BS 5837 is the British Standard for trees in relation to design, demolition and construction. If a UK planning application touches a site with trees on or near it, the local authority will usually ask for a survey against it before they will look at anything else. Get the survey wrong and the whole application stalls behind a planning officer's query, not a design problem.
Here is what the standard actually requires the survey to show, and where surveys most often fall down.
Every tree gets a category, not an opinion
BS 5837 grades each tree A, B, C or U:
- A — high quality, retain if at all possible.
- B — moderate quality, retain where it doesn't conflict with the design.
- C — low quality, retain only if it costs nothing to.
- U — unsuitable for retention regardless of the development, usually on condition grounds.
The category is not a judgement call made at a desk afterwards. It comes from a defined set of criteria — arboricultural quality, condition, safe useful life expectancy, and any physical or aesthetic factors — assessed on site, tree by tree. A schedule where every tree that happens to be in the way is marked C, and every tree that happens to be clear of the site is marked A, is the fastest way to get a survey queried. Planning officers see that pattern constantly, and it reads as the category following the site plan instead of the other way round.
The Root Protection Area is a calculation, not a guess
The RPA is the area around a tree where roots are assumed to be present and construction activity needs a method statement before it can happen. It is calculated from stem diameter, not eyeballed:
RPA radius = stem diameter (measured at 1.5 m) × 12
That number defines a circle (or a modified shape, for trees with an asymmetric root plate) on the constraints plan. Every RPA on the drawing needs to trace back to a stem diameter actually measured on site, at the height the standard specifies — not a diameter estimated from a photo, and not the same round number used for every tree of similar species.
The schedule and the plan have to agree
A BS 5837 submission has two halves: the tree schedule (the table — species, dimensions, category, RPA, recommendations, one row per tree) and the constraints plan (the drawing — every tree, canopy, and RPA overlaid on the site layout). They are two views of the same data, and they are checked against each other.
The single most common reason a survey comes back with questions is a mismatch between the two: a tree on the plan that is not in the schedule, an RPA on the drawing that does not match the stem diameter in the table next to it, or a category letter that changed between the schedule and an earlier draft of the plan without both being updated. None of these are wrong assessments — they are transcription errors, and they are entirely avoidable if the schedule and the plan are generated from the same underlying record instead of drawn and typed up separately.
What a planning officer is actually checking for
Not whether every tree was saved — nobody expects that. They are checking whether the survey is internally consistent and methodologically sound: categories that follow the criteria rather than the site plan, RPAs that follow from measured stem diameters, and a schedule that matches the drawing. A survey that is honest about which trees are low quality and clearly shows its working gets through review faster than one that tries to protect everything and asks the officer to take it on trust.
None of this is legal advice — confirm current requirements with the local planning authority before you submit. This is the exact shape of thing tree·canopy is built to keep consistent: dictate the survey in the field, and the schedule and the constraints plan are generated from the same record, not typed up twice from a clipboard. The BS 5837 report pack is bundled in for UK firms — see how it fits together.