Form Submission: Participation Entry

The influence of climate and demographic change on fire regimes is of great relevance to the international community both from the perspective of ecology and social vulnerability. However, the data sets at the global scale are inadequate in terms of their temporal extent and spatial resolution. For a long-term trend analysis or drawing causal inferences regarding human influence on fire at local-scale, high-resolution fire data extending multiple decades is necessary. In this study, I produced a high-resolution burned area map from Landsat image for Nepal. I modified previous fire mapping algorithms to enhance the accuracy in montane topography. The algorithm uses probabilistic classification followed by rejection tests, which are based on the time-varying characteristics of vegetation. The Landsat-based burned area can go as far back as the 1970s and facilitate fine-scale studies of causal relationships between human and fire.