Predictive Model for Assessing Reservoir Formation Suitability for CCUS in the Niger Delta, Nigeria

There is a need for an efficient way of ranking potential CCUS reservoirs before site characterization, whereas previous Niger Delta CCUS reservoir screenings have been carried out using field-specific criteria. In this work, we ranked 12,734 selected data points of Niger Delta reservoirs provided by the Nigerian Upstream Petroleum Regulatory Commission (NUPRC) in a standardized manner, and we investigated the performance of machine learning (ML) methods in reproducing such ranking. Nine storage-related variables including relative permeability to oil, depth, porosity, pore volume, permeability, reservoir pressure, reservoir temperature, grain size, and water saturation were considered as normalized variables in favor of storage capacity. A suitability index (SI) was defined based on a weighted sum of these nine variables with weights obtained using principal component analysis in line with the composite-indicator development OECD/JRC approach and 12,734 data points were classified into four quartile classes. The ranking was found to be robust against changes in the weights as an equally weighted index showed a Spearman correlation coefficient of 0.97 with the main index, and classified 82% of the data points in the same quartile classes. The FNN performed best in reproducing the index (test R² = 0.9997; cross-validated R² = 0.9999 ± 0.0002) with a classification accuracy of 98.4% for the test set, outperforming XGBoost (R² = 0.9955; 95.1%) and RF (R² = 0.9829; 91.1%); nearly all misclassification occurred near the boundary between classes. As the index depends on the inputs, these metrics represent the ability of the models to reproduce the index as opposed to storage capacity. The present research offers a fully transparent, reproducible screening index with weights validated using independent data, as well as surrogate models for rapid screening of reservoirs.

Authors

Institutions

Publication Details

Journal
International Journal of Latest Technology in Engineering Management & Applied Science
Published
2026-11-16
DOI
https://doi.org/10.51583/ijltemas.2026.151000107
Primary Topic
CO2 Sequestration and Geologic Interactions
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Predictive Model for Assessing Reservoir Formation Suitability for CCUS in the Niger Delta, Nigeria

Sunday Sunday Ikiensikimama, Chibuzor C. Egwuogu, Irene Ijeoma Mbachu
3 links
International Journal of Latest Technology in Engineering Management & Applied Science
CO2 Sequestration and Geologic Interactions
article

Predictive Model for Assessing Reservoir Formation Suitability for CCUS in the Niger Delta, Nigeria

Sunday Sunday Ikiensikimama, Chibuzor C. Egwuogu, Irene Ijeoma Mbachu
article en

Abstract

There is a need for an efficient way of ranking potential CCUS reservoirs before site characterization, whereas previous Niger Delta CCUS reservoir screenings have been carried out using field-specific criteria. In this work, we ranked 12,734 selected data points of Niger Delta reservoirs provided by the Nigerian Upstream Petroleum Regulatory Commission (NUPRC) in a standardized manner, and we investigated the performance of machine learning (ML) methods in reproducing such ranking. Nine storage-related variables including relative permeability to oil, depth, porosity, pore volume, permeability, reservoir pressure, reservoir temperature, grain size, and water saturation were considered as normalized variables in favor of storage capacity. A suitability index (SI) was defined based on a weighted sum of these nine variables with weights obtained using principal component analysis in line with the composite-indicator development OECD/JRC approach and 12,734 data points were classified into four quartile classes. The ranking was found to be robust against changes in the weights as an equally weighted index showed a Spearman correlation coefficient of 0.97 with the main index, and classified 82% of the data points in the same quartile classes. The FNN performed best in reproducing the index (test R² = 0.9997; cross-validated R² = 0.9999 ± 0.0002) with a classification accuracy of 98.4% for the test set, outperforming XGBoost (R² = 0.9955; 95.1%) and RF (R² = 0.9829; 91.1%); nearly all misclassification occurred near the boundary between classes. As the index depends on the inputs, these metrics represent the ability of the models to reproduce the index as opposed to storage capacity. The present research offers a fully transparent, reproducible screening index with weights validated using independent data, as well as surrogate models for rapid screening of reservoirs.

International Journal of Latest Technology in Engineering Management & Applied ScienceVol. 15(10)
University of Port Harcourt (NG)
Climate action, Affordable and clean energy
CO2 Sequestration and Geologic Interactions
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon