This paper describes an automatized multi-step methodology in order to identify the relationships between integrative pesticide quantifications and land-use on a given watershed. This methodology contains multivariate statistical analyses such as hierarchical cluster analysis (HCA) and principal component analysis (PCA), which are commonly used for the interpretation of complex geospatial datasets. A large amount of pesticide concentration data were collected along 1-year monitoring in 2016, for 50 sites located on the Adour Garonne basin (South-West France). For those sampling sites, concentrations of 37 selected pesticides were investigated during six periods of 14-days immersion of integrative samplers. Specifically, the sampling devices used were Polar Organic Chemical Integrative Sampler (POCIS), providing time-weighted average concentration estimates. For each studied site, the associated watershed and its land-use repartition were determined based on the Corine Land Cover 2012 and geographical information system (GIS) aggregation of data. The HCA clustered the 50 sites into five groups with similar main land uses. After that, the datasets of pesticide integrated concentration and land use repartition were analyzed in a PCA. The key variables (pesticide distribution and concentrations) responsible for sampling site discrimination showed consistent patterns of distribution with specific land uses. In order to confirm these observations, pesticide fingerprints (based on the waffle method) of sites with contrasted land use relative to the surface areas were compared. These fingerprints confirmed that there was different and specific patterns, visible at a glance, of pesticide occurrence in surface water, in relation with their initial use at the catchment level. This method allowed identifying sources of contamination that could be interesting to prevent or contain pesticide pollutions beyond simply acting on the most at-risk areas.