STORM wind RP100 — tropical cyclone (present climate)
How hard the wind blows in a once-in-a-century tropical cyclone under today's climate — the only cyclone layer Darwin shows, and the band the loss engine integrates.
Source: STORM v4 (Bloemendaal et al., VU Amsterdam / IVM), IBTrACS-based Year: ~2000 (1980–2017 climatology) Category: Physical risks · Mitigating services · Climate Acute Risks Coverage: Global (all six tropical-cyclone basins) Format: Raster grid (~11.5 km) Used in risk analysis: Yes — gates Climate hazards (Storm mitigation) at ≥ 35.5 m/s, and feeds the Direct Physical Loss cyclone channel
Changed on 2026-09-18. The superseded Tropical Cyclone Wind Hazard (present climate) layer left the platform: it lost the cyclone trigger to this layer in an earlier release but kept writing the Storm mitigation flag from its own 50-year band, captioning it with a 32 m/s threshold that no longer exists anywhere in Darwin. It is gone from the picker, from this catalogue and from the database, and the flags it had written were removed with it. Projects carry no cyclone flag until their next assessment, which then recomputes it against this layer at 35.5 m/s. A missing flag is recomputable; a flag standing on a retired raster cannot be told apart from a real one. This layer is now the only cyclone layer the picker shows, and it is filed under Mitigating services · Climate Acute Risks rather than among the engine inputs.
What it shows
How hard the wind blows at a location during a once-in-a-century tropical cyclone, in metres per second. "Once in a century" is the return period — the average interval between two storms at least that strong — so the map compares places at equal rarity rather than at the worst storm each of them happens to have recorded. All six tropical-cyclone basins are covered.
How it is built
STORM v4 fits a statistical model to IBTrACS, the historical record of observed cyclone tracks, and simulates 10,000 years of storms from it; the wind speed reached at each return period is then read off that synthetic record. Darwin imports the whole set, from the 10-year to the 1,000-year storm. This page documents the 100-year band, the one the app reports. Speeds are 10-minute sustained winds.
How to read it
Higher values mean stronger winds at the same rarity. 33 m/s is where the Saffir-Simpson scale starts category 1, 42 m/s category 3.
Class thresholds
A site is flagged on Climate hazards when its 100-year wind reaches 35.5 m/s — a strong category-1 cyclone, inside the layer's own "High" band (32–42 m/s), so the class reading is unchanged. The flag is captioned High tropical cyclone wind hazard (RP100 ≥ 35.5 m/s).
How this threshold was calibrated
It replaces a ≥ 32 m/s rule read on the 50-year band. The figure could not simply carry over, because 100-year winds exceed 50-year ones everywhere, so the same number would have made the flag more permissive. 35.5 m/s is the value that flags the same share of cyclone-exposed pixels as the old rule did. It is a Darwin calibration, not a published standard.
The measurement behind it, for audit: over 2.11 million cyclone-exposed pixels, the median ratio between the two bands is 1.111, and the 100-year median is 35.24 m/s where the 50-year median sits at 32.
Why this layer
It is the raster the Direct Physical Loss engine integrates to price cyclone damage — STORM return-period wind read through the Emanuel (2011) excess-wind sigmoid, on the property-damage and business-interruption channels alike. Because the screening flag now reads the same raster, the flag and the priced loss explain a site with the same picture.
It is also the only cyclone layer left in the catalogue. Its predecessor carried the hazard's only French name, which moved here with the trigger, so a French search for vent de cyclone tropical still finds it.