Modelling fire hazard in the southern Mediterranean fire rim (Bejaia region, northern Algeria)

Environ Monit Assess. 2019 Nov 13;191(12):747. doi: 10.1007/s10661-019-7931-0.

Abstract

The southern rim of the Mediterranean Basin (MB) has a long fire history but fire hazard is poorly investigated in comparison to the northern rim. We built a fire database using MODIS data (2001-2015) for an area typical of the northern coastal Algeria (Bejaia region) in order to decipher the role of environmental and anthropic controls on the fire frequency and the area burnt. We found a high role of bioclimate, which controls the fuel dryness, ignitability, and biomass. Maximal fire frequency and burnt areas were recorded in northern sub-humid areas with high amounts of forests and shrublands, and fire was limited in the southern sub-arid area. Humans set most fires, and preferentially burn forests, shrublands, pastures, groves, and agricultural lands. The maximal fire frequency and burnt area occurs in wildland urban interfaces characterized by forest-shrublands mosaics with disseminated habitats. Fire activity is low to medium in rural-urban interfaces characterized by agropastoral areas with high habitat density and large habitat patches. Small to large crown fires occur in forests and shrublands, while small surface fires predominate in agropastoral areas and groves. Large fires (> 100 ha) are rare (10%) but contribute for ca. 50% to the total area burnt. These fire features are typical of many rural countries of the southern rim of the MB, and contrast with those on the northern rim. Based on this, we propose to improve the prevention, the detection, and the management of forest fires in the long term and to protect forests that host high biodiversity in Algeria.

Keywords: Fuel limitation; Normalized difference vegetation index; Rural urban interfaces.

MeSH terms

  • Algeria
  • Biodiversity
  • Biomass
  • Ecosystem
  • Environmental Monitoring*
  • Fires / statistics & numerical data*
  • Forests
  • Humans
  • Proportional Hazards Models*