1-in-100 Year Rainfall Event


How much data was this 1-in-100 statistic based upon?  We can only go back as far as records will allow.  Statistically speaking, this may well not be long enough. 

Is past rainfall a predictor of future rainfall, or more generally are past weather patterns predictors of future weather patterns?  We are all aware of the issue of ‘climate change’, so we may not accept the past as a predictor of the future.  Human activity and development may change the course of natural waterways and runoff, so this will also affect the past/future predictor factor over time.

Let’s look more closely at the example from the Bureau of Meterology Website [1]:

“As an example, for Melbourne, a rainfall amount of 48.2 mm in 1 hour can be expected to be equalled or exceeded on average once every 100 years.”

This means that, for each and every year within the 100-year period, the probability of the event occurring is 1%.  The probabilistic nature of the figure (as opposed to certainty) indicates that it is not impossible to have two or more events in the same 100-year period, or even in the same year (but very unlikely).

It should be understood that this statistic holds for one location only (in this example, the “Bureau's official Melbourne rain gauge”).

Is this useful information for you?  It probably would be if the Bureau's official Melbourne rain gauge was located in your front yard.  You would probably be more interested in the equivalent statistic for a broader area (say Melbourne-wide or the whole of Victoria).

Consider that

“In practice, the most intense part of a thunderstorm, or of a cold frontal band, is a small area, across which the rainfall values are almost the same.” [1]


“Values for one hour duration taken at points about 5 km apart would be almost totally independent of each other in some storms but partly related in others.” [1]

What we really want is the probability of this particular event occurring over any one particular hour in one given year (this year) at any location in Melbourne.  This is x% (where x is calculated by combining the figures for each particular location over the whole area).

We do not delve into the details of this calculation.  It should be intuitively clear however that the probability of the event occurring at any location over the whole Melbourne area is greater than the probability of the event occurring at one particular location.

[1] http://www.bom.gov.au/water/designRainfalls/rainfallEvents/why100years.shtml

Brendan Nielsen