Government is presenting surface water flooding as a higher-order resilience issue for England and Wales, with the strongest evidence coming from updated national risk data. The Environment Agency's latest National Flood Risk Assessment for England reports that about 4.6 million properties are now at risk from surface water flooding, a 43% increase since the previous assessment. That shift changes the policy context. The same assessment shows there are now three times as many properties at high risk from surface water flooding as from flooding from rivers and the sea. In the National Risk Register 2023, government raised the impact score for surface water flooding to significant, putting it on a par with river and coastal flooding in national risk planning.
The article links that increase in exposure directly to climate conditions and patterns of urban development. Short, intense summer rainfall is becoming more common, and in densely built areas the effect can move quickly from localised disruption to a threat to life, transport links and essential infrastructure. Recent events help explain why forecasting capability has moved up the policy agenda. Government points to flooding in Coverack in 2017 and London in 2021 as examples of rapid surface water events where warning time, location accuracy and operational confidence all matter to responders on the ground.
The principal programme response has been the three-year Surface Water Flood Forecasting Improvement Project. According to the government article, that work has already moved beyond research and into service design, most notably by supporting the Rapid Flood Guidance service into operational use. That matters because the policy objective is not simply to produce better science but to improve decisions before and during an event. The article states that the Rapid Flood Guidance service will now continue until 2028, giving local responders a clearer planning horizon while government and forecasting partners develop the next round of tools.
A central feature of the current system is impact-based forecasting. The Flood Guidance Statement and the Rapid Flood Guidance service do not only estimate the physical hazard; they also assess the likely consequences for people, property, transport and infrastructure. That combined approach is intended to help emergency responders decide more quickly when to deploy equipment, issue warnings or stand up local response arrangements. To support that judgement, the Flood Forecasting Centre uses the Surface Water Flooding Hazard Impact Model, developed by the National Hazards Partnership in 2020. The model currently provides forecasts from roughly six hours to three days ahead. It draws on runoff forecasts from the Grid-to-Grid hydrological model using Met Office rainfall forecasts, detailed flood maps from the Environment Agency's National Flood Risk Assessment, and a 1 km impact library developed with the Health and Safety Executive. The resulting output is displayed through a dashboard showing county-level likelihood of minimal, minor, significant and severe impacts.
The article also identifies a clear operational gap in the two to six hours before a rapid flood event. User research found that responders want more local detail during that period because it is often the final window for moving assets, warning communities and preparing emergency action. At present, Rapid Flood Guidance relies on detailed short-term storm forecasts from the Met Office's Expert Weather Hub, interpreted with support from hydrometeorologists at the Flood Forecasting Centre. What it does not yet include is real-time hazard and impact modelling of the kind that could strengthen confidence in where the worst local effects are most likely to occur.
To address that shortfall, the Flood Forecasting Centre tested two surface water flood model nowcast tools for the zero to six hour lead time. One was a nowcast version of the Surface Water Flooding Hazard Impact Model. The other was the FOREWARNS surface water flood model developed by the University of Leeds. Both used Met Office ensemble rainfall nowcasts, which combine radar observations with model output to forecast rainfall up to six hours ahead. The trial had three practical aims: to test whether the tools were suitable for operational forecasting, to measure the added value against existing rapid flood forecasting methods, and to gather feedback on what a future operational nowcast product should look like. Phase 1 ran in the Met Office summer testbed in 2024 with more than 50 meteorologists, modellers and academics from the United Kingdom and overseas. Phase 2 in 2025 focused on Flood Forecasting Centre and Met Office staff expected to use the tools in live operations.
The government summary reports a strong result from that testing. It found that use of the two models could have improved Rapid Flood Guidance forecasts on nearly half of the occasions examined. The article also says the models improved identification of impact thresholds for issuing guidance, with gains in location accuracy and, in some cases, the lead time for action. There was also a clear difference in how the tools were valued. Surface Water Flooding Hazard Impact Model performed best when forecasters needed to assess likelihood, severity and the type of impact expected. FOREWARNS, by contrast, was valued for simplicity and for the broader event context it could add during fast-moving situations. In policy terms, that points to complementary functions rather than a simple choice of one model over another.
The recommended next step is a further three-year programme intended to bring new tools into operations by 2029. The article says that work is expected to include performance improvements to both modelling systems, preparation for next-generation Met Office rainfall nowcast data from 2027, priority development of a Surface Water Flooding Hazard Impact Model nowcast product for the zero to six hour period, and possible use of FOREWARNS as an operational prototype. For local resilience forums, emergency services, highways authorities and infrastructure operators, the practical effect is straightforward. If the proposed work is delivered, Rapid Flood Guidance should become faster, more local and more reliable at the point when response decisions are most time-critical. For government, the article amounts to a formal statement that surface water flood forecasting is now being treated as a long-term operational capability requirement rather than a limited-term innovation project.