The weather file for daylight modelling is one of the most important — and most overlooked — decisions in a modern daylight assessment. Under the 2022 BRE guidance, internal daylight to new homes is judged with climate-based daylight modelling (CBDM), and CBDM cannot run without a year of hourly weather data. Choose the wrong file and a compliant room can read as a failure, or a genuinely poor room can look acceptable.
This guide explains what a weather file actually is, why the 2022 rules made it central to daylight reports, how to pick the right one for a UK site, and where assessors go wrong. If you are commissioning or reviewing a report, these are the questions worth asking.
Why weather data suddenly matters for daylight reports
Before 2022, internal daylight to new dwellings was usually checked with the average daylight factor (ADF) — a single geometric calculation that took no account of a building's actual location or climate. The third edition of BRE's Site layout planning for daylight and sunlight (BR 209, 2022) replaced ADF with the two provision methods set out in BS EN 17037. The headline method is spatial daylight autonomy (sDA), which asks how often a point in a room reaches a target illuminance — typically 300 lux — across the year.
"Across the year" is the key phrase. To know how often a room hits 300 lux, the model needs to know how bright the sky is, hour by hour, at that place. That information comes from a weather file. In practical terms, the 2022 switch turned daylight modelling from a fixed sum into a climate simulation — and made the input data a live variable in whether a scheme passes. For the wider picture, see our explainer on climate-based daylight modelling, sDA and daylight autonomy and our summary of the BRE 2022 daylight provision tests.
What a weather file actually is: TRY, TMY and EPW
A weather file is a table of hourly measurements for a full year at a given location — global and diffuse solar radiation, cloud cover, temperature and more. Because no single real year is "typical", meteorologists build a composite year from many years of records:
- Test Reference Year (TRY) — a composite of the most typical months, widely used in the UK and issued for defined locations by CIBSE.
- Typical Meteorological Year (TMY) — the international equivalent, assembled on the same principle.
- EPW (EnergyPlus Weather format) — not a type of year but a file format. Most daylight engines read EPW, so a UK TRY is usually supplied packaged as an EPW file.
So when an assessor says they used an "EPW file", the meaningful question is which underlying dataset and which location it represents. A file labelled for one city can be built from a weather station many miles away, with a noticeably different sky-brightness record.
Which location file should you use?
BS EN 17037 and BR 209 are clear in principle: use the weather data closest to the site's location. In practice that means:
- Pick the nearest representative station, not simply the nearest big-city file. A coastal site and an inland site an hour apart can have measurably different diffuse-light regimes.
- Match latitude sensibly. Northern Scotland receives markedly less winter daylight than the south coast; using a London file for a Highlands scheme flatters the result.
- Record the provenance. A defensible report states the exact file name, source and year-set used, so a local authority's reviewer can reproduce the run.
Urban sites add a further wrinkle: the weather file describes the sky, but surrounding buildings block much of it. That obstruction is modelled geometrically, on top of the climate data — which is why context massing matters as much as the file itself.
How the wrong file changes a pass or fail
Because sDA is a threshold test — the proportion of the room reaching 300 lux for at least half the daylight hours — small shifts in modelled sky brightness move rooms across the line. Two reports on the same building can disagree simply because one used a sunnier location file, an older dataset, or different assumptions about surface reflectance and glazing transmittance fed alongside the weather data. This is exactly why planning officers and neighbouring objectors increasingly scrutinise the inputs, not just the headline percentages. If you have received two conflicting assessments, our guide to the core daylight and sunlight metrics helps you compare like with like.
Common weather-data mistakes to avoid
- Silent defaults. Software often ships with a single national file selected by default; using it for every site regardless of location is the most frequent error.
- No stated provenance. If the report does not name the weather file, its sDA figures cannot be checked or reproduced.
- Mismatched inputs. A precise weather file paired with optimistic reflectance or glazing values undermines the whole simulation.
- Ignoring the alternative method. BS EN 17037 also allows a daylight-factor target route; the two methods are not interchangeable and should not be mixed within one assessment. See our overview of BS EN 17037 for new UK homes.
How Fortress Associates can help
We prepare daylight and sunlight reports to BRE BR 209 (2022) and BS EN 17037, and we document our climate-based daylight modelling in full — the weather file, its source and the reflectance and transmittance assumptions behind every sDA figure — so your local planning authority can rely on the result. Most reports are delivered in four to five working days, with no advance payment. If you are unsure whether your scheme needs CBDM or a simpler assessment, get in touch and we will advise on the right approach. You can also see the full range of what we offer on our services page.
Sources & further reading
Need help with a UK planning project?
Fixed-fee daylight reports and Building Regulations drawings — delivered in 4–5 working days. No advance payment.
Request a free quote