The evolution of fracturing process in fault damagezones and its influence on the rock permeability: case study of the shallow marine and ramp carbonates of Central Apennines (Italy)

Abstract

More than 60% of the world’s oil and 40% of the world’s gas reserves are held in carbonates. For this reason there are significant challenges in term of recovery due to the highly complex internal structure and specificity of carbonate reservoirs. The presence of faults, both seismic and subseismic, compartmentalized the reservoirs and therefore ruled the pathway of deep fluid. This is particularly true in carbonate rocks, where fault-core and the damage-zones show different hydraulic behaviours from the pristine rock as well as between them. This study, by means of an original methodology, proposes a conceptual model of brittle deformation evolution and a predictive computer model of deformation distribution in carbonate damage-zones, starting from the statistical analysis of field data. In particular, this work faces the following main challenges: the field characterization of fault zones and the quantification of the brittle deformation; the analytical study of the field data by means of a Monte Carlo approach; the numerical modelling of fault zones. Since the fracture patterns of petroleum reservoirs in situ are difficult to study in detail, field analogues represent a good opportunity for understanding their fracture-related permeability. The field work was performed in Central Apennine on the Meso-Cenozoic units of carbonate platform and ramp, which represent a very good analogue of rock reservoirs. A total amount of 22 faults have been analysed on the field for this study. The fault database consists mostly on sub-seismic scale faults (scale below the resolution of seismic observation, 10 m); the database was completed with the study of 6 seismic scale faults (scale over the resolution of seismic observation). The results from these study show that the fault-related deformation intensity (H/S) is always significantly higher than the regional one, ranging 9.5-2.7 and 5-2 respectively, so that we can determine the lower threshold of H/S as 2.7 for faults and 2 for regional deformation. The fault kinematic rules the development of the fault deformation patterns. This is true mostly for the faults that showed double dip-slip and strike-slip kinematics, which are more deformed. The evolution of fault-related fracture sets is controlled by the rock anisotropy and the intensity of deformation of the main fracture cleavage is always higher than their suborded. The development of the stratabound and the non stratabound fractures in layered rocks depends on the stress intensity and the fault scale, and is controlled by the mechanical layering. A new methodology was improved to define the boundary between the damage zone and the pristine rock. A good relation exists between the damage-zone width and the intensity of deformation (H/S) for the subseismic scale faults. A power low trend was found between the damage-zone widths and the fault throws, suggesting that the fault damage-zones fracturing process is more intense in the early stage of fault development. The fault-related deformation intensity shows a clear variation through the distance from the master fault plane and three main trends have been detected: a) Exponential decrease; b) Semi-Gaussian decrease; c) Gaussian decrease. The Monte Carlo converging method provides a statistical tool to study the geological factors affecting the fracturing evolution in the fault damage-zone and the results show that the throw affects the deformation distribution mainly in the early stage of fault evolution. A general equation was found to define quantitatively the distribution of the intensity of deformation in fault damage-zones, whose parameters have a geological meaning and are related to the stress intensity. This equation provides a deformation intensity prediction with an error of + 2 H/S, which is the RMS resulting from simulation. The equation was implemented in the fracturing predictive software. The comparison between the models and the field data proves that the resulting models provide a reliable prediction of fracturing distribution in fault damage-zones. Several methodologies and approaches have been proposed to predict the role of faults on brittle deformation distribution and their impact on permeability distribution, for instance indirect method as borehole and seismic imaging technologies, analogical and numerical simulations, micro tectonic studies, or direct method as field studies. The current study starting from real field data is fundamental to fill the gap of information between seismic data and well data, and hence necessary to allow including this information in 3-D reservoir models, simulating fluid flow.