How Increasing Rainfall Is Changing Flood Risk in the UK

Flooding across the UK is becoming more frequent, more widespread and easier to trigger, and increasing rainfall is the reason. The shift is driven less by wilder storms than by a wetter atmosphere and wetter autumns and winters, which together are changing flood risk across the UK. Rivers now rise fast after rain that never looked exceptional. Streets flood in neighbourhoods with no history of it. And once the water arrives, it often lingers for days or weeks after the heaviest rain has passed.

If storms do not obviously look stronger, why does flooding keep getting worse? The answer is not in any single storm. It is in a measured, well-evidenced change in how rainfall behaves, both during storms and across whole seasons. This article sets out that science, and then what it means for anyone building, buying or assessing a site.

Met Office attribution found that storm rainfall in the UK's 2023–24 autumn and winter was around 20% heavier than it would have been without human-caused warming, and that such wet seasons are now at least four times more likely.

The physical baseline has shifted: a wetter UK atmosphere

Start with the air itself. A warmer atmosphere holds more moisture, roughly 7% more for every 1°C of warming, so the weather systems that have always crossed the UK now carry more water and drop heavier rain. This is basic thermodynamics, not a forecast. Warm air lifted by a front, a low-pressure centre or high ground cools and releases its moisture as rain; if the air started wetter, the rain is heavier.

The point that catches people out is that none of this needs more storms, or more violent ones. Most UK rainfall comes from large Atlantic systems that have existed for as long as we have kept records. What has changed is the moisture they draw on. Feed an ordinary weather system wetter air and it produces heavier rainfall when conditions line up.

A wetter atmosphere delivers heavier rain from the same ordinary weather systems, with no extra storms required.

What does attribution science show about UK storm rainfall?

Attribution science asks a precise question: how would the same weather have behaved in a world without human-induced warming? It does not claim climate change "caused" a particular storm. It compares today's climate with a counterfactual pre-industrial one and isolates how warming has changed the rainfall.

The Met Office's analysis of the 2023–24 storm season is a clear worked example. Rather than modelling a future scenario, it examined the rainfall from storms that actually happened between October 2023 and March 2024, using observations alongside large ensembles of climate model runs, and compared today's climate (about 1.2°C warmer than pre-industrial) with a world without human warming. Two results stand out: storm rainfall over that period was around 20% more intense than it would have been in a pre-industrial climate, and rainfall of that intensity has gone from roughly a 1-in-50-year event to something expected about once every five years.

These are not marginal shifts. Catchments are highly sensitive once soils and rivers are near capacity, so a 20% rise in rainfall intensity can translate into a much larger rise in runoff, flood extent and flood duration.

The independent storm-based analysis reaches a closely aligned conclusion by a different route. Instead of named storms it defines "stormy days" from wind severity, then looks at the rainfall on those days across long records. Observations show average rainfall on stormy days up around 30%; once the human influence is isolated, the best estimate is again about 20%. Two independent methods, different datasets, the same answer: storms reaching the UK now deliver materially more rain because the climate they form in is warmer and wetter.

Is it the rain or the wind that is getting worse?

It is the rain. Trends in UK storm winds are genuinely uncertain: depending on the dataset and method, studies show small increases, small decreases or little change at all. Rainfall intensification, by contrast, shows a strong and consistent signal. That distinction matters, because wind and rain are governed by different physics. Wind strength follows atmospheric dynamics and pressure gradients; rainfall intensity follows moisture and temperature.

So flood risk can climb even when a storm does not feel especially dramatic. The wind may be unremarkable while the rain falling with it is well above what the same system would once have produced.

Rainfall, not wind, is now the dominant driver of flooding in the UK.

How do wetter autumns and winters raise flood risk?

Flooding is rarely decided by a single hour of rain. It depends heavily on how wet the ground already is when that rain arrives, what hydrologists call antecedent conditions. This is where the seasonal picture becomes as important as any individual storm.

The Met Office attribution work also looked at total rainfall across the October-to-March period, the UK's main storm season. That season in 2023–24 was the second wettest on record for the UK and the third for Ireland. Attribution puts climate change's contribution at around 15% more seasonal rainfall, and estimates that such wet autumn-winter seasons are now at least four times more likely. In a pre-industrial climate a total like that had a return period of roughly 1 in 80 years; today it is closer to 1 in 20.

Sustained rain of this kind resets the baseline state of the land. Soils saturate, groundwater rises, and rivers run high for weeks. The Met Office has described recent autumns and winters as "waterlogging the soils with virtually no time for them to dry out". Once soils are saturated the system behaves differently: infiltration drops, rainfall becomes runoff almost immediately, and rivers respond faster and peak sooner. That is why flooding increasingly follows rain that, on its own, would not look extreme. The extremity is in the context, not always the storm.

