Assessing daylight to a neighbour's rear extension rooflight
Rooflights serving a neighbour's single storey extension are the hardest daylight case in a terrace, because the standard vertical window tests do not transfer.
Daylight to a neighbour's rooflight cannot be assessed with the vertical sky component, because the VSC method in BRE 209 is constructed for vertical glazing and its 27 per cent reference value has no equivalent for a horizontal aperture. A rooflight sees a large part of the sky dome before anything is built. The right approach is the daylight distribution test, supported where the case is finely balanced by illuminance modelling to BS EN 17037.
This comes up constantly on terraced and semi-detached streets. The neighbour has a single storey rear extension with two or three rooflights over a kitchen-diner, and your client wants a two storey extension on the shared boundary.
Why horizontal glazing behaves differently
A vertical window takes light from a hemisphere split in half by the wall it sits in. Half the sky is behind the wall and unavailable. The 27 per cent VSC reference in BRE 209 is calibrated to that geometry.
A horizontal rooflight faces the whole sky dome. Its unobstructed sky access is several times that of a vertical window of the same area, which is precisely why architects use them in deep plans. It also means a nearby obstruction removes a smaller proportion of the available sky than the same obstruction would remove from a vertical window.
The converse matters too. Because the rooflight starts with so much, a percentage loss that would be alarming for a window can leave the room adequately lit. And because the room below is usually deep, the loss is concentrated at the far end of the plan where the light was already weakest.
So the geometry cuts both ways, and the assessment has to measure rather than assert.
Which BRE 209 test applies to a rooflight
Work through it in this order.
- Daylight distribution, meaning the no-sky line. Plot the area of the working plane that can see the sky through the rooflight before and after. BRE 209 treats a reduction in the lit area to less than 0.8 times its former value as noticeable, and that ratio transfers to a rooflight without modification because it is a geometric test rather than a calibrated one.
- Average daylight factor, where the room is served only by the rooflights and the question is whether the space remains usable. BRE 209 cites reference values by room type for this.
- Illuminance modelling to BS EN 17037, using climate data, where the first two tests are marginal or where the authority asks for a method that handles horizontal glazing properly.
- Sunlight, through annual probable sunlight hours, only where the obstruction lies within 90 degrees of due south of the aperture.
Do not report a VSC figure for a rooflight. Officers who know the standard will query it, and officers who do not will give it a weight it cannot carry. Explain in a sentence why it has been omitted.
Measuring obstruction above a flat roof
The survey is the awkward part. You need the rooflight positions, their size and their height above the flat roof, plus the height of your proposed flank wall and its distance from each rooflight.
Three measurements decide the result.
The distance from the shared boundary to the nearest edge of the rooflight. The height of the proposed wall above the plane of the rooflight glazing, which is not the same as its height above ground. And the length of the proposed wall along the boundary, because a short projection removes far less sky than a full-depth one.
Where access to the neighbour's roof is refused, the rooflight positions can usually be established from the planning history. The neighbour's own extension will have been granted permission or built under permitted development, and the approved drawings are on the council's website. Say in the report that this is the source and that dimensions are scaled rather than measured.
A measured survey of the shared boundary and your own roof level, combined with scaled drawings of theirs, is a defensible basis. An assumption with no stated source is not.
Habitable room status of the space below
BRE 209 directs attention to main living rooms and kitchens, with bedrooms treated as less important. A kitchen-diner under rooflights is squarely within scope.
Two complications arise. First, the space may be open plan, with the rooflit part forming one end of a room that also has rear doors or a window. Where a room is served by more than one aperture, the assessment has to account for all of them, which we look at in when a neighbour's room is served by two windows. A rooflight and a set of rear doors together can leave a room acceptably lit even where one of them is substantially obstructed.
Second, the space may be a utility, a corridor or a bathroom, in which case BRE 209 gives it little weight. Establish the use from the approved drawings rather than guessing from the rooflight positions.
A worked illustrative case on a terrace
Treat this as illustrative only.
