For calendarized and normalized data, the usage of weather sensitive meters is adjusted based on daily degree day data.
Calendarization and Normalization help you compare month-to-month and year-to-year utility bills on a common “apples-to-apples” basis. In both cases, the usage of weather sensitive meters is adjusted based on daily degree day data.
So how do we know when a meter is weather sensitive? A statistical process correlates the usage on each bill with the corresponding degree days in the billing period. The above chart shows the “best fit” regression line results for each heating season. You can think of each line as a mathematical model of the meter’s response to outside temperature.
The vertical axis is the average daily usage and the horizontal axis is the average number of heating degree days per day.
If the line rises to the right, it means that months with more degree days (colder months) have more usage. This suggests that a portion of the usage is weather sensitive. If the statistical correlation is strong enough, we can say that the usage of this meter in the winter is weather sensitive.
If the line is nearly flat or falls to the right, it means that months with more degree days (colder months) do NOT have more usage. This indicates that the usage of this meter in this season is NOT weather sensitive and therefore no weather adjustments should be made.
The above table lists the weather factor (the slope of the line) in units of usage per heating degree day. A factor of 417 KWH/HDD means that for every degree day, the meter consumes about 417 KWH. The table also shows the base load. This is the average daily usage that is NOT weather sensitive and hence NOT weather adjusted.