摘要
The risks and uncertainties related to the storage of high-level radioactive waste （HLRW） can be reducedthanks to focused studies and investigations. HLRWs are going to be placed in deep geological repositories,enveloped in an engineered bentonite barrier, whose physical conditions are subjected tochange throughout the lifespan of the infrastructure. Seismic tomography can be employed to monitor itsphysical state and integrity. The design of the seismic monitoring system can be optimized via conductingand analyzing numerical simulations of wave propagation in representative repository geometry.However, the quality of the numerical results relies on their initial calibration. The main aim of this paperis to provide a workflow to calibrate numerical tools employing laboratory ultrasonic datasets. The finitedifference code SOFI2D was employed to model ultrasonic waves propagating through a laboratorysample. Specifically, the input velocity model was calibrated to achieve a best match between experimentaland numerical ultrasonic traces. Likely due to the imperfections of the contact surfaces, theresultant velocities of P- and S-wave propagation tend to be noticeably lower than those a prioriassigned. Then, the calibrated model was employed to estimate the attenuation in a montmorillonitesample. The obtained low quality factors （Q） suggest that pronounced inelastic behavior of the clay has tobe taken into account in geophysical modeling and analysis. Consequently, this contribution should beconsidered as a first step towards the creation of a numerical tool to evaluate wave propagation innuclear waste repositories. 2014 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting byElsevier B.V. All rights reserved.
The risks and uncertainties related to the storage of high-level radioactive waste （HLRW） can be reducedthanks to focused studies and investigations. HLRWs are going to be placed in deep geological repositories,enveloped in an engineered bentonite barrier, whose physical conditions are subjected tochange throughout the lifespan of the infrastructure. Seismic tomography can be employed to monitor itsphysical state and integrity. The design of the seismic monitoring system can be optimized via conductingand analyzing numerical simulations of wave propagation in representative repository geometry.However, the quality of the numerical results relies on their initial calibration. The main aim of this paperis to provide a workflow to calibrate numerical tools employing laboratory ultrasonic datasets. The finitedifference code SOFI2D was employed to model ultrasonic waves propagating through a laboratorysample. Specifically, the input velocity model was calibrated to achieve a best match between experimentaland numerical ultrasonic traces. Likely due to the imperfections of the contact surfaces, theresultant velocities of P- and S-wave propagation tend to be noticeably lower than those a prioriassigned. Then, the calibrated model was employed to estimate the attenuation in a montmorillonitesample. The obtained low quality factors （Q） suggest that pronounced inelastic behavior of the clay has tobe taken into account in geophysical modeling and analysis. Consequently, this contribution should beconsidered as a first step towards the creation of a numerical tool to evaluate wave propagation innuclear waste repositories. 2014 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting byElsevier B.V. All rights reserved.
