Overshadowing gardens: the two-hour test on 21 March

Overshadowing garden planning arguments are settled by one test in BRE 209: at least half of an amenity area should receive at least two hours of sunlight on 21 March. Where a proposal cuts the sunlit area below 0.8 times what it was before, the loss is likely to be noticed by the people using the space.

Two hours. Half the garden. The spring equinox. Those three numbers decide most rear extension and infill objections that turn on a garden rather than a window.

What the sun on ground test measures

The test looks at the ground rather than at a window. A model of the site and its surroundings is used to work out, for each point on the amenity area, how many hours of direct sunlight reach it on the assessment date.

Points receiving two hours or more are counted as sunlit. The sunlit area is expressed as a proportion of the whole amenity area, and the guideline is that at least half the area should qualify.

Where an existing space already falls below half, the second part of the test applies. The area that remains sunlit after the development is compared with the area sunlit before, and a fall to below 0.8 times the former value is treated as a noticeable loss. That ratio is what catches proposals next to gardens that were already partly shaded.

Both parts are guidance rather than law. BRE is explicit that the numerical guidelines are advisory and should be applied flexibly according to the context of the site.

Why 21 March

The spring equinox is chosen because it is representative rather than extreme. On 21 March day length is close to twelve hours across the country and the sun's path is midway between the June and December extremes.

Assessing on 21 June would flatter almost every scheme, because the sun is high and shadows are short. Assessing on 21 December would condemn almost every scheme in a dense area, because low winter sun casts shadows several times the height of the obstruction.

Some assessments add plots for 21 June and 21 December for context, which is useful for showing how the space performs across the year. The judgement still rests on the equinox figure.

One technical detail that catches people out. On 21 March the United Kingdom is on Greenwich Mean Time, so shadow plots timed in British Summer Time are an hour out. Check which the software used.

Which spaces the test applies to

The guidance is aimed at spaces where sunlight matters to how people use them. In practice that means the main garden of a house, communal amenity space serving flats, parks and playing fields, children's playgrounds, and sitting-out areas.

It is not aimed at every piece of open ground. A narrow side passage, a bin store, a parking court and, in most assessments, a front garden used for parking are outside it. Where a dwelling has both a front and a rear garden, the main amenity area is the one assessed, and that is normally the rear.

Where a garden is long, some assessments consider the part of it closest to the house separately, on the reasoning that this is the part used for sitting rather than storage. That is a judgement rather than a rule, and it needs to be stated and justified rather than applied quietly.

Roof terraces and balconies serving new flats are usually assessed under the same test, and they fail more often than gardens because they sit within the building's own massing.

Small urban gardens and courtyards

The test is hardest on exactly the sites where new homes are most needed, and this is where practitioners disagree.

A small courtyard garden behind a Victorian terrace, enclosed by two-storey walls, may receive very little direct sun on 21 March before anything is built. Applying the ratio strictly, almost any addition fails. Applying it flexibly, the question becomes whether the proposal makes a materially worse space or simply adds to an existing condition.

Decision-makers in dense urban areas do accept departures where the context justifies it. What persuades them is honesty: report the existing figure, report the proposed figure, state the ratio, and then make the contextual argument openly rather than by choosing a boundary for the amenity area that produces a better number.

Choosing the assessed area to flatter the result is the most common criticism of these studies, and it is easily detected by anyone with the site plan.

Producing the shadow plots

The model has to be right before the plots mean anything.

• A measured topographical survey of the site and its boundaries, including levels.
• Existing surrounding buildings modelled in three dimensions to their real heights, including roof forms, parapets and chimneys.
• True north taken from the survey rather than assumed, and the site's latitude and longitude set correctly.
• Boundary walls and fences included, because a 2 m fence casts a long shadow at equinox.
• The proposed development at its full built height.
• The amenity area boundary drawn on a plan and stated.

Trees are the awkward case. Deciduous trees are usually excluded, since they are leafless in March and are not permanent obstructions in the way a building is. Evergreen boundary planting is a judgement call. Say which approach you took either way.

Present a sun hours contour plot over the amenity area for the existing and proposed situations, with the two-hour contour marked, plus the areas and the ratio. Hourly transient shadow diagrams are useful supporting material and are not themselves the test.

Reading a marginal result

A result at 0.79 is not a refusal and a result at 0.81 is not a permission. Both are inputs to a planning judgement about amenity.

What moves the judgement is the character of what is lost. Losing sunlight from the far end of a long garden that nobody sits in matters less than losing it from the paved area outside the back doors, and a study that shows where the loss falls is more useful than one that reports a single number.

Orientation is the other factor. A proposal to the north of a garden takes almost no sunlight from it, whatever its height, and saying so plainly at the start of the assessment saves everybody time.

Where the result is marginal and the scheme matters, test design changes in the model before submitting: reducing height at the boundary, hipping the roof, stepping the plan away, or moving the massing to the northern part of the site. A short Overshadowing Study run at design stage is far less expensive than a refusal, and where windows are also affected the same model supports a Daylight and Sunlight Assessment.

Set the conclusion out in the Design and Access Statement in a few sentences an officer can quote, rather than leaving it in an appendix.

This is general guidance and not legal advice. Check the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.