A COMPARISON OF THEORETICAL ROCK MODELS WITH LABORATORY DATA

A number of rock modelling theories based on single-crystal data were tested in order to assess their effectiveness in modelling actual laboratory rock data. These theories included the Voigt, Reuss, and Hashin-Shtrikman bounds for isotropic single-phase pore free aggregates and composites; and, the self-consistent inclusion theories of Hashin-Shtrikman, Wu, and Walsh, and Thomsen's Biot-consistent approach for porous materials. The results for very low porosity rocks (effectively pore free) indicate that models based on Hashin-Shtrikman theory yield narrow bounds which accurately predict mean laboratory P- and S-wave velocities. However, meaningful comparisons require careful and accurate rock characterization in terms of mineral composition and texture. The theories for porous rock models yielded S-wave velocities which were not as consistent as their counterpart P-wave velocities in matching the laboratory data. Moreover, the discrepancies increase markedly as rock porosity increases. This result reflects primarily, differences in internal microcrack structure and composition between that assumed in the theoretical modelling procedures, and that present in the actual rock samples. Saturated and dry rocks containing cracks were most successfully modelled by the Biot-consistent theory proposed recently by Thomsen (1985).