A Victorian mid-terrace has a neighbour's 4 metre single storey rear extension with two 1.2 by 0.8 metre flat rooflights set 1.5 metres from the shared boundary, serving a kitchen-diner that also has bifold doors at the rear. The proposal is a two storey rear extension, 3.5 metres deep, built to the boundary, with its flank wall standing 2.7 metres above the neighbour's flat roof.
The nearer rooflight loses sky from the whole of one side. The daylight distribution test on the working plane shows the no-sky area increasing along the boundary wall, with the far corner of the kitchen most affected. The bifold doors continue to light the rear third of the room, so the retained lit area stays above the 0.8 ratio. The further rooflight is barely affected, because the obstruction is angled away from it.
Set against that, a design that steps the flank wall in by 600 mm for the rear 1.5 metres of its depth, or reduces the depth to 3 metres, moves both rooflights clearly into compliance. That is the conversation to have with the client before the application rather than after a refusal.
The same geometry applied to a bungalow produces different answers, and we work through that in daylight to a single storey rear extension on a bungalow.
What to offer where the result is marginal
Design changes beat argument, and there are only a few that work.
- Step the flank wall away from the boundary for the rear part of its depth, which is where the obstruction angle is steepest relative to the rooflights.
- Reduce the depth of the upper storey while keeping the ground floor as drawn.
- Introduce a pitched or sloping profile to the flank elevation so the obstruction falls away towards the rear.
- Where neither is acceptable to the client, offer the neighbour an additional rooflight or a widened existing one, carried out at your client's cost under a side agreement. Councils cannot require this, and some officers welcome it as a material offer.
Garden amenity is a separate question and belongs in an Overshadowing Study rather than in the window analysis.
Recording the assessment for the application
The report needs to be readable by a case officer in ten minutes.
State the method and why VSC has been omitted. Give the source of every dimension, separating measured from scaled. Include a plan showing the no-sky line before and after for each affected room, and a section through the rooflight and the proposed wall. Tabulate the results room by room with the ratio against the former value. Then state plainly whether the standard is met, and where it is not, by how much.
A Daylight and Sunlight Assessment that does this is hard to argue with. The design rationale for any stepping or set-back belongs in the Design and Access Statement, cross-referenced rather than repeated.
BRE 209 is guidance, and the Framework and local plan policies on living conditions are what the decision is made under. Say so in the report, because an officer weighing a marginal breach needs to know the standard is advisory rather than a threshold.
PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. This is general guidance rather than legal advice, so check the position with your local planning authority or a qualified professional before relying on it.
Frequently asked questions
Can I use the vertical sky component on a rooflight?
No. The VSC method in BRE 209 is built for vertical glazing and its 27 per cent reference value has no horizontal equivalent, because a rooflight faces the whole sky dome rather than half of it. Use the daylight distribution test, the average daylight factor where the rooflight is the only source, and illuminance modelling to BS EN 17037 for marginal cases.
Which test does apply to a neighbour's rooflight?
Daylight distribution, also called the no-sky line. Plot the area of the working plane that can see sky through the rooflight before and after the obstruction. BRE 209 treats a reduction to less than 0.8 times the former lit area as noticeable, and that ratio transfers because it is a geometric test rather than a calibrated reference value.
What if I cannot get onto the neighbour's flat roof to survey it?
Use the approved drawings for their extension, which will be on the council's website from the original application or prior approval. Combine those scaled dimensions with a measured survey of the shared boundary and your own roof levels, and state clearly in the report which dimensions are measured and which are scaled.
Does it matter what room is under the rooflight?
Yes. BRE 209 directs attention to main living rooms and kitchens and treats bedrooms as less important, with utilities, bathrooms and circulation given little weight. Establish the use from the approved drawings rather than inferring it, because the weight the officer gives the result depends on it.
What design change helps most when the result is marginal?
Stepping the flank wall away from the boundary for the rear part of its depth, because that is where the obstruction angle is steepest relative to the rooflights. Reducing the depth of the upper storey only, or sloping the flank elevation so the obstruction falls away towards the rear, are the other two that reliably change the numbers.
Sources and further reading
This article is general guidance about the planning system in England and, where stated, the rest of the UK. It is not legal or professional advice. Check your local planning authority’s validation requirements and take professional advice for your specific site.