Return period
The average time between events of a given magnitude. A 100-year return period is the equivalent of the event that has a 1 per cent probability of occurring in any given year.
In detail
A return period is a statistical measurement, usually based on past observations, which denotes the average recurrence interval over an extended period of time. They are typically required for risk analysis (i.e. whether a project should be allowed to go forward in a zone of a certain risk) and also to size structures so that they are capable of withstanding an event of a certain return period (with its associated intensity).
Use of a return period does not imply that events will happen regularly. For example, in any given 100-year period, a 100-year rainfall event may occur once, twice, more, or not at all.
The concept of the return period is based on the assumption that there is no trend in the observations used in its calculation. In times when climate is changing, this makes the use of return periods associated with climate variables challenging and potentially misleading.
Theoretical return periods for climate variables can be estimated for any future 30-year time period – however a large dataset of projections consistent with any future single time period is required for reliable estimation of return periods.
The return period is the inverse of the probability that a given event will be exceeded in any given year. The probability of annual exceedance can be a more helpful statistic for denoting the average frequency of events – particularly where there is more than one event a year.
- In UKCP09, one application of return periods is to report changes in extreme storm surges (called return level, or RL). For example, Section 6.3.2 in the UKCP09 Briefing report describes changes in 50 year return levels. More results are provided in Section 4.3 and Section 4.6 of the UKCP09 Marine & coastal projections report.
|
