Answers are grounded in this paper and cite their sources.
Air pollution is a significant global health risk. Accurate spatial and temporal estimates of fine particulate matter (PM 2.5 ) concentrations are therefore essential. However, the high cost of deploying and maintaining comprehensive monitoring networks often limits their spatial coverage, particularly in low-resource settings. This study addresses this limitation by developing a hybrid modeling framework to estimate daily PM 2.5 concentrations between 2004 and 2022 in four Mexican metropolitan areas, integrating satellite-derived Aerosol Optical Depth (MAIAC AOD), meteorological reanalysis data (ERA5), ground-based measure-ments from the National Air Quality Information System (Sistema Nacional de Información de la Calidad del Aire, SINAICA), and spatial covariates derived from OpenStreetMap. In areas where direct PM 2.5 monitoring was insufficient, PM 10 information was used to support subsequent proxy-based PM 2.5 estimation. The proposed approach combines linear mixed-effects models and generalized additive models to obtain spatially continuous PM 2.5 estimates, including days and locations where satellite observations are unavailable. Model evaluation is based on leave-one-out cross-validation and multiple error metrics, allowing comparison between station-based and reanalysis meteorological inputs. In the best-performing region, the Mexico City Metropolitan Area (MCMA), yearly Root Mean Square Error (RMSE) values for PM 2.5 estimates ranged from 3.67 to 9.91 µg/m 3 over the study period. The use of reliable meteorological data supports the generation of PM 2.5 estimates in urban areas that are typically excluded from analyses due to limited monitoring coverage, while maintaining temporal consistency over the study period.
No takes yet. Share an insight, caveat, or question.
Salazar et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: