Scholarlink

<p>The Middle Eastern Gulf Cooperation Council (GCC) states are significant sources of global greenhouse gas (GHG) emissions, primarily due to their rapid economic expansion, increased urbanization, and extensive fossil fuel exploitation. This study examines the demographic (population and urbanization rate) and socioeconomic (net energy consumption, gross domestic product (GDP), foreign direct investment (FDI), and gross national income (GNI)) factors that influenced greenhouse gas emissions in all GCC nations from 1970 to 2019. With modified hyperparameters, the study achieves the best prediction of GHG dynamics using three boosted machine learning models: AdaBoost, gradient boost, and extreme gradient boost (XGBoost). XGBoost, in particular, demonstrates the highest predictive power, with R2 values ranging from 96% to 98% and a low prediction error. Analysis indicates a substantial increase (2–13 times) in GHG emissions over the past 50 years, with projections foreseeing a further rise of 15–35% by 2030. Furthermore, the feature importance of the ensembles indicates a comparatively equal dependency on all the selected variables for predicting GHG. By examining emission trends within a sustainability and low-carbon transition framework, this research contributes to climate mitigation measures and sustainable development planning in the GCC. Combining machine learning-based predictions with an examination of Nationally Determined Contributions (NDCs) offers researchers a policy-relevant perspective on how existing pledges fit with long-term emission paths and sustainability goals set by the UNFCCC.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Greenhouse gas emission modeling for gulf cooperation council countries with boosting techniques

  • 期刊论文

  • 发表日期：

    2026年2月19日

查看全文
  • Muhammad Muhitur Rahman,
  • Md Shafiul Alam,
  • Bijoy Mitra,
  • Mohammad Shahedur Rahman,
  • Aftab Ahmad Khan,
  • Muhammad Tahir Amin,
  • Abdulmoez Al Ismaeel,
  • Syed Masiur Rahman

摘要

The Middle Eastern Gulf Cooperation Council (GCC) states are significant sources of global greenhouse gas (GHG) emissions, primarily due to their rapid economic expansion, increased urbanization, and extensive fossil fuel exploitation. This study examines the demographic (population and urbanization rate) and socioeconomic (net energy consumption, gross domestic product (GDP), foreign direct investment (FDI), and gross national income (GNI)) factors that influenced greenhouse gas emissions in all GCC nations from 1970 to 2019. With modified hyperparameters, the study achieves the best prediction of GHG dynamics using three boosted machine learning models: AdaBoost, gradient boost, and extreme gradient boost (XGBoost). XGBoost, in particular, demonstrates the highest predictive power, with R2 values ranging from 96% to 98% and a low prediction error. Analysis indicates a substantial increase (2–13 times) in GHG emissions over the past 50 years, with projections foreseeing a further rise of 15–35% by 2030. Furthermore, the feature importance of the ensembles indicates a comparatively equal dependency on all the selected variables for predicting GHG. By examining emission trends within a sustainability and low-carbon transition framework, this research contributes to climate mitigation measures and sustainable development planning in the GCC. Combining machine learning-based predictions with an examination of Nationally Determined Contributions (NDCs) offers researchers a policy-relevant perspective on how existing pledges fit with long-term emission paths and sustainability goals set by the UNFCCC.