Spatial and temporal characterization of wildfires, human and biophysical factors in Portugal

Wildfires are uncontrolled fires that can burn the forest and agricultural parcels, semi-natural areas and wildlands (e.g., forests, scrublands and abandoned agricultural areas). The wildfire-prone regions in the world range from tropical savannahs to boreal forests, characterized by factors and conditions required for fire activity. In the last 20 years, wildfires were the 5th costliest and the 6th most frequent type of disaster in the world. In the same period, Europe experienced a high number of wildfires and burnt areas, mainly concentrated in the Mediterranean basin and with increasing trends in countries such as Portugal. The fire incidence presents high spatial and temporal (intra- and inter-annual) variability patterns associated with human activities, including land use/land cover changes, extreme weather conditions, climate variability and climate changes.

The objectives of this study include the identification and characterization of the spatial and temporal variability of wildfire incidence, as well as its main drivers, in Portugal. The study uses the most recent fire and environmental databases, analyses wildfires with different causes (e.g., negligent and intentional wildfires) and, in particular, the influence of weather and climate variability and extremes, namely heat waves (HW) and droughts (D) on the occurrence of large fires.

Obtained results comprise the spatial and temporal patterns of wildfires and the assessment of the main drivers of wildfires. We conclude that extreme weather conditions (HW and D) can explain the spatiotemporal patterns of large wildfires, which are responsible for the vast majority of the total burned area. Nevertheless, other factors (such as vegetation type, topography, distance to roads) can explain part of the variability patterns. Our findings are fundamental for forest, landscape, and wildfire management, as they include the identification and characterization of the areas and frame periods where fires are more frequent and have a greater impact. Additionally, this information, together with the identification of the nature of the main danger factors, can support monitoring and forecasting systems, aiming at the development of strategies for wildfire prevention, preparedness and response activities, as well as adaptation to climate change.

 

Acknowledgements:

This work was financed by the National Funds through FCT - Foundation for Science and Technology under the project UIDB/04033/2020. This work was also supported by the project FRISCO - managing Fire-induced RISks of water quality Contamination (PCIF/MPG/0044/2018), and funding attributed to the CE3C research center (UIDB/00329/2020).