Environ Geochem Health. 2025 Dec 20;48(2):62. doi: 10.1007/s10653-025-02941-8.
ABSTRACT
Air pollution is a major public health concern, particularly for young children who are vulnerable to its adverse effects due to their developing respiratory and immune systems. This study assesses spatial exposure risks for early childhood development centers (ECDs) in Gauteng, South Africa, by analyzing their proximity to major pollution sources. Geographic information systems were used to conduct buffer analysis, kernel density estimation (KDE) and spatial autocorrelation analysis (Global Moran’s I) to identify pollution hotspots and potential clustering of ECDs located in high-risk areas. Statistically, the results showed significant spatial clustering of ECDs around pollution sources (Moran’s I > 0, p < 0.05). Using secondary data, results show that Johannesburg and Ekurhuleni have the highest number of ECDs in high-risk zones, despite Tshwane having the most pollution sources. This highlights that exposure is not solely determined by the number of pollution sources but is influenced by spatial distribution and urban density. Integration of Sentinel-5P satellite NO2 and SO2 data further validated the pollution exposure patterns, with elevated NO2 and SO2 concentrations observed along dense urban and industrial corridors, aligning with KDE-identified clusters. Sedibeng showed lower pollutant levels and more unexposed ECDs, suggesting that urban planning and source dispersion may mitigate exposure risks. A risk map developed by overlaying KDE outputs of pollution sources and ECDs highlighted critical hotspots requiring urgent intervention. Overall, the results suggest that pollution exposure is influenced more by population density and spatial distribution of pollution sources than by the sheer number of pollution sources. These findings underscore the need for targeted environmental health interventions in areas with high exposure risk. Moreover, established buffer zones around pollution sources can support safer spatial planning of ECDs.
PMID:41422363 | DOI:10.1007/s10653-025-02941-8
