Modelling the existing context in a daylight assessment
Most daylight disputes are arguments about the baseline model rather than the calculation. A reviewer's guide to survey sources, assumed room layouts, consented schemes and auditing someone else's geometry.
Most disputed daylight results are not disputes about the calculation. They are disputes about the daylight model existing context: which buildings were included, how tall they were assumed to be, where the neighbouring windows were placed, and whether a consented but unbuilt scheme next door was in the baseline. Get the context wrong and every vertical sky component figure in the report is wrong by the same amount, in the same direction, and usually in the applicant's favour.
So the first thing to audit in any daylight and sunlight assessment is the model, not the results table. A reviewer who checks the geometry and the stated assumptions can usually tell within an hour whether the numbers are defensible.
Why the context model drives the result
Vertical sky component and the no-sky line are geometric measures. They respond to the angle and the extent of obstruction seen from a reference point, so a neighbouring parapet modelled 1.5 metres too low, or a rear projection left out altogether, shifts results across a whole elevation.
Annual probable sunlight hours behave the same way and are more sensitive to the south of the window. A single omitted chimney stack or a misplaced boundary wall can change a winter sunlight figure enough to move a window from pass to fail.
This matters because the retained figures are then compared against the guidance in BRE 209, Site Layout Planning for Daylight and Sunlight, and against whatever local policy applies. The comparison is only as good as the baseline it starts from.
Survey sources and how far to trust them
Four sources get used, and they are not interchangeable.
A measured topographical survey with elevations is the strongest. It gives you real ridge and eaves heights, real window positions on the surveyed elevations, and a coordinate system you can rely on. On a constrained urban site it is worth the fee.
National mapping data with building heights is a reasonable starting point and nothing more. Height attributes are derived, not measured, and they describe the building envelope rather than the facade detail. Aerial survey point clouds sit between the two: good for massing, weak for window reveals and for anything under a canopy.
Planning drawings for neighbouring buildings are useful for window positions, but check the drawing is as-built rather than as-applied, because schemes get amended under section 73 and section 96A.
The honest approach is to mix sources and say so. State for each surrounding building whether its geometry came from measured survey, mapping data, drawings or photographic interpretation, and flag the ones where you have the least confidence. Reviewers accept uncertainty that is declared. They do not accept a model that presents interpolated heights as surveyed ones.
Neighbouring windows and assumed room layouts
You can almost never get inside neighbouring properties, so internal layouts are assumed. That assumption needs to be visible.
Record the window head, cill and width from the survey or from rectified photographs. Then state the assumed room depth, the assumed use of the room, and whether a window is assumed to serve one room or two. A deep open plan layout assumed behind a window will produce a worse no-sky line result than a shallow room, so the assumption is not neutral and should be justified rather than chosen.
Where a property has a conservatory, a rear extension or a rooflight, model it. Where a window appears to be obscured glazed, a bathroom, or serving a stairwell, say which rooms you have excluded from assessment and why. BRE 209 treats non-habitable rooms as less sensitive, and that exclusion is legitimate, but only when it is stated.
| Context element | Usual source | Confidence |
|---|---|---|
| Ridge and eaves heights facing the site | Measured survey | High |
| Heights of buildings behind the first row | Mapping data | Medium |
| Neighbouring window positions | Survey or rectified photographs | Medium to high |
| Internal room depths | Assumed, stated in the report | Low |
| Consented unbuilt schemes | Planning drawings | Medium |
Consented but unbuilt schemes
This is the question that most often changes the answer. If a neighbouring site has permission for a taller building, is the baseline the world as built or the world as consented?
The defensible approach is to report both. Assess against the existing built context, because that is the amenity people currently enjoy, and then run a cumulative scenario including consented schemes that are reasonably likely to be built. Report the two side by side and explain which permissions you included, with application references and dates.
Choosing only the consented scenario looks like an attempt to use someone else's permission to lower the bar. And choosing only the existing scenario ignores a real prospect and leaves the officer with no cumulative picture at all. Where a consent has lapsed or is unlikely to be implemented, say that, and give the reason.
