An average is produced by adding an amount and then dividing by the number of samples. When it comes to rainfall, it is common practice to determine the average rainfall in a given month. For example, March at a particular place having an average of 2.24 inches of rainfall each year. For the average to be more meaningful there needs to be a period of record covering many years. A 30 year, 50 year or 100 year average could be used to get a statistically meaningful record. The longer the period of time, the less the extreme values of precipitation in a month will impact the average. When the sample is too small, such as using a 10 year average, then years with very high or very low precipitation will have a dramatic impact on the average.

One problem that results from an average value of rainfall is that the public will often assume that the average rainfall is what should be expected in a given month each year. For example, when March averages 2.24 inches of rainfall, it can be assumed that there is some sort of climate change or modification to weather if the value is not reasonably close to this value. As with any average, it will be the case that there will be many values that are above the average and many values that are below the average. The average is the average of the extremes and everything in between. Thus, it is not too uncommon for a monthly precipitation value to be significantly different from average. A month with a precipitation very close to the average is often not typical. This problem is particularly evident in semi-arid locations that have a high variation in precipitation from year to year. Some of the years will be drought years while other years will have a high amount of rainfall. This sort of climate can waffle between droughts and floods. The monthly rainfall will often be much above or much below the average monthly rainfall.

Big rain events can have a dramatic impact on the average and what someone sees as the average rainfall per month. For example, the remnants of a hurricane every few years can dump huge accumulations of rainfall on a location. This will inflate the monthly rainfall total and increase the average rainfall. What ends of happening in this case is that most times a month will end up having below normal precipitation since the hurricane rain has inflated the average. For example, suppose a location averages 2 inches of rainfall in August for the last 30 years. Now suppose also that twice in the last 30 years, the remnants of a hurricane have dumped 15 inches on this location in August. This will increase the average rainfall for August. However, on most years the August rainfall will be less than 2 inches. Thus, the average of 2 inches can give the appearance that a location is having an unusual drought year after year when in actuality it is typical for this location to have less than 2 inches of rain in August. One way this problem can be reduced is to “throw out” the 2 highest and 2 lowest rainfall amounts for a given month (thus removing the most extreme values). This can help produce an average that is more typical for a location.