Bicycle streetscapes: a data driven approach to mapping streets based on bicycle usage

Figures & data

Table 1. List of variables and their operationalization along with data sources.

Group	Variable	Operationalization	Data Source
Built Environment	bus_dist	Distance (m) to nearest bus stop	OpenStreetMap highway = bus_stop
	commercial_retail_perc	% of commercial and retail centers within a 30 m buffer of a street segment	OpenStreetMap landuse = [commercial, retail]
	natural_dist	Distance to nearest park, natural land use, or water.	OpenStreetMap natural = * OR leisure = park OR waterway = *
	parking_dist	Distance (m) to nearest motor vehicle parking	OpenStreetMap amenity = parking
	shopping_entertainment_dist	Distance (m) to nearest entertainment and shopping centers	OpenStreetMap shop=* OR amenity = [bar, cinema, cafe, restaurant, fast_food]
Socioeconomic &Demographic	indigenous_perc	% of total population that is indigenous	Canadian Census total aboriginal identity in the population / the total population
	income_med	Median income	Canadian Census median income of economic families in 2015
	minorities_perc	% of total population that belong to visible minority groups	Canadian Census total visible minority population / the total population
	pop_dens	Population density	Canadian Census population density per square kilometer
Bicycling Infrastructure, Behavior, & Safety	bike_mode_share	Bicycle mode share	Canadian Census main mode of travel to work for the working population aged 15 or older / the working population aged 15 or older who commute to work
	Can_BICS	Bike infrastructure comfort class, no infrastructure or non-conforming, low (painted lanes on busy roads), medium (multi-use paths), or high (cycle tracks, bike paths, and local street bikeways).	City of Ottawa open data
	incident_dist	Distance (m) to nearest bike incident	BikeMaps.org, Ottawa police dept.
	Strava_commute_perc	% of activities classified as commutes.	Strava Metro
	Strava_trip_count	Mean daily Strava trips on a street segment on summer weekdays	Strava Metro

Figure 1. The Ottawa streetscape categories. Basemap tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL.

Figure 2. Distributions by street segment (n = 12,170, 1,215 km) for the variables used to classify streetscape category. Scaled values are shown to compare multiple units. Note varying x axis.

Figure 3. Bubble chart showing how the incidents normalized by street length varies with the number of Strava riders for each streetscape category. The bubble number lists the category and the size of the bubble represents the total length in kilometers.

Table 2. Summary of the geometry, length-weighted means (for continuous variables), and proportions (for categorical variables) for the streetscape category.

Variable type	Variables	Type (unit)	Streetscape category
			Bicycling destinations	Wealthy neighborhoods	Urbanized	Lower income neighborhoods	Central residential	Study Area
Geometry	Length	kilometers	143	263	87	262	459	1,215
	Length	% of study area	12	22	7	22	38	100
Built Environment	bus_dist	meters	183	223	58	91	78	123
	comm_retail_perc	%	3	0	75	4	5	8
	natural_dist	meters	17	83	150	112	66	80
	parking_dist	meters	86	178	21	49	50	80
	shop_ent_dist	meters	290	566	101	181	152	260
Socioeconomic & Demographics	indigenous_perc	%	2	2	2	4	2	3
	income_med	$	119,029	148,287	103,811	80,637	113,014	113,712
	minorities_perc	%	20	16	24	31	20	22
	pop_dens	people / km^2	2,538	2,077	4,168	3,868	4,569	3,610
Bicycling Infrastructure, Behavior, & Safety	bike_mode_share	%	8	8	6	4	9	7
	incident_dist	meters	309	532	286	626	202	384
	Can-BICS low comfort	% of category	7.5	1.3	3.6	2.5	4.9	3.8
	Can-BICS medium comfort	% of category	89.5	0	0.5	0	0	10.6
	Can-BICS high comfort	% of category	3	0	0	0	0	0.4
	Strava_commute_perc	%	54	44	56	52	57	53
	Strava_trip_count	count	35	13	16	9	16	16