Tunnel de base du Mont Cenis. Présentation de la coupe géologique de référence (côté France). Apport des reconnaissances à la compréhension de la structuration géologique le long du projet de tunnel

Extended abstract

Introduction

The long (57.5 km) transalpine base tunnel project (“Tunnel du Mont Cenis”) on the new Lyon-Turin line provided an opportunity to carry out a large volume of investigations throughout the linear and over long years (more than 65 km of deep drilling from 1989 to 2007, 4 large multi-kilometer descending galleries, ...). These investigations therefore provide elements for a better understanding of the organization of the upper crustal part (0-1000 m) of a "central" sector of the western Alps chain. This article aims to present the French part of the geological section of the project essentially from a geometric (structural) point of view and by sector, after the investigations.

The main excavation work for the base tunnel will begin in the early 2020s. We are therefore at a turning point in the construction of this large tunnel, which seems to us to be a good opportunity to present a geological section of the French part of the future tunnel.

Geological framework and initial state of knowledge

The tunnel project mainly crosses the internal part of the Alps in a general direction roughly East-West (Figure 1). From France to Italy (W-E sense), the tunnel starts near Saint-Jean-de-Maurienne in the Ultradauphinois Zone and then crosses the Penninic Front, the Subbriançonnais Zone, the Houiller Front, the Briançonnais Houiller Zone, the internal Briançonnais Zone (Vanoise - Ambin) and the Gypsum Nappe (Figure 1); it ends in Italy in the Liguro-Piemontese ophiolitic Zone.

The alpine region crossed by the tunnel project was the subject of numerous studies during the 20th century, which made it possible to establish a solid lithostratigraphy of the region, to delimit the large paleogeographic and structural (or tectonic) zones and to describe the general or detailed structure on the surface. This geological knowledge has been disseminated in the form of numerous theses and scientific articles and is widely recorded in the 1: 50,000 geological map of France and associated explainatory notes.

Evolution of knowledge along the base tunnel

The work prior to the reconnaissance for the base tunnel was based on surface data and not from deep drillings ; the subsurface interpretation made for example in the form of a geological section (Figure 2) are poorly informed and sometimes very hypothetical in depth in comparison with the sections drawn from reconnaissance studies for the base tunnel.

After a general presentation of the changes in the interpretation of the geological structure along the base tunnel as the reconnaissance have progressed (Figure 3), this article describes in more detail the geology of each sector; this one is summarized hereafter.

Through the Ultradauphinois Zone

The Ultradauphinois Zone consists mainly of an Eocene flysch series which presents a regular dip towards the East (Figure 4). This dipping is very probably associated with regular folding, but the fold hinges are very rarely observed. Normal NW dip faults are clearly visible in the landscape and on a satellite image (Photo 1) but are not identified along the alignment.

Several sedimentary sequences are distinguished and dispatched in several flysch formations. The TELT surveys by drilling and detailed complementary field studies made possible to propose a refined lithostratigraphy of the flyschs (Figure 5, see also Photo 2). Large olistolites of Cretaceous age reported in the "schisto-sandstone flysch", they are surrounded by a chaotic facies of "block schist" (or wildflysch) type, the whole constituting an olistostrome (Serre, 1983).

Through the Penninic Front and the evaporites sole associated

The Penninic Front represents a major tectonic contact in the Alpine chain, which marks the separation between the external (to the West) and internal (to the East) domains of the Alps (Figure 1 and Photo 3). This initially thrusting front now reactivated in normal fault is marked on the surface over a plurihectometric thickness by the continuous presence of a megabreccia made of a matrix of gypsum and clastic lenses and blocks of dolomite (+/- shales). The ductile nature of the gypsum matrix (anhydrite in depth) allowed the operation of thrusting as is usually the case in the Alps. No investigations survey has completely intersected the gypsum front in depth and its thickness at the level of the tunnel therefore remains imprecise. The cargneules do not seem to have developed at this front level below the tunnel.

Through the Subbriançonnais Zone

The narrow Subbriançonnais Zone is made up of a stack of Mesozoic layers, which underline a marked plicative structure clearly visible in the landscape (Photo 4, Figure 6); these folds are classically interpreted as a succession of anticlines and synclines overturned towards the West. A different interpretation has been proposed (Ceriani, 2001) but it did not convince us and we stayed with the classic interpretation with a few adjustments. Note, however, that no boreholes crosscut the heart of the Subbriançonnais Zone and thus the precise structural interpretation in depth remains subject to caution.

Through the Houiller tectonic Front

The Front Houiller marks the western boundary of the overlapping Houiller terrains to the West. We also include in this mega-structure the contiguous (supra) hectometric thick "Gypsum Zone" (Figure 6). Like the Penninic Front, this Gypsum Zone consists mainly of surface gypsum and deep anhydrite and contains clasts of different sizes, mainly dolomite and limestone. Two very long TELT core holes intersected the Gypsum Zone. Within the mass of anhydrites, levels of “cargneules” are intersected by drilling F81 in its upper part. More in depth, and in particular along the Saint-Martin-de-La-Porte descending gallery, no cargneules was crossed through the “Gypsum Zone” of the Front Houiller.

