Air pollution poses a great threat to public health and social stability by influencing multiple emotions. In particular, the air quality in developing countries is deteriorating along with rapid industrialization and urbanization, and multiple emotions may change along with regulation updates and air quality trending. Monitoring changes in public emotion is crucial for environmental governance. However, limited evidence exists for long-term effects of air quality on fine-grained emotions. Traditional surveys have the drawbacks of spatial limitations and high costs of time and money. Here, we use deep learning models to identify multiple emotions of over 10 million haze-related tweets and evaluate the effect of air quality on emotional predispositions for 160 cities from 2014 to 2019 in China. We find that sadness and joy are persistently associated with air quality, while anger and disgust are not. Surprisingly, the effects on fear vanished in the last three years. Moreover, air pollution initially had a greater impact on expressed fear in cities with higher income, poorer air quality and a greater percentage of women. Through popularity ranking and dynamic topic model, we interpretively revealed that people are no longer overly panicked and their attention is shifting toward policies and sources of haze. Our findings highlight the temporal evolution in the public's emotional response and provide significant implications for equitable public policies.
Keywords: Air pollution; China; Emotions; Fear; Haze pollution; Social media.
Copyright © 2022 Elsevier B.V. All rights reserved.