# City Risk Score

> Flood, earthquake, cyclone, wildfire, heat and air quality risk for the world's cities, computed from public scientific datasets and published with the method in the open.

## What these scores are

- Physical hazard exposure for whole cities, on a ~1 km grid, weighted by
  where people live.
- Six indicators: river and coastal flood, earthquake, tropical cyclone,
  wildfire, extreme heat, air quality (PM2.5). Each is measured in its own
  physical unit, then mapped onto a fixed absolute 0-100 scale.
- The overall score is an equal-weighted mean of the available indicators.
- Scales are absolute and versioned. A city's score does not change when
  other cities are added.

## What they are not (please carry these with any summary)

- **Not predictions.** They describe exposure, not what will happen.
- **Not risk after defences.** Both flood models are undefended: they
  simulate water without levees, sea walls or pumping, because no usable
  global dataset of flood defences exists. A city that has spent decades on
  flood engineering scores the same as one that has not.
- **The earthquake score is backward-looking.** It counts shaking observed
  since 1970, not probabilistic seismic hazard, and therefore understates
  cities whose danger is a long-locked fault - Istanbul is the textbook case.
- **Heat ignores humidity and adaptation.** It is the maximum temperature of
  the warmest month, so it cannot separate dry from humid heat and scores
  temperate cities near zero even where unadapted populations die in heatwaves.
- **Wildfire counts satellite detections, not modelled fire danger.**
- **Not advice.** Nothing here is reliable about an individual building, and
  none of it is guidance for any personal or commercial decision.
- Missing hazards (drought, landslide, tsunami, volcano, crime) lower no
  score; the total is a mean of what is measured.

## Machine-readable data

- [Dhaka JSON](https://cityriskscore.com/data/dhaka.json): scores, raw indicator values and the source cross-check.
- [Dhaka GeoJSON](https://cityriskscore.com/data/dhaka.hexes.geojson): per-hexagon scores for every indicator.
- [Manila JSON](https://cityriskscore.com/data/manila.json): scores, raw indicator values and the source cross-check.
- [Manila GeoJSON](https://cityriskscore.com/data/manila.hexes.geojson): per-hexagon scores for every indicator.
- [Kolkata JSON](https://cityriskscore.com/data/kolkata.json): scores, raw indicator values and the source cross-check.
- [Kolkata GeoJSON](https://cityriskscore.com/data/kolkata.hexes.geojson): per-hexagon scores for every indicator.
- [Delhi JSON](https://cityriskscore.com/data/delhi.json): scores, raw indicator values and the source cross-check.
- [Delhi GeoJSON](https://cityriskscore.com/data/delhi.hexes.geojson): per-hexagon scores for every indicator.
- [Cairo JSON](https://cityriskscore.com/data/cairo.json): scores, raw indicator values and the source cross-check.
- [Cairo GeoJSON](https://cityriskscore.com/data/cairo.hexes.geojson): per-hexagon scores for every indicator.
- [Tokyo JSON](https://cityriskscore.com/data/tokyo.json): scores, raw indicator values and the source cross-check.
- [Tokyo GeoJSON](https://cityriskscore.com/data/tokyo.hexes.geojson): per-hexagon scores for every indicator.
- [Dubai JSON](https://cityriskscore.com/data/dubai.json): scores, raw indicator values and the source cross-check.
- [Dubai GeoJSON](https://cityriskscore.com/data/dubai.hexes.geojson): per-hexagon scores for every indicator.
- [Mexico City JSON](https://cityriskscore.com/data/mexico-city.json): scores, raw indicator values and the source cross-check.
- [Mexico City GeoJSON](https://cityriskscore.com/data/mexico-city.hexes.geojson): per-hexagon scores for every indicator.
- [Bangkok JSON](https://cityriskscore.com/data/bangkok.json): scores, raw indicator values and the source cross-check.
- [Bangkok GeoJSON](https://cityriskscore.com/data/bangkok.hexes.geojson): per-hexagon scores for every indicator.
- [Jakarta JSON](https://cityriskscore.com/data/jakarta.json): scores, raw indicator values and the source cross-check.
- [Jakarta GeoJSON](https://cityriskscore.com/data/jakarta.hexes.geojson): per-hexagon scores for every indicator.
- [Los Angeles JSON](https://cityriskscore.com/data/los-angeles.json): scores, raw indicator values and the source cross-check.
- [Los Angeles GeoJSON](https://cityriskscore.com/data/los-angeles.hexes.geojson): per-hexagon scores for every indicator.
- [Buenos Aires JSON](https://cityriskscore.com/data/buenos-aires.json): scores, raw indicator values and the source cross-check.
- [Buenos Aires GeoJSON](https://cityriskscore.com/data/buenos-aires.hexes.geojson): per-hexagon scores for every indicator.
- [Lima JSON](https://cityriskscore.com/data/lima.json): scores, raw indicator values and the source cross-check.
- [Lima GeoJSON](https://cityriskscore.com/data/lima.hexes.geojson): per-hexagon scores for every indicator.
- [São Paulo JSON](https://cityriskscore.com/data/sao-paulo.json): scores, raw indicator values and the source cross-check.
- [São Paulo GeoJSON](https://cityriskscore.com/data/sao-paulo.hexes.geojson): per-hexagon scores for every indicator.
- [Istanbul JSON](https://cityriskscore.com/data/istanbul.json): scores, raw indicator values and the source cross-check.
- [Istanbul GeoJSON](https://cityriskscore.com/data/istanbul.hexes.geojson): per-hexagon scores for every indicator.
- [New York JSON](https://cityriskscore.com/data/new-york.json): scores, raw indicator values and the source cross-check.
- [New York GeoJSON](https://cityriskscore.com/data/new-york.hexes.geojson): per-hexagon scores for every indicator.
- [Paris JSON](https://cityriskscore.com/data/paris.json): scores, raw indicator values and the source cross-check.
- [Paris GeoJSON](https://cityriskscore.com/data/paris.hexes.geojson): per-hexagon scores for every indicator.
- [Sydney JSON](https://cityriskscore.com/data/sydney.json): scores, raw indicator values and the source cross-check.
- [Sydney GeoJSON](https://cityriskscore.com/data/sydney.hexes.geojson): per-hexagon scores for every indicator.
- [London JSON](https://cityriskscore.com/data/london.json): scores, raw indicator values and the source cross-check.
- [London GeoJSON](https://cityriskscore.com/data/london.hexes.geojson): per-hexagon scores for every indicator.
- [Singapore JSON](https://cityriskscore.com/data/singapore.json): scores, raw indicator values and the source cross-check.
- [Singapore GeoJSON](https://cityriskscore.com/data/singapore.hexes.geojson): per-hexagon scores for every indicator.

