What is Normal in Weather?

What do the numbers 33, 53 and 43 have in common? How about 12, 20 or 8? Would you be surprised to learn they all represent normal values of snowfall (34, 56 and 45 inches a year) or normal temperature for January 1st (12, 20 and 8 degrees). What the heck is normal? What does it really mean?

First off, let's briefly talk about "weather" and "climate". In the broadest sense of the terms, "weather" is what we experience on a day to day basis. That short term variability is weather. "Climate" is used to describe a longer term period. Be it a few weeks or a month of consistently warm and dry conditions, or rain that seems to fall every weekend. Those week to week variations are smoothed in such a way to give one the sense of what to expect on any given day, at any given time. That would be "climate". There is an old adage "Climate is what you expect, weather is what you get" that applies in this discussion.Climate could also be described as the range of expected conditions in which the 'weather' varies.

In the science of meteorology, the terms normal and average are used interchangeably but really mean two different things. In meteorology, an average (such as an average temperature) is just what you think it is,  the total of all the values in a group divided by the number of values. A normal is different. It is the specific average over a standard 30-year period, currently 1981-2010. For example, a normal temperature is the average temperature over the 30-year period from 1981 through 2010. To explain what we are saying, a basic lesson in statistics is necessary. Don't stop reading, we'll make it easy.

Average, mean, median and mode are some of the most common terms employed in statistics to give a sense of what "normal" is. Climatological data, like daily high temperatures, annual snowfall or total precipitation for a season are all ranges of observed values. The term average is used to describe a measure of the "middle" value of the data set, that is average is a single value that is meant to typify a list of values. So in saying "the average temperature for January 1st in 12 degrees" we speak of a value that represents something taken from a large set of daily values. When we speak of mean we are generally speaking in the same context as average. is the The median  value is a range of numbers calculated by arranging the data from the lowest value to the highest value and picking the middle one. If there is an even number of observations, then there is no single middle value; the median is then usually defined to be the mean of the two middle values. The mode is the most commonly occurring value in our data set. Other terms, such as variance or frequency are used as well, but less common.

Okay, back to the real world of what the heck is the normal snowfall for Grand Forks North Dakota. Remember, the term normal is used interchangeably with average to describe the expected weather conditions on a given day, during a month, season or year. So based on the 30 years [1981 - 2010] of snowfall data for the UND/NWS Climate station, the average snowfall is 45.3 inches. That indicates that during the period from 1981 to 2010, after averaging all the annual snowfall data together and dividing by 30 [the number of years represented by 1981 - 2010], we arrive at an average value of 42.2 inches. But is that really a good way to describe the snowfall in any given year? What happens if we get 39 inches of snow - was it a below normal snowfall year? What happens when we get 52 inches of snow - was it really an above normal snowfall winter?

Below is a table that contains the past 30 year average snowfall for the UND/NWS Climate Station, with several statistical variances used to produce a realistic range of likely seasonal snowfalls.

Average Snowfall 45.5
Standard Deviation 9.2
Maximum Average 54.6
Minimum Average 36.3
Greatest  1996/97
Least 1999/2000

In the table above, there are several key values to look at. The first value, 45.5 inches is the mathematical average snowfall for the 1981 - 2010 winter at the UND/NWS Climate station. The term "Standard Deviation" is the statistical variance in the year - to - year snowfalls. The next two terms, Maximum Average and Minimum Average represent the "normal range" of snowfall one could expect in any given year; this is the average snowfall (45.5 inches) plus or minus 1 Standard Deviation (9.1 inches). In other words, in any given year as little as 36.3 inches or as much as 54.6 inches of snow may fall at the UND/NWS climate station, and be "normal". Another interesting fact is the mode, that is the frequently occurring or repeated seasonal snowfall. Finally, over the past 30 years the greatest snowfall was in the 1996/97 winter when 93.8 inches fell; the least was 16.9 inches during the 1999/2000 winter.

When we look back over the past 30 years of snowfall data for any location, we see a tremendous amount of year-to-year variability. This variability is quite common, and in fact is expected. "Weather" naturally varies from day to day, location to location, and so when describing the normal or average weather we mathematically "smooth out" the variances to give a sense of change that one may expect. The advantages in doing this are numerous; highway engineers use the data to determine the types of materials best suited for a regions climate, or farmers can best estimate when certain crops should be planted, or architects can calculate how much storage is needed to keep a parking lot from flooding during summer rains. Not just the normals, but the extremes - what we call a record event - are important as well.

When all 30 years of snowfall data are collectively filtered using statistical tools, we can arrive at values that describe the overall "climate" of a particular city, state or region. We then call this normal.

So, what is normal? In weather, normal is the average of the extremes - literally.





Return to News Archive

USA.gov is the U.S. government's official web portal to all federal, state and local government web resources and services.