1 Year Post doctoral Position: Analysing the hydrochemical signature of flow paths and deformation processes in mudslides with focus on the preferential flow SCIENTIFIC CONTEXT This work is part of the ANR Project Triggerland (2007-2010) which aims at delivering new technologies (experimental prototypes, numerical tools, models) to better understand the processes governing landslide failures and particularly to identify landslide patterns and possible forerunners that characterize significant changes in landslide dynamics. In current practice, assessment of slope failures is often restricted to empirical or statistical rainfall thresholds. These approaches ignore the physical mechanism by which a landslide is initiated, thus considerably limiting the ability to quantify landslide hazard. Moreover, this approach neglects the influence of slowly-varying factors (e.g. weathering and associated property changes). One specific objective is to improve our understanding of the triggering factors controlling landslide failures or landslide crises as accelerations (failure stage). Experimental sites are located in south eastern France (Draix and Supersauze, See below). This is a joint research project involving among others the Universities of Avignon (France), Utrecht and Delft (Netherlands) and Caen (France). Water flow is long known as the main triggering factor controlling landslides motion. In term of hydrological behaviour, flow processes have a direct hydro-mechanical impact on slope stability. But water flow also implies solute transfer and geochemical processes which have an impact on the strength properties of the material. These processes can also be used as tools for studying flow path, water origin and residence time. Hydrochemistry as tracing technique has been widely used in catchment hydrology (e.g. Kendall & McDonnell, 2003). The tracer information was needed to better understand the processes of streamflow generation. Provided non reactive tracers are used, the information gained may be as important as the residence time of water (Weiler et al., 2003), the improvement of the catchment delineation (Guglielmi et al., 2002) and the mixing proportion between the reservoirs (Cras et al., in press). The demand for increasingly detailed, process-based knowledge is also the drive for applying hydrogeochemistry in landslide research. For example, preferential flows have been identified as major processes in landslide environment but there is no reliable hydrometric technique which can measure such processes. Tracers, either environmental (de Montety et al., 2007) or artificial (Bogner et al., 2004), have long proved to be one of the most relevant tools to get quantitative information on these processes. There is also evidence of influence of pore fluid composition on strength properties. For example, Moore & Brunsden (1996) concluded that there is a strong correlation between the pore water composition and the geotechnical behaviour of the clay. The mineral composition can also influence the sensitivity of the clays (Anson & Hawkins, 1998; Di Maio et al., 2004). In this context, the soil capacity to transfer rapidly water downwards to the saturated zone (preferential flows) is expected to play a major role. Surface discontinuities are known to favour rapid transfer of dilute waters. Detailed analysis on cored borings sometimes shows the lowest pore water salinity at largest depth, indicating fresh water flowing lateral and upwards from the underlying bedrock (e.g. Torrance, 1979, Andersson-Sköld et al., 2005). Therefore the interaction between mineralogy, pore water chemistry and stress state should be considered also in geotechnical research. TASKS At the microscale, laboratory experiments will be carried out to study the relationship between strength, pore water chemistry and mineralogical composition. The results will be analysed in terms of development of preferential flow paths within the landslide body. These results will be related to the results of the study to the influence of preferential flows and unsaturated/saturated

zones on the hydromechanical behaviour as the chemical changes directly influence the physical soil properties such as permeability and strength. At the mesoscale, information upon infiltration processes will be gained in summer 2007 from in-situ simulated rainfalls using input waters enriched in bromide or other non reactive tracer. The results regarding the transit time and the interaction with pre-event waters are expected to constraint the modelling exercise. The applicant will also be asked to propose a geochemical modelling of the system from a long term hydrochemical database (more than 200 analyses of major ions and field measurements from 10 sampling campaigns between 2003 and 2006) and data on the geochemical properties of the material. The main objectives will be to: 1/ Use chemical tracing techniques to quantify the water fluxes and travel time distribution at different scales. 2/ Identify a hydrochemical signature of the deformation processes. 3/ Describe and quantify the impact of pore fluid composition on strength properties. EXPECTED RESULTS 1. Analyses of the relation between fissure dynamics, permeability and changes in water chemical content and their relation to landslide triggering. 2. Assimilating experimental results (such as transit time) in hydrological models of water flows in matrix and in fissures. 3. Quantification of residual shear strength related to pore fluid composition and material type EXPECTED SKILLS The applicant will have a strong background in hydro-geochemical modelling (including solute transfer modelling). Experience and expertise in hydrological processes investigation and hydrological modelling are highly desirable. PRACTICAL DETAILS The applicant will be settled at the University of Avignon (Hydrogeological Laboratory). He will spend at least 2 months at the University of Delft/Utrecht. The gross salary is 27532 euros (net monthly salary of around 1870 euros) for 12 months appointment. Candidates should submit an academic cv and statements of qualification by July, 1st, 2007. Candidates are requested to send their application to Dr V. Marc (Université d’Avignon, UFR Sciences, 33, rue Louis Pasteur, 84000 Avignon, France, email :Vincent.Marc@univ-avignon.fr). The aimed starting date is on October 1st 2007 Avignon is at the crossroads of Provence and Languedoc, near the Mediterranean on a route between Italy and Spain. The climate is Mediterranean, temperate and windy with 300 days of sun a year. For more information, please go to http://www.ot-avignon.fr/pages-en/home.htm DESCRIPTION OF EXPERIMENTAL SITES

SUPERSAUZE

The Super-Sauze mudslide is one of the persistently active landslide (since the 1970’s) occurring in black marls (Malet and Maquaire, 2003). The landslide was triggered at the beginning of the 1970’s at the interface between the moraine and the autochtonous black marls

by using the different discontinuities (fault, bedding plane, joint and schistosity) affecting the bedrock. The mudslide materiel consists of a silty-sand matrix mixed with moraine debris. It extents over an horizontal distance of 850 m and occurs between an elevation of 2105 m (crown) and 1740 m (toe) with an average 25° slope. A detailed morphological description of the mudslide since its genesis can be found in Weber and Herrmann (2000). The paleotopography, corresponding to a succession of more or less parallel crests and gullies, plays an essential role in the behaviour of the landslide by delimiting preferential water and material pathways and creating sections with differing kinematical, mechanical and hydrological characteristics. The total volume is estimated at 750,000 m3 and velocities range from 0.01 to 0.4 m.day-1. The landslide hydrology and kinematics have been monitored since 1996 and a spatio-temporal database on rainfall, temperature, capillary pressure head, soil moisture content, groundwater level and displacement is available (Malet, 2003; Malet et al., 2005). Around twenty open standpipe piezometers with manual recordings, filtered at different levels, are distributed over the landslide body. DRAIX The Draix ERBs (Experimental Research Basins), created by the Cemagref in 1983-1984, are located 13 km north-east of the town of Digne, Alpes de Haute-Provence, France. They consist of Jurassic marine black marls (Bajocian, Bathonian and Callovo-Oxfordian) overlaid by resistant limestone layers which dominate the topography in monoclinal ridges with elevations up to 2000 m. Marls can reach a thickness of 1500 m and are characterized by a dense layering. They are very sensitive to weathering and erosion, very unstable (Antoine et al., 1995), and cover an extensive area in the French South Alps (2200 km²; Brochot, 1999). The basins show a typical badlands morphology with V-shape gullies (Mathys et al., 2003), providing highly loaded floods downstream and silting up reservoirs down to ‘l’Etang de Berre’ on the Mediterranean coast. In the framework of the ANR project, infiltration processes through fractures within unweathered black marls will be investigated at the hillslope scale (see figure below)

References Kendall, C. and J.J. McDonnell (2003). Isotope tracers in catchment hydrology. Elsevier, 3rd

edition. Weiler, M., B. L. McGlynn, K. J. McGuire, and J. J. McDonnell (2003), How does rainfall become

runoff? A combined tracer and runoff transfer function approach, Water Resour. Res., 39(11), 1315

Guglielmi, Y., Vengeon, J.M., Bertrand, C., Mudry, J., Follacci, J.P., Giraud, A., 2002. Hydrogeochemistry: an investigation tool to evaluate infiltration into large moving rock masses (case study of La Clapiere and Sechilienne alpine landslides). Bull. Eng Geol. Env. 61 (4), 311–324.

