Water quality is a major environmental and societal concern, increasingly challenged by pollution, climate change, and evolving management demands. This paper reports on a collaborative initiative between the Community of Municipalities of Sumène-Artense (CCSA) and the University of Clermont Auvergne aimed at developing transparent and accessible methodologies for helping manage water resources. Two complementary approaches were explored: biological tools for assessing quality through biomarkers, and explainable artificial intelligence. Together, these tools provide ecologically grounded indicators of water quality and predictive models that are interpretable by technicians, decision-makers, and citizens. At the heart of the project, and elaborated throughout this paper, is the triptych of Training, Research, and Professionalisation, which served as the guiding framework for linking scientific innovation with societal relevance. Students of different degrees were embedded in field and laboratory work, acquiring technical and transversal skills, while associations, GEMAPI technicians, and local actors participated in data collection and interpretation. This integration strengthens scientific mediation and ensures that research remains both academically rigorous and socially relevant. This study demonstrates how modular, adaptable and explainable methodologies can foster dialogue between science and society while supporting sustainable water governance.