Spatial-temporal evolution characteristics and driving factors of carbon emission prediction in China-research on ARIMA-BP neural network algorithm

TL;DR

This paper analyzes China's carbon emission trends using an ARIMA-BP neural network model, identifying key driving factors and proposing policy recommendations to achieve peak emissions by 2030 and carbon neutrality by 2060.

Contribution

It introduces a combined ARIMA-BP neural network approach to model and analyze the spatial-temporal evolution of China's carbon emissions and their driving factors.

Findings

Energy consumption intensity is the main driver of emission growth.

Per capita GDP and energy structure inhibit emissions.

Industrial structure and population size have smaller effects.

Abstract

China accounts for one-third of the world's total carbon emissions. How to reach the peak of carbon emissions by 2030 and achieve carbon neutrality by 2060 to ensure the effective realization of the "dual-carbon" target is an important policy orientation at present. Based on the provincial panel data of ARIMA-BP model, this paper shows that the effect of energy consumption intensity effect is the main factor driving the growth of carbon emissions, per capita GDP and energy consumption structure effect are the main factors to inhibit carbon emissions, and the effect of industrial structure and population size effect is relatively small. Based on the research conclusion, the policy suggestions are put forward from the aspects of energy structure, industrial structure, new quality productivity and digital economy.