Carbonate Reservoir Facies Modeling: Application of Integrated Workflow from Core, Seismic Inversion to Unsupervised Machine Learning

Carbonate reservoir characterisation is crucial in development stage in order to establish correct understanding of reservoir connectivity and properties such as porosity and permeability in corresponding to the depositional environment facies present. The correct evaluation of reservoir characterisation may improve static modeling and contribute to reserve addition in development plan. Through integration of comprehensive sedimentological, petrophysical and seismic attribute analysis and paired with the inclusion of unsupervised machine learning algorithm, a sensitive depositional environment facies mapping could be identified and established. The updated reservoir depositional environment maps served as crucial input in static modeling for both carbonate fields and the updated static models have enhanced subsurface characterisation for utilization in Full Field Review (FFR) studies. Identification of new opportunities are ongoing to maximise field recovery.