Séminaire Mai-Linh Doan

Abstract

The mechanics of active faults is a young science. A seemingly simple question like the structure of active faults is still an open question. It is difficult to predict the fine structure of an active fault a priori. However, this structuring is a prerequisite for understanding the mechanics of an active fault, its pre-seismic nucleation processes, its co-seismic slip propagation processes,  and  its post-seismic recement processes.

The damage zone represents most of the volume of a fault zone. We aimed to quantify the damage by using a multiscale approach. On a metric to decimetric scale, the borehole logs provide a continuous profile of mechanical properties to characterize the extent and intensity of the damage. An example is  shown on the Alpine Fault in New Zealand, which exhibits a hierarchical damage structure.

A new approach that  considers drilling data also helps  characterize  the hydraulic profiles. In the Nankai subduction zone, offshore Japan, we show that the décollement already develops fluid overpressures even near the trench. The damaged zone was  asymmetric and constituted a hydraulic drain, whereas the fault core was mineralogically differentiated and constituted  an impermeable barrier.

Therefore, mechanical damage is not the only parameter that controls fault evolution. Fault maturation is also associated with  the rock alteration accentuated by fracturing. Laboratory experiments showed that this fracturing was more intense under  co-seismic  loading at high strain rates. At 180°C, the percolation of reactive fluids in these intensely fractured samples could  significantly alter the sample, causing creep and a decrease in permeability. The AlterAction ANR Project aims to better understand the link between alteration and deformation in active faults, with application to the Nojima Fault, which generated the 1995 M_w6.9 Kobe earthquake, and to the geothermal reservoir of Soultz-sous-Forêts.

These multimethod and multiscale approaches open new perspectives to model the structuring of active faults and better understand the diversity of their mechanical behavior.