Fuel type mapping from remote sensing data: Contributions to a sustainable environmental development

Abstract

The conjugation of human development and sustainability of natural resources is a critical issue to our present society. The role of human activity in fire ocurrence is a good example of the difficult dialogue between development and conservation. Recent changes in land use patterns have carried along an increasing rate of fire occurrence, which implies in some cases severe effects on soil erosion and vegetation degradation. For this reason, it is critical to improve current tools to prevent and assess forest fires, so that fire occurrence may be reduced. This paper explores the use of remote sensing methods to derive fuel type maps. These maps are basic sources of information for fire behaviour modelling, since they make it possible to take into account vegetation characteristics for fire propagation. The study area is located in Cabañeros National Park (Central Spain). The fuel type mapping was based in the analysis of two satellite Landsat-TM images and auxiliary information (digital elevation models and texture images). The digital classification worked out to have a global accuracy of 83 %. The larger problems found were to distinguish between the different heights of shrubs and identifying the existence of shrubs under the tress.