Assessing the spatial distribution of nitrogen dioxide in London, Ontario

Figures & data

Figure 1. Map showing the 50 monitoring sites, land use categories, highways, railway, and Ministry of Environment (MOE) monitoring station in London, ON.

Table 1. Categories and corresponding variables considered for inclusion in the LUR model

	Buffer radius in meters (intervals)
Landuse
Industrial area (ha)	200–1600 (200)
Open area and water (ha)	200–1600 (200)
Residential area (ha)	200–1600 (200)
Commercial area (ha)	200–1600 (200)
Transportation
Distance to nearest major  road (km)	n/a
Distance to nearest highway (km)	n/a
Traffic density	50–1000 (50)
Length of major road within buffer (km)	50–1000 (50)
Length of Highway (km)	50–1000 (50)
Length of railway (km)	50–750 (50)
Population and dwelling
Population density within buffer	100–1500 (100)
Dwelling density within buffer	100–1500 (100)
Physical feature
Elevation (m above sea level)	n/a

Table 2. LUR model and cross-validation results for NO₂ based on 50 monitoring locations (R ² = 0.784; adjusted R ² = 0.760)

	Coeff.	Standardized coeff.	Partial R^2	OLS t-value	P > t	Jackknife t-value	P > t	VIF
Intercept				12.03	0.000	13.10	0.000
Industrial land use (1600m)	0.00361	0.201	0.097	2.17	0.035	2.38	0.021	1.754
Length of railways (550m)	0.45028	0.147	0.086	2.03	0.048	2.02	0.049	1.069
Distance to highway	–0.13267	–0.265	0.165	–2.95	0.005	–3.47	0.001	1.647
Traffic density (150m)	0.01412	0.644	0.608	8.26	0.000	6.15	0.000	1.238
Dwelling density (1000m)	2.14373	0.223	0.147	2.75	0.009	2.92	0.005	1.340
Note: VIF, variance inflation factor.

Figure 2. (a) Observed versus predicted NO₂ concentrations based on final LUR model. (b) Observed NO₂ versus standardized residuals from LUR model.

Figure 3.LUR model surface for NO₂ in London, Ontario.

Table 3. Comparison of NO₂ land use regression variables for different cities in Ontario

	Sarnia ^1	Windsor ^2	Hamilton ^3	Toronto ^4	London
Land use categories	Industry 1600 m	Industry 600 m	Industry 200 m	Industry 750 m	Industry 1600 m
	—	—	Open 500 m	—
	—	—	Downtown industrial core	—	—
Roads and traffic	Highways 400 m	—	Highways 50 m	Highways 200m	Distance to highway 401/402
	—	—	—	Major roads 50m	—
	—	Traffic density on C1 roads 400 m	Traffic density 300 m	Traffic density 500 m	Traffic density 150 m
	—	Traffic density on C2 roads 400 m	—	—	—
Railways	—	—	—	—	Length of tracks within 550m
Population and dwelling density	Dwelling density within 2400 m	Sum of population within 700 m	—	Dwelling density 2000 m	Dwelling density within 1000 m
Geographic location	Distance to industrial core	Distance to Ambassador Bridge	Distance to Lake Ontario	X (longitude)	—
Wind	Not used	Not used	Downwind of highway 403	Downwind of all highways 1500 m	Not used
Study area ^a	165 km^2	145 km^2	1372 km^2	633 km^2	422 km^2
Population size	71,419	208,000	660,000	2.6 million	1.6 million
R ^2	0.79	0.88	0.76	0.69	0.784
Number of samplers	39	50	107	100	50
Notes: Sources: ^1 Atari et al. (2008); ^2 Luginaah et al. (2006); ^3 Sahsuvaroglu et al. (2006); ^4Jerrett et al. (2007).
^aBased on city area except for Hamilton, which is based on census metropolitan area.

Atari , D.O. , Luginaah , I. , Xu , X. and Fung , K. 2008 . Spatial variability of ambient nitrogen dioxide and sulfur dioxide in Sarnia, “Chemical Valley,” Ontario, Canada . J. Toxicol. Environ. HealthA , 71 : 1572 – 1581 . doi: 10.1080/15287390 802414158

 PubMed Web of Science ®Google Scholar

Luginaah , I. , Xu , X. , Fung , K.Y. , Grgicak-Mannion , A. , Wintermute , J. , Wheeler , A. and Brook , J.R. 2006 . Establishing the spatial variability of ambient nitrogen dioxide in Windsor, Ontario . Int. J. Environ. Stud. , 63 : 487 – 500 . doi: 10.1080/00207230600802122

 Google Scholar

Sahsuvaroglu , T. , Arain , A. , Kanaroglou , P. , Finkelstein , N. , Newbold , B. , Jerrett , M. , Beckerman , B. , Brook , J. , Finkelstein , M. and Gilbert , N.L. 2006 . A land use regression model for predicting ambient concentrations of nitrogen dioxide in Hamilton, Ontario, Canada . J. Air Waste Manage. Assoc. , 56 : 1059 – 1069 . doi: 10.1080/10473289.2006.10464542

 Web of Science ®Google Scholar