Harnessing AI for smarter water management under a changing climate: A review of machine learning and deep learning applications within EU water framework directive and marine strategy framework directive

The Marine Strategy Framework Directive (MSFD) and the Water Framework Directive (WFD) are key EU policies for safeguarding marine and freshwater ecosystems. They establish targets and evaluation frameworks to ensure good environmental and ecological status. This paper reviews the application of machine learning (ML) and deep learning (DL) in supporting these directives. A dual approach is used: a scientometric analysis to map the research landscape, followed by a systematic review of selected studies. The review focuses on three main areas: (i) use of spatio-temporal data, (ii) integration of climate-related scenario analysis, and (iii) application of ML and DL methods to enhance directive implementation. Results highlight the growing role of AI in water management, demonstrating its potential in handling large datasets, revealing patterns, and supporting predictive decision-making. Nonetheless, significant gaps remain. The underuse of spatio-temporal data hinders predictive accuracy, and scenario analysis methods often fail to capture the full complexity of climate impacts on aquatic systems. These limitations constrain the effectiveness of data-driven policy decisions. The paper calls for further research to better incorporate temporal dynamics in AI models and to refine scenario analysis tools. Such improvements are crucial for advancing adaptive, informed strategies in water resource management aligned with European directives.