Bruce's Blog

Seasonal distribution of cycling

23/3/2021

In order to understand trends in cycling during the pandemic (see other posts) needed to adjust for the normal seasonal shifts in cycling. Could find very little on this so have done an analysis of the data from the National Travel Survey. Have also looked at the London Cycle Hire numbers.

Graph shows how similar the seasonal changes are for data from the National Travel Survey and the data from London bike hires. This is weekday (Mon-Fri) data. Weekend data does not match quite so well, but is not bad. Other sources of data (TfL TLRN counts; DfT urban roads traffic flow) also show much the same relation.
There is a steady rise from around 65% average to a peak of 130% average in June or July. there is a slight fall to September, then a steep fall for the rest of the year to a minimum at Christmas of 40-50% average. At weekends there is less cycling overall, but more seasonal variation. Much the same pattern, a rise from Jan to June/July to a peak of 140% then a steep fall to a low of 40% at the end of December.
Lots of interesting features - small peaks at Easter, significant falls in August (everyone away on holiday?). Intriguing jump from late December to early January - maybe a lot of people resolve to exercise more in the New Year. Much more detail in a full account (guaranteed to cure insomnia) at https://drive.google.com/file/d/1MX_6p-5mkxLX8o3JW66emtz5JgcrlzDt/view?usp=sharing

0 Comments

Cycling down 70% on Cycleway 1 during lockdown.

21/6/2020

0 Comments

Key points

Comparison at Willow Walk with 6 years of previous data.
Midweek peak hour cycling down 70%
Contrasts with motor traffic only down 30%
Weekend cycling as high as weekday with 41% female riders

Weekday Cycle flows

Large reduction in cycling, especially at the morning peak. The drop is probably even greater than 70% as cycle flow has been increasing by 150/year on this route, so we would expect 2020 level to be approximately 1080 at am peak hour. The pattern is still very tidal during lockdown. Flows out of town in the morning and into town in the afternoon were low and similar to pre-lockdown levels (see Table).
Table. Cycle flows (per hour) pre lockdown and midweek and weekends during lockdown. Willow Walk, Cycleway 1.

	am peak,Westbnd	am peak,Eastbnd	Mid day,Weastbnd	Mid day,Eastbnd	pm peak,Westbnd	pm peak,Eastbnd
Midweek pre, 1/5/19	933	35			66	691
Midwk during 16/6/20	223	43	56	60	96	216
Weekend during 14/6/20	54	81	230	110	65	115

Weekend cycling
As is generally the case during lockdown, leisure cycling at weekends is high. The Table gives flows on a sunny Sunday during lockdown. No data is available on weekend cycling pre lockdown. Might be expected to be about 30% of weekday total daily flow (TfL Cordon data, http://content.tfl.gov.uk/traffic-note-3-cordon-and-screenline-surveys-2011.pdf)
To estimate weekday daily flow can use the profile from Travel in London, Report 11, Fig 5.4. For May 2019 this gives 3990 daily flow. 30% of this, the estimate for weekend flow, is 1797. We can estimate daily flow for the lockdown samples by assuming a similar profile to Jamaica Rd where we have enough data to construct an accurate profile. This gives a value of 3088 for the total flow on 15 June 2020. This is 72% higher than the estimate for 2019. This is not as large as the increase estimated for Jamaica Rd, 202%. But there are many assumptions in both calculations and anyway there is no reason to expect all roads to increase by the same amount. What we can say is that, unlike weekday flows that are way down, cycle flows at weekends are up.
Male-female ratio

There are more female cyclists during the week now, under lockdown, but the difference is not statistically significant. This increase is similar to that seen on Jamaica Rd where it did reach significance. There is a markedly higher proportion of female cyclists at weekends compared with midweek. This is probably because there are relatively more female leisure cyclists than there are in the group who cycle to work.
This report in Word document below

lockdown_cw1.docx
File Size:	26 kb
File Type:	docx

Download File

0 Comments

Effect of Lockdown on Cycle Traffic Flows

10/6/2020

1 Comment

Effect of Lockdown on Cycle and Motor Traffic Flows
Report by Prof Bruce Lynn, Healthy Rides Co-ordinator, Southwark Cyclists, London Cycling Campaign. Version 10/6/20.
Key points

Weekend cycling up 200% with 60-70% mode share.
Working week cycling down 33%, but slowly rising. Motor traffic down 35% and stable.
% female cyclists is sig up on pre-lockdown
Commuter “transport” cycling (as opposed to “leisure” cycling) is possibly down by as much as 70%.
So some good, some bad news here. Leaves open the question of whether routine “transport” cycling will even recover to pre-lockdown levels when there is a full return to work and motor traffic increases.

