In an Expert Focus article for WaterBriefing, Vikki Williams, Arup’s Digital Water Leader in the UK, India, Middle East and Africa region, discusses the key role AI and digital solutions play in designing resilient water systems for an age of extremes.

Vikki Williams: Water systems around the world are facing mounting pressure from multiple directions. Ageing infrastructure, designed for more stable patterns of demand, is being stretched by urban growth and new forms of consumption, including the rapid expansion of hyperscale data centres which can sometimes use large quantities of water each day for cooling. For everyone involved in the water sector – from water companies to city authorities – there is no longer a question of whether AI and digital solutions will shape water management, but how they can best be applied to maintain reliability, resilience and system performance.

A critical shift in water systems that requires action now

The operating context for water systems is now evolving rapidly. Climate risk has shifted water challenges from isolated pressures to systemic imbalance, with “too much” and “too little” water increasingly occurring within the same catchment. AI and digital solutions how data analytics, machine learning and satellite imagery can help cities understand how they absorb and manage rainfall – as demonstrated in like the Global Sponge Cities Snapshot. This work established an important foundation for using technology to support nature‑based and system‑wide thinking in urban water management.

At the same time, urban growth and new, continuous sources of water demand are adding pressure at a scale not previously seen. These conditions are increasing the need to move beyond insight alone, and to translate data‑led, system‑wide thinking into operational decisions that can respond dynamically to today’s more complex water realities.

AI for managing inflow and infiltration

One of the clearest opportunities for digital tools in today’s water context is helping utilities move away from broad, asset‑heavy responses towards more targeted, evidence‑led interventions. This is particularly relevant for inflow and infiltration (I&I), where excess water entering wastewater networks can reduce capacity and increase the likelihood of overflows.

In Gloucester, Severn Trent Water faced an 80,000m³ storage requirement but had limited space for the tanks traditionally used to manage excess flows across the catchment. The focus was on improving the quality and usability of existing information by cleaning and updating the utility’s Geographic Information System (GIS) data in just 16 weeks – creating a clearer picture of how the network operates. This enabled AI models to identify where I&I was most likely to occur and interventions could have the greatest impact.

The value of this digital‑led approach lies not only in speed, but in adaptability. Ongoing monitoring allows teams to track changes over time and adjust their I&I reduction strategy as conditions evolve, rather than relying on fixed solutions based on historic assumptions. Importantly, the insights generated are now being used to benchmark and refine approaches across different network conditions, illustrating how digital approaches could support a broader shift across the UK water sector – from reducing storm overflow risk to improving resilience and informing future infrastructure decisions.

Helping nature‑based systems perform under extremes

Grey infrastructure will continue to age and require renewal. By contrast, well‑designed green and blue assets – such as wetlands, storage areas and natural drainage features – can continue to provide storage and attenuation benefits over time, making nature‑based solutions supported by technology increasingly important under extreme climate conditions.

Arup is working with The Nature Conservancy and collaborators in Florida around smart watershed management. Our hydrological modellers are looking at how water behaves in existing stormwater infrastructure under a variety of scenarios. Then we then use AI and machine learning to analyse ways smart infrastructure can hold back, divert, and route water during deluges or during storms to reduce the impact of pollution in the sensitive coastal ecosystems.

The aim is not just to improve a single asset, but to unlock wider benefits. By making natural assets more intelligent, this approach aims to improve the environment and build resilience across entire watersheds. Crucially, because these methods can be tested, refined and transferred, they provide a pathway to scaling climate resilient, nature positive infrastructure in regions facing similar challenges worldwide.

Unlocking the water transition through data and coordination

Technology alone will not deliver the water transition. The next challenge is enabling smarter water management systems to evolve – moving beyond fragmented data, siloed responsibilities and regulatory frameworks that were designed for static assets rather than dynamic systems.

This challenge is evident in Spain and across the EU, where water systems remain highly fragmented and data is often inconsistent or poorly shared.¹ Gaps in coordination between operators, regulators and users limit the impact of digital tools, meaning innovation too often remains confined to pilots instead of informing decisions at a system scale.

Open data is therefore foundational. AI can only support better water management if information on assets, flows, climate risk and demand can move seamlessly across organisations and catchments. Common data standards, interoperable platforms and trusted data‑sharing arrangements are essential to translate insight into action and to support collaboration across sectors.

Designing water systems for adaptation, not stability

The water transition facing global cities is urgent, but it is also full of opportunity. These examples show many of the tools needed to manage greater variability and competing demands already exist – from digital capability to nature‑based approaches that work with water rather than against it. The challenge now is to apply these tools with confidence and intent.

Managing both “too much” and “too little” water cannot be achieved by any one actor alone. It requires cooperation across cities, regions, sectors and governments, aligning water management with decisions on land use, energy, digital infrastructure and urban growth.

With the right policy signals and a shared commitment to long‑term resilience, innovation can move beyond isolated initiatives and begin shaping systems at scale.

1. Spanish Water Industry 2050 by Arup: https://www.arup.com/insights/spanish-water-industry-2050-where-it-stands-and-where-it-is-headed/