Are Flood Maps Accurate? Why a Property Can Flood When the Map Says Low Risk (and Vice Versa)

Are flood maps accurate? At the scale they are built for, meaning a region, a catchment or a whole postcode, broadly yes. At the scale that matters when you are buying a house or designing a single building, often not. National flood maps are screening tools. The same map that flags risk across a plot can be wrong about whether the building on it will ever take in water, and a plot the map calls low risk can still flood. For a buyer, a lender or a developer, that gap is where decisions go wrong.

This is not a fault in the maps. They are doing the job they were designed for. The problem is using a strategic, national dataset to answer a question it was never built to answer: what happens at this property, in this storm, on this ground.

The Environment Agency's latest National Flood Risk Assessment (NaFRA2) puts around 6.3 million properties in England in areas at risk of flooding from rivers, the sea or surface water, with roughly 4.6 million exposed to surface water alone, and the total rising towards 8 million, or one property in four, by mid-century.

When a screening report or a conveyancing search flags one of those properties, the sensible next step is not to panic or to look away, but to get a property-level answer. A site-specific flood risk survey for property purchase reads the national mapping against measured ground levels at the building itself, which is the only way to know what a flag actually means for the house you are buying.

What national flood maps actually show

National flood maps estimate flood risk across the country using large-scale modelling. They are built to identify where risk may exist, so that planning, insurance and awareness decisions can be made. They are not built to replicate the behaviour of water at one address.

In England the Environment Agency publishes several products, each answering a slightly different question:

• The Flood Map for Planning: the statutory map that defines Flood Zones 1, 2 and 3 for river and sea flooding, and the one the planning system relies on. It deliberately ignores flood defences, because defences can be overtopped or fail.
• The Risk of Flooding from Surface Water map: the national picture of pluvial (rainfall-driven) risk, now the most widespread flood source in England.
• The long-term flood risk service: the homeowner-facing tool that reports four risk bands and does factor in existing defences.
• Reservoir inundation mapping: the modelled extent were a reservoir to fail.

Each is built from topographic data, hydrological modelling and catchment analysis applied consistently across millions of properties at once. That consistency is the strength and the limitation. A statutory dataset has to model the whole of England to one method, so it cannot account for the small, local features that decide whether a particular building floods. It also separates flood sources, so a Flood Zone result speaks only to rivers and the sea. If you are unsure which map answers which question, we set the two apart in detail in why you can't use the long-term flood risk map for a planning application, and explain the statutory one in what the Flood Map for Planning is and how to read it.

Are flood maps accurate at the scale that matters?

At national and strategic scale, yes: the maps are a reliable, consistent estimate of where flood risk is concentrated, and they are the right starting point for almost any flood question. At the scale of a single building, they are indicative rather than definitive. A national model cannot see a raised threshold, a step up from the pavement, a private culvert, or a garden that sits half a metre below the house, and those centimetres are what decide whether water reaches the door. So the honest answer to "are flood maps accurate for my property" is that they tell you risk exists in the area, not whether your building is exposed.

A "high risk" flag means risk exists somewhere on the plot. It does not mean the house will flood, and a "low risk" result does not mean it won't.

Why the map and the reality differ at a single property

A national dataset is blind to the very features that govern flooding at one address. The things a map cannot model are usually the things that matter most on the ground:

• Local topography and small depressions that collect water, or high spots that shed it.
• Door and floor thresholds, and the height a building sits above the surrounding ground.
• Private drainage, culverts and highway gullies, and, crucially, their condition.
• Field runoff routes from higher ground nearby.
• Changes a previous owner made to ground levels, hard surfacing or boundary walls.

Surface water is where this matters most. It is now the most widespread source of flood risk in England, with the Environment Agency putting around 4.6 million properties at risk from it, more than from rivers and the sea combined. Surface water rarely respects the tidy boundaries of a map, and it is the source of much of the confusion buyers run into. Since the September 2025 update to Planning Practice Guidance, the risk-based approach formally applies to all sources of flooding, covering surface water and groundwater as well as rivers and the sea. That is also why a Flood Zone 1 site can still face surface water scrutiny: a low river-and-sea rating says nothing about rainfall pooling on the plot.

When a "high risk" flag is misleading: a real Unda investigation

A high-risk flag on a screening report tells you risk exists somewhere on the plot, not where, and not whether it reaches the building. On a recent property purchase in Berkshire, surface water mapping ran from "Low" to "High" across a single plot, and a conveyancing search returned a flood warning that put the sale in doubt. The map was right that risk was present; it could not say where the water actually went.

