POSTFIRE: A model to map forest fire burn scar and estimate runoff and soil erosion risks

Dennis Fox (1) , Y. Laaroussi (2, 3) , L D Malkinson , F Maselli , Julien Andrieu (1) , L Bottai , L Wittenberg

1 ESPACE - Études des Structures, des Processus d’Adaptation et des Changements de l’Espace
2 IES - Institut d’Electronique et des Systèmes
3 NANOMIR - Composants à Nanostructure pour le moyen infrarouge

Forest fires in the Euro-Mediterranean region burn about 450,000 ha each year. Combustion of standing vegetation and the leaf litter leave the soil bare and vulnerable to runoff and erosion thereby increasing downstream risks of flooding. Typically, wildfires near or in the Wildland Urban Interface (WUI) occur on uplands and runoff generated from the burned area concentrates downstream in high density urban areas. Local authorities must decide on a range of post-fire measures to mitigate risks quickly since most large fires occur late in summer shortly before the fall-winter rainy season. A GIS-based model that maps the burn scar and quantifies fire impacts on runoff and soil erosion is a useful tool in defining a strategy. The POSTFIRE model executes 54 routines semi-automatically in IDRISI TerrSet and maps the burn scar using pre and post-fire satellite images, calculates the impact of a fire on total rainfall event runoff, and maps soil erosion rates. In addition to satellite images, the model requires a Digital Elevation Model (DEM), a mask of the general contour of the fire, land cover map, and tables of runoff coefficients and sediment concentration values for the land covers. All other input files are generated automatically by the model. Burn scar classification, runoff and soil erosion maps were compared for two fires in SE France based on SPOT and Landsat images. Differences in burn scar area estimated from unsupervised KMeans classifications of dNDVI and dNBR indices range from 3% to 15% depending on the fire. dNDVI burn scars had slightly greater areas than dNBR derived maps, and SPOT burn scar classifications were slightly greater in area than from Landsat images. Estimated post-fire stream discharge values were about 2-2.5 times greater than pre-fire values and differed by less than 10% between indices and sensors. Similarly, post-fire erosion rates (2.0-2.5 T ha À 1) were consistent among indices and sensors and coherent with field measurements of erosion. Preliminary results indicate the model is simple to use, adaptable to local conditions, and provides realistic outputs for the burn scar, discharge, and soil erosion maps after a forest fire. Model modules, instructions, and sample data can be downloaded here: https://goo.gl/PjDv68.

Domaines: Sciences de l'environnement

Dennis Fox, Y. Laaroussi, L D Malkinson, F Maselli, Julien Andrieu, et al.. POSTFIRE: A model to map forest fire burn scar and estimate runoff and soil erosion risks. Remote Sensing Applications: Society and Environment, 2016, 4, pp.83-91. ⟨10.1016/j.rsase.2016.07.002⟩. ⟨hal-02566494⟩