For schemes where the surrounding massing is still emerging, the cumulative run matters more than the absolute figures. That is particularly true of courtyard schemes where lower floors depend on the eventual height of the opposite block.
Trees, boundary walls and things that move
Deciduous trees are generally not treated as permanent obstructions in BRE 209 terms, because they are seasonal, may be removed, and their canopies are irregular. Dense evergreen screening is different and can reasonably be modelled where it is clearly permanent.
Boundary walls, fences, garden structures and temporary site hoardings need a judgement. A one metre brick wall rarely matters to a first floor window and can matter a great deal to a basement or lower ground floor one. Where a tree is protected by a Tree Preservation Order, note it, because its permanence is then a planning fact rather than an assumption, and the tree survey for the application will record it anyway.
The rule to apply is simple. Model anything whose removal would require consent or would be improbable, and exclude anything that could lawfully disappear next week, with a sentence explaining each exclusion.
Reflectance and material assumptions
Surface reflectance has no effect on vertical sky component, which is purely geometric, and a large effect on illuminance based methods such as those in BS EN 17037, where internal and external surface reflectances feed directly into the result.
Declare the values used for ceilings, walls, floors, external ground and facade materials, and say where they came from. Light coloured courtyard walls modelled at an optimistic reflectance will lift results in exactly the rooms that are most at risk, so this is the assumption a reviewer should test first. The effect of those choices is covered in more detail in the note on reflectance assumptions and daylight results.
Auditing someone else's model
When you are the reviewer rather than the author, work in this order.
- Ask for the model file and the survey it was built from, not just the report.
- Check three measurable heights in the model against the survey or against a rectified photograph.
- Confirm that every building shown on the site plan within the likely zone of influence exists in the model.
- Read the assumptions schedule, and list every assumption that is favourable to the applicant.
- Re-run one or two failing windows with the opposite assumption to see how sensitive the result is.
- Ask what consented schemes were considered and which were excluded.
If the assumptions schedule is missing, that is the finding. A daylight report without a stated list of modelling assumptions cannot be checked, and an officer is entitled to ask for it before determination. The same applies to the outputs: there is a short note on what to include in daylight modelling outputs that sets out the minimum a reviewer needs.
PlanScope can produce a structured Daylight and Sunlight Assessment draft that sets out the method, the assumptions schedule and the reporting framework, with flags where site specific facts are still needed. The modelling itself remains the work of the assessor. 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.
Frequently asked questions
Should consented but unbuilt schemes be in the daylight baseline?
Report both. Assess against the existing built context, then run a cumulative scenario adding consented schemes that are reasonably likely to be built, listing the application references used. Presenting only the consented scenario looks like borrowing a neighbour's permission to lower the test, and presenting only the existing case gives the officer no cumulative picture.
How accurate does the context model need to be?
Buildings facing the site should come from measured survey where access allows, because an error of a metre in parapet height shifts results across a whole elevation. Buildings behind the first row can come from mapping data. What matters is that the report states which source was used for each element and where confidence is lowest.
Are trees included in a daylight model?
Deciduous trees are generally excluded as obstructions because they are seasonal and may be removed. Dense evergreen screening that is clearly permanent can be modelled, and a tree protected by a Tree Preservation Order has a stronger claim to inclusion. Either way, state which vegetation was modelled and which was left out.
What should an assumptions schedule contain?
The source of each building's geometry, assumed neighbouring room depths and uses, rooms excluded as non-habitable, surface reflectances for internal and external surfaces, the treatment of vegetation and boundary structures, and the consented schemes included or excluded with reasons. Without it the model cannot be checked.
Does surface reflectance change vertical sky component?
No. Vertical sky component is a geometric measure of visible sky and is unaffected by reflectance. Illuminance based methods such as those in BS EN 17037 are affected, sometimes considerably, so reflectance values must be declared whenever those metrics are reported.
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.