Data collected on A200, Jamaica Rd, Dockhead. Past data Oct 2012 to January 2020. New data from 10 April to 10th June 2020. Manual counts.
Past data was principally peak hour. Interpeak data calculated from TfL data on time distribution for Central London counts, 2017, Travel in London, Report 11, Fig 5.4
Average Daily Cycle Flows
                        Pre         During   % change
Weekday             6050       4055       -33%
Weekend            2042*    6160       +202%
* Ratio Sat/Sun to Weekday, 0.30. Used to estimate pre weekend flow. 0.3 factor comes from Transport for London Surface Transport Traffic Analysis Centre. TAC Traffic Note 3 – TfL Cordon and Screenline Surveys 1971-2011, http://content.tfl.gov.uk/traffic-note-3-cordon-and-screenline-surveys-2011.pdf
Peak hour cycle flows (per hour)
                     Pre         During      % change
Weekday pm     838         490         -42%
Weekday am      963         240         -75%
Weekend            n/a         650

Note unusual time course of midweek flow. Morning peak almost lost. See last section below for more discussion.
Cycling mode share (% total traffic).
Midweek at peaks pre lockdown, cycle mode share =>50%. Peak mode share during lockdown was only 30% at end April, beginning May, but has risen to 35% am and 55-60% pm. At weekends during the middle of the afternoon cycle mode share is 60-70% on good weather days.
East v West flows.
Weekdays. Pre lockdown flow pattern was VERY tidal, am peak Westbound ca 10 times eastbound flow. At pm peak, eastbound >> westbound. During lockdown, E/W diffs not so marked, but still trend for greatest west flows am and east flows pm.
Trends over time.
Cycle flows. Sig linear slope. Pm peak, eastbound, gone from 35% to 60% mode share. Am peak westbound 30% to 40%. Motor flows show small, just sig upward trend. Includes data to 10th June.

Male/female differences
2018 data am only 18.1% female riders. 7-9 am, weekday.
During lockdown 25.0%, same time weekday, sig higher proportion (p=.02).
For comparison, Q1 (pm peak) pre lockdown was 25.3%, close to lockdown Jamaica Rd value.
Weekend cycling during lockdown 34.6% female, sig higher than weekday (27.1%, all times).
Strong evidence for more female cyclists during lockdown. Increase at busy times weekday indicates new female cyclists OK on busier roads during lockdown. Note motor traffic down 35% (see above), but that is still over 400 veh/hour most of the time during working week.
How much midweek cycling is for transport, how much leisure?
The shape of the time of day profile during lockdown is very different from pre-lockdown (see chart at start). Peak flows are depressed, particularly am. Flows during the day are if anything slightly up. Likely what we are seeing is a lot of “weekend” leisure cycling all week. Tried modelling flow profile for a weighted sum of pre lockdown and lockdown weekend profiles.

Least square fit was for recorded lockdown weekday profile was 33% of weekend flow plus 30% of pre-lockdown flow. Fit is not very good. But indicates that cycling for transport may be only 30% of pre-lockdown levels. Flows at peak are rising, but need to be cautious in interpreting the total weekday flow as transport flow. If cycling to work, to school etc is to contribute significantly to London’s workday travel we need to see a massive rise in weekday cycling. So far the data indicate probably recovery to pre-lockdown levels, but not the 10-fold increase hoped for by some campaigners and politicians.
This report as Word doc below

lockdown_flows_summaryv02.docx
File Size:	63 kb
File Type:	docx

Download File

1 Comment

Serious injuries to cyclists are rising

3/3/2020

0 Comments

Previous blog (January 2020) highlighted the marked fall in cycling fatalities in recent years. At the same time we are seeing an INCREASE in serious injuries. This is not happening in other modes (pedestrians, motorcyclists, car users). It has only been happening since about 2005. I have no explanation.
The report attched below gives the evidence in more detail and briefly discusses reasons. This is a very important issue. A "serious injury" is sometimes life-changing. It is important to work out why the trend has developed and find ways to reverse it. Please comment below so we can develop ideas to investigate the reasons and perhaps find a way to reduce the risk, already small, to the same levels as other transport modes.

cyclists_casualties.docx
File Size:	72 kb
File Type:	docx

Download File

0 Comments

Record low number of cyclists died in collisions in London in 2019

11/1/2020

0 Comments

Five cyclists died in collisions on London’s roads in 2019. This is much lower than the 2018 figure of 12. As shown in the diagram, it is half the average for the previous 5 years. Checking records back to 2002 shows this is the lowest figure over that period; it is probably the lowest figure ever.

A sample of 5 has a confidence interval from 1 to 11. This gives us cause to be cautious. However if we look at the trend it is much more convincing. The graph shows the trend since 2002.

