The no-sky line and what daylight distribution measures

The no-sky line is the contour on a room's working plane that divides the part of the floor able to see a patch of sky through the window from the part that cannot. Daylight distribution is the test built on that contour, and it is the third of the three daylight checks in BRE Report 209, Site Layout Planning for Daylight and Sunlight. If a proposed building opposite reduces the area of that working plane able to see sky to less than 0.8 times its former value, BRE 209 says the loss will be noticeable to the people living there.

That single ratio decides a lot of refusals in dense urban schemes. It is also the test most often skipped in early massing work, because it needs something the other two do not: the inside of someone else's flat.

Where the contour sits and why the working plane matters

The working plane is a horizontal reference surface inside the room, not the floor. BRE 209 puts it at 0.85 m above floor level in dwellings, roughly table height, and at 0.7 m in offices, roughly desk height. Every point on that plane either has a line of sight to some part of the sky dome through the window aperture, or it does not.

Draw a section. Take the top of the external obstruction, run a line over it to the window head, and continue that line into the room until it meets the working plane. Behind that intersection the sky is blocked by the obstruction and by the wall above the window. Plot the same intersection for every point along the window and you get a curve in plan. That curve is the no-sky line.

Raise the obstruction and the line moves towards the window. Raise the window head and it moves away. The room gets no say in the matter beyond its own geometry.

The proportional test applied to the lit area

BRE 209 does not set an absolute floor area that must see sky. It compares before with after. Measure the area of the working plane in front of the no-sky line in the existing condition, measure it again with the proposal in place, and divide.

Below 0.8 and the guidance treats the reduction as noticeable. Above it, the change is unlikely to be apparent to the occupant, whatever the absolute figures look like.

Illustrative example. A first-floor living room in a Victorian terrace is 4.2 m deep and 3.8 m wide, so the working plane covers about 16 m². With the existing low workshop opposite, 12.4 m² of that plane sees sky. The applicant proposes a four-storey block on the workshop site, and the modelled lit area drops to 8.7 m². The ratio is 0.70, which fails the guideline, and the assessment has to say so before arguing about context. The figures here are illustrative and any real room needs measured geometry.

Room depth and window head height

Two variables dominate. The higher the window head above the working plane, the deeper the sky penetrates for a given obstruction angle. And the deeper the room, the larger the proportion of floor that sits in the no-sky zone even before anything is built.

This is why a conversion of a deep warehouse floorplate into flats produces poor distribution results with no new obstruction at all. It is also why replacing a tall Georgian sash with a squat modern opening at the same cill level can push a room from pass to fail, though that change sits outside planning control in most cases.

A single-aspect room lit by one window at the short end is the hardest case. Add a balcony above the window and the effective head height drops again.

Where the contour test bites harder than a window plane measure

Vertical sky component is measured on the outside face of the window, so it knows nothing about what happens once light enters the room. A shallow room with a wide window can lose a lot of VSC and still keep almost all of its distribution. A deep single-aspect room can hold its VSC comfortably and still fail the 0.8 distribution test, because the loss all falls in the part of the plane that was marginal to begin with.

Where the two daylight tests disagree, say which one governs and why, rather than quoting the kinder number. Officers read a lot of these and they notice. Our note on reading the results tables of a daylight assessment goes through the columns in the order a case officer works down them.

Modelling inputs the assessment cannot do without

Distribution needs internal layouts for the neighbouring rooms: room depth, width, use, window positions, head heights and floor levels. On a site surrounded by private dwellings you will rarely have any of that.

The honest position is that assessors reconstruct interiors from external survey, rating lists, historic plans and, where nothing else exists, a mirror-image assumption drawn from the applicant's own building type. That assumption has to be stated plainly in the method section, because it carries most of the uncertainty in the result. We have set out the reasoning behind that baseline in a separate piece on mirror-image baselines and existing obstructions, and the way room use assumptions change results is worth reading alongside it, because a room recorded as a kitchen is treated differently from one recorded as a living room.

The other inputs are easier. Obstruction massing to measured survey, ground levels to a common datum, and the proposed building modelled to its outer face rather than its structural grid.

Reporting it without burying it

Put distribution in the same table as VSC, room by room, with the ratio to two decimal places and a pass or fail against 0.8. Do not report it as a single building-wide average. The test is a room test.

Where rooms fail, the assessment should then do the work that BRE 209 invites: explain the context, the existing condition, the density expected by the adopted local plan, and whether the affected room is a main habitable room or a circulation space. BRE 209 states its numerical guidelines are advisory and should be applied flexibly in different settings, and the London Plan 2021 policy D6 asks for daylight and sunlight to be considered against the character of the area rather than against a single national number. Flexibility is not an excuse to skip the calculation. It is the argument you make after the calculation has been done properly.

If the massing is still moving, run the obstruction angle screen first and only build the full model once the envelope settles. The approach is covered in our piece on obstruction angles and the early massing test.

A full Daylight and Sunlight Assessment draft from PlanScope sets out the method, the tests and the reporting structure, and flags every input that still needs measured survey data with a "to be confirmed" marker rather than filling the gap with an assumption. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. For specialist technical reports, qualified professional input may be required.

This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it.