Uncertainty Quantification in Fractured Reservoirs Based on Outcrop Modelling from Northeast Brazil

This work is aimed at uncertainty quantification of flow performance in fractured reservoirs. The study demonstrates how outcrop data on fracture distribution are integrated into uncertainty quantification workflow. It includes eliciting statistics from the outcrop data to generate ensemble of discreet fracture networks (DFNs), DFNs upscaling into permeability grids, clustering of models according to permeability patterns, choice of training images for the cluster scenarios and generation of multiple realisations, flow simulation to come up with the spread of uncertainty in the recovery performance. The influence of fracture intensity and orientation on recovery efficiency was quantified. Uncertainty was represented by multiple sets of realisations based on training images. The workflow created is applicable to any fractured reservoir studies. The data for the case study is from five outcrops in Jandaira formation from NE Brazil, Potiguar Basin was mapped by Delft University. Results of simulation are of significant importance for southern Brazil hydrocarbon fields which have reserves in similar carbonate formations. This research was focused on geometric properties of fracture networks; further investigation of physical properties of fractures and integration with well and seismic data is essential for accurate properties prediction of fields from southern Brazil.