Predicting fire hazard areas using vegetation indexes, Case Study: Forests of Golestan province, Iran

Every year, hundreds of fires occur in the forests and rangelands across world damage thousands hectare trees, shrubs, plants which cause environmental economic damages. This study aims to establish a real time forest fire alert system for better management monitoring Golestan province. In this study, order prepare hazard maps, required layers were produced based on data MODIS sensor data. At first, natural was divided into two categories training test samples randomly. Then, vegetation moisture stresses greenness considered using six indexes NDVI, MSI, WDVI, OSAVI, GVMI NDWI area category day before occurrence long period 15 years, risk threshold parameters addition selecting best spectral index vegetation. Finally, model output validated occurrences category. The results showed possibility prediction site with more than 80 percent accuracy.