Using satellite data to define exposure to and health effects of extreme weather events in Italy 


Since the 1980s, Italy has become warmer, drier and seen an increase in the number and intensity of flash floods and forest fires. The acute health effects of higher temperatures are well known, including increased mortality from cardiorespiratory disease during heat waves. However, temperatures vary substantially as a result of the urban heat island phenomenon, such that the urban context may be an important factor explaining differential health impacts related to climate change.

In recent years, satellite data has become a valuable tool in epidemiology as an alternative source of environmental exposure data and for the identification of areas at risk of environmental hazards. In comparison to other data sources, satellite data provides better spatial coverage, is relatively cheap (e.g., compared to monitoring campaigns) and its quality is continuously improving.

In Italy, there is currently limited information about which population subgroups are most at risk from the effects of extreme weather events within urban areas, and this may be partly due to limited exposure data availability. The aim of this research is to use satellite data to help bridge this gap and carry out appropriate studies to estimate the short-term health effects of heat waves, with a special focus on the urban heat island phenomenon, and urban characteristics.

This line of research will be integrated with similar research conducted in the UK within the MRC-PHE Centre for Environment and Health (Majid Ezzati).


Project aims

The goals of this project are to:

  • Estimate air temperature for Italy with a 1km x 1km resolution by combining satellite data, temperature data from monitoring networks and land-use characteristics. The analysis comprises of a multivariate random-effects 3-step statistical model.
  • Define the characteristics of the urban heat island for Rome and selected large Italian cities
  • Estimate the effect of heat on mortality and hospital admissions for Italy (by municipal level) thus having effect estimates for urban areas suburban and rural areas.
  • Estimate the differential effect of heat within major cities as a result of the urban heat island phenomenon, the different urbanisation and population characteristics (socio-economic conditions, health status).

The model can be generalised to other countries.

This work is currently funded by the Department of epidemiology Lazio Region – ASLRME Rome, providing funding to support a PhD project (awarded to Francesca De Donato).