QUANTIFICATION DES SOURCES ET DE LA DYNAMIQUE DES SÉDIMENTS DANS DES BASSINS CULTIVÉS CONTRASTÉS DU SUD DU BRÉSIL (AFFLUENTS DU FLEUVE URUGUAY)

par Rafael Ramon

Projet de thèse en Terre solide : géodynamique des enveloppes supérieures, paléobiosphère

Sous la direction de Olivier Evrard.

Thèses en préparation à Paris Saclay en cotutelle avec Federal University of Rio Grande do Sul , dans le cadre de Sciences mécaniques et énergétiques, matériaux, géosciences , en partenariat avec LSCE - Laboratoire des Sciences du Climat et de l'Environnement (laboratoire) et de Université Paris-Sud (établissement de préparation de la thèse) depuis le 03-04-2017 .


  • Résumé

    Quantifier les sources et la dynamique des sédiments dans des bassins versants agricoles contrastés du sud du Brésil. L'intensification de l'agriculture dans le sud du Brésil depuis les années 1970 au détriment des biomes naturels (pampa et forêt atlantique) a conduit à une augmentation de la dégradation des sols. Néanmoins, peu de données quantitatives existent quant aux sources de sédiments qui dégradent les hydrosystèmes et à leur dynamique.

  • Titre traduit

    Quantifying sediment sources and dynamics in contrasted agricultural tributaries of the Uruguay River (Southern Brazil)


  • Résumé

    Water erosion intensified by agricultural activities is one of the main factors of soil degradation in subtropical regions. In addition to the soil loss and the potential decrease of crop yields at the local scale, the excess of surface runoff increases the transfer of sediments and contaminants to the water bodies, resulting in deleterious environmental, economic and social damages. These phenomena propagate over large scales, and the high quantity of sediment produced on the cultivated land of the southernmost part of Brazil (Rio Grande do Sul) accumulates in the Salto Grande Hydroelectric Power Plant Reservoir built on the Uruguay River in 1980 and exposed to excessive sedimentation and eutrophication processes. Despite the associated concerns regarding water quality, there is a lack of quantitative information regarding the quantity of sediment accumulated in the lake (potentially affecting the expected lifespan of the reservoir) and their potential sources. Accordingly, the current research will quantify these sources and their dynamics to guide the implementation of effective soil and water conservation practices across the basin. To this end, the main erosion and sediment transfer processes will be quantified in two tributaries of the Uruguay River, representative of the main contrasting geomorphological regions found in its drainage basin (224,000 km², 70% of this land being located in Southern Brazil, 20% in Argentina and 10% in Uruguay). Land use is predominantly agricultural, draining areas of intensive agricultural production in the states of Rio Grande do Sul (RS-Brazil) and Santa Catarina (SC-Brazil), where small (<50 ha) farming properties dominate in the mountains (with cropland, dairy, intensive breeding of poultry and pigs, natural and cultivated forests), whereas larger (<100 ha) properties are found in the flatter areas (with soybean, maize and wheat production). On the other hand, in the southern region of RS, northern Uruguay and northeastern Argentina, large farms (>100 ha) dominate. In general, most of the areas are used as pasture for extensive livestock and the floodplains are used for irrigated rice, with an increase in the soybean production in the region in recent years. The Conceição River catchment, located in upper parts of the Uruguay River basin, was chosen because it is representative of those areas located on the basaltic southern plateau region of the Serra Geral Formation, where deep clayey Oxisols predominate. In this catchment, the main land use is cropland for grain production and grassland for dairy farming, where inadequate soil management has resulted in very high erosion rates (139.7 t km-2 yr-1). In contrast, the Ibirapuitã River catchment is representative of those environments found on the southern sandstone plateau, with a predominance of shallow sandy soils such as Ultisols and Entisols, originating from the sedimentary rocks of the Paraná basin. In this second catchment, there is a predominance of native grasslands with extensive livestock farming, irrigated rice in the lower relief positions and, more recently, the expansion of soybean production. The objective of this thesis is to characterize the main processes of sediment transfer in these two catchments and to quantify sediment source contributions. The specific objectives are: (i) to quantify the discharge of sediment in bedload and in suspension during rainfall-flow events through a set of hydro-sedimentometric techniques; (ii) to calculate the contribution of the sediment sources for various particle size fractions in both suspended and bedload material; and (iii) to calculate the sediment source contributions considering the main land uses, land management practices, soil characteristics, and the main tributaries as priority source discriminations. The main potential sediment sources were identified by observing the main active erosion processes during preliminary fieldwork campaigns and on satellite imagery. In the Conceição River catchment, the main potential sources are cropland, grassland, unpaved roads and channel banks. In the Ibirapuitã River catchment, four potential sources were identified: channel banks, cropland (grain production areas and cultivated pastures), unpaved roads and degraded native grassland. A minimum of 20 samples will be collected to characterize each source, in each catchment, with the preferential sampling of the uppermost surface layer (0–2 cm), which is the material that will be the most easily eroded. Sediment samples required to quantify the flux of suspended and bedload sediments will be taken during a minimum of two rainfall-flow events of large magnitude, with the coverage of both the rise and recession of the hydrograph. The suspended sediment will be collected with USDH-49 samplers while the bedload sediment will be collected with a Halley-Smith device. Large quantities of sediment will be collected using a continuous flow centrifuge to obtain sufficient material for further analysis. Integrating samplers (e.g. sediment traps deployed during longer periods) will also be used for the tributary tracing approach. The samples will be analyzed for radionuclide activity and biogeochemical, physical (particle size, morphometry and specific surface area) and magnetic properties, before the selection of those tracing properties that will be capable of discriminating between sediment sources. The current research will therefore significantly improve our knowledge of erosion and sediment transfer processes in this subtropical region. Furthermore, the characteristics of sediment transiting in these two contrasted tributaries will be confronted to those of the material accumulated in the Salto Grande Dam Reservoir in order to provide the first tentative estimate of sediment source contributions in this region exposed to accelerated soil erosion.