f97.beERIK SCHMIDT’S WORKSHOP
← All terms

Glossary · Energy industry & energy transition

Heating degree days

Heating degree days heat building weather Energy Consumption Heating load

Heating degree days are an indicator derived from weather data. They show how much heating the weather has notionally required at a location over a given period. For each heating day, the difference between a defined reference temperature and the mean daily outdoor temperature is calculated. These daily values are added up over a month, a heating season or a year. The colder the weather and the longer the heating period, the higher the total. The indicator does not measure an amount of energy consumed. It only describes how strongly the weather determined the demand for heat.

Calculation rules and unit

The unit is the kelvin day (Kd, often also called "degree day"). Without stating the calculation rule, the numerical value says little. Two parameters determine the result: the reference temperature from which the outdoor temperature is subtracted, and the heating day threshold. Only days with a daily mean below this threshold are counted. In Germany, the calculation rule of the German Weather Service (DWD) according to VDI 3807 Part 1 is the most widespread. It uses a room temperature of 20 °C, and only days with a daily mean below 15 °C are counted. This combination is often referred to as G20/15.

An example under this rule: a day with a daily mean of 5 °C contributes 20 minus 5, i.e. 15 Kd. A day at 16 °C does not count at all, because it is above the heating day threshold. Eurostat calculates differently for European energy statistics: 18 °C minus the daily mean, and only on days with a daily mean of at most 15 °C. The same 5 °C day yields 13 instead of 15 degree days there. In the USA, a base of 65 °F without a separate heating day threshold is common. Values from different sources can therefore only be compared after conversion or when the same calculation rule is used.

The heating day threshold describes the outdoor temperature below which a building typically has to be heated. It is not the same as the desired room temperature. How low the actual heating limit of a building lies depends on its condition. A well-insulated house with a lot of solar gain and high internal heat gains can go without heating for longer in mild weather than an unrenovated house. Standard values such as 15 °C are therefore conventions and not a property of a particular building.

Terms: heating degree days, degree days, degree day number

In German, the terms are used inconsistently. The DWD calculates "Gradtage" (degree days) according to VDI 3807 and "Gradtagzahlen" (degree day numbers) using an in-house method based on VDI 2067. According to the DWD, the two values generally differ only slightly. For the degree day numbers, a heating period is determined in addition. There are also variants in which the heating limit itself is the reference temperature, for example the sum of 15 °C minus outdoor temperature. Such values turn out considerably lower than G20/15. What matters is therefore not the name but the exact calculation rule. The entry Degree day number and heating degree days covers this in more detail.

Distinction from heating load and heat consumption

Heating degree days are often confused with the heating load. The heating load is the output a heating system must deliver in an unfavourable design situation at a very low outdoor temperature. It is stated in kilowatts and applies to a point in time. Heating degree days, by contrast, add up the influence of the weather over days, months or years. They show how long and how severe a cold spell was. They say nothing directly about the peak output on a particularly cold morning.

Heating energy consumption is also something different. It is measured in kilowatt hours and does not depend on the weather alone. The building envelope, the heating technology and its efficiency, the controls, ventilation, hot water preparation and user behaviour also play a role. Two buildings at the same location have the same heating degree days but can need very different amounts of energy. Conversely, a building can consume more in a cold year without its efficiency having got worse.

Weather adjustment and comparability

Heating degree days are most useful because they make the influence of the weather on consumption visible. An apartment building can need considerably less gas, district heat or electricity for heating in a mild winter without anything having been improved in the building. In weather adjustment, only the consumption for space heating is converted using the ratio of reference degree days to actual degree days. The result is a consumption as it would be expected under reference conditions. Consumption for hot water hardly depends on the weather. It is therefore separated out and not corrected, as are other components independent of the weather. Anyone who extrapolates the entire consumption including hot water overestimates the influence of the weather.

