The European Parliament recently approved new EU pesticide legislation. Member States must adopt National Action Plans for reducing the risks and impacts of pesticide use and take measures to protect water resources. The French program Ecophyto 2018 targets a 50% reduction in pesticide use. The funding available for these mitigation measures needs to be used in the most cost-effective way. As the characteristics of agricultural production differ across a catchment, policy instruments can lead to various results in terms of environmental effectiveness and implementation costs. Therefore, defining policy priorities requires appropriate tools with results concerning the ecological and social features of agricultural practices. The evaluation ex-ante of scenarios using agro-environmental indicators (AEI) and/or agro-hydrological models needs to implement an effective information system, able to take into account the characteristics of the ecosystem and of agricultural activities at embedded spatial scales. The “trans-time-plot” scale, a homogeneous part of watershed for agricultural practices over time, is used for the calculation of indicators for small agricultural catchments. On a larger watershed scale, other spatially designed entities are used regarding typologies of farming systems and natural conditions, or areas of mitigation measures. The produced data model allows the agricultural practices on these embedded scales to be described in the same database, as well as a range of data relating to river basin characteristics. Agricultural practices (crop rotation, crop management, irrigation, fertilization, pesticide use) are described at the farm level for homogeneous plots or from a typology of farming systems built using surveys and institutional data. A dictionary gives detailed information about all the concepts used. The EIS system generated automatically from the business models offers tools for analysis. It has been tested with a data set collected on the Charente watershed and its sub-basins. Several axes of analyses have been defined (time, amount of pesticides sprayed, spatial aggregation). The tools implemented allow new possibilities for the analysis of data and the calculation of spatial indicators (AEI). The EIS system allows the calculation of a simple indicator of pesticide pressure. We plan to implement the calculation of an institutional indicator at the two embedded spatial scales.