Editorial: partial saturation in compacted soils

As the guest editor of this themed issue of Géotechnique, it is my pleasure and honour to welcome you to the 16th edition of the series of biennial symposia in print, which is devoted to the topic of ‘partial saturation in compacted soils'. In line with tradition, this symposium in print will involve publication of two thematic issues of Géotechnique and will be accompanied by a full-day meeting to be held on 20 June 2011 at the Institution of Civil Engineers in London where all contributions will be briefly presented by the authors and discussed by the audience.

Since 1975, when the first symposium in print was published on the theme of ‘ground treatment by deep compaction', the symposia in print series has become a Géotechnique hallmark and has covered, over time, a variety of research areas and applications ranging from laboratory and field tests to ground improvement, from soft and stiff sedimentary clays to the thermal characteristics of the ground, from suction in unsaturated soils to risk and variability in geotechnical engineering. In all these areas, symposia in print have created momentum in theoretical and empirical research and have become milestone reference for future works. The present edition has the potential to follow in the footsteps of its predecessors as it publishes some of the most recent advances in the area of partially saturated soils with specific reference to the study of compacted fills, covering topics relevant to material characterisation, constitutive modelling, numerical analysis and application to earth structures.

The theme of this symposium blends together two areas of research that have already featured in previous Géotechnique symposia, albeit approached from different angles and with reference to different applications. These two areas relate to the effects of suction and partial saturation on soil behaviour (the symposium on ‘the influence of vegetation on the swelling and shrinkage of clays' in 1983 and the symposium on ‘suction in unsaturated soils' in 2003) and the study of compaction as a method of soil improvement (the symposium on ‘ground treatment by deep compaction' in 1975 and the symposium on ‘ground and soil improvement' in 2000). The recurring interest in these areas of research, together with the recent increase in the number of international conferences dedicated to unsaturated soils, is a clear indication of the continued relevance of the theme of the present symposium to the geotechnical profession.

The study of compacted soils dates back to the early 1930s when engineer Ralph Proctor, from the Los Angeles Bureau of Waterworks and Supplies in the USA, devised a laboratory test for determining the moisture content that enables achievement of the lowest porosity during field compaction of soil fills (Proctor, 1933). Proctor's investigation was motivated by the necessity of developing suitable construction protocols to maximise stability and impermeability of earth dams. In particular, he performed laboratory compaction tests on more than 200 different soils at different moisture contents by using similar compaction energy as that employed in the field during construction of earth dams. In this way, Proctor demonstrated for the first time the existence of what he defined the ‘optimum moisture content' for which the lowest porosity of the compacted fill is achieved. After the seminal work of Proctor in 1933, the study of compacted soils slowly evolved and the effects of capillarity and partial saturation in compacted fills have been progressively better understood. For example, it is now well-acknowledged that the decrease of dry density with decreasing moisture content below the optimum value is not due to a reduction in the ‘lubricant' action of water, as postulated by early researchers, but is rather the consequence of the formation of stabilising capillary menisci at inter-particle contacts which oppose inter-granular slippage during compaction.

This symposium in print has attracted great interest, with a total of 98 papers offered in response to the call issued in summer 2009. All submitted contributions have been subjected to peer-reviewing according to Géotechnique standards, with nine articles and one technical note being finally accepted for publication in this and the next issue of the journal. These ten contributions have been grouped in four sub-themes according to content. Three sub-themes correspond to classical stages of geotechnical research – ‘material characterisation', ‘experimental observation and modelling' and ‘application to engineering problems and case studies' – while the fourth sub-theme – ‘benchmarking of techniques and models' – originates from the work undertaken within the Marie Curie Research Training Network MUSE (Mechanics of Unsaturated Soils for Engineering).

The MUSE Network (MUSE, 2011), funded by the European Commission from 2004 to 2008, has involved a number of universities across Europe in a training programme of early-stage researchers in the area of unsaturated soil mechanics. One of the salient activities of the MUSE project has been the benchmarking of research techniques and models used by different universities, with the objective of contributing to the development of robust methods of analysis for unsaturated soils. The two benchmarking exercises presented in this symposium compare hydromechanical constitutive models and laboratory techniques for the measurement of the soil–water retention curve, respectively. Both benchmarks make use of easily accessible data and commercially available materials so that they can be readily repeated by researchers across the world.

