Land Use Regression of Particulate Matter in Calgary, Canada

Abstract

Two-week integrated samples of particulate matter (PM1.0, PM2.5, PM10) were collected in summer and winter in Calgary, Canada. PM concentrations were higher in summer for all size fractions. In both seasons, spatial variation and clustering were moderate. Land use regression (LUR) models were estimated for each PM size fraction and season, yielding R 2 > 0.75 for PM2.5 and PM10 in summer, and R 2 > 0.45 for PM1.0 in summer and for all winter models. Summer models yielded consistent predictors across size fractions, representing industrial emissions, local traffic, and major arterial traffic. Winter predictors included industrial emissions, major arterial traffic, and distance from open, snow-covered parks. The models suggest industrial pollution covered large areas in both seasons, and was affected by prevailing winds in summer, whereas traffic-related pollution decayed rapidly as distance from roads increased.

Schools, Air Pollution, and Active Transportation: An Exploratory Spatial Analysis of Calgary, Canada

“…In the winter months, air pollution levels increase, owing to residential heating and more intense local driving. Overall, Calgary’s air quality is generally considered safe, yet pollution levels exhibit large spatial variation: highest pollution levels are recorded in the east, the downtown core, and along the river valleys and transportation corridors [ 37 , 38 , 39 ].…”

“…This network provides a sparse sample of pollution records and a poor representation of the spatial pattern of air pollution over the 800 km 2 urban area. High-resolution estimates of air pollution over the city were obtained from land use regression (LUR) models [ 37 , 38 , 39 ] computed by our group based on a large array of spatial data, collected from a network of 50 monitors deployed over two-week campaigns in summer 2010 and winter 2011. LUR estimates for nitrogen dioxide (NO 2 ) and fine particulate matter (PM 2.5 ) were used in this study.…”

“…Using spatially detailed air quality estimates in Equation (1), we calculated a pseudo-AQHI (p-AQHI) for each dissemination area where one or more schools are located. LUR models [ 37 , 39 ] provided estimates for nitrogen dioxide (NO 2 ) and fine particulate matter (PM 2.5 ). NO 2 LUR models used geographically weighted methods, yielding R 2 measures of 0.86 for the summer, and 0.73 for the winter.…”

“…In all models, significant predictors were indicators of industrial activity and traffic. In addition, NO 2 models included wind speed and direction, and population density in the winter [ 37 , 39 ]. Ozone (O 3 ) concentration estimates were not available from the LUR study [ 37 , 38 , 39 ].…”

“…In addition, NO 2 models included wind speed and direction, and population density in the winter [ 37 , 39 ]. Ozone (O 3 ) concentration estimates were not available from the LUR study [ 37 , 38 , 39 ]. Based on the assumption that O 3 is known to be a regional pollutant, i.e., one that exhibits relatively low spatial variation [ 50 ], we used O 3 data from the regulatory passive stations over the corresponding periods.…”