A three-dimensional finite element structural analysis of the Intermediate Scale Borehole Test at the WIPP has been performed. The analysis provides insight into how a relatively new excavation in a creeping medium responds when introduced into an existing pillar which has been undergoing stress redistribution for 5.7 years. The stress field of the volume of material in the immediate vicinity of the borehole changes significantly when the hole is drilled. Closure of the hole is predicted to be larger in the vertical direction than in the horizontal direction, leading to an ovaling of the hole. The relatively high stresses near the hole persist even at the end of the simulation, 2 years after the hole is drilled.
The Waste Isolation Pilot Plant (WIPP) is a research and development facility authorized to demonstrate the safe disposal of low-level radioactive wastes arising from the defense activities of the United States. The WIPP is being developed by the U. S. Department of Energy (DOE) and is located in southeastern New Mexico. The waste will be stored in the bedded salt formation at a depth of about 650 m below the surface. Among the activities underway at the WIPP is the Thermal/ Structural Interactions (TSI) program which consists of a series of in situ experiments designed to address technical issues relevant to the coupled thermostructural response of the underground. Prior to the excavation of the underground test rooms, their responses were computed using what was at that time considered to be the best information on stratigraphy, constitutive modeling, and material properties, all based on knowledge acquired independently of the underground tests. As data from the underground configurations became available, comparisons with the predicted results were made which showed that the predictions of both closure and closure rate were a factor of three less than those measured underground (Morgan et al. 1986; Munson et al. 1986). In the process of addressing the discrepancy, various potential causes were proposed, among them was the question of a "scale effect." This issue was raised by a group of experts in the field of rock mechanics, whose primary experience is in hard or brittle rock. It was also partly driven by the success of an empirical creep model in which merely 1TlUs work supported by the United States Department of Energy (DOE). dividing the elastic Young's Modulus by 12.5 produced a time-dependent response that was in much better agreement with measured closures (Morgan and Krieg 1988). In hard rock, the naturally occurring fracture or joint system in situ (in addition to voids, hard inclusions, etc.) affects the apparent modulus of the rock. The spacing of the joint system in situ typically is so large that the discontinuities are not found in small samples extracted for laboratory use. Consequently, laboratory specimens are typically stronger and suffer, or of a higher modulus, than the large mass of in situ rock. This so-called "scale effect" must be considered when analyzing hard rock structural response.
