How to read the map: white = a municipality unaffected by the selected risks. The deeper the colour, the more exposed the municipality. Colours blend where risks overlap.
Liveability in 2050: with all risks selected, the map displays the overall score: deeper green = more liveable.
0 (less liveable)100
Where to live in France in 2050?
Exposure of 35,000 municipalities to climate risks by 2050 (the official TRACC trajectory: +2.7 °C in France). Select entire categories or expand them to isolate individual indicators. Select everything to see the liveability score.
Find a municipality
Displayed risks
Intensities = relative rankings among French municipalities (percentiles), rather than absolute measurements. A pale municipality is not "risk-free". Public sources and the method appear at the bottom of the page.
/100 liveability (all risks equally weighted)
Sources : Météo-France / DRIAS, TRACC-2023 indicators at +2.7 °C (Explore2-Climat, ensemble median, baseline 1976-2005) · Explore2 / RARE network, future flows by municipality (multi-model median, +2.7 °C) · BRGM / Géorisques, exposure to clay shrink-swell (AleaRG 2025, class estimated from 5 points per municipality) · GASPAR: AZI (atlas of flood-prone areas), DDRM (risks recorded by municipality), approved PPRN · BDIFF, forest fires 2006-2025 · IGN OCS GE, land cover (fuel) · INSEE BPE 2024 & populations · Etalab boundaries. Method : each indicator is converted to a percentile rank (0–100) among municipalities (binary indicators such as PPRN take a value of 0 or 100); the intensity of a category is the mean of its selected indicators; colour ranges from white (p0) to full colour (p100), with multiplicative blending. Liveability = 100 − the mean of the 6 categories. Limitations : projections involve uncertainty; the municipality scale smooths out local contrasts; catnat and PPRN also reflect administrative history; fire risk is based on a model calibrated on observed fires in 2006-2021, with its 2050 climate component applied externally (an aggravation factor), rather than estimated by the model; services exclude provision in neighbouring municipalities. Methodological note : indicator-by-indicator detail, sources and limitations are given in the methodological note (PDF, in French).