EGU24-9597, updated on 08 Mar 2024
https://doi.org/10.5194/egusphere-egu24-9597
EGU General Assembly 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Interactive instant urban climate modelling with AI and LEGO-cities

Andrei Covaci1, Mohamed Firas Kooli1, Hossein Dehghanipour1, Camelia El Bakkali2, Sara Top3, and Lesley De Cruz1,4
Andrei Covaci et al.
  • 1Electronics and Informatics Department, Vrije Universiteit Brussel, Brussel, Belgium
  • 2Community Ecology Laboratory, Vrije Universiteit Brussel, Brussel, Belgium
  • 3Department of Physics and Astronomy, Ghent University, Ghent, Belgium
  • 4Royal Meteorological Institute, Brussels, Belgium

Due to climate change, extreme weather conditions such as heatwaves are becoming increasingly common. Additionally, urbanized areas cause elevated temperatures compared to rural areas, especially during clear and calm nights. This effect is known as the urban heat island. Both climate change and growing cities lead to more intense and frequent temperature extremes causing more frequent and more severe heat stress. Heat stress correlates with cardiovascular diseases and excess mortality [Liu et al., 2020]. A major challenge that cities face today is the implementation of climate adaptation measures to counteract the increased heat stress, for example, by planning more green spaces.  

To spread awareness and for science outreach purposes, we have created the 'Instant urban climate with AI' workshop. In this interactive workshop, participants can build their own simplified city using colour-coded LEGO blocks that represent different land use types.  The participants can insert their city in a box with a camera, which takes an aerial photo of the LEGO city and displays the corresponding temperature map. This allows the participants to investigate the impact of land use on the temperature in their city. 

To increase the game factor, the “Cool your city” game was introduced. The participants start from a default city map, which they then need to adapt with the aim to lower the average temperature as much as possible during hot summer nights. Participants looking for additional challenges can also rebuild parts of the city of Brussels and investigate how land usage in the city impacts the temperature for different weather conditions. 

To make the outcomes of the game realistic, we used a machine learning (ML) model trained on urban weather observations from the regional VLINDER network [Caluwaerts et al., 2021] to compute the temperature for the participants’ LEGO cities. With this project, we established connections from urban climate to other research domains, such as citizen science and artificial intelligence. By providing a low-threshold, fun and tangible way to explore these topics, we hope to make such transdisciplinary research accessible for audiences aged 4 to 104, regardless of socio-economic status, gender or language. The Instant Urban Climate with AI workshop was presented at two outreach events in Brussels, Belgium (the three-day I Love Science Festival and Day of the Sciences) with big turnover rates, over 150 participants per day, and overall positive feedback and engagement. 

Caluwaerts, S., Top, S., Vergauwen, T., Wauters, G., De Ridder, K., Hamdi, R., Mesuere, B., Van Schaeybroeck, B., Wouters, H. and Termonia, P., 2021. Engaging schools to explore meteorological observational gaps. Bulletin of the American Meteorological Society, 102(6), pp.E1126-E1132.  

Liu, J., Varghese, B. M., Hansen, A., Zhang, Y., Driscoll, T., Morgan, G., ... & Bi, P. (2022). Heat exposure and cardiovascular health outcomes: a systematic review and meta-analysis. The Lancet Planetary Health, 6(6), e484-e495.

How to cite: Covaci, A., Kooli, M. F., Dehghanipour, H., El Bakkali, C., Top, S., and De Cruz, L.: Interactive instant urban climate modelling with AI and LEGO-cities, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9597, https://doi.org/10.5194/egusphere-egu24-9597, 2024.