Spatial and temporal dynamics of sediment in contrasted mountainous watersheds (Mexican transverse volcanic axis and French Southern Alps) combining river gauging, elemental geochemistry and fallout radionuclides

International audience; In mountainous environments, an excessive sediment supply to the rivers typically leads to an increase in water turbidity, pollution and a rapid filling of reservoirs. This situation is particularly problematic in regions where reservoirs are used to provide drinking water to large cities (e.g. in central Mexico) or where stream water is used to run hydroelectric power plants (e.g. in the French Southern Alps). Sediment source areas need first to be delineated and sediment fluxes between hillslopes and the river system must be better understood to implement efficient erosion control measures. In this context, the STREAMS (« Sediment Transport and Erosion Across MountainS ») project funded by the French National Research Agency (ANR) aims at understanding the spatial and temporal dynamics of sediment at the scale of mountainous watersheds (between 500 – 1000 km2) located in contrasted environments. This 4-years study is carried out simultaneously in a volcanic watershed located in the Mexican transverse volcanic axis undergoing a sub-humid tropical climate, as well as in a sedimentary watershed of the French Southern Alps undergoing a transitional climate with Mediterranean and continental influences. One of the main specificities of this project consists in combining traditional monitoring techniques (i.e. installation of river gauges and sediment samplers in several sub-catchments) and sediment fingerprinting using elemental geochemistry (measured by Instrumental Neutron Activation Analysis – INAA – and Inductively Coupled Plasma - Mass Spectrometry – ICP-MS) and fallout radionuclides (measured by gamma spectrometry). In the French watershed, geochemical analysis allows outlining different sediment sources (e.g. the contribution of calcareous vs. marl-covered sub-catchments). Radionuclide ratios (e.g. Cs-137/Be-7) allow identifying the dominant erosion processes occurring within the watershed. Areas mostly affected by gully erosion, rill or sheet erosion have been delineated. Furthermore, the measurement of radionuclide content in suspended sediment after the snowmelt suggests that most of this sediment consists in resuspended material rather than on newly eroded soil. In the Mexican watershed, a different contribution of andosols and acrisols to erosion is suspected. Overall, the bulk of erosion is generated by rather small areas of the watershed. In this region characterised by a succession of wet and dry seasons, the Be-7 content in rainfall and sediment has been measured at the scale of a 2.5-km2 sub-watershed in order to better understand the erosion transfer between hillslopes and rivers during the wet season. This outlines the contribution of individual storms to seasonal erosion. Overall, this study brings important insights about sediment sources and fluxes within these watersheds located in contrasted environments. A further step consists in comparing experimental results with model outputs, and to evaluate the impact of on-going erosion mitigation measures.