Litological analysis and reservoir zone identification using the crossplot and k-means clustering methods on well log data
Main Article Content
Abstract
Lithology identification and reservoir zone determination are critical stages in formation evaluation within the petroleum industry. This study aims to identify rock lithology, analyze the relationship between log parameters using the Crossplot method, and determine prospective reservoir zones using K-Means Clustering based on Orange Data Mining applied to open-source well log data from the Kuncherinna-1 well, obtained from the Geoscience Australia database (ECAT 149208), located in the Pedirka and western Eromanga basins, Simpson, Australia. The data used include Gamma Ray (GR), Deep Resistivity (RDEP), Density Log (RHOB), Neutron Porosity (NPHI), Sonic Log (DT), and Caliper Log (CALI) parameters over a depth interval of 2098-2825 m. Crossplot analysis was performed using four parameter combinations: GR vs RDEP, RHOB vs NPHI, GR vs NPHI, and RHOB vs DT. The optimal number of clusters was determined using the Silhouette Score, with k values tested from k=2 to k=6. Evaluation results indicate that k=4 is the optimal number of clusters, yielding the highest Silhouette Score of 0.438. K-Means Clustering produced four lithological groups: Cluster C1 as wet sand, Cluster C2 as a shale zone, Cluster C4 as a transitional shaly sand zone, and Cluster C3 identified as the most prospective reservoir zone candidate based on a combination of very low GR values (15-35 API), very high RDEP values exceeding 30,000 ohm-m, and low NPHI and DT values. The vertical distribution shows that Cluster C3 was identified only within the 2110.74–2161.18 m depth interval, indicating that the prospective reservoir zone is a thin layer located in the shallower part of the analyzed interval. This study demonstrates that the combination of Crossplot and K-Means Clustering implemented in Orange Data Mining provides an effective and objective approach for lithology interpretation and identification of prospective reservoir intervals without requiring prior lithology labels.
Article Details
Section
References
Ali, M., Jiang, R., Ma, H., Pan, H., Abbas, K., Ashraf, U., & Ullah, J. (2024). Data-driven lithofacies prediction in complex tight sandstone reservoirs: A supervised workflow integrating clustering and classification models. Geomechanics and Geophysics for Geo-Energy and Geo-Resources, 10(1), Article 5. https://doi.org/10.1007/s40948-024-00787-5
Asquith, G. B., & Krygowski, D. (2004). Basic well log analysis (2nd ed.). American Association of Petroleum Geologists.
Cao, M., Gao, Z., Yuan, Y., Yan, Z., & Zhang, Y. (2022). A visualization and analysis method by multi-dimensional crossplots from multi-well heterogeneous data. Energies, 15(7), Article 2575. https://doi.org/10.3390/en15072575
Demšar, J., Curk, T., Erjavec, A., Gorup, č., Hočevar, T., Milutinovič, M., Možina, M., Polajnar, M., Toplak, M., Starič, A., Štajdohar, M., Umek, L., Žagar, L., Žbontar, J., Žitnik, M., & Zupan, B. (2013). Orange: Data mining toolbox in Python. Journal of Machine Learning Research, 14, 2349–2353.
Ebraheem, M. O., Ibrahim, H. A., & Senosy, A. H. (2025). A comparative petrophysical evaluation of the Abu Roash, Bahariya, and Kharita reservoirs using well-logging data, East El-Fayoum, Egypt. Scientific Reports, 15, Article 2891. https://doi.org/10.1038/s41598-024-83332-4
Geoscience Australia. (n.d.). The Pedirka and western Eromanga basins well log petrophysics data package (ECAT 149208) [Data set]. Australia’s Future Energy Resources (AFER), Exploring for the Future Program.
Hasan, M. M. U., Hasan, T., Shahidi, R., James, L., Peters, D., & Gosine, R. (2023). Lithofacies identification from wire-line logs using an unsupervised data clustering algorithm. Energies, 16(24), Article 8116. https://doi.org/10.3390/en16248116
Hossain, S., & Rahman, N. (2025). An integrated petrophysical and rock physics characterization of the Mangahewa Formation in the Pohokura field, Taranaki Basin. Scientific Reports, 15, Article 6431. https://doi.org/10.1038/s41598-025-88639-4
Mabiala Mbouaki, A. P., Cai, Z., Sikanyika, E. W., Mwakipunda, G. C., & Konan, H. R. (2025). Petrophysical evaluation of a shaly sandstone reservoir and the effect of clay minerals on reservoir quality: A case study from the Barremian Mengo Sandstone, Kouilou Basin, Republic of Congo. ACS Omega, 10(10), 10081–10106. https://doi.org/10.1021/acsomega.4c08491
Octavia, D. J., Nugrahanti, A., Wiharjo, L., Rahmawan, S., & Kadarusman, S. (2024). Evaluasi litologi dari hasil overlay log densitas dan log neutron pada Sumur Dj-1 dan Dj-2. PETRO: Jurnal Ilmiah Teknik Perminyakan, 13(1), 1–6. https://doi.org/10.25105/petro.v13i1.17386
Rezaei Mirghaed, B., Dehghan Monfared, A., & Ranjbar, A. (2024). Enhanced petrophysical evaluation through machine learning and well logging data in an Iranian oil field. Scientific Reports, 14, Article 28319. https://doi.org/10.1038/s41598-024-80362-w
Rousseeuw, P. J. (1987). Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53–65. https://doi.org/10.1016/0377-0427(87)90125-7
Sobih, N. S., El Nagar, A. A., Abbas, A. E., & Shehab, S. (2025). Integrating petrophysical characterization and reservoir evaluation of the Kafr El-Sheikh Formation in Sapphire field, Nile Delta, Egypt. Scientific Reports, 15, Article 5781. https://doi.org/10.1038/s41598-025-88056-7
Szabó, N. P., Kilik, R., & Dobróka, M. (2023). Robust reservoir identification by multi-well cluster analysis of wireline logging data. Heliyon, 9(5), Article e15957. https://doi.org/10.1016/j.heliyon.2023.e15957
Suyal, M., & Sharma, S. (2024). A review on analysis of K-means clustering machine learning algorithm based on unsupervised learning. Journal of Artificial Intelligence Systems, 6(1), 85–95. https://doi.org/10.33969/ais.2024060106
Tiab, D., & Donaldson, E. C. (2015). Petrophysics: Theory and practice of measuring reservoir rock and fluid transport properties (4th ed.). Gulf Professional Publishing.