Detection of Tree Mortality Induced by Bark Beetle During Drought in Recent Years

0570666 - ÚVGZ 2024 eng A - Abstract
Bárta, Vojtěch - Homolová, Lucie - Hanuš, Jan
Detection of Tree Mortality Induced by Bark Beetle During Drought in Recent Years.
[12th EARSeL Workshop on Imaging Spectroscopy in Potsdam. Potsdam, 22.06.2022-24.06.2022]
Method of presentation: Poster
URL events: https://is.earsel.org/workshop/12-IS-Potsdam2022/ 
R&D Projects: GA MŠMT(CZ) EF16_019/0000797
Institutional support: RVO:86652079
Keywords : bark beetle * remote sensing * hyperspectral data
OECD category: Forestry

The classification accuracy of all employed methods is increasing in time, from 45-55 % in April 8 up to 96-99% in September 15. We also found that data from visible and near infrared area (CASI) separated groups of infested and noninfested trees later than in combination with short wavelength data (SASI). Where the remote sensing data from CASIdetect the infestation 23 after field survey method. And data from both sensors was able to identify infested trees 4 day ahead of field survey. Our results showed that the combination of VNIR and SWIR remote sensing has a potential to detect bark beetle infestation since early phases, being compared or may be better that classic field survey. We are also aware, that the classification accuracy depends on segmentation of tree crown, in our experiment the tree crown was selected manually in each acquisition that is hugely time consuming. Therefore, for success implementation is needed to develop automatic segmentation of tree crown.
Permanent Link: https://hdl.handle.net/11104/0341993