Integrated approach for the assessment of forest fire risk and burn severity mapping using GIS, AHP method, and Google Earth Engine in Western Algeria

Abstract

In Algeria, more than 29% of the forest heritage is located in the western part where forest fires remain the most devastating factor. The assessment of the Forest Fire Risk (FFR) and the mapping of very vulnerable areas seem to be the most effective ways to meet this alarming situation. Remote sensing and Geographic Information System (GIS)-based multicriteria decision analysis (MCDA) methods have improved their efficiency in this domain. In addition, the web platform of Google Earth Engine (GEE) has made the processing of remote sensing data very fast. This study, conducted in the Louza Forest located in Sidi Bel Abbes province in western Algeria, aims to assess FFR using two methods: the Dong model and analytic hierarchy process (AHP) method, which integrate both three effective factor types such as vegetation cover, topography, and human activities. Likewise, the Burn Severity Map (BSM) was made using the Difference Normalized Burn Ratio (dNBR) obtained from Landsat data. The obtained FFR maps from the two models were compared with BSM. Results show a dominance of moderate and high burn severity of 57.68% and 29.13% of the total forest area. The statistical analysis demonstrates that most of the burned areas were located in the moderate, high, and very high risk levels with a rate of 64.27% and 69.41% for Dong and AHP models, respectively. This study shows the effective contribution of GIS and remote sensing to provide a very useful solution for forest managers for sustainable management and the preparation of forest fire protection plans.