Explainer: The New European daylight standard - for architects and planners

Brian T. O’Brien MRIAI first published in Architecture Ireland v336 July/August 2024

20th August 2024

For architects and planners involved in the design of larger-scale housing, an awareness of the implications of the new European standard on daylight, EN 17037 (2018), and the issues that arise as it is applied in Irish practice, is important.

In broad terms, EN 17037, covers three aspects that are relevant to new housing: daylight, sunlight, and view. The sunlight aspects are also applicable to schools and hospital. EN 17037 is now the international standard. In Ireland it was adopted unchanged as IS EN 17037, whereas in Britian changes to harmonise it with previous standards were attached, and it became BS EN 17037 there.

Daylight

In day-to-day practice, IS EN 17037 will have most influence on the question of daylight provision within a new apartment or house. BRE 209 Site Layout Planning for Daylight and Sunlight, 2022 will continue to be the standard for sunlight analysis onto and outside a building (including on neighbours). The EN’s relevance is underpinned by the new Irish national Sustainable Urban Housing: Design Standards for New Apartments (2022), which state that designs should have regard to EN17037 (as well as BRE 209).

So practice in Ireland is now becoming that new housing developments provide daylight, and usually sunlight, analyses using the methodologies of the EN (below) with planning applications. The intention was that they should also adhere to it’s targets but for reasons that will become clear below, many now suggest that targets in the UK version BS EN 17037 should be used for housing projects here in Ireland also, and that analyses against the IS-EN17037 requirements should only be supplied as information. This approach is also rooted in the establised practce prior to the standard, where the language in the accompanying reports made the point that the daylighting targets were advisory only and were not meant to be taken as an instrument of planning policy – usually to justify that some rooms in a development may not reach the levels recommended.

The first significant change mandated by both versions of 17037 on daylight (aka ‘spatial daylight autonomy’ or SDA) is around methodology. Daylight analysis must now be done using validated simulation software (under the previous BS 8206-2, manual methods could be used).

In terms of changes to the daylight performance targets themselves, the previous BS 8206-2 set the requirement as an average illuminance (daylight factor) over the whole space while EN 17037 requires that spaces be provided with both a minimum level of daylight over the vast majority of its area, and also that a ‘target’ (i.e. higher) level of daylight be provided over no less than half of it.

The evaluation must take account of space, time, and using a 3D computer model of the proposed building, the spatial layout inside and the context outside. Time and climate, in the form of daylight intensity through time, are considered using location-specific climate data. The new standard allows for the calculation of daylight levels expected to occur at such a latitude (using a historic weather file in the software) but newly also now allows the use of a file that is an actual record of the historical climate for that specific site if available.

The new analysis procedure sets out a matrix of analysis points across the plan of the room (ignoring a 500mm band along the perimeter) at a height of 850mm over floor level. The study model must include the proposed building’s geometry (facades, roofs, floors) outside and – for any residential or habitable spaces – inside, as well as the window geometry and specification. Reflectances for the materials must also be accounted (to procedures in ISO 15469: 2004).

The results should be in tabular format and ideally also in plans overlain with the relevant figures. Outputs must show what percentage of the room area meets the target and minimum daylight factor thresholds and confirm that these are above the level of performance required.

So what level of daylight is required by EN 17037 ?

The standard sets three illuminance ‘levels of recommendation’: minimum, medium, and high, and while referring to these three levels, it states that the minimum recommendation level of daylight should be provided. Illuminance recommendations have been made for all buildings (regardless of function) across Europe, set out as absolute lux levels rathar than as daylight factors (which are relative) and the fact that sky luminance varies across the EU is accounted for by the standard translating these absolute lux requirements into different daylight factors targets for different EU countries.

