The spatiotemporal evolution of carbon emissions in China using machine learning methods: evidence from remote sensing data

The spatial-temporal distribution and heterogeneity of carbon emissions are important issues in ecological environmental protection and climate change monitoring. Based on DMSP/OLS night light remote sensing data and socioeconomic data, this paper uses machine learning methods to study the carbon dioxide emissions of 281 prefecture-level cities in China from 2005 to 2020, and reveals the characteristics of spatial-temporal distribution, growth trend and agglomeration of carbon emissions. The results show that in recent years, Chinese carbon emissions show an increasing trend, and regional carbon emissions have obvious characteristics of spatial agglomeration and unbalanced distribution. From the spatial dimension, the pattern of carbon emissions gradually evolves to "high in the east and low in the west" and "high in the north and low in the south". This study can provide scientific reference for carbon emission estimation and prediction, energy conservation and ecological environmental protection.