Fracture pattern prediction using geomechanical models incorporating diagenesis, with comparison to outcrop data (Cambrian Eriboll Group sandstones, Northwestern Scotland) and core observations (Tertiary Mirador Formation sandstones, Llanos foothills Colombia)

Effective and accurate characterization of fracture populations is a key for hydrocarbon reservoir assessments. The challenge is that fractures in the subsurface are exceedingly difficult to sample in meaningful ways. Outcrop studies have been an important source of information on fracture patterns providing better knowledge of the distribution, connectivity and geometry of fracture patterns. By establishing an analogy to outcrop observations, limited core data can be used to better understand subsurface fracture patterns, leading to more accurate fracture permeability and flow pattern prediction. Using geomechanical models to simulate fracture pattern development is one way to systematically generalize the detailed observations of outcrop fractures in a specific locality to the general problem of prediction of fractures in the subsurface. Here, I use a combined methodology for fracture pattern description, which includes macro- and microscale observations of fractures and cements in Cambrian Eriboll Formation outcrops and Eocene Mirador Formation core. I use a geomechanical model that simulates fracture growth under the influence of concurrent diagenesis (cement precipitation). I focused on the attributes of fracture size (aperture and length). The Eriboll Formation outcrops are unusually well-preserved examples of opening-mode fractures (veins and joints) with a great exposure of fracture size distributions. Using photographic and conventional mapping techniques, I generated fracture trace maps at the outcrop level at a scale of 1:10. Combining macrofracture observations with measurements along scanlines, petrography, fluid-inclusion analysis and high-resolution automated scanning electron microscopy (SEM), I found three preferred strikes for the fractures at the outcrop: N-S, NW-SE and E-W. From microstructural observations, I found that the current fracture pattern is the result of superimposed deformations that produced N-S striking fractures of different ages that share a common strike but that differ in dip and in cross section show consistent crosscutting relations. Mirador sandstone cores have microfractures that exhibit size scaling that follows a power-law similar to those in the Eriboll. Mirador fracture openness predictions based on diagenesis resemble those found in the outcrop analog. Rock mechanical properties, subcritical crack index, diagenesis, mechanical layer thickness and strains measured from outcrop and subsurface rocks were used as an input for a geomechanical fracture pattern simulator (the geomechanical model Joints, Olson, 2007). I generated plots of fracture patterns that I compared to outcrop patterns. I used the same procedure to predict patterns for Mirador reservoir rocks. My results show how geomechanical models that incorporate fracture and diagenesis can help make outcrop more germane to subsurface fracture prediction