This is what AI means for water management in 2030

The development of artificial intelligence (AI) is accelerating – every week brings new possabilities. What impact will this have on the water sector? This summer, HydroLogic and Wateropleidingen organized a practical masterclass to inspire professionals: with AI, you’re better prepared for extreme weather and can respond more quickly to changing conditions.

With 18 participants from water authorities, municipalities, industry, and consultancy firms, Wateropleidingen director Gabrielle Knufman kicked off the masterclass by asking: “How far along are you with AI?” What united them was a shared interest in the possibilities of AI and the need to apply it in their work. Language models like ChatGPT and Copilot are especially popular – for writing assistance, data analysis, project cost estimation, and Python scripting. While some organizations are exploring AI’s potential, others are thinking about integrating it into their workflows and envisioning the virtual engineer of the future.

Smarter than humans

HydroLogic’s business director Leanne Reichard looked ahead: when AI becomes smarter than humans, how do we ensure it helps us – without replacing us? Her advice: “Use AI as a strategic ally in your work, stay up to date, develop your skills, and share knowledge with each other.”

She asked participants to reflect: what will your role look like in the future? What tasks will AI take over? The consensus: AI will act as an advisor, enabling better and more integrated decision-making. However, trust still needs to grow in the reliability of the data and recommendations AI provides.

Self-learning networks

Colleague Arnold Lobbrecht explained how a self-learning system works: from identifying patterns in data to building a model (or network) that mimics what it has learned and can then make its own predictions. He highlighted two key networks for water management: Convolutional Neural Networks (CNN) and Long Short-Term Memory networks (LSTM). Applications are widespread – from detecting dike cracks and sewer blockages to forecasting drought, rainfall, and drinking water usage.

Fast forecasting

Surrogate or simulation models will make a real difference in the daily practice of water professionals, Lobbrecht said. He demonstrated this with three simulations: a dike breach along the river Lek, flooding in a small polder in the Stichtse Rijnlanden, and an early warning system for potential water overload in Amersfoort. “These AI simulations predict the impact of a dike breach or upcoming flooding much faster than physics-based models – within seconds instead of hours.”

Situation in 2030

According to Leanne Reichard, the water sector in 2030 will look like this:

• Better preparedness for extreme weather: proactive responses thanks to AI-driven decision-making
• Real-time data and model integration will enable adaptive collaboration between water boards, municipalities, and drinking water companies
• Water professionals and citizens will receive tailored information, advice, and actions via their personal AI assistants

The participants

Participants worked on applying AI to real-life cases. In small groups, they enthusiastically shared their experiences based on the examples discussed or tried out new AI tools. With tips from Leanne and Arnold, the world of water management and AI became even more fascinating.

Learn more

Are you interested in this masterclass or in a follow-up session? Let Martine Smallenburg know, and we’ll keep you informed!