The winters since have kept the pattern going. Winter 2025–26 was among the ten wettest on record for England, which saw about 42% more rain than the seasonal average; Worcestershire recorded its wettest February in a series running back to 1836, and England logged its wettest 18 months on record to March 2026.

How does a modern UK flood season unfold?

It builds over weeks, not hours. Once heavier rain and wetter seasons are in place, the route to flooding is mechanical rather than mysterious, and it runs through five stages.

A wetter-than-average autumn sets the baseline, lifting soil moisture and groundwater before winter storms even begin, so rivers enter winter already high. Early storms then arrive carrying more rain than their historical equivalents, with wetter seasons leaving shorter gaps between them and little chance for soils to drain. At some point saturation flips the system: infiltration collapses and rain converts to runoff almost on contact. A storm that is meteorologically ordinary now becomes hydrologically disruptive, driving rapid river rises and widespread surface water flooding. And the impacts persist, because elevated groundwater and waterlogged soils keep rivers high long after the rain eases, so any further rain lands on a system that has not yet recovered.

This is why modern flooding so often looks widespread, prolonged and hard to contain, even when no single storm stands out.

Could this simply be natural variability?

Because attribution is built to separate the two. Rather than leaning on single events, it compares large ensembles of model runs representing climates with and without human warming. Natural variability exists in both worlds; if rainfall is consistently higher in the warmed climate and that difference matches observations, the human contribution can be quantified. It is the same method behind both the 20% rise in storm rainfall intensity and the shift in how often such rainfall occurs.

There is a fair challenge here. UK weather swings hard year to year, largely with atmospheric patterns such as the North Atlantic Oscillation, which can drown out the underlying trend over any short run. The work of Carruthers and colleagues (2025) addresses exactly that. Using dynamical adjustment, which strips out the influence of shifting circulation, they find a clear rise in winter precipitation across mid-to-high-latitude Europe, the UK included. Their striking conclusion is that this non-dynamical signal has emerged around 23 years ahead of the CMIP6 multi-model mean. Multi-model averages smooth out regional extremes, so this does not make the models wrong in direction; it shows the rainfall signal arriving earlier and more strongly than the smoothed average implies.

A 2025 study found the UK-region winter rainfall signal has emerged roughly 23 years ahead of the multi-model average, so the change is arriving faster than smoothed projections suggest.

What does increasing rainfall mean for planning and development?

For anyone building or buying, the practical message is short: the rainfall a site has to cope with is heavier than its historical record implies, and a flood risk assessment has to plan for that. Surface water is now the largest single source of flood risk in England, and "it has never flooded here" carries far less weight than it used to.

That plays out in a few concrete ways on a real scheme. Climate change allowances are the mechanism the system already uses to deal with this: the Environment Agency publishes percentage uplifts for peak rainfall and peak river flow, drawn from UKCP18 projections, and a competent FRA applies them across the development's design life rather than assessing against today's figures. The maps also lag the ground truth, so a low-risk rating on the Flood Map for Planning is not a clean bill of health, particularly for surface water. And drainage has to be sized for the future, not the past: a SuDS scheme built on outdated rainfall intensities will under-deliver on the day it is tested.

• Climate change allowances: a compliant FRA applies EA uplifts for peak rainfall and river flow over the development's lifetime, using the climate change allowances for the site's flood zone and vulnerability.
• Surface water first: with around 6.3 million properties in England at risk today and surface water the largest source, a sound surface water drainage strategy matters as much as the river assessment.
• Maps lag reality: a low rating on the Flood Map for Planning is a starting point, not a verdict, and flood maps can understate real risk, especially for surface water.
• Design for headroom: SuDS and attenuation should be sized on future rainfall, building in capacity that is far more expensive to retrofit later.
• History is no defence: a clean flood record reflects the old climate; saturated ground and higher rivers mean sites can flood for the first time.

A site's flood risk is now set by the climate it will face over its lifetime, not by the rain it has seen in the past.

The joined-up picture

Put together, the evidence is coherent and hard to dismiss. Warming has raised the moisture content of the atmosphere; storms reaching the UK now deliver more rain, quantified across several attribution studies; and autumn and winter seasons are wetter overall, leaving soils, rivers and drains with less room to absorb what falls. Flooding therefore becomes more likely, more widespread and more persistent, without any dramatic change in how often storms come or how hard the wind blows. This is not a future scenario. It has been measured, attributed and, for a growing number of places, already lived through.

Understanding flood risk through increasing rainfall and cumulative wetness, rather than simply "more storms", is now essential to getting a site assessment right. If you are planning a development and need a flood risk assessment that accounts properly for climate change allowances and surface water, our flood risk assessment for planning team can help. For the flip side of the same coin, how one climate produces drought as well as flood, see our explainer on climate whiplash.

Frequently asked questions