An estimation of future county-level cement production and associated air pollutant emissions in China through artificial neural networks

Jiayu Xu,Jinya Yang,Jiaxin Dong,Siwei Li,Jia Xing,Yu Zhao
DOI: https://doi.org/10.1016/j.scitotenv.2024.176036
IF: 9.8
2024-09-13
The Science of The Total Environment
Abstract:Cement production and its air pollutant and carbon dioxides (CO 2 ) emissions in China will be relocated greatly as a joint effect of diverse development of industrial economy and implementation of environmental policies for different regions. The future pathway and spatial pattern of emissions are important for policy making of air quality improvement and CO 2 emission abatement, as well as coordinating regional development. In this study, we developed an artificial neural network (ANN) model to predict cement production at the county level and to calculate the associated emissions of air pollutants and CO 2 at the county level till 2060. Results show that the cement production will decline from 2327 million metric tons (Mt) in 2015 to 704 Mt. in 2060 under the Shared Socioeconomic Pathways 1 (SSP1). Counties closer to provincial capital will experience greater retirement of cement industry. Likewise, the emissions of air pollutants and CO 2 will experience a steady downward trend driven by the declining cement production and the improvement of pollution control technologies. There will be a more significant regional heterogeneity in the reduction of production and emissions at city level compared to the province level. With the clearance for nearly two-thirds of counties, future cement production and emissions will be more intensively distributed in a few cities. The shares of emissions in southwestern regions will grow from 2015 to 2060 while those of eastern regions will continue decreasing. The comparison between the changing spatial distributions of emissions and gross domestic product (GDP) indicates a positive effect of existing policies in reconciling regional economic development and air pollution controls. The outcome could support the analyses on the impact of industrial development on air quality and public health, and the method can be applied widely for other industrial sectors for a more comprehensive understanding of future emission relocation.
environmental sciences
What problem does this paper attempt to address?