Professor Olivier Buzzi

The distribution of cracks strongly affects the hydro-mechanical behavior and strength of soil. In order to characterize the distribution of cracks in undisturbed expansive soil, X-ray computed tomography (X-ray CT) tests were carried out for one type of undisturbed expansive soil from Mile City, China. Seven undisturbed specimens retrieved from depths between 1 and 4.7 m were used to study the evolution of cracks along the drying path. The characteristics of the cracks were quantitatively studied by 3D parameters extracted from CT images by commercial software. The test results suggested that, with an increase in suction, the total volume of the cracks gradually increased, and one large crack formed due to the merging of sub-cracks before the shrinkage limit. When the suction was larger than the suction corresponding to the shrinkage limit, degradation of the cracks caused by the collapse of the soil structure was observed. Meanwhile, due to the development of sub-cracks, the crack anisotropy caused by the typically oriented primary cracks decreased along the drying path. Finally, a conceptual model of crack development for undisturbed Mile clay was proposed based on the evolution of crack volume, number, shape, and orientation.

Reliable shear strength determination of large in situ discontinuities is still a challenge faced by the rock mechanics field. This is principally due to the limited availability of surface roughness and morphology information of in situ discontinuities and the unresolved management of the ¿scale effect¿ phenomenon. Recently, a stochastic approach for predicting the shear strength of large-scale discontinuities was established, encompassing random field theory, a semi-analytical shear strength model, and a stochastic analysis framework. A key aspect of the new approach is the application at field scale, thereby minimising or bypassing the scale effect. The approach has been validated at laboratory scale and an initial large-scale deterministic-based validation showed promising results. However, to date, no large-scale experimental-based validation has been undertaken. This paper presents the first rigorous application of the employed semi-analytical shear strength model and the stochastic approach on a 2¿m-by-2¿m discontinuity surface, with comparison of prediction to experimental shear strength data. The shear strength model was found to generally produce peak and residual predictions within a ± 10% relative error range, with good agreement between predicted and observed damage areas. It was observed that, applying the stochastic approach to seed traces with gradient statistics equivalent to that of the surface, produced predictions that closely resemble the experimental results. Whereas, predicting shear strength from different seed traces results in more variability of predictions, with many falling within ± 20% of the experimental data. The predictions of residual shear strength tended to be more accurate than peak shear strength.

Soil properties, such as wetting collapse behaviour and permeability, are strongly correlated to the soil microstructure. To date, several techniques including mercury intrusion porosimetry (MIP), can be used to characterize the microstructure of soil, but all techniques have their own limitations. In this study, the features of mercury that penetrated and has been entrapped in the pore network of the specimens through MIP testing were investigated by X-Ray microtomography (X-µCT), in order to give an insight into the geometry of macropores and possible ink-bottle geometry. Two conditions of water content and density were selected for the compacted Maryland clay. The distribution and geometry features of mercury entrapped in the microstructure after MIP were characterized and pore size distributions were also reconstructed. The results suggest that, for the two conditions studied in this paper, macropores were evenly distributed within the specimens, and most of them with a non-spherical shape, and with aspect ratio (ratio between the maximum and minimum thickness along a given segment) smaller than three. Different dominant entrance pore size of macropore was obtained from MIP and X-µCT, due to the specific experimental protocol used in tests and the effect of ink-bottle geometry. Only the large pore bodies with high aspect ratio were imaged in X-µCT, due to the extrusion of mercury during the process of depressurization and subsequent sample preparation for X- µCT. But entire pore space was accessible in MIP. The difference in dominant entrance pore size was more significant for specimens with lower void ratio due to a more pronounced aspect ratio.

Soil collapse is a phenomenon triggered by wetting a loaded soil, the structure of which contains large pores. This type of soil response has been studied since the 1970s and models that can predict its occurrence and magnitude were proposed from the 1990s. In particular, the concept of loading collapse (LC) curves has been developed in the framework of unsaturated soil mechanics and it has been validated using low-reactivity soils. Several publications have highlighted that the microstructure of expansive soils evolves significantly during swelling, a phenomenon that may affect the location of the LC curve. With that perspective, some studies mention a need to shift the LC curves for experimental data and model predictions to agree. This paper brings new insight into the significance of microstructure for the collapse of a reactive soil. A series of tests was conducted to characterise the microstructure of soil specimens compacted under different initial conditions. Then, three pairs of wetting tests under constant load and under controlled suction were conducted in order to identify the zone of onset of collapse. It was found that the conventional LC curves cannot adequately predict the occurrence of collapse. In contrast, analysing the swelling response in the light of microstructural characterisation led to the conclusion that, for Maryland clay, occurrence of collapse can be tracked by the evolution of the void ratio associated to the macro porosity during wetting.

