Friday, December 24, 2010

Advocate safer, robust, and efficient transportation for better cycling

What should cyclists advocate?  As an avid cyclist who enjoys such discussions, it is a question that comes up in one form or another.  In many ways, the same discussion arises during conversations with non-cyclists who simply want to get around in a convenient manner, have safe neighborhoods, and enjoy their lives.  That is the direction that I want to take advocacy.  Or more generally, a lot of "us" -- everyone in a neighborhood instead of simply cyclists -- want safe, vibrant, and livable neighborhoods and as Charles Marohn writes,
When you can't let your kids play in the yard, let alone ride their bike to the store, because you know the street is dangerous, then the engineering profession is not providing society any real value. It's time to stand up and demand a change.  It's time we demand that engineers build us Strong Towns.
The point is that instead of talking about what cyclists want, how great cycling is, or the many other benefits that it can give, we should recognize that a lot of what is good for strong neighborhoods is good for cyclists. 

My conclusion in a nutshell: slower road speeds are the easiest and most effective way to improve the cycling/pedestrian environment and a huge improvement for livable neighborhoods in general.  Underlying the idea is that a modest level of uncertainty is good since it requires drivers, cyclists and pedestrians to remain aware of their surroundings to successfully navigate the environment.  Instead of calling this bicycle or pedestrian advocacy, I think that we should push for safer, robust, and more efficient transportation network since we can make transportation safer, give people more acceptable options, as well as make driving trips shorter (make the same trip in less time).  We can make everyone better off regardless of their transportation choice.

Slower is safer for everyone

The claim that slower speeds are safer on local roads is probably accepted by just about everyone.  As one drives faster the distance covered before reacting and stopping increases; controlling a vehicle becomes more difficult; and damage increases in the event of a collision.  How much more dangerous is probably something that remains quite fuzzy for most people.  Given that an alert driver takes about 1.5 second to react and a few assumptions about the rate of deceleration we can determine the following:

Breaking Stopping Distances

Relative to 20 mph, traveling 30 or 40 mph increases stopping distance by ~70% and 160%.  Clearly as one travels faster, the ability of an alert driver to slow their vehicle decreases at an increasing rate. Even a slightly distracted driver -- say it takes three seconds to react doubling the distance in the table above -- would significantly exacerbate the issue.  As expected, vehicle speed is highly correlated with pedestrian and cyclist injury/mortality.
From DOT HS 809 021 October 1999
A pedestrian hit by a vehicle traveling in the low 30s instead of low 20s has almost a three to fourfold increase in mortalityA statistic that is perhaps more surprising for many is that the risk to the driver is significant too

Injuries per 100 Occupants by Change in Speed (deltaV) at Impact
delta VModerate InjurySerious Injury
mi/hAIS 2+AIS 3+
Bowie and Waltz (1994)
Mortality for drivers also follows a similar pattern: velocity changes of 20, 30 , and 40 mph are estimated to have a 0.6, 3.2, and 10.1% fatality rate (Joksch, 1993).  Clearly, cyclists, pedestrians and drivers are all better off at moderate residential speeds in the event of a collision.  Perhaps more important is that the traffic calming literature -- see the same FHWA synthesis -- strongly supports that collisions are reduced as well.  A recent traffic calming success on Lawyers Road in Fairfax County demonstrates the concept.  Consequently, pushing for roads designed for slower velocities and greater user awareness, as opposed to roads built like freeways with lower speed limits, is likely to produce fewer and less serious collisions/crashes. 

A more robust transportation system

A transportation network that gives people several acceptable transportation options is more robust than one that emphasizes driving only and can benefit even those that typically drive.  This is most obvious during major events such as the recent snow storms ("Snowmageddon") or "Tractor Man" where people were forced to choose alternatives due to their cars being trapped by snow or restrictions that made driving unreasonable.  More transportation options, however, are useful for far more mundane problems such as poverty -- cars are expensive to own and maintain: greater than $5K annually for a Ford Focus -- vehicle downtime, or as an alternative for a household second or third vehicle. 

Let me briefly emphasize that last point.  Parents often purchase (or let their teenagers use their vehicle) another car for a teenager because few alternatives exist and chauffeuring him/her around becomes a burden.  In short, teens are terrible drivers.  Giving parents alternatives to letting their teenager drive or greater leverage in limiting driving until a teenager demonstrates greater maturity and skill could literally save their life.  For instance, legislative changes that limit teenager driving have resulted in fewer fatal collisions.    

Naturally, there are people who would actively choose to cycle or walk more if they felt conditions were better.  As I demonstrated above, a strategy that lowers road speeds would make a given road safer and likely more acceptable.  Mind you, I believe virtually any road can be cycled on with a reasonable level of safety by applying a few rules and riding according to the rules of the road.  Based on my conversations with people of varying levels of interest in bicycling, however, when motorized traffic on a road crosses some threshold of volume and speed it is classified as too dangerous (or simply undesirable) regardless of the evidence.  Adopting a paradigm where slower but better flowing traffic becomes the norm clearly makes more roads accessible to average citizens.  

Yet typically one can simply look at a map in combination with some local knowledge to identify "high stress" areas or other obstruction that would limit cycling and walking for many even with some measure of traffic calming.  If these areas represent major connecting points in a local transportation grid, then addressing desirability via some smart engineering is probably worthwhile.  That is, if the cycling and pedestrian transportation network is meaningfully expanded with some thoughtful changes, then it should be seriously considered and pursued.   "Thoughtful" is the operative word here since careless application of bicycle treatments can create perverse situations for cyclists and other road users by increasing the risk of collision at with crossing traffic (intersections) which represents the greatest risk of collision to the cyclist.  While this is supported by observations in the U.S. as well as overseas, there are studies that find little correlation between facilities and collision risk but that facilities are also correlated with increased use and an overall improved safety that many attribute to safety in numbers.  Overall, my take on the literature is that it doubles my emphasis on the "thoughtful" adjective used earlier.  Over-engineering our roads can not only decrease safety directly, but facilities often remove driver/cyclist responsibility from actively thinking about their actions.  But in many cases, the engineering solution is the easiest way to connect sections of the transportation grid for many cyclists and pedestrians.  If one is looking to expand the acceptable cycling transportation grid through high stress areas, pick places with little or no crossing traffic, long ascents on roads with moderate to high traffic volumes, and so on.

More efficient and slower?

Clearly we all expect road diets and traffic calming to be safer however, their effect on vehicle capacity and travel duration is often minimal.
Under most annual average daily traffic conditions tested, road diets appeared to have minimal effects on vehicle capacity because left-turning vehicles were moved into a common two-way left-turn lane. (FHWA)
For instance, the Lawyers Road diet referenced earlier pulled back extreme speeds but only lowered average speed from 45 to 44 mph.  Moreover, distances are generally much shorter for local travel consequently the absolute change in travel duration will be small: a 5-mile trip without any stop lights driven at 30 versus 40 mph takes 10 and 7.5  minutes respectively.  Traffic signals clearly decrease the percentage difference between the two.  But we can actually slow traffic, decrease collisions, and improve travel times with a simple traffic device: the roundabout.

Many Washington DC locals shudder when they think about the high speed traffic circles that litter the area.  Modern roundabouts are not traffic circles nor rotaries!  Roundabouts have much more deflection, are much smaller than circles, and consequently have slower speeds.  Some of the benefits of modern roundabouts include:
  • Reduces injury accidents by 75 percent and fatal accidents by 90 percent.
  • Increases efficient traffic flow up to 50 percent.
  • Helps the environment by reducing carbon emissions by double digits.
  • Decreases fuel consumption by as much as 30 percent.
  • Costs less than traffic signals and does not require expensive equipment or maintenance.
Roundabouts never have a "dead time" like traffic signals where no traffic is in the intersection.  Roundabouts avoid long vehicle waits during off hours.  By design, entering traffic yields to traffic already in the circle and pedestrians traveling on the outside.  Consequently, successfully navigating the roundabout requires that the driver pay attention instead of mindlessly zipping through a green light.  As demonstrated by the red dots, roundabouts have far fewer conflict points than a signalized intersection.

The standard signalized intersection has 32 vehicle-to-vehicle conflict points

The roundabout has 8 vehicle-to-vehicle conflict points.
Not all intersections are appropriate for roundabouts and the literature has mixed reviews for cyclists since some of the early designs put bike lanes in the "circle".  Nonetheless, for the typical local trip, a few well placed roundabouts could easily reduce travel time to offset slower velocities and newer designs either encourage cyclists to behave as a vehicle or provide easy access to behave as a pedestrian


I know several cyclists that think fondly of European facilities.  While they look like a lot of fun and could work in some areas, even Copenhagen facilities show some warts and, to be frank, I have doubts that many would be willing to pay the price to have them.  Moreover, bike safety and ridership in many Netherlands cities likely experience a positive effect from strict liability and slow road speeds.  So the concept of building livable neighborhoods takes into account the European experience.   

Broadly speaking, here in the United States we spent a lot of energy and resources pursuing a strategy of safe crashing.  While seatbelts and other technological achievements are true advancements that help prevent injury and death in the car, we should also consider how people drive when they are in a protective cocoon and the effect on neighborhood livability.  Local streets should be for local travel which includes cycling and walking.  What this means is that for better neighborhoods we need to design our roads for slower travel that require greater driver attention.  Moreover, we need to stop building roads like freeways and simply slapping a 30 mph speed limit on the road.  We should consider innovative and efficient ways to enforce speed limits such as the speed camera lottery or ordinary automated cameras where engineering a slow road proves too costly for various reasons.

Thanks for reading!

Post Script

Just to be clear, I breezed through many details of the safety literature that might be important to particular individuals.  Moreover, I omitted many interesting studies and articles that could be related to various points discussed in the post.  My objective here is to give a broader outline but provide links that an interested individual could follow to obtain a foothold in the literature.  I'm also trying to reach a broader audience that might know little or only one side of the safety literature as well as a much broader community that could support pro-neighborhood policies that would find a more in depth discussion tedious.