Quantification of hygroscopic dead branch movement on boreal forest trees using permanent terrestrial laser scanning
Year: 2026
Type: article
Abstract: Dead branches are an indicator of tree health, offering insights into water relations and stress, structural instability and disease. We hypothesize that dead branches exhibit hygroscopic movements driven by a moisture-sensitive reaction and normal wood bilayer. Thus, these movements can be detected through close-range remote sensing time series to identify dead branches for ecological analysis. We analysed hourly permanent laser scanning data from a boreal forest in Hyytiälä, Finland, monitoring 17 trees (16 coniferous, 1 deciduous) over 3.5 days with < 1 mm/h precipitation and < 3 m/s wind (933 total observations). To detect branch movement, we developed a dedicated method using non-rigid registration, quantifying movement as changes in angle between the branch orientation and the vertical. Linear mixed-effects models tested the relationship between branch angle and modelled wood equilibrium moisture content (EMC), considering time lags up to 8 h. Our results show that all studied branches followed a clear, systematic movement pattern: coniferous branches moved downward during the day and upward at night, with mean movement ranges of 23 ± 11 cm, lagging on average 3.5 ± 1.4 h behind EMC. The deciduous branch showed an inverted pattern, consistent with our hypotheses. Branch movement correlated strongly with EMC, confirming a hygroscopic mechanism. We concluded that systematic, daily dead branch movement is primarily hygroscopic, and can be reliably captured with hypertemporal terrestrial laser scanning under field conditions. Quantifying distinctive movement patterns of individual dead branches on living trees may support developing novel methods for early detection of tree vitality decline.
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,590
Awards: High-performance computing allowing high-accuracy country-level individual tree carbon sink and biodiversity mapping, LS-HYDRO: From forest structure to hydrological function – merging dense Earth Observation data and process-models, Generalisierte Extraktion von Oberflächendynamik aus multimodalen 4DPunktwolken für topographisches Monitoring von Erdoberflächenprozessen und deren Interaktionen (Extract4D), High-performance computing allowing high-accuracy country-level individual tree carbon sink and biodiversity mapping
OpenAlex ID: W7208831447