Estimating PM2.5 concentrations at public schools in North Carolina using multiple data sources and interpolation methods

Air pollution, particularly fine particulate matter (PM_2.5), poses serious health risks to children. Public schools are key locations for assessing air quality exposure. This study estimates PM_2.5 concentrations at North Carolina public schools using EPA and PurpleAir monitor data, applying Kriging and Inverse Distance Weighting (IDW) interpolation. Cross-validation is used to evaluate predictive accuracy. Whereas EPA monitors offer precise but spatially sparse data, PurpleAir monitors provide real-time, widespread coverage with variable accuracy. Combining both improves spatial resolution and predictive power. Kriging, which incorporates both distance and spatial correlation, consistently outperforms IDW in estimating PM_2.5, especially near EPA sites. The addition of PurpleAir data enhances accuracy further, particularly in areas lacking EPA monitors. School-level PM_2.5 estimates reveal suburban schools often face higher pollution levels. Linking these estimates to 8th grade reading scores indicates a negative relationship between PM_2.5 exposure and academic performance, suggesting possible cognitive impacts of air pollution. These findings underscore the importance of integrating PurpleAir monitors at school locations to improve air quality monitoring. Enhanced monitoring would support public health protections and inform education policy. Overall, combining EPA and PurpleAir data using Kriging offers a more accurate, actionable approach to understanding and addressing environmental determinants of student well-being and achievement.