Estimating the shear strength of large in situ rock discontinuities is non-trivial because of the multiscale nature of roughness and the fact that only a very limited extent of discontinuity morphology is visible along traces. Recently, a novel stochastic method based on random field theory and Monte Carlo semi-analytical estimation of shear strength was proposed. The method was validated at laboratory scale and its application to one large natural surface showed that it has the potential to bypass the scale effect. However, a critical issue was reported by the authors of the study: it was found that the random field approach used could not generate the correct distribution of gradients on the simulated surfaces, which translates into an inaccurate prediction of shear strength. The authors had to manually adjust the input of the model to achieve a satisfactory prediction. This paper presents a new multiscale approach using random field theory, which now allows a rigorous generation of large synthetic 3D surfaces with controlled distribution of asperity heights and gradients from the profile of a 2D fracture trace (as might be visible in a rock face) referred to as a seed trace. Each seed trace is first decomposed into three daughter profiles corresponding to three levels of roughness. The statistics of each daughter profile form the input of the random field model at each scale level, allowing synthetic daughter surfaces to be created at each scale level. The synthetic daughter surfaces are then superimposed to obtain a composite rough surface comprising three levels of roughness. The approach was successfully validated with 25 input seed traces, coming from 5 different natural surfaces. This rigorous multiscale approach is essential to apply the stochastic method that was recently developed to predict the shear strength of large in situ discontinuities.

Rockfall is a natural hazard that needs to be rigorously managed along all the major road and railways transport networks by identifying the most appropriate mitigation measures. There has been significant progress in rockfall modelling and rockfall protection systems in recent years but there remains one aspect that is not very well understood and quite challenging to account for in the design of rockfall protection structures, namely the fragmentation of falling blocks upon impact. Rocks often break up upon impact, which leads to a change in size, shape and energy of falling blocks, parameters that affect the design of the protective structures. Before being able to incorporate fragmentation into predictive trajectory models, it is required to better understand the fragmentation process and its likely outcome (number, volume of fragments and their trajectories). To that aim, an innovative experimental setup was developed at the University of Newcastle (Australia) to study rock fragmentation upon impact. The setup was designed to perform controlled vertical drop tests and record the following impact parameters: impact force, impulse, impact duration, velocities (of the block before impact and its fragment after impact) and all components of energy, pre and post impact. Six views (four high-speed cameras and two mirrors) are used for an accurate reconstruction of the 3D trajectory of blocks and fragments, in translation and rotation. This paper presents the validation of the setup via two series of drop tests using mortar spheres. Attention was focused on the evaluation of impact force and impulse from load cells placed under the impacted surface, tracking of translational and rotational velocity and the computation of total kinetic energy (before and after impact) and all components of energy dissipation. The results confirm that the experimental setup and the approach developed can be used to obtain impact force, impulse and to compute the energy balance during the impact and fragmentation and conduct advanced fragmentation testing.

Fragmentation of rocks upon impact during rockfall is a phenomenon that is poorly understood, scarcely researched and difficult to predict. However, to adequately predict the outcomes of rockfall events, it is essential to know whether a given block is likely to fragment given the impact conditions and what will be the outcome of the fragmentation process; that is, the number, size and trajectory of fragments. To date, there is no model or data that can be used to fully answer these questions. This paper presents the first theoretical model that can predict the fragmentation survival probability of brittle spherical blocks upon dynamic impact (i.e. drop tests) from the statistical distribution of material properties, determined from a range of standard quasi-static tests. Considering that survival probabilities tend to follow a Weibull distribution, the model focuses on predicting the two Weibull parameters, commonly known as the shape parameter (m) and the scale parameter (here, the critical kinetic energy). The model is based on theoretically-derived conversion factors used to turn the critical work required to fail disc samples in quasi-static indirect tension into the critical kinetic energy to cause failure of spheres at impact in drop tests. The mechanistic conversion factors specifically account for the shape and size of the specimens tested and the increase of strength under dynamic loading (strain rate effect). Three series of drop tests were conducted (on spheres of three different diameters) and complemented by extensive material characterisation testing in order to validate the novel predictive model. The variability of material properties was characterised, and it was found that the material strength found by the characterisation tests generally follows a Weibull form, but the survival probability distribution of the drop tests seems to be linear. The predicted conversion factors were first compared against their experimental counterparts before validating the prediction of survival probability of the spheres upon dynamic impact (in drop tests). It was found that it is possible to predict the survival probability of artificial rock of three different diameters (50 mm, 75 mm, 100 mm) and two different strengths upon impact solely from the statistical information coming from Brazilian tests and with an average relative error of less than 9%.

