This article presents the first results of hydromechanical tests carried out on a natural fracture submitted to a shear stress after modifying a classical shearing machine. The experiments consist of divergent radial injection tests on cylindrical samples containing a fracture at the midpoint perpendicular to the axis. The device allows a directional analysis of the flow paths, and an analysis of how the flow paths change during shearing. The results show that shearing increases the transmissivity of the fracture, and that increasing shear displacement substantially changes the direction of the main flow paths compared to what they are at the onset of shearing.
The hydromechanical behavior of a fracture depends of several parameters, the effects of which are still poorly understood. The major parameters which have to be considered can be determined from the three following properties: the morphology of the fracture, the infilling of the fracture, and the stress state. These three properties are not completely independent, but it is possible to dissociate them to the degree necessary to gain an understanding of mechanical and hydraulical phenomena. The morphology of the fracture is a generic term which has to be precisely defined. In the hydromechanical sense, a fracture is composed of two walls and the void space demarcated by the two walls. The fissural void depends on the matching of the two walls which each have their own morphology. The matching of the two walls is a function of the stress state that loads the fracture. The geometry of the fissural void also depends on the presence or absence of infilling. Therefore, the hydraulical characteristics of the fracture depend directly on the geometry of the fissural void space which equivalently defines the fracture aperture and the connectivity of void space. The mechanical behavior depends most directly on the geometry of the fracture walls and the presence or absence of infilling. The morphology of the fracture changes with some stress states, particularly in shear. The complexity of phenomena in natural fractures requires a progressive experimental approach that allows different parameters to be dissociated to the greatest degree possible. At present, while there are numerous modeling efforts underway of mechanical and/or hydraulical properties, few experimental studies have been undertaken. However, to both fully understand the phenomena and validate models, it is essential to develop laboratory and in- situ tests. The objective of the work described in this paper is to demonstrate the feasibility of hydromechanical tests under shearing with a modified classical shearing machine. This feasibility study follows a study of the mechanical behavior of natural joints under shearing (Archambault etal. 1996) and puts down a general framework for the study of the relationship between morphology and mechanical (Riss et al. 1996) and hydromechanical behavior of joints (Gentier 1987, Billaux and Gentier 1990). In the case of mechanical behavior under normal stress, these studies show that the topology of the flow network changes with the normal stress until a critical normal stress is reached.
