Groningen: Spatial Prediction of Rock Gas Saturation by Leveraging Selected and Augmented Well and Seismic Data with Classifier Ensembles

TL;DR

This study demonstrates a method for predicting rock gas saturation in the Groningen field by combining well and seismic data using classifier ensembles, significantly improving data augmentation and prediction accuracy.

Contribution

It introduces a novel data augmentation approach and applies classifier ensembles for spatial prediction of gas saturation, showing promising results in a real-world gas field.

Findings

Data augmentation increased training samples by 9 times.

Achieved high prediction accuracy with Matthews correlation coefficient of 0.7689.

F1-score for gas reservoir class is 0.7949.

Abstract

This paper presents a proof of concept for spatial prediction of rock saturation probability using classifier ensemble methods on the example of the giant Groningen gas field. The stages of generating 1481 seismic field attributes and selecting 63 significant attributes are described. The effectiveness of the proposed method of augmentation of well and seismic data is shown, which increased the training sample by 9 times. On a test sample of 42 wells (blind well test), the results demonstrate good accuracy in predicting the ensemble of classifiers: the Matthews correlation coefficient is 0.7689, and the F1-score for the "gas reservoir" class is 0.7949. Prediction of gas reservoir thicknesses within the field and adjacent areas is made.