This paper presents a comprehensive and consistent investigation on three technical aspects associated with mercury intrusion porosimetry (MIP) applied to clayey soils: The effect of intrusion rate, the effect of the freeze-drying method and the influence of sample size. Although a limited number of publications provide recommendations about the MIP technique, the aforementioned effects were largely left unaddressed. This study was performed on four different soils, having different microstructures. In order to obtain a quasi-static condition at the applied pressure, the smallest intrusion rate was suggested to be used in the MIP tests. It was found that the freezing technique does modify the pore size distribution, due to expansion of water on freezing. The more water in the soil specimen, the more pronounced the effect. The specimen size also matters for similar reasons. The outcome of this study provides a technical basis for MIP on clayey soils.

In situ measurements of soil suction and water content in deep soil layers still represent an experimental challenge. Mostly developed within agriculture-related disciplines, field techniques for the identification of soil retention behaviour have been so far employed in the geotechnical context to monitor shallow landslides and seasonal volume changes beneath shallow foundations, within the most superficial ground strata. In this paper, a novel installation technique is presented, discussed and assessed, which allows extension of the use of commercially available low-cost and low-maintenance instruments to characterise deep soil layers. Multi-depth installations have been successfully carried out using two different sensor types to measure the soil suction and water content up to 7 m from the soil surface. Preliminary laboratory investigations were also shown to provide a reasonable benchmark to the field data. The results of this study offer a convenient starting point to accommodate important geotechnical works such as river and road embankments in the traditional monitoring of unsaturated soil variables.

This study aims at providing quality experimental data on the effects of temperature on tensile strength and small strain shear stiffness of two Australian mudstones. The objective is to provide multiscale data in view of developing a numerical model that can capture and simulate the complex multiphysics of underground coal fire propagation. Two mudstones were collected in the Hunter Valley, close to a known underground coal fire, referred to as ¿Burning Mountain.¿ The rock specimens were heated to a range of temperatures (maximum of 900¿°C) for 24¿h, and the materials were comprehensively characterized by X-ray diffraction, thermal gravimetric analyses, optical microscopy and scanning electron microscopy. In addition, mercury intrusion porosimetry was used in order to track changes in pore size distribution with temperature. Investigations at microscale were complemented by testing at the macroscale. In particular, the paper focuses on the evolution of the tensile strength and small strain shear stiffness as the materials are subjected to heating treatment. Results show that both parameters evolve in a non-monotonic manner with temperature. The observed mechanical responses are fully explained and corroborated by microstructural observations.

This paper presents a comprehensive experimental investigation on time-dependent swelling behaviour at both macroscale and microscale of a natural Australian expansive soil in compacted state. A number of one-dimensional swelling tests under different vertical pressures, different initial void ratios and initial water contents were performed. The characterization at macroscale was complemented by extensive microstructural investigations through mercury intrusion porosimetry and scanning electron microscope observation on both as-compacted and swollen specimens. The results were discussed at two different scales within a framework of double-porosity, which was finalized by linking the macrostructural¿microstructural strains ratio with secondary swelling/compression coefficients. The multi-scale correlation appears to be largely independent of the specimen initial conditions. The study showed that the secondary swelling and primary swelling are governed by the same factors and that secondary swelling takes place mainly in macropores, of which the change magnitude depends on the level of confinement applied. The microstructural investigations show that swelling is accompanied by significant microfabric changes.

