Testing the potential of geochemical techniques for identifying hydrological systems within landslides in partly weathered marls

This paper's objective is to determine how useful geochemistry can be in landslide investigations. More specifically, what additional information can be gained by analysing the cation exchange capacity (CEC) and cation composition in respect to the hydrological system of a landslide area in clayey material. Two cores from the Boulc-Mondorès landslide (France) and one core from the Alvera landslide (Italy) were analysed. The NH4Ac and NaCI laboratory techniques are tested. The geochemical results are compared with the core descriptions and interpreted with respect to their usefulness. Both analysis techniques give identical results for CEC, and are plausible on the basis of the available clay content information. The determination of the exchangeable cations was more difficult, since part of the marls dissolved. With the ammonium-acetate method more of the marls are dissolved than with the sodium-chloride method. The NaCI method is preferred for the determination of the cation fractions at the complex, be it that this method has the disadvantage that the sodium fraction cannot be determined. To overcome this problem, it is recommended to try other displacement fluids. In the Boulc-Mondorès example, the subsurface information that can be extracted from CEC analyses was presented. In the Boulc-Mondorès cores deviant intervals of CEC could be identified. These are interpreted as weathered layers (and preferential flow paths) that may develop or have already developed into slip surfaces. The major problem of the CEC analyses was to explain the origin of the differences found in the core samples. Both Alvera and Boulc-Mondorès examples show transitions in cation composition with depth. It was shown that the exchangeable caution fractions can be useful in locating boundaries between water types, especially the boundary between the superficial, rain-fed hydrological system and the lower, regional groundwater system. This information may be important for landslide interventions since the hydrological system and the origin of the water need to be known in detail. It is also plausible that long-term predictions of slope stability may be improved by knowledge of the hydrogeochemical evolution of clayey landslides. From the analysis, it is concluded that geochemistry is a potentially valuable technique for landslide research, but it is recognized that a lot of work still has to be done before the technique can be applied in engineering practice.