Underestimated impact of the COVID-19 on carbon emission reduction in developing countries - A novel assessment based on scenario analysis

Abstract

Existing studies on the impact of the COVID-19 pandemic on carbon emissions are mainly based on inter-annual change rate of carbon emissions. This study provided a new way to investigate the impact of the pandemic on carbon emissions by calculating the difference between the pandemic-free carbon emissions and the actual carbon emissions in 2020 based on scenario analysis. In this work, derived from Autoregressive Integrated Moving Average (ARIMA) method and Back Propagation Neural Network (BPNN) method, two combined ARIMA-BPNN and BPNN-ARIMA simulation approaches were developed to simulate the carbon emissions of China, India, U.S. and EU under the pandemic-free scenario. The average relative error of the simulation was about 1%, which could provide reliable simulation results. The scenario simulation of carbon emission reduction in the US and EU were almost the same as the inter-annual change rate of carbon emissions reported by the existing statistics. However, the scenario simulation of carbon emission reduction in China and India is 5% larger than the inter-annual change rate of carbon emissions reported by the existing statistics. In some sense, the impact of the pandemic on carbon emission reduction in developing countries might be underestimated. This work would provide new sight to more comprehensive understanding of the impact of the pandemic on carbon emissions.

Keywords: Artificial intelligence; COVID-19; Carbon emissions; Developed and developing economies; Pandemic-free scene.