Air pollution due to wood burning produces severe health and environmental problems. Clustering methods are needed to estimate PM2.5 exposures, and identify locations with high PM2.5 concentrations. This study performed a spatial and aspatial clustering analysis of PM2.5 pollutant collected in a mobile campaign in the conurbation of Temuco and Padre Las Casas, Chile. The Getis Ord G∗i statistic was employed to obtain spatial variability of PM2.5 concentrations, and a K-Means clustering method was used to group PM2.5 concentrations with a aspatial perspective. In addition, an integrated spatial and aspatial clustering approach was implemented with the PM2.5 concentration and measurement spatial location. The comparison results suggest that integrating the spatial and aspatial clustering methods yield high quality partitions when considering spatial information.