The present issue of Géotechnique is the first of the two issues dedicated to the symposium in print. It contains five articles, two related to the sub-theme ‘benchmarking of techniques and models' (D'Onza et al., 2011; Tarantino et al., 2011) and three related to the sub-theme ‘experimental observation and modelling' (Alonso et al., 2011; Boyd & Sivakumar, 2011; Romero et al., 2011). The next issue of Géotechnique will also be entirely dedicated to this symposium in print and will contain the five contributions related to the sub-themes ‘material characterisation' and ‘application to engineering problems and case studies'.

The first article of this issue, by D'Onza et al. (2011), compares experimental results from a number of triaxial and oedometer tests with the corresponding predictions calculated by six universities using different hydromechanical constitutive models. The article highlights the sensitivity of predictions to model calibration, which is exemplified by the simulation of one constant water triaxial test where two teams using similar models calculate opposite trends of suction variation because of different calibration choices. In general, predictions from water-retention models show a greater degree of consistency compared to mechanical models. The predicted variation of volumetric strain during shearing registers the largest differences between teams with repercussions on the corresponding variation of degree of saturation due to the dependency of water retention on porosity.

The second article, by Tarantino et al. (2011), compares the water-retention curves measured on the same soil by eight different laboratories using different types of experimental equipment chosen from amongst the pressure plate, the axis-translation oedometer, the high-capacity tensiometer, the osmotic method and the dewpoint potentiometer. It is encouraging to observe that the measurements performed by different research teams appear remarkably consistent not only when the same type of equipment is used but also when different types of experimental techniques are employed. Only the pressure plate seems to overestimate noticeably suction at high values of degree of saturation – when occluded air is present – which is in agreement with previous findings.

The next two articles investigate the effects of the multi-structural fabric of compacted soils on water retention and mechanical behaviour. Both articles highlight the importance of taking into account the interaction between the different structural levels of fabric for an accurate prediction of the macroscopic behaviour of the soil. Most importantly, they emphasise the ‘living nature' of the compacted soil fabric which does evolve significantly during wetting–drying cycles, thus altering the interaction between the different structural levels and consequently changing the macroscopic material properties. Romero et al. (2011) employ a combination of mercury intrusion porosimetry and environmental scanning electron microscopy tests to characterise the evolution of both the pore network and the aggregate skeleton of compacted soils along hydromechanical paths, which is then used as the basis for the formulation of a water-retention model that has a clear physical interpretation. By following a similar approach, Alonso et al. (2011) use laboratory test results to formulate a hydromechanical model of soil behaviour which contemplates the existence of distinct water potentials inside the pores of the different structural levels. The macroscopic response of the soil during wetting/drying can therefore be interpreted in terms of a local (i.e. microscopic) transient phase leading to the achievement of hydraulic equilibrium across the different structural levels.

Boyd & Sivakumar (2011) close this issue with an article that investigates the evolution of confining stress during wetting of unsaturated clay fills under laterally restrained conditions. The authors present a programme of controlled wetting tests performed inside a double-walled triaxial cell on compacted kaolin samples subjected to constant axial load and zero lateral strains with measurement of confining stress. Results indicate that, during wetting, the lateral stress tends generally to increase following a form of variation that depends on the initial specific volume of the sample. In particular, the evolution of lateral stresses during wetting can be interpreted in terms of the changing position of the soil stress state relative to the yield locus.

Contributions to this symposium show that recent developments of experimental techniques and modelling tools have enhanced our understanding of unsaturated soil mechanics and have resulted in the formulation of conceptual material frameworks that are applicable to different categories of soils from high-plasticity clays to coarse-grained materials, rockfills and even rammed-earth structures in residential buildings. Nevertheless, the impact of such advances on engineering practice is yet to be seen and current construction of engineering fills remains based on the specification of the optimum water content and dry density measured from laboratory tests similar to those performed by Proctor back in 1933. Initiatives such as the present symposium can definitely contribute to closing the gap between research and practice by facilitating dissemination of our current knowledge of unsaturated soil mechanics. This knowledge has the potential to enhance the sustainability of current building practices by extending the use of locally sourced soils (often considered unsuitable for construction according to present standards) and, in general, to improve design of earth structures with consequent gains in safety and cost.

I wish you an enjoyable reading of this issue and hope that you will join us at the Institution of Civil Engineers in London on 20 June 2011 to discuss these articles together with their respective authors. Also, remember not to miss the next issue of Géotechnique where the five contributions relating to the sub-themes ‘material characterisation' and ‘application to engineering problems and case studies' will be published. See you then.