A seismic reservoir characterization and porosity estimation workflow to support geological model update: pre-salt reservoir case study, Brazil
L. Oliveira, F. Pimentel, M. Peiro, P. Amaral and J. Christovan
Journal name: First Break
Issue: Vol 36, No 9, September 2018 pp. 75 - 85
Info: Article, PDF ( 1.91Mb )
Price: € 30
Quantitative seismic interpretation plays an important and growing role for reservoir characterization, as seismic data become increasingly reliable as a result of the latest advances in acquisition and processing techniques. With these advances, the multi-disciplinary integration between geology, geophysics and engineering becomes increasingly effective at reducing operational risks relating to reservoir exploration and production. In addition, a multi-disciplinary approach is essential for a better understanding of reservoir features. In this paper, we present an integrated study that combines geophysical and geological approaches to perform porosity estimation and populate the reservoir geological model (static model). For pre-salt oilfields, owing to the complex porosity distribution in carbonate reservoirs, predicting a reliable porosity is a fundamental step for reservoir modelling. After presenting some of the specific challenges associated with pre-salt reservoirs, we will describe pre-stack seismic data preconditioning. This first step is important to improve the seismic data set at the target level in terms of signal-to-noise ratio and resolution before it is used as input to seismic inversion, the second step in this workflow. In the seismic inversion process, reservoir elastic properties are estimated. From these inverted elastic properties, it is possible to perform a Bayesian lithofacies classification and, as the final products for this step, litho-probability volumes are generated in co-operation with the field’s geologists to be subsequently used as input to the geological modelling. For the third and last step, we used facies probability volumes and acoustic impedance to estimate porosity. To use probability volumes in porosity model building, we designed a workflow to transform probability values into 3D porosity trends: first, a categorical facies volume is created by applying cut-offs on lithofacies probability volumes. For each layer of this model, a mean porosity per categorical region (facies), based on porosity logs at wells is calculated. Then, a relationship between acoustic impedance and mean porosity values was determined to create the final trend porosity volume to guide the porosity prediction away from wells. Finally, we ran flow model simulations to quantify the benefits of the integrated workflow compared to a more commonly used 2D porosity map-based approach, showing the improvement in matching static and dynamic reservoir properties, mainly for pressure and Gas Oil Ratio (GOR) predictions.