Pesticide transfers and fate are highly influenced by the presence of discontinuities such as grass strips, slopes, hedgerows or roads that can accelerate or slow down and dissipate water and contaminant fluxes. That is why those landscape elements must be integrated into watershed management plans. It implies taking them into account when modeling water and contaminant fluxes at the small catchment scale. However, if the influence of landscape elements has already been widely explored at field scale, models generally do not reach the catchment scale. The project PESHMELBA aims at developing a new modeling tool of water and contaminants circulation and fate at the scale of small catchments in order to optimize landscape organization. The model explicitly takes into account spatial organization of landscape by representing existing elements, their locations and shapes. The final aim of this modeling tool is to efficiently test and rank different development scenarios in order to assess the influence of agricultural practices, land uses and landscape management strategies on water quality. In PESHMELBA, dominating processes ruling water and contaminants circulation and dissipation for each element type are mainly represented by existing and validated models. New components have also been developed when no suitable model was found in the literature. All these models present different levels of conceptualization and are used as modeling units ensuring a modular structure. Then, the different units are gathered and connected in the OpenPALM coupler (Fouilloux and Piacentini, 1999) in order to implement the spatial and temporal couplings. This innovative approach leads to a spatialized model of the whole catchment. Applications cases are tested with an increasing complexity, from a case with two plots to the hillslope scale with several plots, ditches and rivers. They show that PESHMELBA is a promising tool to compare scenarios considering water and pesticide fate in different complex landscapes.