Fault Mechanics

Multi-scale fault reactivation

In petroleum engineering, fluid production in reservoirs can induce stress changes large enough in the vicinity of nearby dormant faults to reactivate them. Interestingly in some cases, this event can be followed by pressure equilibration between the two compartments delimited by an a priori impermeable fault.

In the particular example of a carbonate reservoir rock under specific conditions (relatively high temperature and pressure), the shear-heating of a creeping fault can activate the dissolution of the rock, producing excess pore pressure, lubricating the fault and causing its reactivation, along with a large increase of permeability (by orders of magnitude) allowing the fault to become a potential flow channel.

We distinguish three scales that are of importance for the aforementioned phenomenon of “chemical fault reactivation”: the reservoir scale that describes the stress perturbations affecting the fault; the fault scale that represents the physical length scale of the system; and the micro-scale where the micro-structure of the rock can be described.

In order to characterise this fully coupled, non-linear Thermo-Hydro-Mechano-Chemical (THMC) behaviour, we designed a three-scale numerical framework using the REDBACK (Rock mEchanics with Dissipative feedBACKs) simulator. This approach links the reservoir (km) scale - implementing a hydro-mechanical model - with the fault at the meso-scale (m) - implementing a THMC reactivation model – and the micro-scale (μm) – implementing a hydro-chemical model on meshed μCT-scan images.

Driving the modelling from the meso-scale provides a physically based modelling approach and assists in reducing the use of empirical relationships, like the Kozeny-Carman law linking permeability with porosity, to model this phenomenon. Using this multiscale framework we are able to reconcile the different physical length scales of the system, from tectonic stresses to grain dissolution in an effort to monitor fault reactivation risks and optimise production strategies in those challenging environments.


Framework for the multi-scale reactivation of faults


Strain localization inside fault zone

All the in-situ observations of fault zones reveal that fault slip occurs within a very thin region of the fault that has accommodated several kilometers of slip. This highly localized zone of shear deformation has a thickness of the order of several micrometers to a few centimeters nested within the fault gouge and is called the Principal Slip Zone. The gouge itself is composed of ultra-cataclasite materials (highly crushed particles).

The thickness of fault slip zones is a key parameter for understanding fault dynamics as it has a major significance for energy dissipation, rupture processes, and seismic efficiency. The temperature build up during a seismic slip is governed by the size of the slip zone. Narrow deforming zones concentrate the frictional heating leading to large temperature rises, and thus more rapid weakening. Therefore, it affects the overall energy budget. Indeed, the localization process is strongly linked with the weakening of the fault: a strong material weakening induces a very thin localization band and, conversely, when the deformation localizes, the response of the mechanical system exhibits a strong softening.

Computing the evolution of strain localization is a challenging task due to the difficulties that arise when handling a softening behavior of the material. The problem becomes mathematically ill-posed  and the results of classical continuum theories predict a localization zone with zero thickness. The response of the numerical simulations of the mechanical system depends thus on the mesh size and so does the energy dissipated by the system.

In order to address this problem, we consider different theories like visco-plasticity and high order continuum theories (Cosserat continuum for example) together with multi-physical couplings. We aim at investigating the role of different weakening mechanisms on the phenomenon of strain localization inside fault zones and the influence of each of them on the energy budget of seismic slip. The goal is to reconcile field observations from exhumed faults regarding the combined roles of grain size evolution and pore-pressure and temperature effects like thermal pressurization and flash heating.