• Energy Research

Abstract Building stocks account for a large share of energy consumption and harbour great potential for reducing greenhouse gas emissions. The field of urban building energy modelling (UBEM) offers a range of approaches to inform climate protection policies, producing output of different granularity and quality. We compare two typology-based (archetype) approaches to urban heat demand calculation in a mixed-use area in Berlin, Germany. The goal is to show challenges and pitfalls and how remote sensing can improve the modelling. The first approach uses 2D cadastral data and specific heat demand values from a typology. For the second approach, we derive a 3D building model from aerial imagery, use parameters from the same typology, and calculate the heat balance for each building. We compare the differences in several geometric parameters, U-values and the heat demand. Additionally, we analyse if window detection on aerial image textures and surface temperatures from aerial infrared thermography can improve the estimated window-wall ratios and U-values. The two heat demand approaches lead to different results for individual buildings. Averaging effects reduce the differences at an aggregated level. Remote sensing can be used to improve some geometric parameters needed for modelling, but still requires additional research regarding U-value estimation.