Article

Machine learning-enhanced seismic attribute analysis for reservoir quality prediction in clastic oil fields

Fahri Maho, Enkelejda Gjinali
Abstract

The aim of the study was to assess machine learning-enhanced seismic attribute analysis for reservoir quality prediction in clastic and Balkan-type oil fields. The validation study was performed on the open-access Project Penobscot seismic and reservoir data for a range of attributes within the predictor space, as well as for the four reservoir quality variables. XGBoost showed the best regression results for all continuous parameters, with the highest accuracy for porosity (R2 = 0.88, root mean square error = 0.024, mean absolute error = 0.018), followed by the sandiness index (R2 = 0.85, root mean square error = 0.054, mean absolute error = 0.039), whereas permeability remained the most difficult parameter to predict (R2 = 0.76, root mean square error = 16.9, mean absolute error = 12.4). Facies classification also demonstrated stable performance, with accuracy of 0.84, precision of 0.82, recall of 0.79, and F1-score of 0.8. The predicted parameter distributions revealed a heterogeneous reservoir system, with porosity as the most stable parameter, sandiness index showing intermediate variability, and permeability displaying the highest contrast. Integrated interpretation of the predicted indicators differentiated the reservoir into high-, moderate-, and low-quality zones with indices of 86, 64, and 38, respectively. In conclusion, reservoir quality is primarily controlled by the factors related to storage and flow potential, sand bodies, and facies, yet machine learning-enhanced methods can effectively assist reservoir engineers in making accurate reservoir predictions in these heterogeneous systems. The practical value of this study is in the potential use of this established workflow to effectively screen for and prioritise oil and gas production within the reservoirs with the help of machine learning, especially those geologically structurally complex clastic systems

Download article

Received 27.10.2025

Revised 20.04.2026

Accepted 29.05.2026

Published 29.06.2026

https://doi.org/10.63341/pdogf/1.2026.16
Retrieved from Vol. 26, No. 1, 2026
Pages 16-29

Suggested citation

Maho, F., & Gjinali, E. (2026). Machine learning-enhanced seismic attribute analysis for reservoir quality prediction in clastic oil fields. Prospecting and Development of Oil and Gas Fields, 26(1), 16-29. https://doi.org/10.63341/pdogf/1.2026.16

