Performance Evaluation of Mechanistic Pore Pressure Prediction Models in Deepwater Niger Delta

Prediction models exist through which pore pressure estimation can be made, however, there exists some discrepancies between measured values and predicted values. This study explores the performance of mechanistic pore pressure prediction models rooted in Terzaghi’s principle (Eaton, Bower, and Miller’s models) in four (4) Deepwater wells in the Niger Delta region. The study presented the predictions and quantitatively measured their performance using mean absolute error and percentage average absolute deviation. Quantitatively, Bower’s model had the lowest mean absolute error (MAE) and percentage average absolute error (PAAD) for the 4 Deepwater wells, followed by Eaton, then Millers. In one of the wells (Well3), The mean absolute error was 761.62, 1709.22, and 2759.11 for Bowers, Eaton, and Millers models respectively. For the same well, the percentage average absolute error was 17.75%, 38.14%, and 62.79% for Bowers, Eaton, and Millers models respectively. Caution should be exhibited in the application of the models with proper fine tuning, if they are to be used in predicting pore pressures in Deepwater Niger Delta. This study has significantly advanced the understanding of pore pressure prediction in Deepwater Niger Delta, by revealing in a quantified manner, the deviations of pressure predictions with these models from measured values.