As part of the International Conference on Urban Climate (ICUC12), the World Bank-GFDRR team under the Global Program on Urban Heat Resilience and the City Resilience Program hosted a hands-on hackathon focused on urban heat resilience. We designed a set of challenges across three spatial scales: city, neighborhood, and building, and provided curated datasets for Timbuktu (Mali) and Davao (Philippines).
Participants were invited to explore these datasets using starter code based on our own workflows. The goal: develop scalable and actionable tools to support climate adaptation and urban planning. Whether it's identifying locally effective cooling interventions or helping a mayor make the case for heat shelters, this repository holds everything used in the hackathon:
- Ready-to-run Python notebooks
- Open-source urban datasets
- Real-world challenge descriptions
- Tools for modeling, analysis & visualization
-
Start with the documentation, under the
docs/folder. Begin with:- The motivation behind the project
- A general overview of the hackathon setup
- Details about the input datasets used
-
Explore the notebooks
Notebooks are organized by spatial scale and model type (TARGET - city, SOLWEIG - neighborhood, ENERGYPLUS - building/class room). Each notebook loads helper functions from the functions/ folder.
- Running the notebooks
All notebooks are designed to run smoothly in Google Colab. They should also work in other environments, just make sure to adjust the BASE_DIR path in the section labeled “Section to adjust” to reflect your own file structure, pointing to the data that is available as .zip per city under data/
We hope this repo inspires further innovation in tackling urban heat stress. Feel free to fork, adapt, and contribute!