NCDC Frequently Asked Questions
How are degree day normals computed?
[ Return to FAQs ]
The following white paper (United States Climate Normals, 1971–2000: Degree Day Computation Methodology) [PDF] is available regarding the two-tiered approach to computing degree day normals.
The 1971–2000 degree day normals are computed using a new methodology. Previously, degree days were computed using the Thom rational conversion formulae (Thom, 1954, 1966). The Thom method allows a monthly degree day total to be estimated from input average temperature means and standard deviations.
For the 1971–2000 normals, degree day totals were computed in two distinct ways. For stations that are not first-order National Weather Service locations, the rational conversion formulae developed by Thom (1954, 1966) was modified by using inputs of daily spline-fit (rather than monthly) means and standard deviations of average temperature. This modification improved consistency of the estimated degree day totals by eliminating monthy-by-month 'steps' in the inputs. For first-order stations, where daily data sets are largely devoid of missing values, monthly degree day totals were derived directly from daily values.
What are 'spurious' degree day values in the daily normals?
In the daily normals files, values of '-99' (or an asterisk in non-digital printouts) in the daily heating/cooling degree days represent values of 1 that have been designated to be 'spurious'. Such values are designated as spurious because of their separation from the major rise and fall of non-zero degree day values over the course of a heating/cooling degree day season, yet their presence assures consistency between the monthly total and the sum of the daily total (when values are considered equal to 1).
Thom, H.C.S., 1954: "The rational relationship between heating degree days and temperature," Monthly Weather Review, Vol. 82, pp. 1-6. [PDF Copy]
Thom, H.C.S., 1966: "Normal degree days above any base by the universal truncation coefficient," Monthly Weather Review, Vol. 94, pp. 461-465. [PDF Copy]