Simulation de mesures environnementales pour réduire le nitrate dans les eaux de surface avec le modèle hydrologique SWAT sur un bassin versant agricole du Sud Ouest de la France

[Departement_IRSTEA]Territoires [TR1_IRSTEA]DTAM [Axe_IRSTEA]DTAM1-REPRO [Axe_IRSTEA]DTAM2-ACPUBInternational audienceAgriculture is known to have a great impact of nutrients enrichment on continental water resources. In south-West of France, water resource are essentially surface water and shallow aquifer. Nitrogen dynamic in river is complex and highly variable throughout season and year, depending on hydrology, landuse, removal in stream. In this context, agricultural impacts on nitrogen concentration are a matter of concern for agricultural decision-maker. In order to introduce sustainable land use concepts in this hilly, clayey and agricultural shallow soil context, the hydrological simulation model SWAT2005 has been tested as a valuable tool to evaluate the consequences of such land use changes on water and nutrient balance components. This semi-distributed hydrological model coupled with agronomical model EPIC is able to simulate the impact of each agricultural landuse at the outlet of the Gers catchment (500 km²). Nitrogen losses have been measured during 10 years (1996-2005) at the outlet and are used to validate the model calibration. Agricultural data at communal scale coupled with Spot image analyses have been used to evaluate agricultural distribution and pressure in SWAT. The aim of this modelling exercise is to simulate nitrogen load in surface water depending on plant growth, culture rotation and management practices. The ability of SWAT to reproduce nitrogen transfer and transformation at this scale and in this agricultural context will be evaluated by a discussion of importance of each nitrogen cycle process in nitrogen losses in surface water. Finally, SWAT were used to test 53 scenario based on new agricultural management practices to improve the nitrogen concentration in river. The best scenario shows a reduction of 30% of the nitrate concentrations in river water