that ridership on Dunsmuir Viaduct and Burrard Bridge, and particularly on Dunsmuir Street, has increased since the installation of the Hornby lane: the first trickle of evidence of the Hornby Effect.

We can also see hints of the Hornby Effect by looking at ridership on Burrard Bridge and Dunsmuir Viaduct before and after Hornby was installed. The data shows us that before Hornby, it was hard to predict whether Dunsmuir Viaduct would have a busy day by looking at Burrard Bridge. After Hornby, this prediction is much easier. (Nerd cred: the correlation between Burrard Bridge ridership and Dunsmuir Viaduct ridership improved after Hornby was installed.)

Figure 7. Click on the graph for a larger version.

Figure 8. Click on the graph for a larger version.

But what does this mean? Well, it could mean that the installation of Hornby led people to use both Dunsmuir Viaduct and Burrard Bridge on the same day — perhaps on a sunny weekend afternoon. Or it could mean that Hornby removed a barrier to using Burrard Bridge or Dunsmuir Viaduct. Or, it could mean that Hornby got people more used to the idea of using separated lanes, so people started using Dunsmuir Viaduct instead of other routes into downtown.

Either way, Hornby has clearly had an effect on ridership on the other lanes. What’s more, the Hornby Effect looks to be a positive one.

Female Ridership

As we’ve discussed before, female ridership can serve as an indicator for bikeability. Luckily, City of Vancouver staff collected data on riders’ gender:

Studies in summer 2011 also found that a broader demographic of cyclists were cycling into and through downtown. The portion of women cycling on Hornby Street had grown from 28% to 32% of cyclists with the introduction of separated bicycle lanes. Dunsmuir Street had 35% women cyclists. 2.5% of cyclists on the Burrard Bridge on summer weekends were children. Women make up 51% of Vancouver’s population and children 5-10%. Follow-up demographics studies are scheduled for Summer 2012.

There is still a ways to go, but here we have another study showing an increase in the proportion of female riders at the same time as an overall increase in ridership.

Safety

The City’s report contains some very positive and interesting information on safety. Separated bike lanes are often lauded for increasing the appearance of safety, but there are sometimes doubts about whether actual safety is increased. In our case, at least, we have good news:

Collisions of all types (involving vehicles, bicycles and pedestrians) are down 19% on Dunsmuir Street and down 18% on Hornby Street (2008 and 2009 vs. 2011).

It looks like these bike lanes have made us all safer.

Conclusion

The City’s report contains good news about ridership on the separated lanes downtown. If we take a closer look, we can see that things are even better: accounting for temperature, ridership is up even more; Hornby appears to have had the desired effect of creating a network downtown; these bike routes are now safer; and the overall bikeability of the city has gone up.

Now we wait for the decision from council and start taking a look at what we can learn from these lanes to make the next ones even better.

This concludes our wonderful journey. That’ll be $6.50, please.

___________________________________________
The fine print:

• The ridership data was downloaded from the City of Vancouver’s web site.
• The temperature and precipitation data were downloaded from Environment Canada’s web site for the weather station at YVR airport.
• There were some small gaps in the ridership data, which were filled in using a regression equation day of the week and ridership on other lanes. If you’d like more details on this methodology, feel free to get in touch.
• You can download the source data for the graphs here by choosing the graph you are interested in and hovering over the Export icon below the graph.