A biopurification system (BPS) is used on-farm to clean pesticide-contaminated wastewater. Due to high pesticide loads, a BPS represents a hot spot for the proliferation and selection as well as the genetic adaptation of discrete pesticide degrading microorganisms. However, while considerable knowledge exists on the biodegradation of specific pesticides in BPSs, the bacterial community composition of these systems has hardly been explored. In this work, the Shannon diversity, the richness and the composition of the bacterial community within an operational BPS receiving wastewater contaminated with various pesticides was, for the first time, elucidated over the course of an agricultural season, using DGGE profiling and pyrosequencing of 16S rRNA gene fragments amplified from total community DNA. During the agricultural season, an increase in the concentration of pesticides in the BPS was observed along with the detection of significant community changes including a decrease in microbial diversity. Additionally, a significant increase in the relative abundance of Proteobacteria, mainly the Gammaproteobacteria, was found, and OTUs (operational taxonomic units) affiliated to Pseudomonas responded positively during the course of the season. Furthermore, a banding-pattern analysis of 16S rRNA gene-based DGGE fingerprinting, targeting the Alpha- and Betaproteobacteria as well as the Actinobacteria, indicated that the Betaproteobacteria might play an important role. Interestingly, a decrease of Firmicutes and Bacteroidetes was observed, indicating their selective disadvantage in a BPS, to which pesticides have been introduced.