Predicting the Influence of Porosity on CO2 Injection Efficiency in the Selected Niger Delta Reservoirs, Nigeria: A Machine Learning Approach
Authors
Department of Petroleum and Gas Engineering, University of Port Harcourt, Choba, Rivers State, Nigeria (Nigeria)
Department of Petroleum and Gas Engineering, University of Port Harcourt, Choba, Rivers State, Nigeria (Nigeria)
Department of Petroleum and Gas Engineering, University of Port Harcourt, Choba, Rivers State, Nigeria (Nigeria)
Article Information
DOI: 10.51583/IJLTEMAS.2026.151000090
Subject Category: Machine Learning Approach
Volume/Issue: 15/10 | Page No: 1114-1130
Publication Timeline
Submitted: 2026-10-26
Accepted: 2026-10-31
Published: 2026-11-10
Abstract
Carbon sequestration in geological formations is one of the methods through which emissions could be reduced in petroleum operations. There are depleted and mature hydrocarbon reservoirs in the Niger Delta basin that should be evaluated for their capability in sequestration. In this research, machine learning models are developed to predict a set of CO₂ injection-efficiency index and the impact of porosity on the predictions. Analysis was conducted using 12,000 secondary sources of reservoir data with regard to porosity, permeability, CO₂ injection rate, injection pressure, brine saturation, reservoir pressure, and efficiency response. Feedforward neural network (FNN), Random Forest, XGBoost, and stacked ensemble were assessed with R², RMSE, MAE and mean error (ME). The FNN gave the best test-set performance (R² = 0.728, RMSE = 5.08, MAE = 4.05), followed by XGBoost (R² = 0.723) and Random Forest (R² = 0.712); the stacked ensemble (R² = 0.715) did not improve on the individual models. Pearson correlation shows that the highest recorded correlation was between porosity and the efficiency index (r = 0.67), brine saturation (r = 0.53) and permeability (r = 0.43). SHAP showed that porosity positively contributed to model predictions. The above observations justify the use of comparative and interpretable machine learning as a screening tool, as long as the efficiency criterion is correctly understood as defined in the source data points, and if the predictions of the machine learning model are independently verified against experimental measurements or simulations. It does not by itself prove the existence of any causality between porosity and any production performance.
Keywords
porosity; CO₂ injection efficiency; carbon capture, utilization and storage (CCUS); machine learning; ensemble modeling; SHAP; Niger Delta Reservoirs
This article is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
Downloads
References
1. Adeogba, A. A., Mchargue, T. R., & Graham, S. A. (2005). Transient fan architecture and depositional controls from near-surface 3-D seismic data, Niger Delta continental slope. AAPG Bulletin, 89(5), 627–643. https://doi.org/10.1306/11200404025 [Google Scholar] [Crossref]
2. Ahmed, T. (2010). Reservoir engineering handbook (4th ed.). Gulf Professional Publishing. [Google Scholar] [Crossref]
3. Al-Fattah, S. M. A. (2013). Artificial neural network models for determining the relative permeability of hydrocarbon reservoirs. Saudi Arabian Oil Company. [Google Scholar] [Crossref]
4. Arigbe, O. D., Oyeneyin, M. B., Arana, I., & Ghazi, M. D. (2019). Real-time relative permeability prediction using deep learning. Journal of Petroleum Exploration and Production Technology, 9, 1271–1284. [Google Scholar] [Crossref]
5. Bachu, S. (2015). Review of CO₂ storage efficiency in deep saline aquifers. International Journal of Greenhouse Gas Control, 40, 188–202. [Google Scholar] [Crossref]
6. Bahonar, E., Ghalenoei, Y., & Salmani, S. (2022). New relative permeability correlations for carbonate reservoirs through data-driven modeling. Journal of Chemical and Petroleum Engineering, 56(1), 15–35. [Google Scholar] [Crossref]
7. Bennion, B., & Bachu, S. (2005). Relative permeability characteristics for supercritical CO₂ displacing water in a variety of potential sequestration zones in the Western Canada Sedimentary Basin [Paper presentation]. SPE Annual Technical Conference and Exhibition, Dallas, TX, United States. [Google Scholar] [Crossref]
8. Burnside, N. M., & Naylor, M. (2014). Review and implications of relative permeability of CO₂/brine systems and residual trapping of CO₂. International Journal of Greenhouse Gas Control, 23, 1–11. [Google Scholar] [Crossref]
9. Chadwick, A., Arts, R., Eiken, O., Williamson, P., & Williams, G. (2009). Geological storage of carbon dioxide: Monitoring and verification of CO₂ storage in the Sleipner project. Elsevier. [Google Scholar] [Crossref]
10. Davies, O. A., Israel-Cookey, C., Uko, E. D., & Alabraba, M. A. (2020). Reservoir capacity and injectivity characterization for carbon dioxide (CO₂) geo-sequestration in the Niger Delta. International Journal of Research and Innovation in Applied Science, 5(10), 19–26. [Google Scholar] [Crossref]
