Information about residential heat demand is crucial for sustainable climate mitigation and adaptation as approximately 30% of the heat demand stem from residential heating which contribute to one third of the nationwide CO2 emissions. This service provides information on heat demand in kWh/a based on three different refurbishment scenarios (no refurbishment, advanced refurbishment, usual refurbishment).

Residential heat demand is calculated based on a method using the constructional elements of the buildings, as well as on building type and construction period and climatic properties. The constructional elements (exterior wall surface, roof surface, ground surface and volume) are extracted from LoD2 data, whereas building type (single-family house, terraced house or multi-family house) and construction period are predicted using the machine learning algorithm Random Forest. For the climatic properties degree days are used from the German Weather Service stations and are derived area-wide using a linear regression. Three different heat demand scenarios, differing based on the assumed amount of refurbishment conducted are calculated. The three heat demand scenarios are referred to as: no refurbishment, usual refurbishment and advanced refurbishment. The building-specific heat demand is aggregated to a 100x100 m grid and due to privacy protection, cells containing less than 3 residential buildings are not depicted.

This is a product of the project "Entwicklung eines standardisierten Analyse- und Ergebnisrasters für Wärmepläne zur Umsetzung der Energiewende im kommunalen Bereich" (ANSWER-Kommunal) funded by the Federal Ministry for Economic Affairs and Climate Action (Bundesministerium für Wirtschaft und Klimaschutz (BMWK)). Grant number: 020E-SY 48192