Integration of artificial intelligence and machine learning with isotope hydrology to support sustainable water management

Introduction The integration of artificial intelligence (AI) and machine learning (ML) with isotope-based approaches has demonstrated a strong potential for enhancing the analysis of hydrological processes, tracer dynamics, system responses, and strengthening water resource management under increasing environmental pressures. However, this integration has not been fully explored and formulated yet. Methods To fill this knowledge gap, this paper conceptualizes perspectives on the application of AI/ML tools in isotope hydrology to promote sustainable water management. Results were developed based on the expert consultations and group work following the modified Delphi and Nominal Group Technique method. Fourteen international experts with diverse background worked together during the technical meeting coordinated by International Atomic Energy Agency (IAEA) and United Nations Educational, Scientific and Cultural Organization (UNESCO) (Vienna, Austria, 2025). Results and discussion Diverse applications of AI/ML methods to enhance interpretation of isotope data, improve isotope-enabled modeling accuracy, and support decision-making were conceptualized in this paper. The results showed the need for developing interoperable data platforms, fostering interdisciplinary collaboration and hybrid physical-AI modeling, ensuring ethical AI/ML practices and the advancement of IAEA Global Water Analysis Laboratory (GloWAL) Network. The approach presented here has broader applicability and could be implemented across other networks and related programmes to support resilient and sustainable water resource management.