Winding its way from Morgex to Entrèves, the second section of the Aosta-Mont Blanc motorway runs for most of its 17km length underground, emerging briefly onto four short viaducts and a few embankments. Three twin-tube tunnels – Morgex, Pré Saint Didier and Dolonne – together run 12km, with a maximum ruling gradient of 3.2%, the result of a commitment by the design engineers to respect the environment. Siting the route underground also reduces maintenance problems, which are particularly severe in winter because of the very low temperatures and heavy snowfalls.
The adjacent tubes of the Dolonne tunnel, each with a cross section of 100m², run along the right bank of the Dora Baltea river for 2,900m through extremely heterogeneous lithological formations with overburdens varying between 10m and 340m.
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
In drawing up a final detailed design, including connecting passageways, shelter recesses and vehicle stopping points, the motorway designer Spea Ingegneria Europea (SIE) had to comply with the recently introduced Merloni Law promoting transparency and reliability in contracting of public works, especially on forecasts of construction times and costs.
SIE brought in consulting engineer Rocksoil of Milan to revise the final design using the “Analysis of Controlled Deformation in Rocks and Soils” (ADECO-RS) approach, which has proven its reliability in predicting construction times and costs even under very difficult conditions (Lunardi, 2000).
Survey phase
The area through which the Dolonne tunnel passes is in the upper part of the Valle D’Aosta between Courmayeur and Dolonne at the point where the valley divides into the Val Ferret towards the north east, and the Val Veny towards the south west. It is one of the most geologically complex parts of the Alpine range, lying on the border zone between the Pennine systems which prevail in the central and western Alps and the Elvetico-Delfinese system that extends across the outer sector of the western Alps. The two systems are separated by the Pennine Front, a very large thrust fault of a regional character which in the area in question runs north east-south west with the dip to the south east.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataThe front runs roughly along the line of the bed of the River Dolonne. As a result the ground through which the tunnel passes is divided into two parts with very different characteristics:
The geological survey to reconstruct the lithostratigraphy, the tectonic structure and hydrogeology of the area in detail, involved continuous core drilling and RQD evaluations on the samples, SPT tests, seismic refraction, and piezometric and permeability surveys. It also included a study of aerial photographs and numerous field investigations. It found:
Study of the geomechanics and the geological structure was carried out to characterise the materials and describe the main characteristics of the discontinuities (type, attitude, spacing, shape, opening, roughness, filling, strength, water conditions). A summary of the geotechnical and geomechanical characteristics of the different types of ground is given in Table 1.
The hydrogeology is characterised by a number of springs of varying size and quality scattered above the tunnel alignment along the River Dolonne valleys the right bank of the River Dora Baltea. Given the structural characteristics and permeability of the ground, a constant presence of water is probable during excavation. In the Helvetic area this should just show as small drips of water, but in the detritus ground at the portal on the Mont Blanc side and in the lowermost Pennine system there could be diffused and concentrated inflow.
Diagnosis phase
In the diagnosis phase the geological, geomechanical and hydrogeological information and predictions made using simple analytical methods of calculation (characteristic lines) were used to divide the tunnel alignment into sections with uniform stress-strain behaviour determining the stability of the face. These were:
ategory A comprised those sections of tunnel in which mathematical calculations predicted that:
This is the case of the granite and porphyroid sections, which have excellent strength characteristics compared with the stresses brought into play by excavation under the overburden in question.
Category B included all those sections of tunnel where mathematical calculation predicted that:
This category was found in the sections to be driven through Ultra-Helvetic ground (the schists and clayey schists of Aaeliano and of the Upper Lias, metamorphic crystalline limestones) as long as the overburdens and the hydrogeological conditions permit an arch effect to be created naturally.
Category C included the sections of tunnel where:
These conditions will be found at the portals and along most of the section on the Aosta side under shallow overburdens where the only way an arch effect can be created is artificially. The predictions made for each tube of the tunnel have been summarised in Figure 1. From the diagnosis it was clear that, apart from the initial sections at the portals, the tunnel could be divided from a stress-strain viewpoint into two sections characterised by very different behaviour: generally stable to the north of the River Dolonne and generally unstable, with the additional problem of water under pressure to the south of the river.
