Ambient PM_2.5 Exposure Modeling in LMICs: An Example from Peru

Purpose of Review: Fine particulate matter (PM_2.5) poses a public health risk, disproportionately impacting low- and middle-income countries (LMICs). In Peru, where ambient concentrations in urban areas significantly exceed the World Health Organization’s annual guideline of 5 µg/m³, lack of air pollution monitoring hinders exposure assessment, health effect research, and policy development. Here, we review efforts to create a national database of estimated ambient PM_2.5 in other LMICs, and then discuss our efforts in Peru. Recent Findings: We highlight the Peru-based NIH-funded GeoHealth Hub’s efforts to establish a nationwide low-cost sensor (LCS) network of 176 PurpleAir monitors. We then describe a hybrid approach for modeling ambient PM_2.5 exposure across Peru, leveraging data from LCS, satellite remote sensing, chemical transport models, and advanced machine learning methods. The ground-monitoring network includes sensors in both urban (62.5%) and rural (37.5%) areas, in the 24 Regions of the country, set up in collaboration with national environmental agencies. Initial application of our hybrid approach in Lima demonstrated good prediction for the years 2010–2023, with an R² of 0.88 with existing regulatory ground monitors. We are working to extend the model across Peru at a daily level and at a 5-km² resolution for 2024–2026. Summary: The sustainability of these efforts will depend on building local capacity, securing long-term funding, and integrating the LCS network within the current regulatory environmental monitoring network. The hybrid approach offers a scalable solution to address data scarcity and enable high-resolution exposure modeling in Peru and other LMICs.