Clustered spatially and temporally resolved global heat and cooling energy demand in the residential sector.

• A high-resolution spatio-temporal approach for estimating global heat and cooling energy demand. • K-means clustering for deriving energy density bands of each heat and cooling energy demand. • Open-access data for spatial energy density bands for 165 countries covering 99.96% global energy users. • 5% of heat demand is at very high energy densities worldwide, while >50% is at very low density. Climatic conditions, population density, geography, and settlement structure all have a strong influence on the heating and cooling demand of a country, and thus on resulting energy use and greenhouse gas emissions. In particular, the choice of heating or cooling system is influenced by available energy distribution infrastructure, where the cost of such infrastructure is strongly related to the spatial density of the demand. As such, a better estimation of the spatial and temporal distribution of demand is desirable to enhance the accuracy of technology assessment. This paper presents a Geographical Information System methodology combining the hourly NASA MERRA-2 global temperature dataset with spatially resolved population data and national energy balances to determine global high-resolution heat and cooling energy density maps. A set of energy density bands is then produced for each country using K-means clustering. Finally, demand profiles representing diurnal and seasonal variations in each band are derived to capture the temporal variability. The resulting dataset for 165 countries, published alongside this article, is designed to be integrated into a new integrated assessment model called MUSE (ModUlar energy systems Simulation Environment) but can be used in any national heat or cooling technology analysis. These demand profiles are key inputs for energy planning as they describe demand density and its fluctuations via a consistent method for every country where data is available. [ABSTRACT FROM AUTHOR]