Valutazione dello stato sanitario delle piante forestali: quali prospettive dal telerilevamento nel visibile e nell’infrarosso?

Tree health monitoring: perspectives from the visible and near infrared remote sensing. Based on a comprehensive literature analysis, we present a critical review of those optical remote sensing techniques operating with the visible (VIS) and near infrared (NIR) bands for the assessment of health in forest trees. Physical, biological and physio-pathological issues of VIS-NIR reflectance of leaves are described pointing out that a decrease of NIR reflectance is highly influenced by stress conditions on tree caused by abiotic and biotic fac - tors. In many cases the NIR spectral band is more sensitive than the VIS one, allowing to detect plant stress long before the appearance of visible symptoms. A description of the main remote sensing methods is provided, including radiometric measurements and multispectral imaging approaches. False colour infrared (FCIR) images collection and their photointerpretation and processing are shown as they represent the most relevant means to acquire information of canopy from its reflectance properties. The amount and the quality of the obtainable data depend on: (i) field conditions; (ii) the type of the adopted instrument (camera, radiometer); (iii) the recording system position (ground platforms, aircraft, satellite); (iv) the format of the data (analogical, digitalised or digital); and (v) the photointerpretation technique. Results from literature are discussed stressing the limits of remote sensing methods. Remote sensing in VIS and NIR spectral bands is generally a powerful classification tool to detect and score tree stress. Nevertheless, it is not a diagnostic tool in that it does not provide information on the cause of stress. Moreover, the method should be adequately tested at single tree level for many important pathogens, in particular root rot, butt rot and stem rot fungi. In perspective, new high spatial resolution satellite images and their GIS software elaboration might be suitable to improve remote sensing analysis.