AI presents opportunities and challenges for sustainable water management
Expert(s) |
Antonio Moreno Rodenas
Artificial Intelligence (AI) offers significant opportunities to enhance sustainable water management worldwide, yet integrating these technologies also presents a series of challenges. In partnership with UNESCO’s Intergovernmental Hydrological Programme (UNESCO-IHP), Deltares has released a new publication that explores the current state of AI and Machine Learning (ML) applications in water management.
The publication Applications of AI for water management serves as a reference guide for policymakers, water technicians, and professionals without specific backgrounds in AI or data science.
Opportunities to improve water management
AI is proving to be a powerful tool for addressing a broad range of water-related challenges, such as surface water management, groundwater modelling, irrigation, flood risk assessment, climate adaptation, and water quality. These technologies provide new possibilities for creating more efficient and flexible systems that can quickly adapt to the evolving challenges in water management.
Dr Antonio Moreno-Rodenas, lead author and Senior Researcher at Deltares, emphasises the potential of AI, but also underscores the importance of careful and responsible implementation: "AI can be a game changer for water management, but only if we deploy it thoughtfully and responsibly. We must bridge the gap between hydrology and data science, and work together to develop applications that are both reliable and auditable."
AI can be a game changer for water management, but only if we deploy it thoughtfully and responsibly
Dr Antonio Moreno-Rodenas, Deltares
Promising paths
While AI applications in water management have grown exponentially in recent years, the integration of AI into operational systems with a high Technology Readiness Level (TRL) remains limited. However, promising examples are already emerging, such as computer vision techniques for water infrastructure inspection and data-driven methods for detecting anomalies in water distribution networks. It is expected that AI technologies, including multimodal large language models and advanced neural networks, will provide a new generation of tools for water managers.
Dr. Koen Verbist, Programme Specialist at UNESCO-IHP and co-organizer of the publication adds: “It was crucial to create a resource that speaks directly to water managers and policymakers, bridging the gap between cutting-edge AI developments and the practical application on the ground."
This publication aims to make knowledge more accessible and bring AI-concepts and potential applications closer to the water management field
Dr. Koen Verbist, UNESCO-IHP
Ethical considerations and collaboration
The publication also highlights the ethical considerations associated with the use of AI in water management. It is critical that these technologies are applied responsibly to avoid inadvertently harming vulnerable communities and ecosystems.
In the publication's foreword, Abou Amani, Director of the Division of Water Sciences at UNESCO, highlights that “by integrating AI into water management systems, communities and industries can achieve a more sustainable and resilient approach to managing this vital resource.”
"Yet", he adds, “there is a clear need for further training and capacity building, and closer collaboration between AI practitioners and hydrologists.” Sharing knowledge and ensuring access to high-quality data are essential for developing unbiased AI tools that support the water management field in addressing global water challenges.
New publication by UNESCO-IHP and Deltares
The UNESCO-Deltares publication Applications of AI for water management - on the state of use of AI in water management - has been written by: Antonio Moreno-Rodenas, Koen Verbist, Annelies Mertens, Isabel Gerritsma, Jing Deng, Arjen Haag, Ümit Taner, Jonathan David Nuttall, Ruben Dahm, Ali Meshgi, Hans Korving, Silvia Bianchini, Veronica Tofani, Nicola Casagli, Patrick Ray, Saiful Haque Rahat, Jianzhao Guan, Yin Chen, Lei Zhang, Hongling Shi, Julia Kaltenborn, Kim Bente, Andrew McDonald, Chris Derksen, and Giriraj Amarnath.
