Modeling Psychological Toll in the Context of Urban Air Pollution Risk: Geography and Validation, 2020 versus 2021

Air pollution degrades urban air quality and affects public health and social well-being. While emotional responses to pollutants have been documented, few studies have examined the cross-year spatiotemporal distributions of these effects. This study analyzed data from over 14,000 urban park visitors in 48 Chinese cities in 2020 and 2021. Facial expression scores for happiness, sadness, and neutrality were regressed against same-day concentrations of particulate matter (PM 2.5 , PM 10 ), nitrogen dioxide (NO 2 ), and ozone (O 3 ) using hybrid machine learning models. The concentrations of PM 2.5 , PM 10 , and NO 2 peaked during winter in northern and central China, whereas those of O 3 increased during spring and summer. High positive emotion indices were most frequent during autumn in southern cities, whereas sadness was more prevalent in northern winters. The machine learning results indicated that particulate matter had a stronger influence on emotional shifts than socioeconomic factors. Air pollutants consistently drive negative emotional perceptions, particularly in northern Chinese cities during winter.