Machine learning application in forecasting tire wear particles emission in China under different potential socioeconomic and climate scenarios with tire microplastics context.

Little information is available on different contribution of TMPs from tire wear particles (TWPs), recycled tire crumbs (RTCs) and tire repair-polished Debris (TRDs) in the environment at national scale and their potential tendency. In this study, the TWPs were predicted using machine learning method of CNN (Convolutional Neural Networks) algorithms under different potential socioeconomic and climate scenarios based on the estimation of TMPs in China. Results showed that TWPs emission exhibited the most important part of TMPs, followed by RTCs and TRDs in China. The three mentioned tire microplastics largely distributed in Chinese coastal provinces. After machine learning applied in CNN using the dataset of estimated emission of TWPs from 2008 to 2018, the express delivery volume and education funding at the current increased rate would not have significant impacts on TWPs emissions; Additionally, TWPs emissions were also sensitive to changes of economic and transportation development; Low temperature conditions would further promote TWPs emissions. Accordingly, the rational development of logistics and green economy, the equilibrium improvement of education quality, and the increase of public traffic with new energy would be helpful to mitigate TWPs emissions. The obtained findings can enhance the understanding TMPs emission at particular scale and their corresponding precise management.