Geosyntec REELab reposted this
If you have been my LinkedIn contact for a while, you would know the only thing I find more interesting than subaqueously deposited silts is the uncertainty around in situ stresses below slopes, and how this uncertainty affects static liquefaction triggering assessments. We looked into it a bit numerically in 2022 (https://lnkd.in/d7f8U6vC), and it was a factor in some of the outcomes of our slope stability benchmarking exercise (https://lnkd.in/dBi8Avac). I am happy to announce we just submitted the manuscript “Blind Constitutive Model Predictions of Stress Conditions in Tailings Slopes: A Round-Robin Study” for consideration in the Géotechnique tailings Symposium in Print. We have also posted an unreviewed preprint https://lnkd.in/d2dTznB9, something that while common in other scientific disciplines is less so in geotechnics (although increasing), in the hope of getting feedback from the community in parallel to the formal review. In the current work a suite of triaxial testing on a platinum tailings was made available in a public repository, and numerical modellers were asked to predict the response of the same material to two stress paths - K0 axisymmetric consolidation and drained simple shear (plane strain) - with the later carried out using a bespoke HCTS module developed by Riccardo largely for this purpose (https://lnkd.in/d_crAJsS). Sixteen groups submitted entries to some or all of the scenarios outlined. The results indicated significant variation of K0, principal stress angle and intermediate principal stress ratio between the different entries, with some closely matching the experimental results and others diverging significantly. In some cases, the same constitutive model gave quite different outcomes depending on the calibration prepared by different groups (from the same experimental dataset). As with all our work, electronic data is freely available and can be found in the online repository linked at the end of the preprint. Some of the tailings used in the study may also be possible to share, depending on how the review process goes (e.g. if additional experiments are required, or not). I would like to thank all my co-authors (33!) for their contributions, and the Future Tails program which enabled this work to be carried out. A particular thanks to Riccardo Fanni for developing a HCTS module to let this happen, Gertraud Medicus for providing motivation and encouragement when it all seemed a bit hard to close out and Andy Fourie for making Future Tails happen. The attached figures show: 1) The various K0 assumptions made by participants in our previous slope stability benchmarking exercise, which formed part of the motivation for the current work 2) The CSL from the experimental dataset 3) The range of results in the current work for drained simple shear simulations of an AD sample, compared to the experimental results