D46 Implementation of EQ-LS-HY coupling model on type landslide in Alps – ACRI

The « La Clapière landslide » is located near the village of Saint Etienne de Tinée. It is bordered to its north-western side by the Tenibres river and to its south-eastern side by the Rabuons river, both rivers flowing down to the Tinée river at the bottom of the landslide. This unstable slope is typical of rock slopes deformations and is well documented. It is characterized by a series of trenches (see Fig. 1 and Fig. 2) that form a perched water-saturated zone connected by large fractures to a basal aquifer. Because of the high heterogeneity of permeabilities of the medium (Fig. 3), the mechanics of water infiltration and its effect on the landslide displacements are very complex. Moreover, permeabilities of the slope are likely to change with time. This is linked to the coupling between slope deformations and interstitial pressures. For example, a change in the groundwater flow with an increased fluid pressure in the slope will induce mechanical deformations, which can then change the hydraulic properties, thus affecting the groundwater flow. Within the frame of this deliverable, the La Clapière site will be taken as a reference case to study hydromechanical coupling effects on the stability of rock slopes.

Therefore the work first consists in retrieving the values of the different permeabilities of the medium. For that, we will elaborate a series of numerical models at different scales from the trench scale (see for example modelled area on Fig. 1) to the entire slope scale using PORFLOW software (developed by ACRI) to determine the permeability of the trenches (K_trench in Fig. 3) and of the landslide (K_landslide in Fig. 3). Preliminary analytical calculations allowed to give some ranges of the different permebalilities of the slope (in Fig. 3). At a chosen scale, calculations will be repeated for different state of deformations deduced from in situ multi-scale and multi-temporal measurements. This should give a rough relashionship between permeability and deformation of the slope. The input data of these models are the slope geometry and the geochemical properties of the rocks. The output is a pressure field, the chemical and discharge variations at the springs. The aim of the modelling is to test the permeability parameter by varying its value until the results of the model fit the hydrogeochemical observations (springs waters chemical and discharge variations versus time) that are continuously acquired since December 1995. The model is loaded with precipitations events that are continuously monitored on the site.

Fig. 1: Cross-section of La Clapière landslide: two main satured zones: a perched saturated zone mainly located in the trenches deposits and a basal saturated zone located within the gneissic massif discontinuities. The rectangles show the various scales of modelled areas of the slope.

Fig. 2: Example of 2D simplification for modelling tests (vertical cross section).

Fig. 3: La Clapière slope hydrolgeological map (red and blue circles are springs)

Secondly, once the permeabilities have been determined with PORFLOW, the groundwater flow within the slope can be restored and a permeability versus slope deformation law is implemented in the models. Hydromechanical numerical models are then runned to understand the evolution of the landslide under precipitations. Two cases using two different codes are used:

- the slope is taken as a poroelastic continuous equivalent medium. FLAC3D code is used.

- the slope is taken as an heterogeneous medium with preexisting persistent discontinuities. The Universal Distinct Element Code (UDEC) is used.

The comparison of the results of the two models allow to emphasize the fracture role on the medium behavior. Simulations are made in a saturated-unsaturated transient state. The model is loaded with precipitations events. Calculated surface displacements are compared to measured ones with GPS.

Main scientific results will be establishing a coupling law between permeability and deformation of rock slopes deduced from in-situ measurements (GPS, hydrochemistry and springs water discharge). Sensibility of safety factors to such coupling effects will then be tested on a simple theoritical case. Finally, a method to better take into consideration such effects will be proposed.