Extracting Burned Areas Based On Remote Sensing And Spatial Analysis Of Burn Severity Using Landsat TM

This paper was to present a new method of identifying burned areas which chose Huzhong region of the Great Xing'an Mountains as research area,September 2010 post-fire TM image and September 2007 pre-fire TM image were as the basic data.DEM image and forest types map as the auxiliary data.We used NDVI,NDSWIR,MNDWI,dNBR and other RS indices to build a decision tree classification model,then science systematically analyzed the spatial distribution of fire severity with topographic factors such as slope,aspect and elevation.This paper could provide theoretical basis and data support for forest fire prevention and management of Great Xing'an Mountains.Used this model to identify ten burned areas of Huzhong in 2010.Fire severity was divided into four classes according to the threshold value of dNBR,then overlaying analyzed between fire severity map and slope,aspect,elevation using SuperMap.The overall accuracy and Kappa coefficient of decision tree classification were 97.97% and 0.9432.Compared with the Parallelepiped method and ISODATA method,the total classification accuracy was 7.56% and 17.32% higher,respectively.The Kappa coefficient was also increased.In the decision tree method,the producer's accuracy and user's accuracy were 97.51% and 97.54%,the Parallelepiped method were 90.43% and 96.52%,the ISODATA method were 94.35% and 95.68%.Fire severity was divided into four classes according to the threshold value of dNBR:unburned,low,moderate and high.Moderate severity burned area was 46.6% of total,high severity burned area was 33.2%.After overlaying analyzing,64.4%(4177ha)of burned area located at the elevations from 1000 m to 1500 m.and 45.9% of burned area located at level ? slope(15°~24°).The burned area at the southern slope occupied 21.4%(1391ha)of the total.The decision tree classification model presented in this paper could identify burned areas accurately and the total classification accuracy was higher than Parallelepiped method and ISODATA method,and the burned area was closer to the method of visual interpretation.Moderate and high severity burned areas occupied most of the total burned areas,there was a Correlation between the burn severity and slope,aspect,elevation.