The Project

About us

Humans are the main cause of wildfire ignitions worldwide. Human activities related to fire occurrence have been previously studied but in a fragmented and regional way. In addition, humans are also affected by burnings. Fires have important impacts on human lives and resources, particularly when severe fire seasons arise. The Global Analyses of Human Factors of Fire Risk (Anthropofire) project aims to provide an integrated and global view of the main drivers of human fire occurrence and vulnerability to better understand causes and impacts of wildfires on peoples’ lives and values.


Several studies have tried to understand factors behind human-caused wildfires. They are focused on particular regions, and mostly consider fire causes but not fire effects. The Anthropofire projects aims to derive a global database of fire occurrence and human factors previously identified as related to fire danger or fire vulnerability, to create a more comprehensive assessment of fire risk conditions.


Anthropofire aims to identify the main human drivers of fire occurrence at global scale, thus improving the mitigation of the negative effects of fire on humans and ecosystems. It aims to answer the following questions:

  • What is the actual magnitude of human caused fires at global scale?

  • What factors explain human fire ignitions at regional and global scale?

  • What recent climatic and socio-economic changes have affected historically human-driven fire regimes?

  • What are the main impacts of fires on people’s lives and infrastructures?

  • How human factors should be included into fire models and fire risk assessment systems?

  • How human factors of fire ignition and damages should be better communicated?


The project is structured around two geographical scales: global and regional. Both will be structured around a dedicated GIS, which will include the main variables related to fire occurrence and explicative factors, on one hand, and vulnerability factors on the other (figure 1).

The global scale will make it possible to obtain an overall view of the human factors associated to fire ignition, based on satellite burned area products and existing geographical databases.

The regional analysis will be based on sites characteristic of specific fire regimes and will be surveyed from high-resolution satellite images and existing national databases. Once these databases are created, fire occurrence metrics will be estimated from the explicative variables using different statistical and machine learning methods. Regional models will also incorporate the impact of local cultural and economic conditions on fire occurrence. The impacts of national and EU legislation on fire trends will also be considered.