The 2019 value looks like the continuation of a very real trend. The trend line estimate for 2019 would be 8, still the equal lowest value on record. The downward trend in fatalities is statistically very robust. Also remember that numbers cycling have increased substantially over this period, The trend shows lives lost falling steadily by 0.6 per year.
There are about 500,000 regular (several times a week) cyclists in London. So the risk of dying last year was 1:100,000. Is that a large risk? There are about 5 million people regularly walking about in London. Last year 71 were killed in road traffic collisions. This is a risk of 1:70,000. OK, last year had rather low cycle fatalities while the figure for pedestrians was higher than expected. But you can see that the risk of being a pedestrian is not much different from the risk of being a cyclist.
There is a clear message :

The risk of being killed while cycling is very low
This risk is falling steadily, year on year.

0 Comments

Cycle flows along Quietway 1 – Pattern of arrival and departures

6/6/2019

0 Comments

Cycle flows along any cycle route are dynamic. Cyclists join the route and leave the route. Here some data is presented about a short stretch of Quietway 1 in Bermondsey, South London.
Counts were mostly made in May 2019 at the time of peak am and pm flows. For some junctions we have large counts (around 1000) while at others only small sample counts were made (at least 53). It is best to take any values as liable to an error of at least +/- 20%. Flows are given as % max flow segment (Willow Walk). Actual peak flows at this site were: am, 1050; pm 650. For more information on flows and time trends, see https://southwarkcyclists.org.uk/quietway-1-cyclist-numbers-still-increasing/.
Key data is summarised in the Diagram above.
Patterns of joining and leaving are reversed in direction but otherwise quite similar for am and pm, presumably because many cyclists use the same route to work and homeward. Flows towards central London fell after Willow Walk. It is at this point that cyclists start to move off towards specific destinations often via the nearest bridges (London Bridge, Tower Bridge). Flows further out than Willow Walk increase as cyclists join Q1. For example at Oxley Close cyclists join from the south in the morning and leave in the evening, presumably taking routes to and from Peckham and Nunhead.
At the main junctions on this stretch between 12 and 36% of cyclists either leave or join. Q1 always keeps at least 64% of cyclists. Further out (not shown on the diagram) at the Surrey Canal Rd/Bridgehouse meadows junction the same pattern is seen. At this junction total flows in the evening approaching the junction are about half those on Willow Walk (312/hour). Around 77% of riders continue on Q1 along Surrey Canal Rd while 23% leave Q1 and continue across Bridgehouse Meadows.
A few counts are available for 2015-16. At the most central junction (Pages Walk) at the am peak 57% of cyclists left Q1 in August 2016, just a few months after the “official” opening. This is much higher than the 16% recorded in 2019. This probably indicates that, as cyclists found more of Q1, they switched routes, preferring to stay on Q1 longer. Or it may be that the many additional cyclists who now use Q1 (flows have doubled since 2016) do so because it continues for longer towards their destination. At the Oxley Close junction number joining the Q1 route has increased a small amount in 2019 compared with September 2015 (before Q1 created). This may be surprising as a feeder junction (Glengall Rd/Old Kent Road) was improved for cyclists in 2017. Although we should remember that the total numbers joining will have increased, just that as a proportion of the Q1 flows they have not changed.
Perhaps the most important general point to emerge from this analysis is to demonstrate how much joining and leaving there is. People use a designated route for the part of their journey where it suits them. But leave it as the near their destination, and join it taking routes from their personal starting point. In designing cycle routes it is essential to consider these aspects. We need to think about a network, not just single routes. It will never be possible for every origin and destination to have a designated route between them. Providing links to longer Quietways and other labelled routes is essential if we are to get maximum use from them. So consider the network, not just the A to B.

0 Comments

A little more cycling, but same number of cyclists

15/1/2019

0 Comments

Latest road traffic numbers for cycling in the UK
There is a clear analysis in the latest Road Traffic Estimates:Great Britain 2017. Have extracted page 1 and the cycling bits and attach them below. A valuable professional view that contrasts with the very optimistic statements of British Cycling, Cycling UK and others.
Total mileage is going up linearly, but at just 30% in 20 years, so about 1.5% each year. But numbers of cyclists are not going up. It is just the same number cycling a bit more each year.
We need to face up to the failure of the many initiatives to promote cycling and maybe think a bit outside the box. Bit more information and a chance to comment in the linked blog.
Looking more closely at the data shows that the increase in mileage is largely in Inner London. Away from Inner London the trend is very weak or zero. More on this in a later blog. Population surveys agree with the traffic data. Numbers of cyclists actually fell significantly over the first 2 years of Sport England's Active Lives Survey.
Seems to me important to face up to the failure of the many initiatives to promote cycling over recent years. Time to start thinking "outside the box".

road-traffic-estimates-in-great-britain-2017_cycling.pdf
File Size:	316 kb
File Type:	pdf

Download File

0 Comments

42% more female cyclists on quiet routes

24/10/2018

1 Comment

Counts at peak time show that female cyclists comprise just 23% of the total. On quiet routes with little traffic the proportion is 42% higher than on routes with a lot of traffic. Counts were made in South Inner London. In Bermondsey on the main radial routes towards the City there were 18% female cyclists on the busy A200 route compared with 26% on the parallel Quietway 1. At the Elephant there was a difference between the 2 main roads where we counted. On Newington Causeway going south there is some segregation and a safe junction. Here there were 23% female cyclists. On London Road where there is no protection and a still dangerous junction there were only 16% female cyclists.

