Submitted as: development and technical paper
26 Jul 2023
Submitted as: development and technical paper |  | 26 Jul 2023
Status: this preprint is currently under review for the journal GMD.

High-resolution multi-scaling of outdoor human thermal comfort and its intra-urban variability based on machine learning

Ferdinand Briegel, Jonas Wehrle, Dirk Schindler, and Andreas Christen

Abstract. As the frequency and intensity of heat waves will continue to increase in the future, accurate and high-resolution mapping and forecasting of human outdoor thermal comfort in urban environments is of great importance. This study presents a machine learning based outdoor thermal comfort model with a good trade-off between computational cost, complexity, and accuracy compared to common numerical urban climate models. The machine learning approach is basically an emulation of different numerical urban climate models. The final model consists of four sub-models that predict air temperature, relative humidity, wind speed, and mean radiant temperature based on meteorological forcing and geospatial data on building form, land cover, and vegetation. These variables are then combined into a thermal index (Universal Thermal Climate Index – UTCI). All four sub-model predictions and the final model output are evaluated using street-level measurements from a dense urban sensor network in Freiburg, Germany. The final model has a mean absolute error of 2.3 K. Based on a city-wide simulation for the city of Freiburg we demonstrate that the model is fast and versatile enough to simulate multiple years at hourly timesteps to predict street-level UTCI at 1 m spatial resolution for an entire city. Simulations indicate that neighborhood-averaged thermal comfort conditions vary widely between neighborhoods, even if they are attributed to the same local climate zones, e.g. due to differences in age and degree of urban vegetation. Simulations also show contrasting differences in the location of hot spots during the day and at night.

Ferdinand Briegel et al.

Status: open (until 18 Oct 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Ferdinand Briegel et al.

Ferdinand Briegel et al.


Total article views: 258 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
187 66 5 258 6 3
  • HTML: 187
  • PDF: 66
  • XML: 5
  • Total: 258
  • BibTeX: 6
  • EndNote: 3
Views and downloads (calculated since 26 Jul 2023)
Cumulative views and downloads (calculated since 26 Jul 2023)

Viewed (geographical distribution)

Total article views: 246 (including HTML, PDF, and XML) Thereof 246 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 03 Oct 2023
Short summary
This study presents a new approach to model heat stress in cities using artificial intelligence (AI). We show that the AI model is fast in prediction but accurate when evaluated with measurements. The fast-predictive AI model enables several new potential applications, including heat stress prediction and warning, downscaling of potential future climates, evaluation of adaptation effectiveness, and, more fundamentally, development of guidelines to support urban planning and policy-making.