The objective of this study is to improve the understanding of load transfer mechanism of Geogrid-Reinforced Pile-Supported Embankments (GRPS) via a new 3D analytical approach and comprehensive field tests. A full-scale embankment was built over a silty clay of medium compressibility as a part of the Liuzhou-to-Nanning High-speed Railway (LNHR) in China. Six sections of the embankment have been heavily instrumented producing comprehensive data of high quality. Field measurements evidence the existence of soil arching, membrane contribution and ground reaction, phenomena that are all contributing to load transfer mechanism. The new 3D analytical arching model accounts for a triangular arrangement of piles and, unlike existing methods, accounts for all relevant components of load transfer mechanisms. In addition, two key parameters were introduced in the model: an elastoplastic state parameter of soil arching (a) and a coefficient of equivalent uniform stress (ß). The former was used to satisfy the load equilibrium in case of partial arching while the latter was adopted to allow possible nonuniform vertical stress acting on the ground surface. The so-called ground reaction method was incorporated in an innovative manner to take into account the reactive support of the subsoil beneath geogrid-reinforced layer when estimating the tension development in the geogrid. Finally, the performance of the proposed model was assessed against several existing models and field measurements. Results showed that the new model presented herein outperforms existing models and satisfactorily predicts both the pile efficiency and tension development within the geogrid.

This study falls in the context of underground coal fires where burning coal can elevate the temperature of a rock mass in excess of 1000°. The objective of the research is to experimentally characterize the change in mechanical behaviour, mineralogy and microstructural texture of two sedimentary rocks when subjected to temperatures up to 1200¿°C for 24¿h. Specimens of local sandstone and mudstone were comprehensively characterized by X-ray diffraction and thermal-gravimetric analysis. These analyses were complemented by optical microscopy and scanning electron microscopy on polished thin sections. In addition, pore size distributions of these heated rocks were inferred by means of mercury intrusion porosimetry. These results were extended to an estimation of the intrinsic permeability using the Katz¿Thompson model. Investigations at micro scale were followed by mechanical testing (both unconfined and confined compression tests) on cylindrical specimens of heated rocks. Results show that the unconfined compressive strength (UCS) of both rock types tends to increase when the temperatures increases up to 900¿°C, beyond which the UCS tends to slightly decrease. As for the permeability, a clear increase in intrinsic permeability was observed for both rocks. The macroscopic behaviour was found to be fully consistent with the changes observed at micro scale.

In rockfall science, the bullet effect refers to the perforation of a rockfall mesh by a small block traveling at high speed. To date, there is still no comprehensive experimental data set investigating the underlying mechanisms of such effect. The bullet effect illustrates the fact that the capacity of a rockfall mesh depends on the size and speed of the impacting block. This paper presents the results of an experimental study on the effect of block size and mesh geometry (aperture and wire diameter) on the mesh performance. The results clearly show that the amount of energy required to perforate the mesh drops as the blocks get smaller. They also suggest that the mesh performance reaches a maximum and reduces to zero when the mesh cannot sustain the static load imposed by very large blocks. The outcome of the first series validates an analytical model for mesh perforation, making it the first simple model capturing the bullet effect. A second series of tests focused on the effect of mesh geometry and it was found that decreasing the mesh aperture by 19¿% improves the performance by 50¿% while only an extra 30¿% could be gained by increasing the wire diameter by 33¿%. The outcomes of the second series were used to discuss and redefine a dimensionless geometrical parameter G* and to validate a simple power type equation relating the mesh characteristics and the mesh performance.

This paper presents a new coating material [hand-spray plaster (HSP)] used to measure the bulk volume of soil specimens having irregular shapes. The new method has been validated against two benchmark methods, namely, the wax and plastic bag methods, by conducting swelling and shrinkage tests on intact Maryland clay. The results show that the new method yields similar values of volume but with much reduced data scattering. The HSP method is also far easier to use than the other two methods. Finally, the stiffness of the coating has been measured and its restraining effect has been found to be negligible. Some of the benefits of using the HSP method are: (1) limited fluid retention by the specimen postimmersion for volume measurement, (2) reduced water-intake rate with elimination of cracking upon swelling caused by high-suction gradients, (3) absence of the restraining effect on specimens upon swelling, and (4) accurate determination of the swelling and shrinkage curves with only one specimen per curve. Copyright by ASTM Int'l (all rights reserved).

The high-capacity tensiometer developed by Ridley and Burland in 1993 is a milestone in experimental unsaturated soil mechanics. This device, which relies on development of tension in an enclosed water body, permits direct measurement of negative water potential. Many tensiometers have been built since 1993, all being based on the same principle although the design may differ slightly from the original. In particular, the common characteristic is a very small water reservoir that is believed to be the location of bubble nucleation, a phenomenon referred to as cavitation that impedes the sensor from functioning properly. Many have studied cavitation and different explanations or cavitation mechanisms have been proposed. However, all the considerations put forward were derived without being able to capture what happened inside the water reservoir at cavitation. This is now achieved with the new tensiometer specifically designed to "see" inside the water reservoir during suction measurement. For the first time, cavitation has been captured via high-speed photography and the mechanisms of cavitation can be explained using physical evidence. The first outcome is that it is possible for a high-capacity tensiometer to function to its full range with a large water reservoir. Then, the analysis of high-speed photographs reveals that the bubbles triggering cavitation are nucleated in the ceramic and make their way to the water reservoir. Cavitation occurs only when the air phase reaches the water reservoir.

