Editorial: partial saturation in compacted soils

Welcome to this second issue of Géotechnique devoted to the symposium in print on ‘partial saturation in compacted soils'. The previous (April) issue was also devoted to the symposium in print and featured five articles on fundamental topics related to the sub-themes of ‘benchmarking of techniques and models' and ‘experimental observation and modelling'. The present issue has, instead, a more applicative focus with contributions covering barrier systems for underground disposal of nuclear waste, earth dams and embankments, as well as a study of lime-treated compacted clays. The first three contributions in the present issue are dedicated to the sub-theme ‘application to engineering problems and case studies' (Alonso et al., 2011a; Gens et al., 2011; Thorel et al., 2011) while the last two contributions are devoted to the sub-theme ‘material characterisation' (Taibi et al., 2011; Tang et al., 2011).

The first article, by Gens et al. (2011), investigates the hydromechanical behaviour of a compacted mixture of bentonite powder and pellets used as a fill around nuclear waste in underground disposal schemes. The mixture is characterised by a double structure that includes micropores inside the pellets and macropores in the powder between pellets. The behaviour of the mixture is modelled by assuming a ‘virtual' overlap of two distinct materials corresponding to the two pore structures. Changes in the stresses and degree of saturation of the fill are obtained by superimposing the corresponding predictions obtained from the constitutive laws of each of the two overlapping materials. At the core of the proposed formulation there are the assumptions that the deformation of one structural level can induce deformation in the other one and that local hydraulic gradients govern flow across the two pore classes. The approach enables prediction of complex patterns of swelling pressure (showing a peak followed by a fall and a subsequent increase towards a stationary value) as observed during confined hydration tests. This work confirms the importance of taking into account multi-modal material fabric during the formulation of constitutive laws for compacted soils and is in line with results published in the April issue of Géotechnique by Romero et al. (2011) and Alonso et al. (2011b).

In the second article, Alonso et al. (2011a) present a numerical model of a 40 m high earth and rockfill dam recently built in eastern Spain. The model incorporates three main material zones, namely the clay core, the upstream/downstream rockfill shells and the foundation soil. For each of these three zones, different sets of constitutive laws are adopted to describe the mechanical, retention and permeability behaviour. The adopted constitutive laws reproduce generally well the response of the corresponding material as observed during laboratory tests on representative samples. Only the dilatancy of the rockfill samples during shearing is poorly predicted. This is consistent with results published in the April issue of Géotechnique by D'Onza et al. (2011), who highlighted the difficulty of accurately calculating volumetric strains during shearing for a range of different constitutive models. The dam was instrumented with load cells, piezometers and displacement gauges to measure total stress, pore water pressures and settlements, respectively. Such measurements agree well with the predictions of the numerical model during construction. The model is also used to forecast the response of the dam during the impoundment that will take place later in 2011.

The third article, by Thorel et al. (2011), presents the results from a centrifuge test in which a loosely compacted model embankment is subjected to a rise of the water table while settlements and changes of void ratio are measured along the embankment height as flooding advances from the base. Collapse-compression takes place in the deeper soil layers that undergo full saturation while little deformation is measured in the upper soil layers that stay above the water table and, hence, remain unsaturated. Consistent with current understanding, the largest collapse occurs at the very base of the embankment because of the existence of high stress levels due to the soil self-weight. The authors propose an approximate methodology for estimating crest settlements from simple double oedometer tests. Such a methodology is capable of giving realistic estimates of settlement due to flooding, provided that the changing position of the water table inside the embankment can be tracked with relative accuracy.

In the fourth article, Tang et al. (2011) explore the reasons why the small-strain stiffness of lime-treated compacted soils tends to be lower for samples retrieved from the field compared to similar samples prepared in the laboratory. The authors identify two possible causes for such behaviour, namely the degradation of inter-granular cementation due to wetting–drying cycles induced by the in situ climatic conditions and the coarser grain size of field samples in comparison with laboratory-prepared samples. The authors also show that suction has no significant influence on the stiffness of lime-treated compacted soils (unlike untreated compacted soils) indicating that inter-granular cementation dominates over capillarity in governing material behaviour. After extreme drying, the stiffness of the lime-treated material reduces significantly suggesting that a dramatic increase of suction can damage inter-particle cementation with the additional consequence of making the effect of capillarity on soil behaviour more evident.

The technical note by Taibi et al. (2011) closes this issue by presenting wetting/drying tests performed on four different soils used for the construction of earth dam cores in southern France and north-western Algeria. For each soil, multiple samples are compacted with the same effort at different water contents on both sides of the optimum value and then subjected to suction changes under unconfined conditions. Results show that the compaction water content has little influence on volumetric deformations during wetting/drying but it has a rather noticeable effect on water retention behaviour. Larger differences are also observed in terms of both volumetric deformations and water retention behaviour for samples compacted with variable effort at constant water content. The constitutive parameters describing swelling/shrinkage and water retention behaviour are then correlated to the soil liquid limit.

All ten contributions published in this and the April issue of Géotechnique will be presented by the authors during a full-day symposium meeting to be held on 20th June at the Institution of Civil Engineers in London. The meeting is open to all interested people and those wishing to take part in it are invited to register by visiting http://ice.org.uk/geotechniquesip. The ten contributions will be presented in four thematic sessions, corresponding to the four sub-themes of the symposium. Each session will also feature a panel discussion, chaired by a leading researcher in the area, which will provide the audience with the opportunity to engage with the authors in a debate of key issues. In addition, delegates will be presented with two keynote lectures delivered by Prof. Eduardo Alonso (Universitat Politécnica de Catalunya, Spain) and Mr Tony O'Brien (Mott MacDonald, UK). The symposium will therefore provide both researchers and practitioners with fruitful and stimulating opportunities to exchange knowledge and to network. This is expected to be a very well-attended meeting so I would encourage you to register as soon as possible.

As is inevitable in the organisation of any special issue, strict deadlines had to be set during peer-reviewing so that articles could go to print according to the planned schedule. I would therefore like to acknowledge the enormous amount of work which has been expended by referees and members of the symposium sub-committee in reviewing manuscripts within a very tight timescale and to express my sincere thanks to all authors for responding to requests of revision with very short deadlines. Last but not least, I am indebted to staff at Thomas Telford and ICE for efficiently managing the review process and organising the symposium meeting. This symposium in print is the fruit of the work of a team without which it could not have happened.