A vehicle-level analysis of public transport emissions and sociodemographic disparities using mobile phone data for group characterization

Public transport systems are a major source of urban air pollution, yet conventional approaches often fail to capture the spatial and temporal dynamics of population exposure to bus emissions. This study proposes a framework to estimate public transport emissions and assess population exposure to emissions with high spatial and temporal resolution. The approach integrates GPS-based bus trajectories, vehicle-level emission factors, and dynamic population data derived from mobile phone records. By capturing real-world driving conditions and hourly population flows, our method enables a detailed mapping of where emissions occur and who is exposed. We apply this framework to Santiago, Chile, revealing that traditional static models, i.e., those that do not account for driving conditions, significantly underestimate emissions by 37.5% for PM2.5 and 36.8% for NOx. The results also show pronounced disparities in exposure: working-age adults and lower-middle-income groups face the highest emission levels, while wealthier and younger populations are consistently less exposed. These differences reflect commuting patterns, spatial segregation, and the allocation of cleaner vehicles.