What is Normal?

When it comes to the meteorological community, a ‘daily normal’ has certain meaning – the average of the past X years for a parameter for a given calendar day, smoothed to fit a normal curve. However, when working to communicate that information to the energy industry – and many other industries – it is easily misconstrued and misunderstood. This includes even explaining the difference between ‘normal’ and ‘average’. Many individuals outside of the meteorological community have the expectation that “normal” is what is supposed to happen on that day when in reality, normals tend to actually occur irregularly. In other words, normal is often times anything but average. "Daily average” temperature values in the US are derived from two instantaneous observations, maximum and minimum, which combined with inherent instrument error can result in a number of misunderstood and sometimes non-sensible values exposed to the industries that exploit this information.. Analyzing methods used in other countries such as Japan, UK, and France, we see there is no wrong answer to determining how to derive a “normal”, but ways that better capture true representations of average for a given day, especially as it pertains to energy demand. Using more data points – such as an average of 24 hourly values – might better represent the weather over the course of a day. Alternatively, energy industry specific normals could be an alternative where different hours are weighted differently based on energy demand and industry specific sensitivities. Furthermore, using a uniform method of “normal” calculations throughout the world – at least in specific industries - would greatly aid in improving the overall understanding of normal as well as the ability to communicate its meaning.