Predicting fluid flow in fractured reservoirs by application of lacunarity analysis of natural fracture analogues

Abstract

Researchers have used lacunarity, a parameter that quantifies scale-dependent clustering in patterns to distinguish fracture networks that belong to the same fractal system. Also, in a previous study, the authors showed that lacunarity is efficient in representing the connectivity and fluid flow in synthetic fractal-fracture models having a single fractal dimension. The objective of this research is to investigate if the concepts thus developed is applicable to outcrop analogues which are representative of sub-surface fractured reservoirs. A set of nested fracture networks belonging to a single fractal system but mapped at different scales and resolutions is considered in this study. Lacunarity and connectivity values of these maps are evaluated using geospatial data analysis techniques. Fracture continuum (FC) models are built from these fracture maps and a streamline simulator, TRACE3D is used to flow simulate these maps. Results show that, although, the fractal dimension of these maps is same, but there exist stubble differences in the values of lacunarity, percolation connectivity, and also the fluid recovery values. It is further noted that the clustering, connectivity, and fluid recovery values can be pairwise correlated very well for these natural fracture maps. Thus, the overall results indicate that connectivity in fracture maps and hence in turn their flow properties are controlled by lacunarity or scale-dependent clustering attributes. Therefore, there could be novel applicability of lacunarity parameter in calibrating discrete fracture network (DFN) models with respect to connectivity of natural fracture maps and prediction of flow behavior in fractured reservoirs. Thematic collection: This article is part of the Digitally enabled geoscience workflows: unlocking the power of our data collection available at: https://www.lyellcollection.org/topic/collections/digitally-enabled-geoscience-workflows