From a consideration of the concepts of geological weathering and structure, it can be expected that rockfall hazards should be characteristically different in different geological environments. This paper tests this idea by looking at the geometric characteristics of rock fragments formed on natural slopes in four different geological environments in Eastern Australia, where rockfall phenomena are often characterised by rolling of pre-detached debris. By measuring the three principle dimensions and making a systematic assessment of the shape characteristics of samples of rock debris in significant geological environments, it is found that the distributions of size and shape for the surface debris are statistically different. From the results, it is shown that the size and shape of debris is directly controlled by the rock type, its weathering characteristics and the structure of the parent rock mass. The severity of rockfall hazards is shown to be relatively lower in areas of Tertiary basalt, as the size of rolling fragments is limited by closely spaced fracturing inherited from its formation and the tendency to deteriorate further as it weathers deeply and rapidly. It is also lower in areas of Palaeozoic volcanics, since these tend to produce relatively angular fragments with higher proportions of fragments that are inherently more resistant to rolling. By contrast, thickly bedded sandstones form larger blocks with a larger proportion of shapes that are more prone to rolling. The size distribution of fragments is shown to be well approximated by a log-normal statistical distribution, and using the data provided in this study, it is possible to generate the size and shape data needed to undertake a stochastic assessment of rockfall trajectories in different geological environments. © 2013.

Hydromechanical compression tests have been performed on rock-mortar interfaces representing the contact between a host rock and a concrete bulkhead within an underground nuclear waste repository. The rock used in the tests is a Toarcian argillite. Most published studies concerning rock-concrete interfaces involve concrete-concrete contacts in which rock replicas are used instead of real rock samples. As a result, the effect of rock features, such as bedding planes and changes of the rock interface zone, on the hydromechanical behaviour of the interface cannot be investigated. The tests discussed in this paper demonstrate that the mechanical response is not affected when changing the parameters of the samples even not mismatching both walls. On the contrary, an initial monitored lateral displacement modifies the hydromechanical behaviour by limiting the interface ability to be hydraulically closed. The latter ability has been quantified by a simple evolution law and the difference of behaviour between the two kinds of samples has been confirmed. The analysis led to determine the hydraulic aperture shows that the values obtained are much lower to the classical values available in the literature. Finally, the application of the experimental results to the confinement by a bulkhead showed the localization of the flow within the interface. © 2006 Elsevier Ltd. All rights reserved.

Variation of the shear strength of rock joints due to cyclic loadings is studied in the present paper. Identical joint surfaces were prepared using a developed moulding method with special mortar and shear tests were performed on these samples under both static and cyclic loading conditions. Different levels of shear displacement were applied on the samples to study joint behaviour before and during considerable relative shear displacement. It was found that the shear strength of joints is related to rate of displacement (shearing velocity), number of loading cycles and stress amplitude. Finally, based on the experimental results, mathematical models were developed for evaluation of shear strength in cyclic loading conditions. © 2003 Elsevier Ltd. All rights reserved.

Reliable prediction of shear strength for engineering scale in-situ discontinuities is still problematic. There is currently no consensus or a satisfactory method to estimate shear strength, account for surface variability and manage the effects of the recognised ¿scale effect¿ phenomenon. Shear behaviour of a discontinuity rock mass greatly depends upon the rock joint roughness, which is generally concealed within the rock mass. As such, the limited amount of accessible surface information complicates even further the exercise of shear strength prediction. Often, small size specimens (e.g. rock core) are recovered to conduct experimental tests and predictions can be made from analysing traces. In this paper, three different shear strength prediction approaches were followed and their relative performances were compared. The results reveal that the prediction of shear strength from recovered sub samples can be significantly variable. The application of Barton¿s empirical model to four selected traces, produced a large scattering of results, with the prediction highly dependent on the trace used. Conversely, the application of Casagrande and co-workers¿ stochastic approach on the same traces, produced the least scatter and provides statistical data to quantify variability and uncertainty.