Therapy phase
After formulating reliable predictions of the stress-strain behaviour of the ground as a result of excavation, appropriate stabilisation techniques were chosen to control, contain or anticipate and eliminate deformation for each section of tunnel with uniform deformation behaviour.
The guiding principles on which design of the tunnel section types was based are:
The tunnel section types designed were:
Type B2 involved ground improvement of the advance core using fibreglass structural elements and drainage in advance where necessary, a preliminary lining of 200mm-thick fibre-reinforced sprayed concrete reinforced with double steel ribs, a layer of waterproofing consisting of non-woven fabric and PVC, a 600mm-thick final lining in concrete closed by a 700mm tunnel invert cast at a distance of not less than 1.5 tunnel diameters from the face;
Precise specifications were made in terms of systems, advance rates and procedures including admissible extrusion and convergence values expected for each geological formation depending on the size of the overburden along the route. Also carefully specified were all factors required by the design engineer in deciding measures for preconfinement and confinement of the cavity and for modulating work during tunnel advance to adapt to the actual geomechanical conditions.
These detailed specifications also made it possible to adopt a quality assurance system conforming to ISO 9000 in the construction phase.
Drainage tunnel
To deal with the expected high levels of flow in the area close to the bed of the Dolonne River on the Aosta side, the design specified a drainage tunnel, 200m long and 50m² in cross section be driven in advance of the right hand tube starting at chainage 3+650. This valved drainage tunnel is to reduce water heads which would otherwise reach 700mm to 800mm above the crown of the tunnel. In the long term it will ensure that groundwater can be captured and channelled back to the aquifers it came from.
Statics calculations
The statics and deformation behaviour of tunnels, both in the construction phases and the final service phase were analysed and verified by calculations on 3D and 2D models in the elastic-plastic range.
For grounds that respond to excavation with stone type behaviour (stable face or face stable in the short term) an important part of the analysis was focused on gravitational instability (wedges of rock isolated from the network of discontinuities) using the software package UNWEDGE 2.3 developed by the Rock Engineering Group of the University of Toronto. For grounds that respond to excavation with cohasive or loose soil type behaviour (face stable in the short term or face unstable) the investigation was performed using finite differences (FLAC 3.3) or finite element (ADINA 6.1.4) analysis. The models were developed to simulate the behaviour of the tunnel-ground system at different phases of tunnel advance as realistically as possible with particular attention paid (in the 3D models) to the effect of preconfinement treatment of the cavity and ground reinforcement in the advance core, to deformation values to be expected during excavation and to stresses on the preliminary and final linings.
Monitoring phase
The adequacy of design hypotheses was verified by geostructural mapping of the face and monitoring of the stress-strain behaviour of the face and the cavity, observed during construction at each of the stages and sequences specified in the design.
To achieve this, a complete monitoring system was devised that included as many as 145 systematic and 10 main stations for each tube.
The systematic monitoring stations (set at 50m intervals in category A sections of tunnel, 27m intervals in category B sections and every 9m in category C sections) involved:
The main monitoring stations (placed in critical zones and at least one for each lithological section) were designed above all to measure probable stress-strain interactions between the two tubes and changes in the stress state of final linings. In addition to monitoring performed by the systematic stations they also included:
Given the considerable heterogeneity of the ground, systematic sub-horizontal core drilling (either continuous or with destruction of the core) was performed, 30m to 40m long, in advance of the face, on the Aosta side to obtain direct advance knowledge of the material to be excavated in zones with poor geomechanical conditions.
Contracts
Construction of the tunnel was awarded on the basis of the final design specifications as an all-in, lump-sum contract in which the contractor accepts all types of risks including geological risk. Work started a few months ago from both portals and has advanced hundreds of metres.
Related Files
Figure 1
