Detecting mistletoe infestation on Silver fir using hyperspectral images

Silver fir is one of the most important commercial and ecological species in Croatia, and is currently the most damaged tree species in Croatian forests. It is increasingly being infested by white mistletoe, whose irrational consumption of water and nutrients contributes to physiological weakening of the tree. Mistletoe may indicate the presence of stress factors in the site and may even cause premature death. The increment of mistletoe-infested trees is lower compared to that of non-infested trees. Hence, these trees should have a cutting priority, while their timber still retains its commercial value. Areas of mistletoe-infested fir forests are expanding; therefore, there is a high correlation between silver fir forest damage and mistletoe distribution, with an almost exponential increase in mistletoe populations. The purpose of this research was to develop an efficient and reliable method of mistletoe detection. Surveying was performed in the mountain area of beech-fir forests, using a hyperspectral scanner. The Spectral Angle Mapper (SAM) classification for 5° proved to be the best classification method. The results were also confirmed by terrestrial validation, i.e. aerial surveying with a non-pilot aircraft from a height of ~30 m above the crowns. The results of this research do not differ significantly from the results obtained with standard ground methods. Limitations of standard ground methods further highlight the importance and efficiency of hyperspectral imagery in mistletoe detection. Furthermore, the first “spectral signatures” - endmembers for fir and mistletoe, were determined, and entered into a spectral library database. The results of this research are applicable primarily in the field of forest management, protection and harvesting, but they can also supply experts in other fields with a basis for explaining various observed phenomena. It outlines some of the possibilities of the application both in scientific and operative fields.