This paper presents an extension of the local second gradient model to multiphasic materials (solid particles, air, water) and including the cavitation phenomenon. This new development was made in order to model the response of saturated dilatant materials under deviatoric stress and undrained conditions and possibly, in future, the behaviour of unsaturated soils.

The Hunter Expressway is a $1.7 billion road project in the Hunter Valley, NSW, Australia to provide a new east-west connection between Newcastle and the Lower Hunter. A large part of the materials encountered on site are expansive tuffaceous soils that would normally be discarded. However, because of increasing financial, environmental and space constraints, it is required to make use of these swelling materials in embankments. These materials include highly weathered claystone containing about 20% in mass of smectite. The highly plastic claystone is present in layers alternating with layers of nonplastic coal. Hence, when excavating the material, the claystone is mixed with the coal and it becomes relevant to consider the swelling potential of the mixture instead of that of the claystone alone. Some typical coal and claystone materials used in embankments were characterized by compaction curves and Atterburg limits. Then, a series of swelling tests were conducted on compacted specimens, of mixtures of coal and claystone, prepared in the laboratory with comparable initial conditions (suction and void ratio) so that any change in swelling potential could be solely attributed to the coal content. The results show that the addition of the coal has an effect in reducing the swelling potential of the claystone beyond a simple dilution effect. Some type of swell inhibition is implied. The addition of coal also significantly reduces the time taken for final swell strain to be reached. © 2014 Taylor & Francis Group, London.

In Australia there has been a significant increase in areas affected by dryland salinity over the 200 years since European settlement. As a result of land use and climatic changes this trend is expected to continue, with areas at high risk of dryland salinity predicted to more than double over the next 40 years. This presents a significant risk to road asset managers as concentration of salts within road pavements has been shown to cause damage to pavements and surfacings. The present study explores the impact of dryland salinity on road pavements by assessing the effect of naturally occurring salts on the tensile and shear strength of a sealed granular pavement. Tensile and shear strength testing was conducted on a granular pavement material at a range of salt concentrations. This testing confirmed that the presence of natural salts can have a positive effect on bitumen-sealed granular pavements in contrast to the negative impact salt crystallisation can have on bitumen surfacings. While in solution, salts were found to have a very small positive effect on tensile and shear strength, which is likely related to clay chemistry interactions. Upon crystallisation, salts were observed to form cubic bonds within the pores of the granular material which led to a significant increase in tensile strength. However, at interface between the surfacing and the granular material whisker-shaped crystals were observed to grow from the bitumen. The results suggest that increases in dryland salinity, while presenting a risk to the performance of road surfacings, may actually increase the tensile and shear strength of road pavement materials.

Results from laboratory experiments on natural fracture samples from the medium (roughly 20 m by 20 m by 20 m) sized superficial limestone bedrock reservoir, namely Coaraze natural site, France, is presented here. The BCR3D (3D direct shear box) with its hydraulic sectorized device, and its portable laser beam from 3S (Soils, Solids, Structures) Laboratory, France, was used. The interpretation of the cyclic hydro mechanical compression tests shows an apparent residual intrinsic transmissivity of Tr = 3 × 10-15 m3, the evolution of the intrinsic transmissivity versus the normal stress is anisotropic. The comparison of the fracture void spaces with hydro mechanical factors show that: the back computed method to obtain the void spaces is verified for low normal stress values, and the anisotropy of the outflow is explained by the modeled evolution of the void spaces with the normal stress. The Dunat (1996) law that depends on the contact surface could integrate a spatial distribution and damage factor and/or a plastic parameter. © 2006 Taylor & Francis Group.

A wide study of the medium (roughly 20m by 20 m by 20 m) sized superficial sandstone bedrock reservoir, namely Coaraze natural site, France, is planned. Laboratory and site experiments will be compared and the data set will be used for simulations with 3DEC/3Flow ITASCA code implemented with later models. A part of the laboratory experiments is presented here. The BCR3D (3D direct shear box) with its hydraulic sectorized device, and its portable laser beam from 3S (Soil, Structure, Solide) Laboratory, France, is used. The interpretation of the rock matrix characterization tests, of the cyclic mechanical compression test, and the cyclic hydro mechanical compression tests show that the normal stiffness is stress dependent, the initial normal stiffness value, kn0 = 122 MPa/mm, dependent on the initial fitting of the two joint walls, and the maximum normal stiffness value, knimax = 164.3 MPa/m, is reached at dmax =0.5 mm of normal displacement. © 2005 Taylor & Francis Group.