China has become the world's largest emitter of carbon dioxide, putting significant pressure on the government to reduce emissions. This study analyzes the driving factors of carbon emissions in 281 prefecture-level cities in China from 2003 to 2019, based on carbon emission data matched with the locations of thermal power stations and nighttime light data. Firstly, we compare the accuracy of multivariate linear regression and random forest models, finding that the random forest regression yields superior results. Then, we rank the impact of various factors using the random forest method, revealing that population, economic development, and industrialization are the top three influencing factors. The interaction between population and economic development explains 68.5% of carbon emissions, with regional variations in the ranking of influencing factors. The main policy implications of this study are as follows: firstly, there is no need to overly concern about the impact of population growth on carbon emissions, and policies regarding fertility can be adjusted flexibly; secondly, controlling urbanization to a certain extent is conducive to achieving efficient low-carbon cities; thirdly, during the process of industrialization, carbon emissions inevitably increase, and it is advisable to accelerate industrialization to reach a turning point as soon as possible.
Keywords: CO2 emissions; Emission reduction strategies; Environmental impact; Random forest.
© 2024 Published by Elsevier Ltd.