Sun on ground: testing a garden at the equinox

The sun on ground test asks whether at least half of an amenity area still receives two hours of sunlight on 21 March. That is the criterion in BRE 209, it is the one a case officer will quote back at you, and running the sun on ground test for a garden properly takes about as long as arguing about it afterwards.

It is a separate exercise from the window-based tests. Vertical sky component and annual probable sunlight hours measure what reaches a window. This measures what reaches the ground, which is what neighbours actually complain about when a three storey side extension goes up on their south boundary.

What the test measures

The assessment models the shadow cast by existing and proposed built form across a defined ground plane, then reports the proportion of that plane receiving at least two hours of direct sun on the assessment date.

Two outputs come out of it. The first is a sun on ground plot: a plan of the amenity area shaded to show the area meeting the two hour criterion before and after development. The second is the percentage. Officers read the percentage and look at the plot, in that order, so both need to be legible at the scale they will be printed.

The guidance also gives a relative test for areas that fall below half. Where an amenity area does not meet the two hour criterion after development, the area that does meet it should not be reduced to less than 0.8 times its former value. That second limb matters far more in practice than the first, because in dense urban contexts a great many gardens fail the absolute criterion before anyone builds anything.

Why 21 March and not another date

The spring equinox gives a sun path roughly midway between the winter and summer extremes, and it is the same on 21 March and around 21 September. Using it avoids the two obvious distortions: a June assessment flatters almost any scheme, and a December assessment condemns almost all of them.

It is a convention, not a statute. Nothing prevents a scheme also presenting 21 June and 21 December plots, and on a site where the geometry is unusual it can be worth doing. Where the design relies on summer use of a terrace, showing the June condition is honest rather than promotional. The choice of dates and what each one shows is set out in choosing assessment dates for sunlight studies.

What you should not do is quietly substitute a different date because 21 March produces an awkward number. Reviewers check the date on the plot first.

Which amenity areas get assessed

The guidance covers gardens serving existing dwellings, and also parks, playing fields, children's playgrounds, outdoor sitting-out areas and open spaces where sunlight matters to the use.

Deciding the assessed extent is where judgement enters, and where reports diverge. A long rear garden serving a terraced house is usually taken as a whole. But an area occupied by a substantial outbuilding, a mature hedge line or a shed is not usable amenity space, and including it inflates both the baseline and the result. State what you have included, show it hatched on a plan, and say why anything is excluded.

Front gardens are normally excluded unless they are used as sitting-out space in fact. Driveways, bin stores and side alleys are excluded. Where a garden serves flats in multiple occupation, the communal area is assessed as one space rather than divided notionally between units.

An Overshadowing Study that defines its assessed areas clearly is much harder to argue with than one that presents a single percentage for a whole site.

Producing the plots

Three deliverables usually accompany the numbers, and they do different jobs.

1. Transient overshadowing diagrams: hourly shadow positions across the assessment date, typically from an hour after sunrise to an hour before sunset, existing and proposed side by side.
2. Sun on ground plots: the two hour contour mapped over each assessed amenity area, existing and proposed.
3. A tabulated summary: area assessed, area meeting two hours before, area after, percentage before and after, and the ratio against 0.8.

The model has to include the things that actually cast shadows. Existing boundary walls and fences at their real heights. Mature trees, at least where they are evergreen or protected, with a note on how deciduous canopies have been treated. Neighbouring buildings beyond the application site, modelled from survey data rather than from an aerial photograph.

Ground levels matter on sloping sites and are frequently flattened in the model. A terrace cut into a slope behaves very differently from the same terrace on level ground, and a reviewer with the topographical survey will notice.

Gardens that already fail before development

Plenty of them do. A small north-facing garden behind a terrace in a Victorian suburb may receive almost no direct sun on 21 March whatever happens next door.

This is the situation the 0.8 relative test exists for, and it should be presented as the primary test in those cases rather than buried after a failed absolute comparison. Say plainly that the amenity area does not meet the two hour criterion in the existing condition, give the baseline figure, and then show the retained proportion. The relative test and where a loss becomes noticeable is covered in the 0.8 relative test.

Be careful with very small baselines. Where an existing garden has only a few square metres meeting the criterion, the ratio becomes volatile and a loss of one square metre can breach 0.8 while meaning almost nothing to the occupier. Report the absolute areas alongside the ratio so the decision maker can see the scale of what is at stake. That is a judgement an officer is entitled to make, and it is better made with the numbers in front of them than after a refusal.

Communal amenity space in flatted schemes

New flatted development creates its own amenity areas, and those are assessed too, against the scheme's own massing rather than against a neighbour's.

The problem is usually self-inflicted. A courtyard enclosed on three or four sides by five storeys of flats will fail comfortably, and no amount of landscaping fixes the geometry. The design responses are structural: reduce height on the southern arm of the block, open a corner, step the section back, or move the amenity space to a roof terrace where it gets sun.

Play space is a particular case, because local policy often sets its own sunlight expectation on top of the BRE criterion. Check the adopted standard before you fix the layout, and check whether the authority counts a roof terrace towards the requirement at all.

Design changes that recover sunlit area

When a scheme fails, the useful question is which part of the massing casts the shadow at which hour. The transient diagrams answer that directly.

• Reducing height at the southern or south-western end of a building usually buys more sunlit ground than the same reduction elsewhere.
• Setting an upper storey back from the boundary moves the shadow edge measurably at low sun angles.
• Pitching a roof away from the affected boundary, rather than using a flat parapet at full height, recovers early and late hours.
• Narrowing a projecting rear wing is often more effective than shortening it.
• Relocating the neighbour's usable amenity space is not a design change you get to make, so do not present it as one.

Test the options in the model before drawing them. An hour of sun path work at the right moment saves a resubmission, and it gives the Daylight and Sunlight Assessment something to say beyond a pass or fail. The interior tests changed in the 2022 revision as well, and the practical differences are covered in BRE 209 2022 changes for new interiors.

Enter your postcode on PlanScope to see which amenity and daylight documents a scheme is likely to need.

This is general guidance, 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.