References

  1. Abd-Elfattah, N., Dahroug, A., El Kammar, M., & Fahmy, R. (2025). Machine learning and AVO class II workflow for hydrocarbon prospectivity in the Messinian offshore Nile Delta Egypt. Scientific Reports, 15(1), article number 3566. doi: 10.1038/s41598-025-86765-7.
  2. Al Hasan, R., Saberi, M.H., & Riahi, M.A. (2025). Seismic facies classification and TOC prediction using seismic attributes and Hjorth parameters, in the Groningen field in northeastern Netherlands. Discover Applied Sciences, 7(10), article number 1202. doi: 10.1007/s42452-025-07794-5.
  3. Alexandridis, I., Oikonomopoulos, I.K., Carvajal-Ortiz, H., Gentzis, T., Kalaitzidis, S., Georgakopoulos, A.N., & Christanis, K. (2022). Discovery of a new source-rock interval within the Pantokrator Formation, Ionian Zone, western Greece: Insights from sulfur speciation and kinetics analyses. Marine and Petroleum Geology, 145, article number 105918. doi: 10.1016/j.marpetgeo.2022.105918.
  4. Ali, M., Changxingyue, H., Wei, N., Jiang, R., Zhu, P., Hao, Z., Hussain, W., & Ashraf, U. (2024). Optimizing seismic-based reservoir property prediction: A synthetic data-driven approach using convolutional neural networks and transfer learning with real data integration. Artificial Intelligence Review, 58(1), article number 31. doi: 10.1007/s10462-024-11030-8.
  5. Arkadiusz, D., Tomasz, T., Anita, L.Ś., & Krzysztof, S. (2025). Machine learning approaches for predicting lithological and petrophysical parameters in hydrocarbon exploration: A case study from the Carpathian foredeep. Energies, 18(17), article number 4521. doi: 10.3390/en18174521.
  6. Babayeva, T., Guliyev, A., İslamzade, T., İslamzade, R., Haciyeva, X., Ashurova, N., Aliyeva, A., & Maksudov, S. (2024). Impacts of irrigation with Cd-contaminated water from Sugovushan Reservoir, Azerbaijan on total cadmium and its fractions in soils with varied textures. Eurasian Journal of Soil Science, 13(2), 145-152. doi: 10.18393/ejss.1424421.
  7. Badr, S.M., Mohamed, F.H., Mansour, A.S., & Elessawy, A.A. (2026). Machine learning integration with seismic attributes for lithofacies prediction and distribution in Abu Madi reservoir, onshore Faraskour gas field, east Nile Delta, Egypt. Journal of Petroleum Exploration and Production Technology, 16(2), article number 24. doi: 10.1007/s13202-025-02125-1.
  8. Chen, Y., Zhao, L., Pan, J., Li, C., Xu, M., Li, K., Zhang, F., & Geng, J. (2021). Deep carbonate reservoir characterisation using multi-seismic attributes via machine learning with physical constraints. Journal of Geophysics and Engineering, 18(5), 761-775. doi: 10.1093/jge/gxab049.
  9. Chenin, J., & Bedle, H. (2022). Unsupervised machine learning, multi-attribute analysis for identifying low saturation gas reservoirs within the deepwater Gulf of Mexico, and Offshore Australia. Geosciences, 12(3), article number 132. doi: 10.3390/geosciences12030132.
  10. Cristea, V.-M., Baigulbayeva, M., Ongarbayev, Y., Smailov, N., Akkazin, Y., & Ubaidulayeva, N. (2023). Prediction of oil sorption capacity on carbonized mixtures of Shungite using artificial neural networks. Processes, 11(2), article number 518. doi: 10.3390/pr11020518.
  11. El-Dabaa, S.A., Metwalli, F.I., Maher, A., & Ismail, A. (2024). Unsupervised machine learning-based multi-attributes analysis for enhancing gas channel detection and facies classification in the serpent field, offshore Nile Delta, Egypt. Geomechanics and Geophysics for Geo-Energy and Geo-Resources, 10(1), article number 185. doi: 10.1007/s40948-024-00907-1.
  12. Farfour, M., & Foster, D. (2022). Machine learning and seismic attributes for prospect identification and risking: An example from offshore Australia. In SEG international exposition and annual meeting (article number D011S011R004). Houston: OnePetro. doi: 10.1190/image2022-3752064.1.
  13. Farfour, M., Hedjam, R., Foster, D., & Gaci, S. (2025). Machine learning and seismic attributes for petroleum prospect generation and evaluation: An example from offshore Australia. Interpretation, 13(4), 805-815. doi: 10.1190/INT-2024-0177.1.
  14. Ferreira, D.J., Dias, R.M., & Lupinacci, W.M. (2021). Seismic pattern classification integrated with permeability-porosity evaluation for reservoir characterization of presalt carbonates in the Buzios Field, Brazil. Journal of Petroleum Science and Engineering, 201, article number 108441. doi: 10.1016/j.petrol.2021.108441.
  15. Fociro, A., Fociro, O., Prifti, I., Muçi, R., & Pekmezi, J. (2023). Petrography and mineralogy of the Oligocene flysch in Ionian Zone, Albania: Implications for the evolution of sediment provenance and paleoenvironment. Open Geosciences, 15(1), article number 20220442. doi: 10.1515/geo-2022-0442.
  16. Hussain, M., Shakir, U., Naseer, Z., Amjad, M.R., Arshad, F., Kamal, M., & Radwan, A.E. (2025). Integrating conventional and machine learning aproaches with seismic inversion for optimized porosity estimation in heterogeneous siliciclastic rocks. Physics and Chemistry of the Earth, Parts A/B/C, 142, article number 104248. doi: 10.1016/j.pce.2025.104248.
  17. Jia, W., Liu, D., Zhang, G., Lan, T., Sun, Z., & Hu, H. (2026). Lithology prediction based on well-log data and seismic inversion result: A machine learning approach. Journal of Applied Geophysics, 247, article number 106131. doi: 10.1016/j.jappgeo.2026.106131.