Through the Briançonnais Houiller Zone

The Houiller Zone covers a width of around 15 km; it consists mainly of monotonous alternations of sandstones and pelites / shales deposited during the Upper Carboniferous in a fluvial and lacustrine context. Coal lenses are sometimes present in quite large proportions. Dolerite veins intersect the series but have not been spotted along the course of the base tunnel.

Four units are distinguished within the Houiller Zone. These are from West to East (Figure 7) : the Encombres, Bréquin-Orelle, La Praz and Fourneaux units. The stratigraphic relationships between the units are still unclear.

The Encombres, Bréquin-Orelle and Fourneaux units correspond to the “Houiller productif” big unit identified by previous authors and all contain coal intercalations. The Encombres Unit has been individualized mainly because of its large intimate deformation and very significant convergence phenomena that were observed during the digging of the first descending gallery of Saint-Martin-de-La-Porte. The Unit of La Praz corresponds to the “Houiller stérile” of the authors. Of small extent along the tunnel, it consists mainly of fine coarse sandstones and minor black pelitic schists.

The entire Briançonnaise Houiller Zone is regularly folded. The folds, from decimetric to hectometric scale, are visible on outcrops or in the landscape (Photo 5), they are mostly associated with a synschistose regional deformation. The precise deciphering of the global structure of the Houiller zone is quite difficult due to the absence of continuous well-identified strata. In addition, the detailed structuration is often very complex in the Houiller Zone.

Through the internal Briançonnais Zone

The internal Briançonnais Zone corresponds to the largest linear section of the French part of the base tunnel. We distinguish the Sapey-Modane-Avrieux area in the West and the Ambin massif in the East (Figure 8). Between the two, the tunnel intercepts the Gypsum Nappe and also a discrete unit newly identified and considered to represent the Piemontais Zone. The Schistes Lustrés Nappe is superimposed on these units at a high position above the tunnel and is not intersected by it.

Sapey-Modane-Avrieux sector (SMA)

The terrains of the Sapey-Modane-Avrieux sector (SMA) are surmounted to the west by the Houiller terrains in high position (Figure 8); they begin with an orthogneiss slice (Gneiss du Sapey) which follows the reverse contact with the Houiller series. The gneisses have a mylonitic texture over several decameters along the contact, due to the early overlapping towards the West of the gneisses on the Houiller Zone.

These gneisses, the origin and age of which have been debated for a long time now appear clearly as an ante-alpine basement slice of the Briançonnais. The prototype of the Sapey orthogneiss is dated to 452 +/- 5 Ma, that is to say to the Upper Ordovician (Bertrand et al., 2000).

The structure of the SMA area is complex ; several deep boreholes, new detailed surface surveys and excavation front surveys of the Villarondin-Bourget-Modane descending gallery have, however, made it possible to offer a coherent and generally well-supported structural interpretation (Figure 9). Two contrasted structural zones are however distinguished which exhibit very different geometric styles on either side of a major anticline (Râteau d'Aussois anticline) (Figure 8).

West of the Râteau d'Aussois anticline, the SMA terrains are generally inclined towards the West (Figure 8A) and consist, under the Houiller Zone and the Sapey Gneiss in a succession of folded scales separated by deep descending transpressive faults dipping towards the West. These faults have ben intersected by several long boreholes (1000 m or more) and detailed surface surveys. These tectonic scales and associated folds show a generalized vergence towards the East. The Râteau d’Aussois anticline shows the same eastern vergence and thus appears contemporary with the structuration of the western part of the SMA sector.

East of the Râteau d'Aussois anticline, the structure is very different (Figure 8B). From numerous core drill holes and a very long directed borehole (SDA, 1,822.5 m, of which 609 m are core drills) and land surveys, we observe a superposition of basement and cover entities separated by thrusting planes showing a geometry that is similar to a “ramp and flat” system. The dip of the layers and thrusting planes, which is generally rather weak, straightens up (folded) at the level of the Râteau d’Aussois anticline, which reveals its posteriority.

The Middle Triassic carbonate terrains (Photo 6) present at the eastern termination of the SMA sector, constitute two tectonic scales inclined towards the East in abnormal contact with the quartzitic terrains of the Lower Triassic along the Aussois detachment fault (Figure 8B).

Some thrusting planes are marked by cargneules at the surface and by brecciated anhydrite at depth (intersected by the VBM descending gallery). Cargneules thus represent a weathered facies of brecciated anhydrites. The cargneules which cover the slope of the right bank of the Arc in the sector of Aussois-Sardières more likely mark the trace of the initial major contact of the Schistes Lustrés Nappe in the Briançonnais Zone (thrust having probably been reactivated in detachment).

