Cooling Degree days and Heating degree days

Cooling Degree days and Heating degree days

Cooling Degree-days (CDD) and Heating Degree-days (HDD) are meteorological indices defined as integrated temperature deviations from a base temperature over time[1]. Formally, degree-days are defined as a summation of the differences between the outdoor temperature and some threshold (or reference base) temperature over a specified time period (such as at annual time scales).

Put simply, a Cooling Degree-day is the outdoor temperature  below which a building would not require cooling.

Mean Cooling Degree Days 1986-2016 (Data source: GLDAS[5])

Conversely, a Heating Degree-day is the outdoor temperature above which a building would not require heating.

Mean Heating Degree Days 1986-2016 (Data source: GLDAS[5])

Naturally, the choice of the threshold outdoor temperature becomes important while calculating the CDD and HDD for a particular region. Furthermore, because of the non-linear relationship between degree Celcius (˚C) and degree Fahrenheit (˚F), degree-days computed on either of the units is not convertible to the other.

Mathematically, HDD and CDD are represented as temperature sums in ˚C (or ˚F)*day:

The degree-days methodology is commonly applied in the energy sector for planning energy systems and predicting seasonal load demands. Moreover, traders and economists also utilize degree-days for market instruments such as weather derivatives and insurance, to iron out short to seasonal time scale weather related variations in energy demand. Because the thermal comfort in buildings relate to both cooling and heating systems, the degree-days have been developed with the corresponding dual concepts of CDD and HDD.

A key issue in the application of degree-days is the definition of the reference base temperature, which should ideally differentiate the region specific thermal factor. For instance, a widely used common base temperature of 18.3 ˚C (65 ˚F) for computing degree-days over Texas (U.S.A) and Siberia (Russia) may not account for the underlying heterogeneous mean climate in the two regions, wherein the residents would have adapted to differential thermal comfort level. As the choice of base temperature can be both subjective and through statistical reasoning, defining an appropriate region-specific threshold often creates a dilemma, especially in large countries with diverse temperature ranges (such as the United States of America, India and China).

Additionally, accounting for population weightage for degree-days at regional or country scales is equally important. For instance, population weighted degree-days for Beijing (China) with an urban population of ~21 million and Philadelphia (U.S.A) with a similar annual climate but relatively sparse population (~1.5 million)[2] would be better representative of the true energy demands for heating and/or cooling compared to un-weighted degree-days.

Data and methodology used in ENERGYA for assembling a historical global gridded degree-day dataset

The degree-day methodology is widely used as tools for assessing weather related energy requirements in buildings[3][4]. A measure of changes in both duration and magnitude of degree-days under future projected climate can provide important basic information for formulating energy policies. Understanding these changes at both regional and global scales requires assembling a comprehensive spatio-temporal dataset of CDD and HDD at fine-scale gridded resolution, weighted by the population density at equivalent spatial scales.

For assembling a historical global dataset of CDD and HDD, we utilize the daily minimum and maximum surface temperature (˚C) from the Global Land Data Assimilation System Version 2.1 (GLDAS-2.1) dataset[5]. The degree-days are computed at the native 0.25˚ (~ 27 x 27 km) global gridded resolution, for years 1971-2016, using Climate Data Operators (CDO) ver 1.9.0[6]. As discussed above, employing a constant reference base temperature across the global domain is both implausible and of little practical application. In our preliminary analysis, we therefore employ a range of widely used thresholds adapted in the literature, ranging from 15-23 ˚C.


  1. [1]Day, T. Degree-days: Theory and Application [Butcher, K. (ed.)] [1–98] (The Chartered Institution of Building Services Engineers, London, 2006).
  2. [2] and
  3. [3]Scott, M. J. & Huang, Y. J. [Effects of Climate Change on Energy Use in the United States] Effects of Climate Change on Energy Production and Use in the United States [7–28] (CCSP, Washington, 2008).
  4. [4]Collins, M. et al. [Long-term Climate Change: Projections, Commitments, and Irreversibility] Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T. F. et al. (ed.)]. [1029–1136] (Cambridge University Press, Cambridge, 2013).
  5. [5]Rodell, M. et al. [The Global Land Data Assimilation System (GLDAS)], (Bulletin of the American Meteorological Society 85: 381-394, 2004).
  6. [6]Climate Data Operators (CDO):