At the Elephant and Castle, after the changes, there were 19% female cyclists on 2 of the main roads compared with 28% on the quiet bypass route.

At the Elephant there was a difference between the 2 main roads where we counted. On Newington Causeway going south there is some segregation and a safe junction. Here there were 23% female cyclists. On London Road where there is no protection and a still dangerous junction there were only 16% female cyclists.
None of this is actually very surprising. But it does demonstrate clearly that if we want more cyclists from a wider demographic, improving the quality of cycle routes is an effective way to do this. And an important aspect of quality is not cycling alongside heavy motor traffic and not competing with it at junctions.
Raw counts in the Excel sheet below

proportion_female_busy_quiet.xls
File Size:	89 kb
File Type:	xls

Download File

1 Comment

Cycling to work is as good for you as 5 gym sessions a week

7/11/2017

2 Comments

2017 has seen 2 publications from the Danish GO-ACTIWE randomised control trial on the relative health benefits of cycling and gym exercising (refs at the end).
Overweight or obese, but otherwise healthy, 20-45 year-olds were recruited. They were randomly assigned to 4 groups: BIKE, taking up cycling to work (9-17km depending on baseline fitness); VIG and MOD, 2 groups doing 5 gym sessions a week at moderate or vigorous intensity; CON, a control group asked to carry on normal behaviour, but who were promised a free gym membership at the end of the trial. 130 people stayed the course for 6 months, of whom 21 were in the cycling group.

Not surprisingly aerobic fitness (VO2max) improved in the active groups, by 11% for the cyclists and 15% for the vigorous gym group (NB not a statistically significant difference between groups). Body weight fell slightly by 1.8% (2.9kg) in the cyclists and by a similar amount in the VIG group. Good news was that all the weight loss was fat. As shown in the diagram, fat mass fell by 2.1kg in the BIKE group, a 7% reduction. The Copenhagen research group also looked at markers of inflammation that are risk factors for coronary heart disease. One of these, C-reactive protein (CRP), fell by 34% in the BIKE group (and a similar amount in the VIG group).
These data are in line with previous work from longitudinal and cross-sectional surveys (e.g. see 21/4/17 and 2/12/16 blogs below). But they are important as this was a well-designed trial with tightly monitored interventions. So it adds significantly to the growing body of data demonstrating the health benefits of cycling. And although gym session were as effective as cycling, there is no question which option is cheaper and is easier to fit into a busy schedule
Sources
J S Quist, M Rosenkilde,M B Petersen, A S Gram, A Sjodin, B Stallknecht, Effects of active commuting and leisure-time exercise on fat loss in women and men with overweight and obesity: A randomized controlled trial, International Journal of Obesity accepted article preview 10 October 2017; doi:10.1038/ijo.2017.253.
A S Gram, E-M Bladbjerg, J S Quist, M B Petersen, M Rosenkilde, B Stallknecht. Anti-inflammatory effects of active commuting and leisure time exercise in overweight and obese women and men: A randomized controlled trial. Atherosclerosis 265 (2017) 318-324
M Rosenkilde, M B Petersen, A S Grama, J S Quist, J Winther,
S D Kamronn, D H Milling, J E Larsen, A P Jespersen, B Stallknecht. The GO-ACTIWE randomized controlled trial - An interdisciplinary study designed to investigate the health effects of active commuting and leisure time physical activity. Contemporary Clinical Trials 53 (2017) 122–129

2 Comments

Southwark's Try-and-Ride Scheme

30/9/2017

0 Comments

Southwark council's cheap bike loan scheme is not found by searching the Southwark web site or the site of Cycle Confident who run the scheme. This is very puzzling. This is a very good scheme for getting people started or restarted in cycling. Please let us know your experience with the scheme by Commenting on this blog post.

0 Comments

<<Previous

Seasonal distribution of cycling

Cycling down 70% on Cycleway 1 during lockdown.

Effect of Lockdown on Cycle Traffic Flows

Serious injuries to cyclists are rising

Record low number of cyclists died in collisions in London in 2019

Cycle flows along Quietway 1 – Pattern of arrival and departures

A little more cycling, but same number of cyclists

42% more female cyclists on quiet routes

Cycling to work is as good for you as 5 gym sessions a week

Southwark's Try-and-Ride Scheme

Author

Archives

Categories