Methodology
LocalRisk is built on official UK government datasets published under Open Government Licence v3.0. This page documents the source, vintage, spatial resolution and nation coverage of every dataset that informs a postcode result, and how raw values become risk bands.
LocalRisk is a free UK climate risk checker. Enter a postcode and the site returns plain-English risk bands across six hazards: flood, heat, air quality, subsidence, coastal erosion and green space resilience. Every band traces back to an official UK government dataset and every dataset is published under the Open Government Licence v3.0.
The result is an area-level screening indicator, not a property-level survey. Where data is published at postcode resolution, LocalRisk uses it directly. Where the underlying data is coarser (council, 1km grid, LSOA) the wider value is displayed on the postcode page and labelled. The site is designed to make official data easier to read, not to replace a professional flood report, structural survey or environmental search.
Sources: Environment Agency (England), SEPA (Scotland), Natural Resources Wales (Wales). DfI Rivers (Northern Ireland) is not yet integrated.
In England the Environment Agency NaFRA2 postcode product aggregates property-level flood scores to a postcode label, taking the highest property risk in the postcode as the postcode band. This means a single high-risk address can place an entire postcode in the Higher band, so the label is a screening signal for the area, not a count of affected homes. For property-specific detail use the EA's Check Your Long-Term Flood Risk service.
For Scotland, SEPA does not publish a postcode product. LocalRisk derives a band by spatially querying SEPA's published flood extent polygons (river, coastal, surface water) against each postcode centroid using a 25 metre buffer; the highest likelihood band found becomes the postcode's band. The Wales method uses NRW national flood zone classifications under the same centroid-to-polygon approach. Northern Ireland flood data is published by DfI Rivers but uses a different methodology and has not yet been integrated; NI postcode pages explicitly say so rather than show a placeholder band.
Bands: Higher (more than once every 30 years, ≥3.3% annual chance), Medium (once every 30-100 years), Lower (once every 100-1,000 years), Very Low (less than once every 1,000 years or outside modelled flood zones).
Long-term probability only. Property elevation, drainage, and defences not included. Scotland and Wales results are area-level indicators derived from postcode centroids, not address-level data.
Source: Met Office UKCP18 Regional projections, plus ERA5 reanalysis for the recent observed weather panels. Resolution: 12 km grid, aggregated to Local Authority District. Coverage: UK-wide.
Heat risk counts projected days per year above 25 degrees Celsius across the 2021 to 2040 near-term period under the RCP8.5 high-emissions scenario, taking the 50th percentile (central estimate). All postcodes inside a given Local Authority District share the same projected count because UKCP18 is published at council resolution.
These are probabilistic projections. The 50th percentile represents a central estimate, but the full range of modelled outcomes is wide. RCP8.5 is the high-emissions scenario conventionally used in UK property and infrastructure risk assessment to avoid underestimating long-term exposure; it is a planning assumption, not a prediction of the most likely outcome. Outcomes would be materially lower under lower-emissions scenarios.
The "Recent observed weather" panel on postcode and council pages uses ERA5 reanalysis (2023 to 2025) and shows actual recent conditions rather than projected ones.
Bands: Higher (31+ hot days/year), Medium (23-30 hot days/year), Lower (0-22 hot days/year). Thresholds based on the statistical distribution of projected values across all UK councils.
Climate projection, not a forecast. Urban Heat Island effects (often ~2-5 degrees Celsius in dense urban areas) are not explicitly modelled in these regional projections.
Source: Defra UK-AIR Modelled Background Pollution Data (PCM). Resolution: 1 km grid, aggregated to council on a population-weighted basis. Coverage: UK-wide.
Air quality uses Defra's annual mean PM2.5 (fine particulate matter) concentration. The WHO 2021 annual guideline is 5 µg/m³; the current UK legal target for 2040 is 10. Most UK councils sit between 6 and 12, with London boroughs and parts of the urban West Midlands at the top.
Council-level averages smooth out neighbourhood variation. Readings beside busy roads, in industrial areas, or downwind of major sources will typically be higher than the council figure shown.
Bands: Lower (below 6 µg/m³, ~30% of councils), Medium (6-8 µg/m³, ~60% of councils, the typical UK level), Higher (above 8 µg/m³, ~10% of councils).
Most UK councils exceed the WHO 5 µg/m³ guideline.
Source: British Geological Survey GeoClimate UKCP18 combined with the BGS Soil Parent Material Model. Resolution: 1 km soil grid combined with 2 km climate grid.
Subsidence risk assesses ground movement from clay shrink-swell under future climate conditions. The Soil Parent Material Model identifies the underlying soil type, and the GeoClimate layer models how soil moisture is expected to change through the 2021 to 2040 near-term period. Clay-rich soils expand when wet and contract when dry; that seasonal movement is what stresses foundations.
Bands map directly to BGS shrink-swell classifications: Probable (Higher), Possible (Medium), Negligible (Lower). The output reflects the geology and climate signal only; it is not a measure of past insurance claims, building age, foundation design, drainage condition or proximity to trees.
Coverage: England and Wales confirmed. Scotland and Northern Ireland coverage is being verified - BGS publishes several related shrink-swell products covering different parts of Great Britain, and LocalRisk is clarifying which specific products underpin the Scotland and Northern Ireland values returned by the API.
Based on soil composition, not claims history.
Sources: Environment Agency NCERM 2024 (England), Natural Resources Wales (Wales), NatureScot Dynamic Coast (Scotland). Northern Ireland not yet integrated.
