The inability to accurately resolve subseismic-scale structural discontinui­ties such as natural fractures represents a significant source of uncertainty for subsurface modeling practices. Fracture statistics collected from outcrop analogs are commonly used to fill the knowledge gap to reduce the uncertainty related to fracture-induced permeability anisotropy. The conventional methods of data collection from outcrops are tedious, time consuming, and often biased due to accessibility constraints. Recent advances in virtual outcrop-based methods in fracture characterization enhance conventional methods by streamlining data collection and analysis. However, certain limitations and challenges exist in virtually obtained fracture data sets. The ability to identify fractures that are both exposed as lineations and as planes from a digital outcrop model depends heavily upon the fidelity and resolution of its surface display of RGB color, reducing the capacity of light detection and ranging (­lidar) to the resolution of the scanner-attached camera. In the present study, we adopted a hybrid approach, combining lidar-based digital outcrop models and georeferenced high-quality photomosaics, providing improved texture maps in terms of pixel density compared to maps generated from on-scanner camera images. With this approach, the effects of truncation on digital outcrop models were limited, giving the ability to detect fractures that would otherwise be aliased from on-scanner camera imagery. The fracture system developed within the exposures of the Mississippian Boone Formation, an outcrop analog for age-equivalent reservoir objectives in Mississippi Lime hydrocarbon play, was characterized using conventional and virtual outcrop-based techniques. To test the fidelity of the virtual fracture extraction approach, fracture orientation statistics generated from lidar are compared with equivalent data sets collected using traditional surveys. The results suggest that terrestrial lidar, coupled with referenced gigapixel photomosaics, provide an effective medium for fracture identification with the capacity of resolving fracture characteristics with sufficient fidelity to potentially act as conditioning data for discrete fracture network models, making it an attractive alternative tool for fracture modeling workflows.

You do not currently have access to this article.