A Geomechanical Approach to Predicting Fracture Orientation, Density and Conductivity: An Example From the Gorm Field, Danish North Sea
Abstract
We examine the combined use of Elastic Dislocation (ED) theory to predict fracture orientation/density and in situ stress analysis to predict fracture conductivity. Effective use of ED theory is
well
documented in both the structural Geology and petroleum Geology literature but it is not routinely applied in the oil and gas industry. In situ stress analysis is more commonly used in
well
design, mitigating fault reactivation risk or to predict fracture orientations that are most likely to transmit fluids. Combining ED forward modelling with in situ stress analysis can provide a useful framework for understanding fracture behaviour in the development/appraisal phase. In producing fields where stress indicators and logged fractures are available to calibrate the geomechanical modelling, the process is valuable for
planning
future wells and estimating fracture permeability modifiers for use in simulations. The Gorm field is a salt-cored, faulted chalk dome, central to the infrastructure of the Danish North Sea. Production in recent years has waned. To this end a geomechanical study was commissioned to derisk critical
well
infill. Core and image log data from 13 wells provide a first order, independent assessment of the predictive method. We run forward ED models (using FaultED) of mechanical response to the net effects of fault slip, underlying salt uplift/withdrawal and regional strain. These models predict the stress perturbations around the set of seismically resolved faults. At each (arbitrary) observation point in the reservoir a wide range of mechanical attributes are calculated (e.g. fracture mode, orientation, proxies for fracture density). Comparison between observations of orientation and density from 13 wells and predicted fractures yields encouraging correlations. In situ stress analysis estimates the orientation-dependent proximity of a fracture to failure and hence likely hydraulic signature. A critically stressed fracture is more likely to transmit fluid than a stable fracture. Again, we find a positive correlation between observed conductive/resistive fractures in the
well
reports and our geomechanical predictions. Recently a
well
was drilled in an area where our ED models predict high fracture density. “It had much higher productivity than predicted by matrix properties modelling, and is now the most productive
well
in the field accounting for over 25% of daily production, clearly enhanced by natural fractures” (Arnhild pers comm.).
AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014