Research status /
Methods and validation status
Shaping Cool Cities brings several ways of reading urban heat into one spatial frame. This note separates the available research inputs and figures from the model results being rechecked.
What existsSix-city voxel models and comparative research figures, a 40,344-cell dataset, a Python pipeline and explorable input maps.
The research inputs
The study covers Amsterdam, Athens, Barcelona, Berlin, Madrid and Paris. A Python pipeline brings Landsat temperature and vegetation, VoxCity 3D city models, Urbanity street networks, Global Streetscapes image semantics and EUBUCCO footprints onto a common grid with stable city and cell identifiers.
The 40,344-cell dataset, six-city voxel models, comparative figures and selected public input maps show how the sources were assembled. They provide material for inspecting urban form and data coverage.
Validation in progress
Model validation is being rebuilt after my own re-audit found problems in the temperature target and in how the spatial cross-validation folds were separated. Until that work is finished, no model score, explanation or cooling scenario here is a validated claim.
The rebuild addresses corrected temperature inputs, coordinate leakage and complete spatial holdouts. Earlier XGBoost and SHAP explorations describe associations; they do not establish the cooling effect of an intervention.
Present limits on use
The inputs and figures can support discussion of source alignment, urban geometry and the research method. The study does not yet provide validated predictions or evidence for ranking cooling interventions.
This is a portfolio status note, not a completed validation report. Earlier performance and cooling figures in the original materials remain unvalidated.
Read the original work
Research repository (opens in a new tab)
Code and methods for the six-city study. Earlier performance and cooling claims remain unvalidated while model validation is rebuilt.
Original research presentation (opens in a new tab)
Visual presentation of the six-city study, data sources and methods. Its earlier performance and cooling figures remain unvalidated while model validation is rebuilt.