Trend-based multi-period decomposition and decoupling methodology for energy-related carbon dioxide emissions: A case study of Portugal

Governments worldwide are pursuing public policies to reduce greenhouse gas emissions while sustaining economic growth. Several methodologies, including the Logarithmic Mean Divisia Index (LMDI) decomposition, Tapio decoupling analysis, and the decoupling effort method, have been employed to analyze energy-related carbon dioxide emissions. These approaches have been applied across various time frames: single-period, year- by-year, and multi-period analyses. However, previous studies have often overlooked significant trend changes in the indicators. This study introduces a methodology that integrates decomposition and decoupling analysis within a multi-period time frame, explicitly accounting for major trend shifts in the carbon dioxide time series. The time frame is defined using a total mean squared error (TMSE) minimization approach. The decomposition analysis employs the additive LMDI method, while the decoupling analysis utilizes the Tapio and decoupling effort models. A case study of Portugal's carbon dioxide emissions from 1995 to 2020, disaggregated into six energy-consuming sectors, demonstrates the effectiveness of this methodology. The results highlight the substantial impact of carbon intensity, particularly in the electricity and heat sectors. This study demonstrates that accounting for trend changes in period selection provides critical insights, enabling amore thorough and accurate analysis of carbon dioxide emissions.