For England and Wales the EA's National Coastal Erosion Risk Mapping (NCERM 2024) models shoreline change over 20, 50 and 100-year horizons under the "no active intervention" scenario. For Scotland, NatureScot's Dynamic Coast future erosion extents are used.
A postcode is flagged for coastal erosion risk when its centroid falls within 200 metres of a projected erosion zone, with the band reflecting how soon the modelled retreat is expected to reach that area. Postcodes that do not intersect any erosion zone are shown as Very Low.
Bands: Higher (fast erosion within the short-term horizon), Medium (medium-term), Lower (slow or negligible), Very Low (stable, accreting or inland).
Shoreline modelling is indicative; outcomes depend on future sea-level rise, storm intensity, and coastal management decisions.
Source: ONS Access to Gardens and Public Green Space (with Natural England ANGSt inputs). Resolution: LSOA (Lower Layer Super Output Area) mapped to postcode. Coverage: England only. Scotland, Wales and Northern Ireland are not covered because equivalent address-level access statistics are either not published or use a different methodology that is not directly comparable.
Green space resilience reports the percentage of addresses in the surrounding LSOA that sit within 200 metres of publicly accessible green space - parks, nature reserves, playing fields. It is a positive indicator (Higher is better), not a risk score, included because access to green space is linked to better health and to urban-heat buffering.
Bands: Higher (25% or more of LSOA addresses within 200m), Typical (10-25%), Lower (under 10%).
Private gardens and farmland are not counted; only formally classified, publicly accessible green space.
Coverage varies by data source. This table summarises which hazards are returned for each nation.
| Hazard | England | Scotland | Wales | Northern Ireland |
|---|---|---|---|---|
| Flood | EA NaFRA2 | SEPA | NRW | Not yet integrated (DfI Rivers) |
| Heat | UKCP18 | UKCP18 | UKCP18 | UKCP18 |
| Air quality (PM2.5) | Defra UK-AIR | Defra UK-AIR | Defra UK-AIR | Defra UK-AIR |
| Subsidence | BGS | Coverage being verified | BGS | Coverage being verified |
| Coastal erosion | NCERM 2024 | NatureScot Dynamic Coast | NCERM / NRW | Not yet integrated |
| Green space resilience | ONS / Natural England | Not covered | Not covered | Not covered |
Where a hazard has no integrated data for a nation, the postcode page will say so explicitly rather than display a placeholder band.
Each hazard converts its underlying numeric value into a plain-English band using either a national-percentile cut or a threshold defined by the source agency.
Recent weather data on postcode and council pages uses ERA5 reanalysis (2023 to 2025), produced by ECMWF and accessed via the Open-Meteo Historical Weather API. Temperature and soil moisture use ERA5-Land at ~9 km resolution; precipitation and wind use the ERA5 atmospheric model at ~25 km. ERA5 is a modelled area estimate, not a station observation.
Live conditions (current temperature, wind gusts, today's rainfall) use the Open-Meteo forecast API based on Met Office weather models. The air quality outlook uses CAMS atmospheric modelling. Live flood alerts come directly from the EA, SEPA and NRW APIs.
Long-term climate context in city insight articles uses Met Office station 1991-2020 normals for rainfall and sunshine hours, because ERA5 systematically overstates UK city rainfall (by approximately 35-45%) and sunshine hours (by approximately 28-39%) compared to station observations. This is a known limitation of using a coarse-grid reanalysis to describe city-scale climate.
Each source dataset is refreshed when its publishing agency releases a new version. The Environment Agency, SEPA, NRW, Defra, BGS, ONS and the Met Office set their own publication schedules; LocalRisk pulls new versions in within a few weeks of release. The version date is shown on every postcode page. UKCP18 projections were published in 2018 and remain the current UK Government national climate projection. The recent-weather (ERA5) window is refreshed daily.
LocalRisk is an area-level screening tool. Postcodes typically contain ten to a hundred addresses, and properties at different ends of the same postcode can face different exposure. Several hazards are published only at council or LSOA resolution and that wider value is displayed on the postcode page.
The English flood band reflects the highest property risk in the postcode under the EA's NaFRA2 aggregation method. A single high-risk address can label an entire postcode as Higher even when most homes are at Lower risk.
Northern Ireland flood data and coastal erosion data are not yet integrated. NI postcode pages explicitly say so rather than show a placeholder band.
Future climate projections rest on the Met Office UKCP18 RCP8.5 scenario at the 50th percentile. The 50th percentile is the central estimate within that scenario, not the most likely outcome across all scenarios. The full probabilistic range is wider in both directions.
LocalRisk does not constitute financial advice, an insurance assessment, a structural survey, or a regulated environmental search. Always verify your property using official sources like the EA flood service, SEPA, NRW or via a qualified surveyor for formal decisions.
LocalRisk takes UK government climate data that already exists and turns it into a plain-English risk check for any UK postcode. Flood data comes from the EA for England, SEPA for Scotland, and NRW for Wales; Northern Ireland uses a different system that LocalRisk has not yet plugged in. Heat and air quality data covers every UK postcode and comes from the Met Office and Defra. Subsidence data comes from the British Geological Survey. Coastal erosion data covers the coastlines of England, Scotland and Wales. Green space data is England only because the equivalent statistics are not published for the other UK nations.
The result is an indicator for the area around the postcode, not a property-level survey. If you are buying a house or making an insurance decision, use LocalRisk to get the lay of the land and then get the property-specific check that fits your situation. All of LocalRisk's data is free and public; the site exists to make it easier to read.