Cras, A., Marc, V. and Travi, Y., in press. Hydrological behaviour of sub Mediterranean alpine headwater streams in a badlands environment. doi:10.1016/j.jhydrol.2007.03.004

de Montety, V., V. Marc, Ch. Emblach, J.-P. Malet, C. Bertrand, O.Maquaire, T.A. Bogaard, 2007. Identifying origin of roundwater and flow processes in complex landslides affecting black marls in southern French Alps: insights from an hydrochemical survey. Earth Surface Processes and Landforms, 32, 32-48.

Bogner, C., Marc V., Di Pietro L., Doussan Cl., Ruy S., Gaudu J.C., Perrin P., Cognard-Plancq A.L., Emblanch Ch., Daniel M., Simler R., 2004. Tracer study of infiltration in a field soil – simulated rainfall experiment on a large lysimeter. International Workshop on the Application

of Isotope Techniques in Hydrological and Environmental Studies, 6 – 8 September 2004, Paris, pp.43-44.

Moore, R. and Brunsden, D., 1996. Physico-chemical effects on the behaviour of a coastal mudslide. Geotechnique, 46(2); 259-278.

Anson, R.W.W. and A.B. Hawkins, 1998. The effect of calcium ions in pore water on the residual shear strength of kaolinite and sodium montmorillonite. Géotechnique 48, 6, pp.787–800.

Di Maio,C., L. Santoli, P. Schiavone, 2004. Volume change behaviour of clays: the influence of mineral composition, pore fluid composition and stress state. Mechanics of Materials 36, pp.435–451

Berry, R.W., Torrance, K., 1998. Mineralogy, grain-size distribution and geotechnical behaviour of Champlain clay core samples, Quebec. Canadian Mineralogist 36, 1625– 1636.

Torrance, J.K., 1979. Post-depositional changes in the pore water chemistry of the sensitive marine clays of the Ottawa area, eastern Canada. Engineering Geology 114, 135– 147.

Andersson-Sköld, Y., J.K. Torrance, B. Lind, K. Odén, R.L. Stevens, K. Rankka, 2005. Quick clay – A case study of chemical perspective in Southwest Sweden. Engineering Geology, 82, pp.107-118.

Malet, J.-P. and Maquaire, O., 2003. Black marl earthflows mobility and long-term seasonal dynamic in southeastern France. In: Picarelli, L. (Ed). Proceedings of the International Conference on Fast Slope Movements: Prediction and Prevention for Risk Mitigation. Patron Editore, Bologna: 333-340.

Weber, D. and Herrmann, A., 2000. Contribution de la photogrammetrie numerique a l'etude spatio-temporelle de versants instables; l'exemple du glissement de terrain de Super-Sauze (Alpes-de-Haute-Provence, France). Bulletin de la Société Géologique de France, 171(6): 637-648.

Malet, J.-P., 2003. Les ‘glissements de type écoulement’ dans les marnes noires des Alpes du Sud. Morphologie, fonctionnement et modélisation hydro-mécanique. Doctoral Thesis, University Louis Pasteur, Strasbourg, 364p.

Malet J.-P., Van Asch, Th.W.J., Van Beek, R., and Maquaire, O., 2005. Forecasting the behaviour of complex landslides with a spatially distributed hydrological model. Natural Hazard and Earth System Science, 5: 71-85.

Antoine, P., Giraud, A., Meunier, M. and Van Asch, T., 1995. Geological and geotechnical properties of the "Terres Noires" in southeastern France : weathering, erosion, solid transport and instability. Engineering Geology, 40: 223-234.

Brochot, S., 1999. Les bassins versants de Draix, expérimentaux... et représentatifs, Les bassins versants expérimentaux de Draix, laboratoire d'étude de l'érosion en montagne. Cemagref Editions, Draix, Le Brusquet, Digne, pp. 223-234.

Mathys, N., Brochot, S., Meunier, M. and Richard, D., 2003. Erosion quantification in the small marly catchments of Draix (Alpes de Haute Provence, France). Calibration of the ETC rainfall-runoff-erosion model. Catena, 50: 527-548.