Developing a Historical Building Classification for Mapping Urban Thermal and Morphology Parameters in Urban Climate Models

Historic variations in building regulations and construction practices have shaped the thermal properties of today’s urban fabric, yet these differences are often neglected or oversimplified in urban weather and climate models. Building materials, insulation levels, and construction techniques vary strongly across construction periods, leading to spatial differences in heat storage and release, energy demand, and vulnerability to temperature extremes. Most current urban mesoscale models rely on generic classifications, such as Local Climate Zones (LCZs), which limits their ability to capture this heterogeneity and reduces the accuracy and reliability of weather and climate forecasts.

Using Amsterdam (The Netherlands) as a case study, we present a novel geospatial modeling framework that explicitly incorporates historical building characteristics into numerical weather and climate simulations. Based on detailed cadastral data and an extensive review of historical building regulations and practices in The Netherlands, we define ten “heritage building classes” representing distinct construction periods and their typical thermal properties. These classes are mapped across the city using GIS techniques, producing high-resolution heritage building maps. For each class, representative thermal parameters are derived and implemented into the Weather Research and Forecasting (WRF) model.

We assess the sensitivity of simulated urban temperatures to these period-specific building properties and evaluate model performance against in-situ meteorological observations from the Amsterdam Atmospheric Monitoring Supersite (AAMS). This approach provides a scalable pathway for integrating historically informed building characteristics into urban climate models, creating a foundation for future improvements in urban climate simulation.