Scale effect and impact of discontinuities in the dynamic elastic constants of the Campanian chalk of Bougival (France)

Abstract

Abstract Physical properties of in situ rock mass are usually estimated from results obtained through laboratory tests on intact rock samples because the access to in situ rock may be quite challenging. This approach however raises some questions concerning the number of samples needed for reliable result, the validity of the extrapolation of the parameters from centimetre scale to a large rock mass and finally the effect of discontinuities contained in the rock mass. An underground quarry in Bougival with easy access to metre-scale pillars and the possibility to collect large number of samples has been chosen to analyse the scale effect and the anisotropy of the Campanian chalk. Different experiments have been designed to determine the dynamic elastic properties (Young’s modulus and Poisson’s ratio) based on geophysical approaches: ultrasonic measurements on laboratory samples, and “hammer” seismic measurements in situ. The static Young’s modulus and Poisson’s ratio have been determined through uniaxial compression tests on centimetre core samples. Pillars with and without visible discontinuities, as well as with various overburden rock thicknesses, have been chosen in order to analyse the possible impact of different heterogeneities on the elastic properties. Core samples of intact chalk, with 40mm to 100mm diameters, have been studied in laboratory. The high dispersion observed on the different results suggests that if only a few tests are analysed, the conclusions may not be representative. A statistical approach is more appropriate to analyse the mechanical properties of the chalk. The dynamic Young’s Modulus and Poisson’s ratio calculated from laboratory samples (centimetres) and in situ rocks (about ten metres) do not reveal any clear impact of size on these elastic properties. The presence of discontinuities has a major impact on both the dynamic Young’s modulus and Poisson’s ratio. Decreasing values of these properties have been observed where discontinuities (fractures, flints) have been detected. Finally, the overburden rock thickness above the underground quarries (from 14m to 50m) seems to have no effect on the mechanical properties; the uncertainty of the measurements, partly due to the heterogeneity of the chalk mass, is likely to be more important than the effect of load on the pillars.