Energy system models on country level usually contain multiple energy carriers at different granularity. While data is comparably rich in terms of temporal and spatial resolution for the electricity part, much less is known for heat. Especially the true demand for heat as a function of usage and time is difficult to obtain. In many cases, energy consumption data (fuel oil, natural gas, district heating etc.) is taken as approximation for the final energy end-use of heat. Different heat distribution technologies bring their own bias on temperature levels and heating hours, like with ground floor heating vs. radiator, Therefore, historic consumption data is not an appropriate base for modelling of energy systems with long prospect. The present research work proposes a novel top-down methodology for generating aggregated load curves on heat demand, with a focus on residential space heating. Maps of population density distribution combined with norm temperature proffles and the definition of heating degree days provides a tempo-spatial map of heating demand. The knowledge of total residential space heating demand is used to identify the aggregated demand curve, suitable for energy system modelling. (C) 2018 The Authors. Published by Elsevier Ltd.