Predicting carbon dioxide emissions in the United States of America using machine learning algorithms

• Published in: Environmental science and pollution research international 2024-05, Vol.31 (23), p.33685-33707
• Main Authors: Chukwunonso, Bosah Philip, AL-Wesabi, Ibrahim, Shixiang, Li, AlSharabi, Khalil, Al-Shamma’a, Abdullrahman A., Farh, Hassan M. Hussein, Saeed, Fahman, Kandil, Tarek, Al-Shaalan, Abdullah M.
• Format: Article
• Language: English
• Subjects: Accuracy; Air pollution; Algorithms; Alternative energy; Aquatic Pollution; Artificial neural networks; Atmospheric Protection/Air Quality Control/Air Pollution; Carbon dioxide; Carbon dioxide emissions; Climate change; Climate change mitigation; Decomposition; Earth and Environmental Science; Ecotoxicology; Emissions; Emissions control; Energy efficiency; Energy policy; Energy sources; Environment; Environmental Chemistry; Environmental Health; Environmental policy; Filters; Forecasting; Fossil fuels; Global warming; Greenhouse effect; Greenhouse gases; Human influences; Learning algorithms; Machine learning; Neural networks; Nodes; Pollution monitoring; Recurrent neural networks; Renewable energy; Renewable resources; Research Article; Waste Water Technology; Water Management; Water Pollution Control
• ISSN: 1614-7499; 0944-1344; 1614-7499
• DOI: 10.1007/s11356-024-33460-1

Carbon dioxide (CO 2 ) emissions result from human activities like burning fossil fuels. CO 2 is a greenhouse gas, contributing to global warming and climate change. Efforts to reduce CO 2 emissions include transitioning to renewable energy. Monitoring and reducing CO 2 emissions are crucial for mitigating climate change. Strategies include energy efficiency and renewable energy adoption. In the past few decades, several nations have experienced air pollution and environmental difficulties because of carbon dioxide (CO 2 ) emissions. One of the most crucial methods for regulating and maximizing CO 2 emission reductions is precise forecasting. Four machine learning algorithms with high forecasting precision and low data requirements were developed in this study to estimate CO 2 emissions in the United States (US). Data from a dataset covering the years 1973/01 to 2022/07 that included information on different energy sources that had an impact on CO 2 emissions were examined. Then, four algorithms performed the CO 2 emissions forecast from the layer recurrent neural network with 10 nodes (L-RNN), a feed-forward neural network with 10 nodes (FFNN), a convolutional neural network with two layers with 10 and 5 filters (CNN1), and convolutional neural network with two layers and with 50 and 25 filters (CNN2) models. Each algorithm’s forecast accuracy was assessed using eight indicators. The three preprocessing techniques used are (1) without any processing techniques, (2) processed using max–min normalization technique, and (3) processed using max–min normalization technique and decomposed by variation mode decomposition (VMD) technique with 7 intrinsic mode functions and 1000 iterations. The latter with L-RNN algorithm gave a high accuracy between the forecasting and actual values. The results of CO 2 emissions from 2011/05 to 2022/07 have been forecasted, and the L-RNN algorithm had the highest forecast accuracy. The L-RNN model has the lowest value of 1.187028078, 135.5668592, and 11.64331822 for MAPE, MSE, and RMSE, respectively. The L-RNN model provides precise and timely forecasts that can help formulate plans to reduce carbon emissions and contribute to a more sustainable future. Moreover, the results of this investigation can improve our comprehension of the dynamics of carbon dioxide emissions, resulting in better-informed environmental policies and initiatives targeted at lowering carbon emissions.