By Jenny Connelly-Bowen, CBN Graduate Research Assistant

This week, we’re sharing maps on the bicycle network in the St. Louis area. These were compiled by Bram Boettge, Damon M. Hall, and Thomas Crawford from the Center for Sustainability at Saint Louis University. Their recent paper, “Assessing the Bicycle Network in St. Louis: A Place-Based User-Centered Approach,” was published in Sustainability last week. You can download a copy of the full report here.

In their paper, Boettge, Hall, and Crawford discuss the benefits of bicycle commuting and the ways in which urban planners can encourage cycling with on-street bicycle infrastructure, including shared-use lane markings (“sharrows”), on-street painted bike lanes, and buffered bike lanes. Because data is mixed on whether these features actually increase bicycle commuting, the authors argue that cities should assess their bicycle infrastructure using a data-driven, user-centered approach.

Local cyclists can provide expert insight on how current street infrastructure affects their commuting patterns and stress levels. Past Bicycle Route Choice (BRC) studies have tapped participant cyclists for their preferences regarding route characteristics (such as speed limit, traffic level, and bike facilities present) and stress levels experienced on a given route. Boettge, Hall, and Crawford’s report adapts BRC approaches to gauge cyclist stress levels on shorter route segments in the St. Louis area’s bicycle network.

The maps below cover a stretch of St. Louis’ Central Corridor between East St. Louis County and Downtown. They were compiled using data from a survey administered to 89 St. Louis-area cyclists in August and September of 2015. The authors used the Level of Traffic Stress assessment method (developed by prior researchers) that gauges the amount of stress a cyclist feels while riding a particular route segment. After reporting on their level of cycling confidence, participants were given maps of the study area and asked to trace a recent bike route in red (lots of stress; LTS = 3), yellow (some stress; LTS = 2), and green (very little stress; LTS = 1). They were asked to weigh a variety of factors when reporting perceptions of stress, including traffic volume and speed, number of lanes, lane width, and presence of bicycle infrastructure.

Participant-supplied maps were then combined into a single layer for data analysis using GIS, and participant responses were used to generate two dependent variables: number of times ridden (NTR) and average level of traffic stress (avgLTS). Independent variables accounted for environmental factors on each route: street functional classification, traffic speed, number of lanes, and bicycle infrastructure type. (See pages 4-8 of the report for full details regarding the authors’ methodology.)

The first map displays the number of times each road segment in the study area was ridden by participants, as well as the bicycle infrastructure available along each segment. Yellow roads are used the least often, and red roads are used the most. Cyclists appear to use roads running east and west more often than those running north and south. Routes that run along the central core of the study area are also more heavily trafficked than those on the periphery. Among bicycle infrastructure types, “sharrows” are the most common, and buffered bike lanes are the least common:

Bicycle Infrastructure and Number of Times Ridden

The authors note that this map allows users to compare parallel roads according to the bicycle infrastructure available:

Many of these occurrences show that cyclists chose roads with bike lanes over roads with sharrows but chose roads with sharrows over roads with no infrastructure at all. One example is where Washington Avenue, Locust Street, and Olive Street run parallel just west of Downtown. Olive Street contains a bike lane and was the most heavily used of the three parallel streets (p. 8).

The next map displays the average Level of Traffic Stress values for the roads used by study participants; it also retains the bicycle infrastructure that was shown on the first map. Red segments are the most stressful for cyclists, and green segments are the least stressful. Collector and arterial roads appear to generate higher stress levels among cyclists than local roads (see page 7 of the report for road functional classification definitions), as do Downtown routes when compared to routes in other areas. Interestingly, many north-south roads—which were used less often by participant cyclists—were rated with higher stress scores than east-west roads:

Bicycle Infrastructure and Average Level of Traffic Stress Values

The authors note that although there is less consistency between bicycle infrastructure and stress scores than there is between infrastructure and use frequency, some comparisons are still discernible:

There are several occurrences where roads with bicycle lanes have higher avgLTS values than roads with sharrows or no infrastructure. Focusing on the same location as in the section above, we find an example of this where Washington Avenue, Locust Street, and Olive Street run parallel just west of Downtown. Olive Street, with a bike lane, is rated the most stressful out of the three streets (p. 10).

In their analysis, Boettge, Hall, and Crawford find that roads with higher speed limits, higher functional classifications, and more lanes draw more cyclists and produce higher stress levels. However, while robust bicycle infrastructure correlates with higher cyclist use rate, it does not have a statistically significant relationship with cyclist stress level. In their full report, the authors provide a detailed analysis on the individual effect of each independent variable on cyclist use rates and stress levels. They conclude with a discussion on how this data can be used by both officials and cyclists to improve bicycle network planning and cycling experiences in St. Louis.

To access maps from previous editions of the Community Builders Exchange, click here.