11. Eigbe, P. A., Ajayi, O. O., Olakoyejo, O. T., Fadipe, O. L., Efe, S., & Adelaja, A. O. (2023). A general review of CO₂ sequestration in underground geological formations and assessment of depleted hydrocarbon reservoirs in the Niger Delta. Applied Energy, 350, Article 121723. https://doi.org/10.1016/j.apenergy.2023.121723 [Google Scholar] [Crossref]
12. Ganguli, S. S., & Dimri, V. P. (2023). Reservoir characterization: State-of-the-art, key challenges and ways forward. In Developments in structural geology and tectonics (Vol. 6, pp. 1–35). Elsevier. [Google Scholar] [Crossref]
13. Honarpour, M., & Mahmood, S. M. (1988). Relative-permeability measurements: An overview. Journal of Petroleum Technology, 40(8), 963–966. [Google Scholar] [Crossref]
14. Jika, H. T., Onuoha, K. M., & Dim, C. I. P. (2020). Application of geostatistics in facies modeling of Reservoir-E, “Hatch Field” offshore Niger Delta Basin, Nigeria. Journal of Petroleum Exploration and Production Technology, 10(3), 769–781. https://doi.org/10.1007/s13202-019-00788-1 [Google Scholar] [Crossref]
15. Juanes, R., Spiteri, E. J., Orr, F. M., Jr., & Blunt, M. J. (2006). Impact of relative permeability hysteresis on geological CO₂ storage. Water Resources Research, 42(12), Article W12418. https://doi.org/10.1029/2005WR004806 [Google Scholar] [Crossref]
16. Kim, Y., Jang, H., Kim, J., & Lee, J. (2017). Prediction of storage efficiency on CO₂ sequestration in deep saline aquifers using artificial neural network. Applied Energy, 185, 916–928. [Google Scholar] [Crossref]
17. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems 30 (pp. 4765–4774). [Google Scholar] [Crossref]
18. Michael, K., Golab, A., Shulakova, V., Ennis-King, J., Allinson, G., Sharma, S., & Aiken, T. (2010). Geological storage of CO₂ in saline aquifers: A review of the experience from existing storage operations. International Journal of Greenhouse Gas Control, 4(4), 659–667. [Google Scholar] [Crossref]
19. Mikhaylov, A., Moiseev, N., Aleshin, K., & Burkhardt, T. (2020). Global climate change and greenhouse effect. Entrepreneurship and Sustainability Issues, 7(4), 2897–2913. https://doi.org/10.9770/jesi.2020.7.4(21) [Google Scholar] [Crossref]
20. Ojo, A. C., & Tse, A. C. (2016). Geological characterization of depleted oil and gas reservoirs for carbon sequestration potentials in a field in the Niger Delta, Nigeria. Journal of Applied Sciences and Environmental Management, 20(1), 45–55. https://doi.org/10.4314/jasem.v20i1.6 [Google Scholar] [Crossref]
21. Osinowo, O. O., Ayorinde, J. O., Nwankwo, C. P., Ekeng, O. M., & Taiwo, O. B. (2018). Reservoir description and characterization of Eni field offshore Niger Delta, southern Nigeria. Journal of Petroleum Exploration and Production Technology, 8, 381–397. [Google Scholar] [Crossref]
22. Pham, Q. C. (2025). Constrained CO₂-EOR optimization considering impurities, CO₂-EOR type, volume, and oil recovery vs. CO₂ storage [Master’s thesis, Memorial University of Newfoundland]. [Google Scholar] [Crossref]
23. Raza, A., Rezaee, R., Gholami, R., Rasouli, V., Bing, C. H., Nagarajan, R., & Hamid, M. A. (2015). Injectivity and quantification of capillary trapping for CO₂ storage: A review of influencing parameters. Journal of Natural Gas Science and Engineering, 26, 510–517. https://doi.org/10.1016/j.jngse.2015.06.046 [Google Scholar] [Crossref]
24. Short, K. C., & Stauble, A. J. (1967). Outline of geology of Niger Delta. AAPG Bulletin, 51(5), 761–779. [Google Scholar] [Crossref]
25. Usen, W. S., & Akpabio, I. O. (2023). Evaluation of reservoir rocks in Gabo Field, Niger Delta for CO₂ storage and retention. World Journal of Applied Science and Technology, 15(2), 332–339. https://doi.org/10.4314/wojast.v15i2.26 [Google Scholar] [Crossref]
26. Vo Thanh, H., Sugai, Y., & Sasaki, K. (2020). Application of artificial neural network for predicting the performance of CO₂ enhanced oil recovery and storage in residual oil zones. Scientific Reports, 10, Article 18204. https://doi.org/10.1038/s41598-020-73931-2 [Google Scholar] [Crossref]
27. Wang, S., Tian, H., Zhao, X., Yan, Y., Yang, X., & Wang, X. (2024). Assessment of the geological storage potential and suitability of CO₂ in the deep saline aquifers in the northwest plain of Shandong Province, China. Energies, 17(24), Article 6387. https://doi.org/10.3390/en17246387 [Google Scholar] [Crossref]
28. Yasemi, S., Khalili, Y., Sanati, A., & Bagheri, M. (2023). Carbon capture and storage: Application in the oil and gas industry. Sustainability, 15(19), Article 14486. [Google Scholar] [Crossref]
29. Zhong, Z., Carr, T. R., Wu, X., & Wang, G. (2022). Application of a convolutional neural network in permeability prediction: A case study in the Jacksonburg-Stringtown oil field, West Virginia, USA. West Virginia University, Department of Geology and Geography. [Google Scholar] [Crossref]
30. Zou, J., Huo, S., Liu, E., Shu, M., Shan, X., & Li, W. (2019). Prediction of relative permeability curve based on machine learning driven by logging data. Key Laboratory of Petroleum Resource Research, Institute of Geology and Geophysics, Chinese Academy of Sciences. [Google Scholar] [Crossref]
This work is licensed under CC BY 4.0