  18. Kumar, P.C., Bedle, H., Kumar, J., Goswami, T.K., & Sain, K. (2026). Seismic facies characterization: Integrated subsurface-outcrop analysis for complex depositional systems in northeast India. Artificial Intelligence in Geosciences, 7(1), article number 100196. doi: 10.1016/j.aiig.2026.100196.
  19. Li, W., Yue, D., Colombera, L., Duan, D., Long, T., Wu, S., & Liu, Y. (2023). A novel method for seismic-attribute optimization driven by forward modeling and machine learning in prediction of fluvial reservoirs. Geoenergy Science and Engineering, 227, article number 211952. doi: 10.1016/j.geoen.2023.211952.
  20. Liu, J., Lin, C., Elders, C., & Faris, A. (2026). Sandbody prediction based on fusion of seismic multi-attributes and machine learning under sedimentary facies constraint – a case study of Chenguanzhuang area in Dongying depression, Bohai Bay basin. Applied Sciences, 16(7), article number 3341. doi: 10.3390/app16073341.
  21. Murtaza, G., Ali, N., Hussain, W., Iqbal, S.M., Azhar, M.U., Ahmad, T., Iltaf, K., & Ahmed, A. (2025). Integrating petrophysical, seismic and rock physics analyses for precise reservoir characterization. Earth Systems and Environment, 9(3), 2165-2187. doi: 10.1007/s41748-025-00654-7.
  22. Narayan, S., Sahoo, S.D., Kar, S., Pal, S.K., & Kangsabanik, S. (2024). Improved reservoir characterization by means of supervised machine learning and model-based seismic impedance inversion in the Penobscot field, Scotian Basin. Energy Geoscience, 5(2), article number 100180. doi: 10.1016/j.engeos.2023.100180.
  23. Oitseva, T. (2024). Geological structure and mineralogical composition of the Kara-Ayak rare metal ore occurrence (East Kazakhstan). In International multidisciplinary scientific geoconference surveying geology and mining ecology management SGEM 2024 (pp. 71-78). Vienna: SGEM. doi: 10.5593/sgem2024/1.1/s01.10.
  24. Olutoki, J.O., Elsaadany, M., Siddiqui, N.A., Haque, A.E., Ali, S.H., Rashid, A., & Akinyemi, O.D. (2024). Estimating petrophysical properties using geostatistical inversion and data-driven extreme gradient boosting: A case study of late eocene McKee formation, Taranaki Basin, new Zealand. Results in Engineering, 24, article number 103494. doi: 10.1016/j.rineng.2024.103494.
  25. Owusu, P., Raef, A., & Sharaf, E. (2025). Carbonate seismic facies analysis in reservoir characterization: A machine learning approach with integration of reservoir mineralogy and porosity. Geosciences, 15(7), article number 257. doi: 10.3390/geosciences15070257.
  26. Pradhan, A., & Mukerji, T. (2022). Consistency and prior falsification of training data in seismic deep learning: Application to offshore deltaic reservoir characterization. Geophysics, 87(3), 45-61. doi: 10.1190/geo2021-0568.1.
  27. Rahma Putra, M.H., Hermana, M., Yogi, I.B., Hossain, T.M., Abdurrachman, M.F., & Kadir, S.J. (2024). Reservoir porosity assessment and anomaly identification from seismic attributes using Gaussian process machine learning. Earth Science Informatics, 17(2), 1315-1327. doi: 10.1007/s12145-024-01240-7.
  28. Reda, M., El-Gendy, N.H., Raef, A., Elmashaly, M.M., AlArifi, N., & Barakat, M.K. (2024). Advancing neogene-quaternary reservoir characterization in offshore Nile Delta, Egypt: high-resolution seismic insights and 3D modeling for new prospect identification. Environmental Earth Sciences, 83(20), article number 581. doi: 10.1007/s12665-024-11846-1.
  29. Ren, K., Li, W., Zhao, L., Wang, W., Wang, J., Wang, H., Li, Y., Qu, L., & Yue, D. (2026). Seismic-attribute optimization for improving reservoir prediction integrating the spectral decomposition and automated machine learning. Journal of Applied Geophysics, 248, article number 106167. doi: 10.1016/j.jappgeo.2026.106167.
  30. Ren, T.F., Feng, Z.B., Zhang, Y., Zhang, X., Jiang, L., Ning, Y.L., Wang, J., Ding, J., & Qi, Z.S. (2025). A seismic multi-attribute sandbody identification method based on the LightGBM-RFECV coupling algorithm. Applied Geophysics, 22(3), 757-769. doi: 10.1007/s11770-025-1292-y.
  31. Riyadi, Z.A., Olutoki, J.O., Hermana, M., Latif, A.H., Yogi, I.B., & Kadir, S.J. (2024). Machine learning prediction of permeability distribution in the X field Malay Basin using elastic properties. Results in Engineering, 24, article number 103421. doi: 10.1016/j.rineng.2024.103421.
  32. Shevko, V.M., Zharmenov, A.A., Aitkulov, D.K., & Terlikbaeva, A.Z. (2021). Complex processing of oxidized copper and zinc oxide ores with simultaneous production of several products. Physicochemical Problems of Mineral Processing, 57(1), 226-249. doi: 10.37190/ppmp/131091.
  33. Soto, J.I., et al. (2024). Contrasting styles of salt‐tectonic processes in the Ionian Zone (Greece and Albania): Integrating surface geology, subsurface data, and experimental models. Tectonics, 43(1), article number e2023TC008104. doi: 10.1029/2023TC008104.
  34. Wang, J., Guan, D., Huang, X., He, Y., Li, H., Xu, W., Liu, R., & Feng, B. (2025). Integrated SOM multi-attribute optimization and seismic waveform inversion for thin sand body characterization: A case study of the paleogene lower E3d2 sub-member in the HHK depression, Bohai Bay basin. Applied Sciences, 15(9), article number 5134. doi: 10.3390/app15095134.