Shaftesbury is not a river flood town, and the mapping is unusually blunt about it. None of it falls inside the river-and-sea flood zones. Nothing here is recorded as having flooded before. About one postcode in ten still rates high on at least one flood measure. The risk in a place like this is water running down a hill, not water rising up one.
Start with the zones. The Environment Agency puts 0% of the area's postcodes in flood zone 3 and 0% in flood zone 2. Those are exclusive bands rather than nested ones, with zone 2 marking the additional medium-probability fringe beyond zone 3, so two zeros means neither band reaches the town at any point. For a settlement built on top of a greensand hill that is the expected answer, and it is still worth having in writing rather than assumed from the view.
What that leaves is surface water, which is the risk that actually applies here and the one a zone map was never built to show. The postcode indicators used on this site fold surface water in alongside rivers and sea, and for a town laid out on a slope they are the measure to read. Rain landing on a hill does not stay on the hill, and the gradient decides where it ends up.
The climate-adjusted zone share is also 0%, so the allowance the Environment Agency applies for future conditions does not pull any part of the town into a river-and-sea zone. That is a forward look at rivers and the sea alone. It carries no view on how hard it will rain.
For insurance, that combination is the easy one to quote. No zone coverage and no recorded flood history means cover should price as ordinary rather than as a negotiation. Surface water is still where a quote can catch you out, because insurers price an individual address rather than a town, so get an indicative figure before you instruct a solicitor instead of after.
On the broader postcode indicators, which rate each postcode against at least one flood measure including surface water, about 10% come out high, about 11% medium and about 29% low. Those bands overlap the zone picture rather than adding to it, so a postcode appearing here is not a second, separate hazard stacked on the first. Roughly one in ten rating high is the figure to take to a specific address, because the town-wide answer is never the answer for one front door.
On the modelled high-risk flood extents, the area shows 0% undefended and 0% defended. Worth understanding what the second of those means elsewhere: defended only records that a defence exists in the model, not that the house behind it is safe. Here the distinction never comes up, which is a better position to be in than being on the correct side of a wall.
Sea level is a background item this far inland. The nearest tide gauge, at Portsmouth, has recorded a rise of about 2 millimetres a year across its record from 1962 to 2023. That is a measured historic rate over six decades, not a projection, and a hilltop town in north Dorset is not where it arrives first.
No historic flooding is recorded anywhere in the area. That is real reassurance, with one limit worth keeping in mind: records capture what was reported and mapped, not every garden that has ever gone under after a summer downpour.
Flood risk here sits in the middle of the range, and the split is what puts it there. Nothing from rivers or the sea, enough surface-water flagging across the postcodes that the address matters far more than the town, so check the one you are buying rather than the one on the sign at the top of the hill.