For those of us used to BS 8206, it is counter intuitive that the EN does not set different illuminance targets for different room functions(bedoom, kitchen etc. It was for this reason when introducing 17037 into the UK, the comittee there attached a national ‘annex’ to it overwriting its targets with ones that are i) lower and ii) specific to each room function. In effect thay reset the targets at those of the previous BS standard while transposing the rest of the new European standard into force (unchanged otherwise) as BS EN 17037.

Translating the EN’s requirements to Ireland gives us the following requirements:

Applied to a notional residential space, that has vertical windows only, aiming to meet EN 17037’s minimum level of recommendation in Dublin, one would need to provide a daylight factor of 2% over half of its floor area, and no less than 0.7% over 95% of it, for half the possible annual daylight hours (for both metrics) of the year.

So what of the alternative and easier to meet BS EN 17037 ? The surprising fact that no distinction is made between building or room function in the targets in EN 17037 has led to the situation where most analysts, and some guidance suggest it is not suitable for usage in dwellings here. Dublin City Council’s own development plan for instance and a number of FI requests this author has seen from other planning authorities, suggests planners accept that it is unsuitable for housing in Ireland and that BS 8206 and BS EN 17037 ‘have relevance’ -indicating it would seem that the target illuminances in the latter (BS EN17037) are acceptable.

These targets are as follows:

Only time will tell whether these lower requirements for bedrooms and living rooms in the BS EN, that we have long relied on gain traction as accepted practice (albeit using the new simulation procedures) or whether they are superceded by the requirements of the new EN which would make developing dense housing more difficult if applied rigidly by planners.

The other recent development is the high court judgement that for rooms combining kitchen, dining, or kitchen, dining and living – as is common in new apartment designs, the higher illuminance level, that of a kitchen at 2% DF, must be followed for the whole combined space.

In terms of assessing the daylight impact on neighbouring dwellings where we can’t include their internal layout in our computer models, the vertical sky component method in BS 8206-2, applied to their windows, is still to be used.

Sunlight

The second area where EN 17037 comes into contact with design is in the area of sunlight exposure (SE), a new term that supplants previous ones like overshading, solar availbility etc.

On sunlight, at least one habitable space per dwelling must meet the criterion. The previous BS 8206-2 referred to annual probable sunlight hours and set a minimum target for the new dwelling while putting a limit on the reduction a neighbouring building could be subjected to.

The new EN (both the European/ Irish and the UK versions) take the simpler approach of combining a minimum number of hours when sunlight must be able to enter a room (assuming a clear sky) and a reference day of the year on which this must be able to happen. This approach can be applied both to the new building and used as a way to evaluate whether the impact on a neighbour’s sunlight is ‘noticeable’ and thus presumably acceptable or not.

The minimum requirement in EN 17037 is for 1.5 hours of sunlight on a selected date between 1 February and 21 March. As the sun is higher in the sky in March than in February, the later date is becoming our de facto selected date. EN 17037 also defines a medium target (three hours) and a high target (four hours). The sunlight can be at any time of the day and does not need to be continuous over the 24 hours of the reference date.

In terms of the evaluation itself, EN 17037 allows for both manual evaluation using sunpath diagrams as well as simulation with computer models. With the software approach, specific points on a computer model are tested against location-specific solar geometry information. The model should include the building’s external geometry (facades, roofs, and windows) and that of the context, but as the sunlight exposure evaluation is carried out at the window of a dwelling, internal floors and walls don’t need to be represented.

For neighbour impact tests, The same annual probable sunlight hours methodolgy is used and the same evaluation as before is appled. This comprises a cascading set of tests: post development, does the neighbour still receive 25% of annual and 5% of winter sunlight, if not does it still receive 80% of its previous sunlight and the annual loss is not more than 4% ? - all of which would be deemed ‘notceable’.

There is more to say about view, which is also covered in the standard but we will leave that for another article.

If you need a daylight or sunlight analysis for your housing or school project, please get in touch at brian@solearth.ie or 087 2357530

Brian T O Brien MRIAI M.Arch Berkeley