Unsupervised Electrofacies Classification and Porosity Characterization in the Offshore Keta Basin Using Wireline Logs
A study applied unsupervised machine learning to classify electrofacies and characterize porosity in the offshore Keta Basin, Ghana, using wireline logs from a single well. K-means clustering identified four distinct electrofacies with moderate separation supported by silhouette analysis, reflecting geological variations from shale to sandstone. The method demonstrates a reproducible, log-based approach for subsurface characterization in data-scarce frontier basins.
- ▪The study used six standard wireline logs from Well C, analyzing approximately 11,195 depth samples in the offshore Keta Basin.
- ▪K-means clustering in multivariate log space identified four electrofacies, supported by an average silhouette coefficient of about 0.50.
- ▪The resulting electrofacies show depth-continuous patterns linked to changes in clay content, porosity, and rock framework, forming a geological continuum.
- ▪The proposed unsupervised workflow provides a robust, reproducible method for early-stage formation evaluation in areas with limited core data.
- ▪The research was accepted for presentation at ICECET 2026 and is categorized under artificial intelligence, machine learning, and geophysics.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2604.27126 (cs) [Submitted on 29 Apr 2026] Title:Unsupervised Electrofacies Classification and Porosity Characterization in the Offshore Keta Basin Using Wireline Logs Authors:Hamdiya Adams, Theophilus Ansah-Narh, Daniel Kwadwo Asiedu, Bruce Kofi Banoeng-Yakubo, Marcellin Atemkeng, Thomas Armah, Richmond Opoku-Sarkodie, Rebecca Davis, Ezekiel Nii Noye Nortey View a PDF of the paper titled Unsupervised Electrofacies Classification and Porosity Characterization in the Offshore Keta Basin Using Wireline Logs, by Hamdiya Adams and 8 other authors View PDF HTML (experimental) Abstract:This study presents an unsupervised machine learning workflow for electrofacies analysis in the offshore Keta Basin, Ghana, where core data are scarce.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
