Rufaioglu Süreyya Betül, Ismael Amjed Mohamed, Kaplan Fatma, Bilgili Ali Volkan
Department of Soil Science and Plant Nutrition, Faculty of Agriculture, Harran University, Sanliurfa, Turkey.
Environ Monit Assess. 2025 Sep 2;197(10):1075. doi: 10.1007/s10661-025-14496-3.
Accurate forecasting of greenhouse gas (GHG) emissions is essential for assessing climate change dynamics and developing evidence-based environmental policies. This study aims to comparatively evaluate the prediction performance of various machine learning algorithms using annual GHG emission data (CO, CH, and NO) for Turkey from 2012 to 2021. The dataset was split into 80% training and 20% testing subsets. The input variables consist of the year and emission category codes, while the output variable represents the annual emission value for each gas. The machine learning algorithms applied in the analysis include random forest, decision tree, ensemble regressor, LightGBM, gradient boosting, and XGBoost. Model performance was assessed using error metrics such as R, MAE, RMSE, and MSE. The results indicate that gradient boosting algorithms particularly gradient boosting (R = 0.995) and XGBoost (R = 0.994) achieved the highest accuracy, significantly outperforming other models. LightGBM and ensemble regressor also delivered strong predictive performance, whereas the decision tree model showed the lowest accuracy. The analysis further reveals that more than 90% of total GHG emissions are attributable to CO, and all three gases exhibited a consistent upward trend over the study period. This study is among the few to focus on the annual level forecasting of greenhouse gas emissions in Turkey using machine learning algorithms. It offers a comparative evaluation of random forest and gradient boosting methods, highlighting their performance across different emission categories. The study contributes to data-driven decision-making processes in regional climate policy. Furthermore, the findings suggest that integrating AI-based forecasting tools into GHG monitoring systems can significantly enhance transparency, accuracy, and response capacity in climate governance.
准确预测温室气体（GHG）排放对于评估气候变化动态和制定基于证据的环境政策至关重要。本研究旨在使用2012年至2021年土耳其的年度温室气体排放数据（CO、CH和NO），对各种机器学习算法的预测性能进行比较评估。数据集被分为80%的训练子集和20%的测试子集。输入变量包括年份和排放类别代码，而输出变量表示每种气体的年度排放值。分析中应用的机器学习算法包括随机森林、决策树、集成回归器、LightGBM、梯度提升和XGBoost。使用R、MAE、RMSE和MSE等误差指标评估模型性能。结果表明，梯度提升算法特别是梯度提升（R = 0.995）和XGBoost（R = 0.994）实现了最高的准确率，显著优于其他模型。LightGBM和集成回归器也具有很强的预测性能，而决策树模型的准确率最低。分析还表明，温室气体排放总量的90%以上归因于CO，并且在研究期间所有三种气体都呈现出一致的上升趋势。本研究是少数几个使用机器学习算法关注土耳其温室气体排放年度水平预测的研究之一。它对随机森林和梯度提升方法进行了比较评估，突出了它们在不同排放类别中的性能。该研究有助于区域气候政策中的数据驱动决策过程。此外，研究结果表明，将基于人工智能的预测工具集成到温室气体监测系统中，可以显著提高气候治理的透明度、准确性和响应能力。