In Germany, weather adjustment is legally required for energy consumption certificates. Under § 82 paragraph 3, the final energy consumption for heating must be weather-adjusted, so for residential buildings not the share for hot water. On gesetze-im-internet.de, the law now appears under the title Gebäudemodernisierungsgesetz (GModG), previously Gebäudeenergiegesetz. The basis is consumption data from at least 36 consecutive months. The method must comply with the generally recognised rules of technology. For simplified methods that the responsible federal ministries have published in the Bundesanzeiger, this is presumed. This obligation applies to consumption-based certificates, not generally to every consumption analysis.

For these certificates, the DWD provides climate factors by postcode. A climate factor is the ratio of the degree days of a reference time series at the Potsdam location to the actual annual degree days at the building location. It is calculated for rolling periods of twelve months. This removes two influences at the same time: a milder or colder year and a milder or harsher climate at the location. A building in Freiburg and one in Hof can thus be compared at least roughly. Anyone who only wants to compare their own years with each other, by contrast, often refers to the long-term mean of the same location. The two approaches lead to different figures and should be disclosed.

Weather adjustment does not make the values perfectly comparable. It does, however, remove the largest distortion, the difference between a mild and a cold year. A low heating bill can result from good efficiency, but also from mild weather, a small living area, a low room temperature or vacancy. The indicator helps to tell these causes apart.

Significance in the power system

With the electrification of heat supply, heating degree days are also becoming more important for the electricity market. Every heat pump shifts heat demand from oil, gas or district heating systems into electricity consumption. In cold periods, the electrical load therefore rises, especially when many buildings are heating at the same time. Heating degree days explain a large part of this fluctuation over the seasons. They are therefore a common input variable for load forecasting over months and years.

For grid operation, however, a sum of degree days is not sufficient. Three very cold days burden grids, storage and power plants differently than a moderately cold but long month, even if both yield a similar number of degree days. The first dimension is better described by load peaks and design temperatures, the second by heating degree days. In addition there is the timing: cold high-pressure situations can coincide with little wind. Heat demand then rises while weather-dependent generation is weak. Heating degree days do not show this coincidence. It must be considered together with wind and solar generation, residual load, grid bottlenecks and available flexibility.

Typical misinterpretations

  • Heating degree days as an explanation of heating costs: Costs also depend on energy and base prices, tariffs, metering concept and efficiencies. In a mild year, costs can still rise if prices climb sharply. Heating degree days only explain how the weather influences the quantity consumed.

  • Comparing values from different calculation rules: G20/15 according to DWD and HDD according to Eurostat (18/15) differ systematically. Mixing them creates apparent climate differences or trends.

  • Single years as climate evidence: In the long term, falling heating degree days can point to milder winters. Individual years, however, fluctuate strongly. Reliable statements require long time series and a consistent method.

  • Applying standard values to every building: Solar radiation, internal heat gains and ventilation behaviour overlay the relationship with the outdoor temperature. In very efficient buildings, the actual heating limit lies below 15 °C. Standardised values represent these buildings only approximately.

  • Extrapolating the electricity demand of heat pumps linearly: In cold weather, it is not only the heat demand that rises. For air-to-water heat pumps, efficiency usually also falls, because the temperature difference between heat source and heating circuit grows. Defrosting cycles and possibly an electric heating element come on top. With the same heating degree days, electricity consumption can therefore turn out very different depending on the system and flow temperature.

What the indicator achieves

Heating degree days make the weather-dependent part of heat demand tangible. They prevent changes in consumption from being hastily interpreted as a gain or loss in efficiency. They do not fully separate climate, building and use. But they provide the necessary correction variable to tell these influences apart. They say nothing about peak loads, reserves or security of supply. They show how strongly the weather shaped heat demand over a period, and thus make consumption, efficiency and electrification more comparable.

Sources

Related glossary entries

Edited by eCMS on 12.11.2026 05:05

© 2026 Erik Schmidt. All rights reserved. Quotations with a link to the source are welcome.