Characterizing human exposure to traffic-related air pollution through daily activities using an agent-based...

Characterizing human exposure to traffic-related air pollution requires accounting for both the spatio-temporal variability of pollutant concentrations and the diversity of human activities. Agent-based mobility models coupled with synthetic populations offer promising opportunities to integrate these dimensions within a unified modelling framework. In this study, we propose an integrated modelling chain combining the EQASim synthetic population generator, the MATSim agent-based transport model, the COPERT emission model, the SIRANE dispersion model, and a dedicated exposure module. We show that dynamic mobility patterns can be integrated into exposure analyses to characterize exposure according to daily human activities. The proposed framework is applied to the Lyon metropolitan area and compared with a conventional static exposure approach. Results show that cumulative daily exposure indicators remain relatively similar between the static and agent-based approaches at the metropolitan scale. However, the dynamic framework provides additional information by capturing intra-day exposure variability, activity-specific exposure patterns, and the redistribution of populations throughout the day. Individual trajectories illustrate how work, education, and leisure activities modify exposure profiles in ways that cannot be represented by static approaches. Moreover, we find that many individuals experience high pollution levels at home and exposure follows daily traffic cycles. The proposed framework provides a basis for future developments integrating travel exposure, multi-day mobility patterns, more comprehensive assessments of environmental inequalities and other traffic-related nuisances, and coupling with well-established health impact assessment frameworks.