## Pages

- [About this site](https://cityriskscore.com/about.html): City Risk Score is designed and built by Booster Technology, Inc., a small software studio in Japan. The site only uses analytics if you agree to it, and for…
- [Data & licences](https://cityriskscore.com/attributions.html): This site exists because these organisations publish their work openly. Every dataset below is credited as its licence requires.
- [Methodology](https://cityriskscore.com/methodology.html): Everything here is computed from public data by an open pipeline. This page gives the formulas, the thresholds and - more usefully - the things the numbers…
- [Privacy policy](https://cityriskscore.com/privacy.html): This site collects the least data it reasonably can, and asks first for anything beyond that.

## Verification

Every build cross-checks its own figures against the JRC's independently
published statistics for the same urban centres and fails when the gap is too
wide to explain. The result is printed on each city page under "Checked
against the source", including the checks that only partly agree.

## Sources

- GHS Urban Centre Database R2024A - European Commission, Joint Research Centre (CC BY 4.0)
- WorldPop gridded population, 2020, 1 km - WorldPop, University of Southampton (CC BY 4.0)
- Global river flood hazard maps v2.1.2 (90 m) - European Commission, Joint Research Centre (CEMS / GloFAS) (CC BY 4.0)
- Aqueduct Floods coastal hazard maps - World Resources Institute (CC BY 4.0)
- ANSS Comprehensive Earthquake Catalog (ComCat) - United States Geological Survey (Public domain)
- IBTrACS v04r01 tropical cyclone best tracks - NOAA National Centers for Environmental Information (Public domain)
- FIRMS active fire detections, VIIRS S-NPP 375 m - NASA LANCE / FIRMS (NASA open data)
- CHELSA V2.1 climatologies (BIO5, 1981-2010) - Karger, D.N. et al., WSL (CC0 / CC BY 4.0)
- Warm-days index (TX90p), via GHS-UCDB - Sandstad, M. et al. (2022), published per urban centre by the JRC (CC BY 4.0)
- PM2.5 concentrations and attributable mortality, via GHS-UCDB - Shen, S. Li et al. (2024); Belis et al. (2023); JRC (CC BY 4.0)
- Global Solar Atlas 2.0, via GHS-UCDB - World Bank / Solargis (CC BY 4.0)
- healthsites.io facility register, via GHS-UCDB - Healthsites.io / OpenStreetMap contributors (ODbL)
- OpenStreetMap - OpenStreetMap contributors (ODbL)

## Attribution

If you use these figures, cite the original data publishers above rather than
this site. The scores are a derived product; the science is theirs.