Reflectance assumptions and how they move a daylight result

Two daylight assessments of the same room can report results a third apart without either author making an arithmetical mistake. The difference sits in the reflectance assumptions: how bright the walls, ceiling and floor are assumed to be, how much light the building opposite bounces back, and what the ground in between is taken to be. Those values are inputs the author chooses, they are rarely visible in the summary tables, and they move the answer more than most reviewers expect.

Where reflectance enters a daylight calculation

Not every daylight test is affected, which is the first thing to establish when reading a report.

The vertical sky component is purely geometric. It measures the proportion of sky visible at the centre of a window from a vertical plane, so no reflectance value changes it. The same is true of the no sky line, which maps where a patch of sky is visible at working plane height. If a report's neighbour assessment rests on those two tests and the retained ratio approach described in the 0.8 rule for retained daylight, reflectance is not the argument to have.

Reflectance matters where interior illuminance is being predicted. The average daylight factor formula in the BRE guidance divides by one minus the square of the area weighted average room surface reflectance, so the assumed brightness of the room appears directly in the denominator. Climate based daylight modelling, which BS EN 17037 and the current BRE guidance both use for assessing daylight inside new dwellings, is more sensitive still, because it simulates interreflection between every surface in the model over a full year of sky conditions.

That split explains a pattern in submissions. Neighbour impact sections tend to be reflectance free and hard to argue with. Sections on daylight inside the proposed flats, of the kind discussed in daylight inside new flats compared with neighbours, are where the assumptions do the work.

The default values and where they come from

A conventional set of interior reflectances has settled into use through CIBSE lighting guidance and the European daylight standard: around 0.7 for the ceiling, 0.5 for walls and 0.2 for the floor. Ground outside is usually taken at 0.2. External obstruction surfaces default to 0.2 as well, on the basis that brickwork, weathered concrete and dark cladding sit in that region.

Those numbers describe a light, freshly decorated room with a mid tone carpet. They are reasonable for a generic assessment and they are the values a reviewer will assume were used unless told otherwise.

The trouble starts when a scheme's actual design departs from them in one direction while the model stays at the defaults, or the reverse. A dark stained timber floor sits nearer 0.1. Exposed fair faced concrete soffits are well below 0.7. Dark grey feature walls, which appear in a great many approved interior schemes, pull the area weighted average down sharply because walls dominate the surface area in a normal room. A model at default reflectance predicting a room that will be finished dark is optimistic, and an experienced consultee knows it.

External obstructions and the surfaces you cannot control

External reflectance is the assumption most worth interrogating, because it concerns surfaces the applicant may not own.

A pale render or light stone facade 12 m away returns considerably more light into a room than a dark brick one. So does a fully glazed commercial frontage, whose behaviour a single diffuse reflectance value describes only crudely. Where a model has been set with obstruction reflectance at 0.35 or higher, the result depends on a neighbouring building staying that colour, and nothing in the planning permission secures that.

The defensible approach is to model the obstructions at the conventional 0.2, then, if the geometry does benefit from a bright facade opposite, present that as a sensitivity run rather than as the headline result. Doing it the other way round invites an officer to discount the whole assessment.

Ground reflectance deserves a sentence too. Pale paving in a courtyard does raise illuminance in lower storeys, but courtyards get wet, dirty and shaded, and a value above 0.2 needs a reason.

Glazing transmittance, frames and maintenance

Transmittance is not reflectance, but it sits in the same part of the calculation and gets the same loose treatment.

Clear double glazing has a diffuse visible transmittance in the region of 0.68 to 0.7. Triple glazing is lower. Solar control coatings, which the overheating strategy may require to satisfy the cooling hierarchy, can drop transmittance to around half that, and a scheme that specifies solar control glass in the energy assessment while modelling clear double glazing in the daylight assessment is internally inconsistent. Check the two documents against each other.

Two further reductions apply. The frame occupies part of the opening, so the effective glazed area is smaller than the structural opening, and a maintenance or dirt factor accounts for accumulated soiling. Where a report states a single transmittance figure with no frame or maintenance correction, ask which of them is included.

How far the result moves

The published average daylight factor formula lets the sensitivity be demonstrated without any modelling software. This is an illustrative calculation, not a result for any real room.

Take a room where everything except reflectance is fixed. At an area weighted average reflectance of 0.4, the denominator term is one minus 0.16, so 0.84. At 0.6 it becomes one minus 0.36, so 0.64. Dividing by the smaller number raises the predicted daylight factor by a little over 30 per cent. A room reported at 1.5 per cent average daylight factor on the higher assumption falls to about 1.15 per cent on the lower one, which in a borderline case is the difference between a compliant table and a refusal reason.

Climate based modelling behaves in the same direction and is harder to sanity check by hand, because the interreflection is calculated rather than approximated. The order of magnitude is similar: interior reflectance and obstruction reflectance are usually the two most influential non geometric inputs in the model.

Disclosing the values you used

The fix is unglamorous and takes half a page. Every daylight report should carry an assumptions schedule listing interior surface reflectances, external obstruction and ground reflectance, glazing transmittance, frame factor, maintenance factor, working plane height, grid spacing, and the climate file where climate based modelling has been used.

Say where each value came from: a standard, a published guide, or the architect's material specification. Where a value departs from convention because the design justifies it, show the specification clause that supports it. And where a room will be sold with an unknown finish, use the conventional values rather than the ones the marketing render suggests.

Rooms behind an assessment also need identifying by plan reference and use, since the applicable target depends on whether a space is a bedroom, a living room or a kitchen, and a north facing room has a different starting point again, as daylight to a north facing habitable room sets out.

How a reviewer tests an optimistic model

If you are checking someone else's assessment, four moves find most of the problems.

1. Ask for the assumptions schedule. Its absence is itself informative, and a report that cannot state its reflectances has not been through internal review.
2. Compare the reflectances against the architect's material palette and the interior finishes on the drawings, then against the glazing specification in the energy and overheating documents.
3. Ask for a sensitivity run at conventional values, and at 0.2 obstruction reflectance where a higher figure has been used. A well run study will already have one.
4. Check which tests actually depend on the assumptions. Where the neighbour case rests on vertical sky component and no sky line, the reflectance argument is a distraction, and the geometry is where the scrutiny belongs.

None of this requires re-running the model. It requires the author to show their inputs, which is a reasonable thing to ask of any technical document that a decision will rest on.

PlanScope drafts a Daylight and Sunlight Assessment structured to the tests councils apply, with the modelling inputs listed as fields to be completed rather than invented, so the assumptions schedule exists from the start. Councils normally expect this kind of assessment to be authored or reviewed by a suitably qualified specialist, and modelling itself remains their work. Reports should be reviewed before submission, and PlanScope does not guarantee planning approval.

This is general guidance, not legal advice. Check the current BRE guidance and any local policy requirements with your local planning authority or a qualified specialist before relying on it.