Faulting and healing of the crust throughout the seismic cycle: From microscale physico-chemical processes to a global rheology

In the Earth's upper crust, geological deformation is primarily accommodated by fracturing
and faulting. Fault motion can be catastrophic and generate earthquakes, but these are rather rare events in time: most faults in the crust are either very slowly moving (creeping) or completely locked between earthquakes. During this so-called interseismic period, fractures heal (i.e., they regain strength) and seal (i.e., they close paths for fluid flow) due to chemical lithification processes. The competition of healing and sealing processes with fracture growth conditions the location and timing of future earthquakes along faults. Hence, understanding how rocks heal and fracture, and the feedbacks between these processes, is an essential step towards a better knowledge of earthquake generation and the associated
hazards. Despite the considerable attention this problem has received since the late 1970's,
mostly through the development of phenomenological frictional constitutive laws, very little is
known in terms of actual microscale mechanisms, from which the observed empirical macroscopic laws could
arise. In particular, a unifying micromechanical framework encompassing fracture gowth, healing and
sealing remains to be determined. Therefore, I propose to tackle this problem from a different point
of view, by (1) identifying and quantifying experimentally the physico-chemical processes controlling
ultra-slow deformation and fracture healing, and (2) theoretically input those processes into a rigorous
micromechanical framework that can be used to extrapolate from the laboratory to field scale.

The major outputs of the project will be: (1) quantitative experimental results documenting fracture
growth and healing rates in limestone and granite, (2) a unified model, based on microscopic mechanisms,
allowing predictions of the mechanical and transport properties of those rocks at large scale.
These scientific outputs are fundamental in nature and the primary target is to improve our knowledge
of faulting and earthquake processes in the Earth. Althought the biggest users of the data and models
obtained during this project will be academics, aspects of the project, such as a better understanding
of fault reactivation, will also greatly benefit to engineers in the energy industry (geothermal, oil/gas)
and civil protection agencies.

The new laboratory data obtained during the project will be of great help to the community studying fault mechanics, as they can be used to model the mechanisms of post- and inter-seismic deformation, as well as the early stages of earthquake nucleation. In addition, the unified fracturing/healing constitutive model I intend to develop will impact the modellers community, which has been focused almost solely on semi-empirical (``rate-and-state'') friction laws in the past 30 years. The process-based theoretical model I suggest will indeed bring a complete understanding of the physical processes from which macroscopic friction laws arise.

My research will also impact the academic community working on earthquake hazard assessment. The outputs of the project, detailed experimental data and mechanical model, can be used to model seismicity rates and earthquake recurrence times along active faults, which are two essential ingredients in earthquake hazard maps. In addition, the improved understanding of the crustal deformation processes as a function of physical conditions (pressure, temperature, time, presence of fluids) will be used to better assess risks associated with volcanic eruptions.

The "upstream" knowledge obtained during the project will bring will have an impact on industrial
problems related to (1) induced seismcity in geothermal/oil/gas reservoirs, and (2) reservoir monitoring
and modelling.
When reservoirs (geothermal or hydrocrabon) are exploited in regions with complex tectonic history
and ancient faults, an important question is whether fluid injection for hydraulic fracturing can reactivate
those faults. My work will bring essential and quantitative knowledge on fault healing/sealing
processes, and hence will help to determine the reactivation stresses of healed faults and fractures. This will constitute a basis for hazard mitigation and impact reduction in and around geothermal and gas fields.

In addition, the model I plan to develop will be based on and tested against experimental observation
on porous carbonates, which constitute a major reservoir rock type. The mechanical model,
presented in the form of a constitutive law (i.e., stress-strain relations) will thus be useful for reservoir
monitoring and modelling applications in carbonate basins.