Portal de Eventos, Congreso Colombiano y Conferencia Internacional de Calidad del Aire y Salud Pública

Tamaño de la fuente: 
A source-scaling method for PM source apportionment in CMAQ simulations of Bogota air quality
James East

Última modificación: 04/06/2019


Concentrations of fine and coarse particulate matter (PM2.5 and PM10) in Bogota frequently exceed Colombia’s national ambient standards and contribute to negative health impacts for the city’s 8.5 million residents. Air quality modeling of the Bogota urban area completed during the past several years has improved understanding of the air quality problem in Bogota. Recent advances in the emissions inventory for Bogota have led to improved modeling of PM2.5 and PM10 for comparison with observations and refined guidance for pollution control measures. However, uncertainty remains in the chemical composition of modeled PM and in the source apportionment of specific emissions sectors to the total PM mass at observation sites, limiting inferences that can be drawn from observations and existing model data. We close this gap by applying a brute-force source-scaling method to reveal source apportionment at 5 sites where chemically speciated observations are available. In addition, we apply source specific chemical emission profiles to infer a greater degree of chemical composition in model output for comparison to observations. We show that resuspended particulate matter (RPM) from unpaved roads contributes 48% of PM10 and 27% of PM2.5 at observation sites, on average. At observation sites, on-road emissions contribute 27% of total PM2.5, averaged across all sites, which is equivalent to the proportion contributed by RPM. RPM contributes to 44% and 70% of PM2.5 and PM10, respectively, at an observation site colocated with unpaved roads. Applying chemical emissions profiles to model results decreases the unspeciated portion of PM10 from 64% to 27% at an urban background site. The application of this method extends previous modeling work and allows for useful comparison to observed, speciated PM data.