Remote Sensing of Forest Fire: Approaches in Burned Area Detection
Writer : Phua Mui How
Date : 5 Dec 2011
Publisher : Seminar Bencana Alam 2011, Universiti Malaysia Sabah, Kota Kinabalu.
Forest fire is an threatening disaster in the world. Recurrent fires associated with El-Nino events have been destroying vast areas ofBorneo’s forests. Accurate quantitative information about the frequency and distribution of the burned areas are imperative to fire management but are lacking in the tropics. This paper provides a brief review on widely used burned area detection methods. The remote sensing approach for detecting change can be broadly classified into post-classification comparison and spectral change detection methods. The post-classification comparison method involves the analysis of differences between two classified images. On the other hand, the spectral change detection method requires the changes in land cover result in persistent changes in the spectral signature of the fire-affected land surface. A recently developed hybrid method that enabled the detection of burned areas of multiple fire events is also explained. The hybrid method shows superior performance to the post-classification comparison and spectral change detection methods.