Deriving forest fire probability maps from the fusion of visible/infrared satellite data and geospatial data mining

摘要: Information on fire probability is of vital importance to environmental and ecological studies as well management. This study aimed at comparing two forest mapping techniques, one based primarily freely distributed EO (Earth observation) data from Landsat imagery, another purely GIS modeling. The Normalized Burn Ratio (NBR) computed was used detect the high severity area NBR difference between pre- post-fire conditions. GIS-based modeling a multi criterion evaluation technique, into which other attributes like anthropogenic natural sources were also incorporated. ability both techniques map evaluated for region in India, suitable ancillary had been previously acquired support rigorous validation. Subsequently, conceptual framework prediction zones an newly introduced herein fusion technique constructed. Overall, EO-based found be most option, since it required less computational time resources comparison approach. Furthermore, approach offered appropriate path developing identification model long-term pragmatic conservation forests. potential these approaches may provide information that can useful mitigation policy makers, assist resilience practices.