“One in a hundred years” means a 1% chance in any given year. It does not mean one per century, it does not mean you are safe for ninety-nine years, and it says nothing whatever about how long it has been since the last one.
over n years, P(at least one) = 1 − (1 − p)n
Over a 30-year mortgage a 1-in-100 hazard is 26%. Over a 50-year building life it is 39%. Over a century it is 63% — not 100%, which is the other thing people assume. None of those is 1%, and the arithmetic is not linear while everybody's intuition is.
Nothing is uploaded and nothing is stored. This page has no server: it is arithmetic in your browser on numbers a council hazard map or an engineer gave you. It knows nothing about your property and it is not advice about it.
1 · Over how long?
The map says it is rising
Councils now publish hazard against climate scenarios, so the annual chance is a function of the year rather than a constant. Put the end-of-horizon figure in and the years are compounded one at a time — which is not the same as holding today's number, and not the same as using the end number throughout.
It is not the only hazard
One per line: name, return period. A property is not exposed to one thing, and the arithmetic for several is the same arithmetic run again.
2 · The same arithmetic, backwards
The question an engineer asks is the other one: given how long this has to last, and how much risk is acceptable over the whole of that, how rare an event do I design for?
p = 1 − (1 − risk)1/n
What it does not know
It assumes the years are independent. They are not, quite: wet years cluster, because the climate drivers behind them last longer than a year. Clustering makes both the very quiet stretches and the two-in-a-decade stretches more common than the arithmetic here says, without changing the long-run rate.
It assumes the number you typed is right. A 1-in-100 line on a hazard map is an estimate from a record that is usually shorter than a century, extrapolated. The uncertainty on it is often wider than the difference between the bands, and none of that uncertainty is anywhere on this page.
A return period is for a place, not for a region. “A 1-in-100 event somewhere in the catchment” is a much more frequent thing than “a 1-in-100 event at this property”, and the two get reported in the same sentence constantly.
Several hazards are treated as independent, and they are not. The storm that floods the river brings the slip. Treating them as independent understates the chance of a very bad year and overstates the chance of exactly one thing going wrong.
It says nothing about depth, damage or cost. An exceedance is a line being crossed. Whether that means water in the garden or water in the ceiling is not in the return period, and it is the thing that actually decides what happens to you.