Air pollution poses a significant threat to human health, especially for the vulnerable groups such as children. Given that schools are central to their daily lives, ensuring good air quality in these environments is crucial. This study evaluates the impact of traffic restriction interventions around schools by integrating citizen science monitoring data with advanced modeling techniques. From February 4 to March 4, 2023, within the framework of a citizen science project called "NO2, No Grazie!", NO2 concentrations were measured in Milan and Rome (Italy), Italy's two most populated cities, both affected by high traffic-related pollution, using passive samplers. The spatial distribution of NO2 across entire city territories was estimated using Land Use Random Forest (LURF) models. Four traffic restriction scenarios were developed alongside a business-as-usual one; furthermore, each school was characterized by the social vulnerability of its area. In total, 486 samplers were analyzed in Milan and 407 in Rome, with NO2 levels averaging 47.1 μg/m3 and 42.6 μg/m3, respectively. LURF models explained 64 % and 53 % of the measured variability, with traffic proximity as a major predictor. Among 659 schools in Milan and 1595 in Rome, all traffic restriction scenarios led to significant NO2 reductions. The most effective scenario reduced NO2 by 2.7 μg/m3 in Milan and 1.9 μg/m3 in Rome on average, with maximum observed decreases of 11.1 μg/m3 and 16.1 μg/m3, respectively. Schools in socioeconomically deprived areas had lower NO2 levels and were less impacted by the restrictions. The study underscores the value of traffic policies in improving air quality around schools.
Keywords: Air pollution; Children; Citizen science; Machine learning; Schools; Social vulnerability; Sustainability; Urban health.
Copyright © 2025 The Authors. Published by Elsevier Inc. All rights reserved.