Explorant la petjada dels incendis amb drons en els ecosistemes forestals

Abstract

Forest fires episodis are something frequent and occur naturally in the territory; although they can also occur due to anthropogenic causes. With the advance of climate change, an increase in its risk is expected, as well as in the intensity and extent of these episodes. In the case of Catalonia, the increase in the risk of fire is also due to afforestation, due to the abandonment of rural life that occurred around the sixties of the last century. There are different techniques for recognizing burnt areas: in the field (on site) or with remote systems. These remote systems are based on obtaining images from satellites or unmanned aerial vehicles (drones). The use of drones has increased recently and, compared to satellites, the main advantage is the high resolution of the images. This work explores the use of remote sensing with drones and multispectral cameras as recent tools for the exploration of vegetation and burned areas. Based on the data collected, different vegetation indices and surface models have been calculated, which allow an assessment of the state of the post-fire vegetation. The results obtained show that within the wide range of vegetation indices, some are more suitable than others depending on the characteristics of the study area. The vegetation index maps show differences in coloration within the burn area when comparing the scenarios before and after the fire. The resolution of the drone images has made it possible to delimit each of the peaks and study the dNDVI value tree by tree. Although the burns carried out by the firefighters only burn the herbaceous and shrub layer, it has been seen that the tops within the burning area have been partly affected by the fire. Finally, the comparison of satellite images with drone images shows that the former have less spatial resolution than the latter and, therefore, less useful in studies where precision is required