< Back to Data Catalog
< Back to Data Catalog

Extreme Cold - Human Risk, Projections, Decadal, Global (v2.0)


The extreme cold projection product uses the same method as the historical product but is forward looking. This data is intended to be used to assess the projected extreme cold conditions under two future scenarios (SSP2-4.5 and SSP5-8.5) from 2030-2090.

thermal, global, risk, cold, wind chill, projections, regional, hazard, ssp245, future, temperature, cmip6
Temporal Extent:
2025-01-01 to 2095-01-01
11 km
Product Version:

Technical Description

The global Extreme Cold Risk Projection products capture projected risk of extreme cold across the globe using well-established thermal stress metrics. These risk maps reflect the future risk from extreme cold in global land regions at decadal time steps for Shared Socioeconomic Pathway scenarios SSP2-4.5 and SSP5-8.5. These maps use future CMIP6-based projections of temperature and wind speed to calculate likely changes to extreme cold conditions into the future. The spatial scale for this dataset is 11 km/pixel and data is available for the years 2030 - 2090 (where windows centered around the decadal years are used to calculate the projected extreme cold conditions). Extreme cold frequencies (between 0-1) are provided for the combinations of SSP and time horizon.

This dataset contains the following fields:

  • extreme_cold_frequency: Frequencies of extreme cold occurrence at pixel locations over the future time period. This looks at the frequencies of extreme cold occurrence at pixel locations over the relevant future time period used for calculation. For instance, a value of 0.5 would indicate that half of the days were under extreme cold conditions for that location. This is a value between 0.0 and 1.0.
  • extreme_cold_risk_class: The extreme cold risk classification for each value. Values are integers from 1 to 5 representing the following classes: Very Low Risk (1), Low Risk (2), Moderate Risk (3), High Risk (4), and Very High Risk (5). Values are categorized using the 44th, 67th, 84th, and 94th percentiles of the
  • extreme_cold_frequency field, exclusive of '0' values.