Résumé
The Callovo-Oxfordian claystones were chosen for their low permeability. This feature marked by the clay content of the rock is inevitably accompanied by a low tensile strength. This generates the development of an excavation damaged zone. Seismic elastic wave propagation analysis methods are potentially valuable for the characterization of the damaged zone. These methods are more or less adapted to clay environment mainly due to the strong wave attenuation. The easiest seismic method to implement is probably the seismic refraction. It is used from the surface of the wall. It involves analyzing the waves refracted at the interface between the fractured zone and the bedrock. Sato et al. (2003) showed with this method the influence of excavation mode on the extension of the damaged zone in the underground laboratory in Japan Tono (sedimentary clay rock). Similarly, Lagarde et al. (2006) used the method in the Callovo-Oxfordian claystone. The application of seismic refraction in this context is at the limit of resolution because it assumes a multilayer tabular model without velocity reverse and the error on the arrival time measured is very close to time variations due to the damaged zone. The analysis of the surface waves is based on the dispersive property of these waves. They have the particularity to propagate at different depths depending on their frequency. The objective is therefore to measure the dispersion curve of the wave model and find the velocity profile that best matches. This method called MASW (Multi-Channel Acquisition of Surface Waves) is typically used to find the velocity profile of a tabular field where the inversion is simplified. As part of the thesis of Lagarde (2007), the MASW method was extended to the concave geometry of deep underground excavations to characterize the velocity profile in the damaged zone. It makes it possible to reveal any velocity profile including those with one or more velocity reverse. This method was used at the Meuse Haute Marne URL to characterize the damaged zone at the bottom of a drift. Seismic methods can also be implemented in borehole, for example ultrasonic logs. They are well suited to the scale of the damaged zone and numerous surveys in deep underground excavations. In the Meuse Haute Marne URL, this method has been used on numerous occasions (Schuster et al, 2001) during the different phases of the excavation of shafts and galleries. These velocity measurements are used to describe both the evolution of the gradient of damage / deconfinement from gallery wall and isolated fractures at greater depth. Acoustic emissions (AE) occur during initiation and expansion of cracks. Quantifying and locating those AE can bring information in determining the initiation and expansion of the damaged zone. Monitoring these AE has been successfully applied around structures in very resistant rock under high stresses such as granite host rocks of the Lac du Bonnet, Manitoba (Meglis et al, 2005). In claystones, however, the strong wave attenuation make them almost undetectable. The only opportunity to detect them is to place acoustic sensors in close proximity to sources. Forney (1999) developed a compact device capable of recording AE reflecting the creation of the damaged zone during the excavation of a gallery in Mont Terri. Claystones might show a natural variability that must be taken into account in the analysis of velocity variations induced by damage. A solution to overcome this problem is to measure the velocity of the waves in the rock before it was damaged and monitor its evolution during the excavation of the build. This technique called "velocity survey" has been implemented several times during the excavation of shafts (Balland et al, 2009), galleries or slots. It allowed characterizing the intensity and extension of the damaged zone, as well as the phenomena occurring during sealing or reconfinement of a fractured rock vicinity. Using a specifically designed 3D device with S and P waves back analysis in the stiffness matrix changes.