Scientific Reports, Год журнала: 2024, Номер 14(1)
Опубликована: Сен. 2, 2024
A comprehensive approach is essential in India's ongoing battle against air pollution, combining technological advancements, regulatory reinforcement, and widespread societal engagement. Bridging gaps involves deploying sophisticated pollution control technologies addressing the rural–urban disparity through innovative solutions. The review found that integrating Artificial Intelligence Machine Learning (AI&ML) quality forecasting demonstrates promising results with a remarkable model efficiency. In this study, initially, we compute PM2.5 concentration over India using surface mass of 5 key aerosols such as black carbon (BC), dust (DU), organic (OC), sea salt (SS) sulphates (SU), respectively. study identifies several regions highly vulnerable to due specific sources. Indo-Gangetic Plains are notably impacted by high concentrations BC, OC, SU resulting from anthropogenic activities. Western experiences higher DU its proximity Sahara Desert. Additionally, certain areas northeast show significant contributions OC biogenic Moreover, an AI&ML based on convolutional autoencoder architecture underwent rigorous training, testing, validation forecast across India. reveal exceptional precision prediction, demonstrated evaluation metrics, including Structural Similarity Index exceeding 0.60, Peak Signal-to-Noise Ratio ranging 28–30 dB Mean Square Error below 10 μg/m3. However, challenges persist, necessitating robust frameworks consistent enforcement mechanisms, evidenced complexities predicting concentrations. Implementing tailored regional strategies, technologies, strengthening frameworks, promoting sustainable practices, encouraging international collaboration policy measures mitigate
Язык: Английский