Return period


Return period describes the expected (mean) time (usually in years) between the exceedence of a particular extreme threshold. Return period is traditionally used to express the frequency of occurrence of an event, although it is often misunderstood as being a probability of occurrence.


The use of return period is not helpful in communicating risk to the public as an average period between events, the return period involves the idea of periodicity, which can mislead non-experts [*1].

For example, the probability of exceedance of the flood defence system is set at 1% in any year. This event is usually translated into having a return period of 100 years. Many people will regard it as most unlikely that such an event will occur again in their lifetime.

However, the same probability (1%) stands for any year! Within the period of 70 years, there is 50% chance for the event to occur at least once, and 15% chance to occur at least twice. So, discussion of a defence standard based on the return period of the load can make the public believe defences are more secure than that may be.

Moreover, the return period typically refers to the hazard intensity and not the consequence. The probability of loss is often considered the same as the return period level of defences. However, this assumption does not match the likely performance of the system, excluding any information about the probability and magnitude of consequence during more frequent or more severe events.

Headers and keywords
Page content

Back to:

Flood risk

Abstract terms



*1 Huntingdon, MacDougall (2002)



Further Information