Greenhouse gas emission forecasting for sustainable climate decision making in GCC countries using ARIMAX, ETS, segmented regression, and hybrid ensemble models

Greenhouse gas (GHG) emissions have become a major global environmental challenge due to their contribution to climate change. Given the GCC countries’ high dependence on fossil fuels, rapid industrialization, and among the world’s highest per-capita greenhouse gas emissions, accurate forecasting is essential for effective climate planning and sustainable development. This study aims to develop, compare, and validate time-series and hybrid forecasting models for annual GHG emissions (Mt CO₂e) across the GCC. The study utilized the data of GHG emissions (Mt CO₂e) from 1970 to 2024 from the World Bank data. The study used different modelling approaches with structural-break regressors, segmented regression, Exponential Smoothing State-Space (ETS) models, linear regression with structural breaks, and inverse-RMSE-weighted hybrid ensembles were fitted to the annual GHG emission series. Model performance was evaluated using RMSE, MAE, MAPE, and NRMSE. Sample entropy was employed to quantify the structural regularity of each series. A rolling-window cross-validation was conducted to validate model stability, and forecast uncertainty was estimated using a 2,000-replication parametric bootstrap. Model performance was evaluated under a leakage-free out-of-sample design (training 1970–2019; testing 2020–2024), with model orders and hybrid weights fixed from the training window and benchmarked against random-walk-with-drift and linear-trend models. The hybrid ensemble achieved the lowest RMSE for Saudi Arabia (22.89, versus 23.40 for the random walk with drift) and Kuwait (6.99, versus 9.60 for ETS). For the other four countries, single or naive benchmark models performed as well or better: Oman (ARIMA, 3.39), Qatar (ARIMAX, 4.58), the United Arab Emirates (linear trend, 6.13), and Bahrain (ETS and random walk with drift, both 1.79). The hybrid ensemble is therefore a country-specific option rather than a generally superior framework. Coverage of the 95% prediction intervals on the 2020–2024 hold-out was 80% for four countries, 100% for Bahrain, and 0% for the UAE; with five hold-out observations per country, this is a diagnostic rather than a calibration guarantee. Sample entropy is reported as descriptive context only. Forecasts for 2025–2029 indicated continued increases across all six countries, with Saudi Arabia projected to rise from 869.0 to 927.4Mt CO₂e. The study concluded that sustained growth in regional greenhouse gas emissions necessitates a shift from reactive environmental management to proactive, data-driven decision-making. Integrating reliable forecasting into national climate monitoring systems can facilitate early detection of emerging trends, timely evaluation of mitigation strategies, and informed policy interventions to support net-zero commitments in addition, sustainable development goals.

Authors

Institutions

Publication Details

Journal
Discover Sustainability
Published
2026-11-18
DOI
https://doi.org/10.1007/s43621-026-05356-2
Primary Topic
Energy, Environment, Economic Growth
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Greenhouse gas emission forecasting for sustainable climate decision making in GCC countries using ARIMAX, ETS, segmented regression, and hybrid ensemble models

Khulud Al Salman, Muhammad Daniyal, Hibah Abdulrahim Bahri, Abbas Al Mutair et al.
3 links
Discover Sustainability
Energy, Environment, Economic Growth
article

Greenhouse gas emission forecasting for sustainable climate decision making in GCC countries using ARIMAX, ETS, segmented regression, and hybrid ensemble models

Khulud Al Salman, Muhammad Daniyal, Hibah Abdulrahim Bahri, Abbas Al Mutair, Alya Almutairi, Ali Hassan
article en

Abstract

Greenhouse gas (GHG) emissions have become a major global environmental challenge due to their contribution to climate change. Given the GCC countries’ high dependence on fossil fuels, rapid industrialization, and among the world’s highest per-capita greenhouse gas emissions, accurate forecasting is essential for effective climate planning and sustainable development. This study aims to develop, compare, and validate time-series and hybrid forecasting models for annual GHG emissions (Mt CO₂e) across the GCC. The study utilized the data of GHG emissions (Mt CO₂e) from 1970 to 2024 from the World Bank data. The study used different modelling approaches with structural-break regressors, segmented regression, Exponential Smoothing State-Space (ETS) models, linear regression with structural breaks, and inverse-RMSE-weighted hybrid ensembles were fitted to the annual GHG emission series. Model performance was evaluated using RMSE, MAE, MAPE, and NRMSE. Sample entropy was employed to quantify the structural regularity of each series. A rolling-window cross-validation was conducted to validate model stability, and forecast uncertainty was estimated using a 2,000-replication parametric bootstrap. Model performance was evaluated under a leakage-free out-of-sample design (training 1970–2019; testing 2020–2024), with model orders and hybrid weights fixed from the training window and benchmarked against random-walk-with-drift and linear-trend models. The hybrid ensemble achieved the lowest RMSE for Saudi Arabia (22.89, versus 23.40 for the random walk with drift) and Kuwait (6.99, versus 9.60 for ETS). For the other four countries, single or naive benchmark models performed as well or better: Oman (ARIMA, 3.39), Qatar (ARIMAX, 4.58), the United Arab Emirates (linear trend, 6.13), and Bahrain (ETS and random walk with drift, both 1.79). The hybrid ensemble is therefore a country-specific option rather than a generally superior framework. Coverage of the 95% prediction intervals on the 2020–2024 hold-out was 80% for four countries, 100% for Bahrain, and 0% for the UAE; with five hold-out observations per country, this is a diagnostic rather than a calibration guarantee. Sample entropy is reported as descriptive context only. Forecasts for 2025–2029 indicated continued increases across all six countries, with Saudi Arabia projected to rise from 869.0 to 927.4Mt CO₂e. The study concluded that sustained growth in regional greenhouse gas emissions necessitates a shift from reactive environmental management to proactive, data-driven decision-making. Integrating reliable forecasting into national climate monitoring systems can facilitate early detection of emerging trends, timely evaluation of mitigation strategies, and informed policy interventions to support net-zero commitments in addition, sustainable development goals.

Discover Sustainability
Princess Nourah bint Abdulrahman University (SA), Queen's University Belfast (GB), University of Sargodha (PK), Taibah University (SA)
Climate action
Energy, Environment, Economic Growth
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon