Spatial regression modelling of particulate pollution in Calgary, Canada

Area of Science:

• Environmental Science
• Public Health
• Spatial Analysis

Background:

• Particulate pollution poses significant health risks.
• Black carbon (BC) is a key component of PM2.5 and a growing concern.
• Understanding spatial distribution of pollutants is crucial for targeted interventions.

Purpose of the Study:

• To develop and validate land use regression (LUR) models for PM2.5 and PM10 concentrations.
• To analyze the spatial distribution of black carbon (BC) using the δC index.
• To identify key land use predictors of air pollution in Calgary.

Main Methods:

• Developed LUR models for PM2.5, PM10, and δC (BC index) in Calgary (Summer 2015, Winter 2016).
• Employed spatial autoregressive models (SARlag and SARerr) due to spatial autocorrelation.
• Validated models against observed data and mapped fine-scale pollution concentrations.

Main Results:

• SARlag models showed good fit, outperforming previous studies.
• Industrial activities, traffic, and elevation were consistent predictors.
• Predicted concentrations generally aligned with observed patterns, highlighting road networks and industrial zones.

Conclusions:

• LUR models effectively capture spatial variations in particulate pollution.
• Seasonal patterns and land use associations for PM fractions remained consistent over a 5-year interval.
• The study provides a high-resolution pollution map for urban planning and health assessments.