From Avrieux to the Etache Vallon

East of Avrieux and up to the valley of Etache valley (almost 7 km long), the planned tunnel continues under the Schistes Lustrés Nappe in a superficial position, especially in gypsum-Anhydrite terrains constituting the "Gypsum Nappe".

The geology in depth of this sector, under the Longe Côte Massif, was very poorly understood before the investigations for the tunnel were carried out. Several deep vertical boreholes (largely cored) and a very long directed borehole (SDET) (2,950 m including 172 m cored) as well as detailed surface surveys around the Longe Côte massif allowed to partially decipher the complex geology of the massif (Figure 9).

Three lithostratigraphic and tectonic Mesozoic units are identified from West to East (with local repetitions) along this section (Figure 9 and Figure 10): the Gypsum Nappe (mainly gypsum or anhydrite), the newly defined Lozes Unit (gypsum, dolomite, limestones and marbles) and the Briançonnais unit “de l'Esseillon” (conglomeratic micaschists, quartzites and carbonate cover).

Normal or detachment faults have been highlighted within this set of units (Figure 8 and Figure 9, Photo 8).).

The Ambin Massif

From the Etache valley and up to the Italian border (almost 10 km), the tunnel crosses the Ambin Massif under a rock cover locally greater than 2,000 m) (Figure 11). This Massif represents the eastern part of the internal Briançonnais Zone, it forms a structural dome which appears in window (15 km x 20 km approximately) under the Ligurian-Piedmontais schists (Figure 1).

From the West, the tunnel first crosses the Groupe d'Ambin series for less than 2 km; these terrains were studied on surface and from drillings (especially the long directed drilling of Etache, SDE). They are essentially quartzitic micaschists and albitic gneisses, long attributed to the Permian, but at least “permo-triassic”. These terrains of the Group of Ambin mark the western flank of the dome of Ambin and present a fairly simple overall structure on the scale of the longitudinal profile, according to a strong dip towards the West. Contact with the Clarea basement is marked by a zone of more or less intense deformation.

The Clarea basement constitutes the major part and the core of the Ambin Massif. It is fairly well known on the surface but it has only been investigated in depth by two boreholes and thus appears to be little recognized in borehole in view of the volume it represents and the length of the tunnel concerned (nearly 8 km in France). But this observation must be balanced by the fact that this basement appears relatively homogeneous.

This basement consists mainly of micaschists fairly dark and often bluish gray (presence of glaucophane amphibole) with quartz veins and in which lenses of chlorito-albitic gneisses are inserted and levels of boudins like amphibolite. Ante-alpine mineral parageneses (biotite and garnet) are partially preserved away from alpine shear zones. The main foliation of the basement is generally subhorizontal.

No major fault has so far been highlighted or suspected within the Ambin Massif. The presence of one or more major tectonic contacts is very likely at the base of the Ambin Massif, but probably at very great depth, below the level of the base tunnel.

Conclusions

The investigations carried out on behalf of TELT along the planned Mont Cenis tunnel made it possible to reinterpret the geological structure of a wide section of the Western Alps, on the scale of the first hectometers and up to 1-2 km of the upper part of the earth's crust.

If along a part of the transect, the structure represented is quite close to that reported by the previous authors (Ultradauphinois, Subbriançonnais and Briançonnais Houillère zone), in several sectors, the proposed reinterpretation proves to be entirely new and supported by compared to previous interpretations. This is the case for the internal Briançonnaise Zone and the Gypsum Nappe Zone:

• within the internal Briançonnais Zone, a complex polyphase structure is highlighted from West to East in the Sapey-Modane-Avrieux sector. On either side of the major anticline of the Râteau d’Ausssois in the central position in this sector, we identify in the East, an early westward phase according to geometries similar to a “ramp and flat” thrust system, and in the West, a succession of tectonic scales and related folds associated with a following retrotranspressive eastward tectonic phase;

• in the east of the Sapey-Modane-Avrieux sector, a global structural interpretation is proposed for the section of the project corresponding to the Gypsum Nappe zone structurally very little known until now. A new unit (Lozes unit) reported to the Piemontais domain is individualized in this sector.

Late extensive tectonics is also highlighted or specified within the internal Briançonnais and Gypsum nappe zones.

Investigations along the transect have also made it possible to complete or clarify the lithostratigraphy of the Ultradauphinoise Zone, the Gypsum Nappe and the newly identified Piemontais Unit (Lozes Unit).

Finally, surface observations coupled with observations on cores and along descending galleries have made it possible to confirm the fact that the cargneules that highlight overlapping accidents on the surface give way in depth to an anhydritic breccia and thus correspond to a facies resulting from the meteoric alteration of the latter.

Keywords: Western Alps, New Lyon-Turin Line, Mont-Cenis Tunnel, Geological investigations, Geological cross-section, Ultradauphinois Zone, Penninic Front, Subbriançonnais Zone, Houiller Front, Briançonnais Houiller Zone, Briançonnais internal Zone, Nappe des Gypses, Piémontais Zone (Lozes Unit), Ultradauphinois Flysch, Cargneules.