Estimating the area burned by agricultural fires from Landsat 8 Data using the Vegetation Difference Index and Burn Scar Index

Obtaining an accurate estimate of the area of burned crops through remote sensing provides extremely useful data for the assessment of fire-induced trace gas emissions and grain loss in agricultural areas. A new method, incorporating the Vegetation Difference Index (VDI) and Burn Scar Index (BSI) models, is proposed for the extraction of burned crops area. The VDI model can greatly reduce the confounding effect of background information pertaining to green vegetation (forests and grasslands), water bodies and buildings; subsequent use of the BSI model could improve the accuracy of burned area estimations because of the reduction in the influence of background information. The combination of VDI and BSI enables the VDI to reduce the effect of non-farmland information, which in turn improves the accuracy and speed of the BSI model. The model parameters were established, and an effects analysis was performed, using a normalized dispersion value simulation based on a comparison of different types of background information. The efficacy of the VDI and BSI models was tested for a winter wheat planting area in the Haihe River Basin in central China. In comparison with other models, it was found that this method could effectively extract burned area information.