When we read the mapping against LiDAR data and the levels we measured on site, the picture changed. The highest predicted depths sat in the rear garden, a shallow bowl where water flowing off higher ground to the south, east and west collects, rather than draining to the culverted watercourse to the north-west. That ponding zone is separated from the house by a series of steps up in level: the dwelling sits roughly 360mm above the patio, which sits roughly 360mm above the garden. Door thresholds ranged from about 100mm at the front to around 520mm at the garage, with services set well above ground.

Set against measured site levels, the dwelling fell within "Low" risk today, rising to "Medium" with climate change, while the genuine "High" risk stayed in the garden, with somewhere to go.

The buyer got an evidence-based answer in place of a worrying but unspecific search result, and something they could put to a lender or insurer. The full project is written up here. The map was not wrong. It simply could not tell them what they needed to know.

When a "low risk" property still floods

The error runs the other way too. A property the map shows as low risk can still flood, because the national datasets do not capture every local mechanism. This is the more dangerous case, because a reassuring result discourages a closer look.

Properties outside mapped flood zones flood through several routes the maps do not model in detail:

• Surface water runoff off neighbouring land or roads, arriving faster than local drainage can cope.
• Blocked or undersized drainage, so ordinary rainfall backs up.
• Groundwater rising from below, which neither the Flood Map for Planning nor the long-term service maps.
• Sewer surcharge, where a combined system overflows in heavy rain.
• An ordinary watercourse, a ditch or minor stream the national maps do not model.

A defended site can still flood if a defence is overtopped or fails, which is exactly why the statutory Flood Map for Planning ignores defences. A clean map result is a reason to check the local detail, not a guarantee. It is the same reason a house can sit in a flood zone yet have never flooded, and the reverse.

What the 2026 map updates changed

The national maps are getting sharper, but they are still national. Two recent changes matter for anyone reading a flood map in 2026.

First, NaFRA2, the Environment Agency's second-generation National Flood Risk Assessment published in December 2024, redrew the national picture: it put 6.3 million properties in areas at risk today and around 4.6 million at risk from surface water, a 43% rise on the previous assessment. Our explainer on understanding and managing flood risk with NaFRA2 covers what changed and why. Second, on 28 May 2026 the Environment Agency added surface water climate change extents and banded depth data in the Flood Map for Planning's May 2026 update, and withdrew the older long-term surface water layers from planning use. Both make the maps more useful, but neither turns a national model into a property-level one.

The takeaway for accuracy is unchanged: newer data narrows the error, but a national map still cannot read the ground at one address.

How to check whether your property is really at risk

Start with the free screening tools, then commission a site-specific assessment if the screening, or your own experience, raises a question. Screening tells you whether to look harder; only a property-level assessment tells you what the risk means for the building. A sensible order:

1. Run the postcode through a free checker. Unda's flood risk map screens the Environment Agency datasets for any postcode in England and returns a one-page report. Treat it as an indicative first pass, not a verdict on the building.
2. Gather local evidence. Historic photos, dates and depths of past flooding, and what neighbours have seen are information no desk-based map holds.
3. Note the property's own defences. Ground levels, door thresholds and existing drainage all change the picture.
4. Commission a site-specific flood risk survey where a flag, a search result or a lender's question needs a real answer. This is the stage that reads the modelling against measured site levels, as it did on the Berkshire purchase.

If you are earlier in the process, our step-by-step guide to checking flood risk before buying a house walks through the free tools, the conveyancing searches and the TA6 disclosure in order.

What a flood map means when you're buying or selling

For a buyer, a flood flag is a prompt to investigate, not a verdict on the property. Environmental and conveyancing searches use the same national datasets and broad-scale modelling, with no site visit and no assessment of the building; they are a screen, not the final word. Since 30 March 2026 the TA6 Property Information Form asks sellers about flooding from any source, so the subject now comes up in most transactions. A mapped flag met with property-level evidence is far easier to put to a lender or insurer than a flag left unexplained.

For a seller or owner, the same logic is reassuring: a flood flag does not have to reduce a property's value once the real risk is understood. Research from Bayes Business School found flood-exposed homes sell at a discount that climbs to around a third, roughly 31%, for the highest-risk properties, so the difference between "the map flags this postcode" and "this building is safe to the door" is worth real money. For a developer, the map is only the trigger: a low rating still has to be evidenced in a flood risk assessment for planning, and a high rating rarely means a site cannot be built on.

Frequently asked questions

Get a property-level answer, not just a map

A national flood map is the right place to start and the wrong place to stop. It will tell you whether risk exists in the area; it will not tell you whether your building is exposed, or where on a plot the water actually goes. When a search flags flood risk, a lender asks a question, or a property has flooded before, a site-specific flood risk survey for property purchase gives you the evidence-based answer. Get a quote and we will respond within 60 minutes.