KNOWLEDGE HUB | SERIOUS GAMES
A 1 in 100 year event, or 1% annual exceedance probability (AEP), has a 1% chance of occurring in any single year. Over a 30 year mortgage, the chance of at least one such event is about 26%. Over 100 years, it is about 63%. Choose an event size and a decision timeframe to see how likely it is, then simulate 1,000 possible futures.
Last reviewed October 2026 by Dr Tom Logan.
Why is it called a 1 in 100 year event?
The name comes from the average recurrence interval (ARI): the average time between events of a given size over a very long record. A 100 year ARI means an event of that size occurs once every 100 years on average. The term suggests the event arrives once a century, when in reality each year carries the same independent chance. This is why scientists and engineers increasingly prefer annual exceedance probability (AEP), the chance that an event is equalled or exceeded in any given year. A 100 year ARI event has an AEP of about 1%.
Why numbers matter in hazard communication
Research by Peters et al. (2025) found that while experts tend to describe risk with words such as “unlikely” or “rare”, most people prefer to receive numbers when making decisions. Information that includes numbers is seen as clearer and more useful, and is more likely to prompt appropriate action than verbal labels alone. Pairing a number with a plain-language label, such as “a 26% chance over 30 years, or roughly 1 in 4”, reaches the widest audience.
How climate change affects return periods
Return periods come from historical records, such as stream gauges, rainfall observations, tide gauges and earthquake catalogues. They assume the past is a reliable guide to the future. For many climate and natural hazard events, particularly flooding, extreme rainfall and coastal inundation, that assumption is weakening as events once considered rare become more frequent. Treat return period estimates as a baseline, and review them as climate projections and hazard modelling are updated.
Choose an event size, from a 2 year to a 500 year return period.
Choose a decision timeframe, from 1 year to 100 years. Presets include an election cycle, a Long-Term Plan, a home mortgage, an asset life and a human lifetime.
Read the chance of at least one event over that timeframe, with its plain-language likelihood label.
Roll one possible future year by year, or compare 1,000 futures to see how outcomes vary.
The simulator applies to any hazard expressed as a return period or AEP, including floods, earthquakes, storms, tsunami and extreme wind.
The chance of at least one event over n years is P = 1 − (1 − AEP)ⁿ. For a 1% AEP event over 30 years, that gives 1 − 0.99³⁰, or about 26%.
3 years
5.9%
3.0%
1.5%
10 years
18.3%
9.6%
4.9%
30 years
45.5%
26.0%
14.0%
50 years
63.6%
39.5%
22.2%
100 years
86.7%
63.4%
39.4%
A return period describes average frequency, not a schedule. A 1 in 100 year event has a 1% chance every year, independently, so it can occur twice in a decade or not at all for 200 years. Because most planning, investment and design decisions span decades, the annual figure understates the risk people actually face. The more useful question is how likely the event is within the timeframe of the decision.
FAQs | QUESTIONS CLIENTS ASK
How do I calculate the chance of a flood over several years?
What does 1% AEP mean?
What do the likelihood labels mean?
Does the simulator work for earthquakes and storms?
Does climate change affect the results?
Have another question? Talk to our team
What a 1 in 100 year event means over the life of a decision
Peters, E., Blow, A., Chapman, D. A., and Shoots-Reinhard, B. (2025). The power of numbers in natural hazard communication. Journal of Risk Research.
Understanding return periods and AEPs (Resilience Explorer® subscribers)
Why numbers matter in hazard communication (Resilience Explorer® subscribers)
If your organisation is managing infrastructure, land, or community assets and needs to move from risk visibility to risk-informed decisions, we would like to show you how the platform works.
