Numerical-Mechanical Modelling of Stress Distribution and Natural Fractures Under Variable Paleostress Conditions, Lisbon Valley, Paradox Basin, Utah

Abstract

The degree to which the stress field in the crust is variable has long been debated. Whereas some studies suggest that the stress field in the earth’s crust is highly regionally coherent, many field observations indicate substantial local variations associated with the impact of mechanical perturbations, such as stratigraphy, structural position, and proximity to faults. In addition to field stress data such as borehole data and earthquake focal mechanisms, stress orientation and magnitude can be modeled using numerical mechanical approaches. The validity of the assumptions made in these modeling approaches and the sensitivity of the results to uncertainty in these assumptions requires conditioning the models with empirical observations from case studies of natural structures. In this study, we focus on the geologic structures in Lisbon Valley, within the Paradox Basin in eastern Utah, to evaluate the degree to which natural fractures in the area reflect the relative contributions of regional stresses and local stress perturbations due to proximity to larger geologic structures. The field site is comprised of a doubly-plunging salt-cored anticline with gentle limb dips that has been breached by a system of normal faults. These faults exhibit significant variations in total offset and fault zone architecture, providing us with the ability to evaluate the impact of model assumptions relative to structural position on the anticline and proximity to fault segments. We constructed a detailed 3D geologic model of the area based on surface geology, well tops, and a series of 19 balanced cross sections across the structure. The model was then used as the framework for 3D geomechanical modeling using a boundary element elastic dislocation method. Empirical observations of the paleostress field were provided by mapping the spatial distribution and orientation of fracture sets across the study area from satellite data. Several fracture sets are observed in the study area; one set is sub-parallel to strike on the fold limbs while the other set is sub-perpendicular to the faults. Qualitative comparison between observed fracture orientations and those predicted by the model-derived stress field provide a basis for interpreting the timing of fracture formation relative to the timing of growth of the larger-scale geologic structures, as well as an ability to evaluate the relative importance of model assumptions such as relative paleostress magnitudes and the relative contribution of regional stresses and local stress variations due to modifications associated with the presence and local geometry of fault systems. The well exposed NW striking, fault-parallel fractures along with its crosscutting fault-oblique fractures are most consistent with having formed under Laramide stress conditions. The younger, NE-oriented fractures are consistent with formation in a post-Laramide to modern state of stress. The stress perturbation due to faults and salt deformation is highly dependent upon the stress shape and magnitude. The ability to predict modern stress is essential to borehole stability, reservoir management, and subsurface storage applications, while accurate understanding of paleostress conditions is necessary for predicting sub-seismic natural fractures that could impact permeability or reservoir compartmentalization.