The international DECOVALEX (DEvelopment of COupled models and their VALidation against EXperiments) research project has been at the forefront of advancing the understanding of coupled thermal-hydrological-mechanical-chemical (THMC) processes as applied to the safe geological disposal of radioactive waste since 1992. Quintessa has been actively involved in this research project since 2008 in support of Radioactive Waste Management Limited (RWM).
DECOVALEX brings together organisations from industry, government and academia to tackle some of the most technically challenging problems in the geological disposal of radioactive waste. Each phase of DECOVALEX (approximately 4 years long) consists of a series of ‘Tasks’ each considering a particular issue, as exemplified by the results of one or more experiments. Research Teams working under each Task then follow an organised work structure, coordinated by a Task Leader, to attempt to gain system understanding and build confidence in model predictions. Elements of the work include: inter-code benchmarking; interpretive modelling of experiments (both at laboratory and field scale); performing generalised predictions on the basis of the developed models; and, most challengingly, performing blind predictions of experimental results.
Quintessa participates in this project at two levels; as a Research Team and as part of the central organisation of the project. On the project organisation aspect, Alex Bond was the leader of Task C1 during DECOVALEX-2015 (2012-2015) and is now the Scientific Coordinator (also known as the Technical Secretary) for DECOVALEX-2019 (2016-2019). Quintessa’s research participation in the project started in DECOVALEX-2011 (2008-2011); key outcomes and summaries of the research work are outlined below.
Quintessa, in conjunction with the University of Edinburgh, contributed to Task A of DECOVALEX-2011, which was concerned with the evaluation of numerical modelling capabilities for simulating coupled THMC processes in argillaceous rocks. The work was focussed on using coupled model to better understand the complex results of the Ventilation Experiment at the Mont Terri Rock Laboratory in Opalinus Clay close to the Swiss-French border.
The primary accomplishments from DECOVALEX-2011 were:
- Multi-phase flow modelling of laboratory experiments with successful reproduction of experimental observations (Bond et al. 2013a, 2014).
- Multi-phase flow modelling and fully coupled mechanical deformation of a large-scale field experiment, including successful blind predictions of experimental responses (Bond et al. 2013a, 2014; Garitte et al. 2013; Millard et al. 2013.)
- Excellent reproduction of observations of the non-reactive geochemical evolution associated with the field experiment through tracer transport modelling based on the variably saturated hydro-mechanical response (Bond et al. 2013b, 2014).
- Good reproduction of the observed reactive geochemical evolution associated with the field experiment through reactive transport modelling based on the coupled hydro-mechanical response (Bond et al. 2013b, 2014).
- Successful PhD achieved by Myles English at the University of Edinburgh.
Following on from DECOVALEX-2011, Quintessa contributed to Tasks A and C1 of DECOVALEX-2015, on this occasion working in conjunction with both the University of Edinburgh and Amec Foster Wheeler. Task A was based on the SEALEX in-situ experiments in a tunnel in argillite at Tournemire in France. The objective of Task A was to predict the hydro-mechanical performance of bentonite-based seals in horizontal boreholes. Task C1, meanwhile, focussed on the analysis of coupled THMC processes in experiments on rock fractures, in particular changes to fracture permeability through fluid-rock reactions under hydrothermal conditions.
The primary accomplishments from DECOVALEX-2015 were:
- Numerical modelling of a water injection test designed to measure permeability of the Toarcian argillite in which the SEALEX experiments are conducted, combined with an analysis of the data collected over the first year of the experiment (Thatcher et al. 2016a).
- Development of a novel approach to representing the hydro-mechanical behaviour of the bentonite which worked well for the experiments in the SEALEX programme (Thatcher et al. 2016b).
- Development of novel techniques for practicable representation of physical and chemical processes occurring on very different spatial and temporal scales (McDermott et al. 2015).
- Successful coupled modelling of the novaculite and granite single fracture experiments, exposing and evaluating key uncertainties that had not been established previously in the literature (Chittenden et al. 2016).
- Detailed comparison of the modelling approaches and outcomes of the modelling teams participating in Task C1, hence gaining considerable insight into the strengths and weaknesses of differing representations (Bond et al. 2016a and 2016b).
- Successful PhD achieved by Andrew Fraser-Harris at the University of Edinburgh.
The overall project outcomes are summarised in a peer-reviewed RWM report (Bond et al. 2015) and associated underpinning reports.
Quintessa is currently developing and applying models for two tasks in the current phase (DECOVALEX-2019):
- Task A on advective gas movement through artificial and natural clays, which is being led by the British Geological Survey; and
- Task E on up-scaling Thermo-Hydro-Mechanical modelling for Callovo-Oxfordian claystone, led by Andra.
Work is ongoing, but early results are very positive with initial results from Task E having been presented at the Clay Conference 2017 and an initial synthesis of results from Task A is expected to be published in 2018. An overview of the technical work completed by the RWM team during the first year of DECOVALEX-2019 was published in May 2018