A heat health hazard index based on ECMWF data

Claudia Di Napoli, Christopher Barnard, Christel Prudhomme, Florian Pappenberger


ECMWF has begun to produce pre‐operational forecasts of a heat health hazard index in real time using meteorological forcing data from its Integrated Forecasting System (IFS). The high‐resolution and probabilistic forecasts, which predict the Universal Thermal Climate Index (UTCI) at the global scale, will become available to users in the second half of 2020.

The production of human thermal stress data from ECMWF’s meteorological products answers the call, issued by the World Meteorological Organization in 2004, to incorporate biometeorological forecasts into the suite of products and services offered by meteorological and hydrological services. Biometeorological forecasts consist in the prediction of weather‐related conditions, such as thermal stress due to extremes of heat and cold, that might negatively impact human health. Extreme heat, for instance, is responsible for a variety of risks, including dehydration, cramps, and even death, especially during sustained periods of high temperatures (heatwaves). Mortality and other health problems (frostbite, hypothermia) can also occur as a result of exposure to extreme cold.

Calculating the index

The UTCI is a state‐of‐the‐art indicator representing the thermal (heat and cold) stress of the human body. It is calculated using an advanced model of human thermoregulation coupled with a clothing insulation model.
The models estimate the effect of air temperature, wind speed, water vapour pressure and short‐ and long‐wave radiant fluxes on human physiology. The UTCI takes values on a stress category scale ranging from extreme cold stress to extreme heat stress. A pre‐operational system has been developed at ECMWF to produce UTCI forecasts in real time. The system uses as forcings the forecasts of air temperature, humidity, wind speed and mean radiant temperature (MRT) from ECMWF’s Integrated Forecasting System (IFS). MRT is a critical physical quantity representing how human beings experience radiant fluxes. Its automatic computation is fully integrated into the system. UTCI and MRT outputs are generated at the same time as ECMWF’s extended‐range forecasts up to 46 days ahead.

Reanalysis data

The system has also been used to generate historical datasets of UTCI and MRT based on the ERA5 reanalysis. The datasets currently span the period from 1979 to the present and provide a record of thermal stress in past extreme events. Reanalysis UTCI gridded data for the June 2019 European heatwave, for instance, show widespread conditions of thermal stress across the continent. As can be seen in the figure, extreme heat stress hazardous to human health occurred in parts of Spain, France and Italy.

Reanalysis UTCI and MRT data are now available for download through the Climate Data Store run by the Copernicus Climate Change Service (C3S) implemented by ECMWF. Forecasts and reanalysis data will be archived in ECMWF’s Meteorological Archival and Retrieval System (MARS) from the second half of 2020.

Heat stress during the June 2019 heatwave. The map shows locations where conditions of heat stress, as represented by UTCI reanalysis data, occurred between 24 June and 2 July 2019 in Europe.