Professor Qiuhua Liang at Loughborough University has won a UKRI Turing AI Fellowship to develop more localised flood forecasts using artificial intelligence and high-performance computing. The project aims to give communities, emergency responders and policymakers information tailored to specific places, rather than a forecast that treats every postcode as much the same.
Watch Desk analysis
What happened
Loughborough University says Liang is one of 20 new Turing AI Fellows, part of a £39 million UKRI investment in AI research. His three-year project, HiPA-Flood, will explore how AI and high-performance computing can support personalised, real-time information about surface-water flooding.
The university says the aim is to improve forecast accuracy and help people prepare for flood risks. That describes the project’s intended work, not a result already demonstrated in practice.
Why it matters
Flood warnings are most useful when they tell people what may happen where they are, and in time to act. More localised forecasts could help residents, emergency services and policymakers make better-informed decisions. The fellowship also puts AI and computing to work on a practical climate challenge, rather than leaving the technology’s public value at the level of a rather polished promise.
Our read
This is a substantive research project with a clear public purpose and a useful test ahead: whether its forecasts become more accurate and actionable. The funding supports the work; it does not, by itself, establish that the system will improve warnings. The interesting measure will be what it can tell people about their local risk that existing forecasts cannot.
What to watch
- How Liang’s team combines AI with high-performance computing for local forecasts.
- Whether the project publishes evidence that forecasts are more accurate or useful.
- How forecast information reaches communities and emergency responders in time to act.
Discussion spark: For flood warnings, what would matter most to you: a more precise forecast for your area, or clearer advice about what to do when risk rises?
Sources and evidence
- Source update (17 November 2026, 00:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.