One of the major hurdles in structural Geology and geomechanics is the ability to accurately model the evolution of complex geologic structures in a reproducible and efficient manner. Whether modeling large-scale crustal deformation or outcrop-scale localized deformation, the ability to create realistic, predictive models remains constrained by the limitations of the modeling approach. Typically, complex geologic problems have been characterized by one of two modeling approaches: physical analog modeling and numerical modeling. Despite their individual strengths, there remains a gap in integration of the two approaches. Physical modeling is generally more capable of simulating 3-D structural complexity, including discontinuous deformation such as faulting. However, analog modeling is cumbersome for conducting multiple parametric analyses, and is not amenable to extraction of quantitative stress information. In contrast, numerical modeling can record complex stress and strain fields during model evolution and is well-suited to parametric analyses but often suffers from being too perfect due to the lack of natural flaws and heterogeneities. To overcome the inherent limitations of each approach and to leverage their relative strengths, we have undertaken a series of carefully designed experiments to simulate fluid-induced deformation (i.e. hydraulic fracturing). The impact of factors such as initial stress state and mechanical stratigraphy can be considered. The physical models use analog materials (e.g. sand) that can be numerically represented with established constitutive relationships (e.g. elastoplasticity) that permit inelastic (permanent) deformation to be captured. Likewise, the physical modeling loading conditions (e.g. liquid wax injection) translate to numerical fluid flow or pressure loads. The physical and numerical model results both suggest that complex non-planar, branching fractures should be the expected deformation response to fluid injection into geologic media. Integration of the contrasting modeling approaches provides more robust interpretations of both the geologic process and resulting deformation than can be obtained from either approach individually.
AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014