DECODING FUTURE CO₂ EMISSIONS IN THE AGRI-FOOD SECTOR: INSIGHTS FROM SCENARIO ANALYSIS

As concerns over climate change intensify, the need for accurate and efficient monitoring of carbon dioxide (CO₂) emissions has become increasingly urgent, particularly in sectors like the agri-food industry, which contribute significantly to global emissions. This work combines Principal Component Analysis (PCA), Random Forest regression, and second-degree polynomial modeling to forecast CO₂ emissions under three strategic scenarios: Sustainable, Aggravation, and Efficient. Using historical data, the models project emissions through 2030 and 2050, incorporating scenario-specific policy adjustments. Results show that mitigation strategies in the Sustainable and Efficient scenarios could reduce emissions by up to 46.5% and 62.5%, respectively, by 2050 compared to 2019 levels. In contrast, the Aggravation scenario predicts a near doubling of emissions. These findings highlight the complexity of environmental data and the value of scenario-based modeling for informing sustainable policy and strategic decision-making in the agri-food sector.