Post-fire forest disturbance monitoring using remote sensing data and spectral indices

Wildfires are recurring in many terrestrial ecosystems all over the world. Accurate assessment of the forest ecosystem, affected by fire is of great importance for the fires spread predicting and modelling of the post-fire activities for recovery of the affected territories. High spatial and spectral resolution satellite data were used to evaluate the vegetation variation on a fire-affected territory, located on the northwest slopes of the Rila mountain, considering its spatial heterogeneity. The forest fire was spread on the area of deciduous forests Turkey oak (Quercus cerris L.), and coniferous: Scots pine (Pinus sylvestris L.) and European larch (Larix desidua, Mill.). Different spectral indices like Disturbance index (DI), Normalized difference greenness indices (NDGI) and Normalized Difference Vegetation Index (NDVI) and derived from remote sensing methods (satellite data from different sensors Landsat and Sentinel) as well as the Geographical Information System (GIS) were applied for the forest disturbance assessment in two periods after forest fire occurrence. The results of the applied integrated model provide a quantitative information about the fire effects for distinct forest types. The documented spatial distribution of the territory based on the obtained DI values shows clear differences between the fire-affected forest types, thus demonstrating the usefulness and accuracy of the approach followed.