Determination of fault damage zones in sandstone rocks using numerical models and statistical analyses

Abstract

The width of fault damage zones is usually assessed through scaling relationships using fault displacements measured from outcrops. However, field data show a large dispersion, indicating that other variables, such as rock properties, can affect the development of damage zones. This article proposes a novel, practical and efficient methodology to predict the width of fault damage zones in sandstone reservoirs, including geomechanical properties’ impact. Firstly, a deterministic model based on elastoplastic relationships and the Finite Element Method (FEM) is defined to assess the damage zone width. Next, empirical correlations between the porosity and the geomechanical parameters are established. Then, planned numerical experiments considering specific domains for each input parameter are generated using the Design of Experiments (DOE) technique. The corresponding results are used to build response surfaces. Finally, a statistical analysis is carried out to define a regression model to express the relationship between the damage zone width, the rock porosity, and the fault displacement. The fit quality is adjusted considering the ANOVA table, Pareto charts, and regression analysis. The results from the regression models indicate a perfect fit with the information predicted by the deterministic model and field data.