What the Study Found
- Hourly laser scans of 17 dead branches in a Finnish forest showed every one moving daily; the 16 conifers drooped by day and lifted at night.
- Branch tips moved 4โ43 cm over 3.5 days (23 cm on average in conifers), 6โ112 times the scanner’s measurement uncertainty.
- Branch movement trailed modelled wood moisture by 3.5 hours on average, with individual lags of 1 to 5.5 hours.
- One birch branch moved the opposite way, and all branches were picked by eye with no ground-truth health data.
A LASER scanner mounted 30 meters above the ground in a Finnish forest takes the same hourly portrait of the trees below it, and has done since 2020. Dead conifer branches tracked modelled wood moisture, drooping by day as the air dried and lifting at night as it damped, a pattern that fits wood swelling and shrinking unevenly from top to bottom. Dead wood has a reputation for sitting still. Seventeen dead branches, picked out by eye from the scanner’s archive, did anything but.
Vanessa Guthier, a research assistant at Heidelberg University in Germany, and colleagues there and at the Finnish Geospatial Research Institute went back through that archive for the branches of 14 living trees and 3 dead ones, and reported what they saw in Agricultural and Forest Meteorology. They worked retrospectively on 65 hourly scans from a June 2020 stretch of light wind and little rain, which gave 933 observations of branch position in all. The authors say theirs is the first quantification of daily movement in dead tree branches, building on a creosote-bush study in which dead shrub branches tracked humidity.
Wood is hygroscopic, meaning it takes up water from damp air and gives it back to dry air, swelling and shrinking as it goes. Conifer branches carry a band of compression wood along their underside, and that band changes length more than the wood above it. When the two sides change length by different amounts, the branch bends, the way the bimetal strip in an old thermostat curls when its two metals expand unequally. In broadleaf trees such as birch, a similar band called tension wood grows on top instead, which should flip the direction of the bend.
A forest is not paper: branches hide behind one another, the wind picks up, and any movement has to be teased out from measurement noise.
A Droop That Follows the Damp
Every one of the 17 branches swung on a daily cycle, and the 16 conifer branches (Scots pine and Norway spruce) all drooped through the day and lifted overnight. Across the 17, the tips moved through a range of 4 to 43 centimeters over the 3.5 days, and for the conifers the average range was 23 centimeters, on branches that averaged 0.82 meters long across the whole sample. Those ranges were between 6 and 112 times larger than the measurement uncertainty, so they are not scanner jitter.
Guthier explains that the upper and lower parts of a branch expand at different rates depending on the species, so that โeven dead branches can move significantly up or down as humidity fluctuates during the day,โ as she puts it. In pine and spruce the dip came as the day dried out; the single birch did the opposite, rising by day and sinking by night.
The tracking step used an existing program that matches each hourly 3D scan of a branch to the first scan, so a bent branch can be followed even when the scanner sees only part of it. The moisture side was modelled, not measured. Air temperature and humidity from a nearby research station went into a standard wood-moisture equation, which returned a daily cycle from roughly 7 percent to roughly 25 percent. Branch movement trailed the modelled moisture by 3.5 hours on average, with individual branches lagging between 1 and 5.5 hours, a delay that fits water diffusing slowly through wood. Within each drying or wetting spell the link looked close to straight: in pine, each percentage point of moisture tilted the branch up 0.49 degrees on average.
What a Laser Can’t See
The branches were picked by eye, because the existing dataset came with no ground-truth health data: a branch counted as dead when the scans showed no foliage on it, which is shakier for needled conifers than for leafy trees. Three of the 17 sat on trees that were dead outright, and those behaved like the rest, which is some comfort.
The Heidelberg press release says the movement signatures of living and dead branches clearly differ, but the team scanned no living branches; that contrast leans on earlier work, in which living creosote branches tracked the air’s thirst for water as well as humidity while dead ones tracked humidity. Then there is the sample: a single birch, a single 3.5-day stretch of June weather, one forest in Finland.
Why a Dead Limb Is Worth Watching
Dead branches are a health signal. They can point to drought, disease or crowding, so knowing where they are is useful. Picking them out of a living canopy takes effort. A scanner that never blinks could, in principle, do the looking.
Bernhard Hรถfle, who leads the Heidelberg group, says the approach could let forest and city managers track tree health automatically and catch decline early. The paper itself is more modest, calling the work a proof of concept that may support such methods one day.
Reference
Guthier, V., Weiser, H., Campos, M., Puttonen, E., & Hรถfle, B. (2026). Quantification of hygroscopic dead branch movement on boreal forest trees using permanent terrestrial laser scanning. Agricultural and Forest Meteorology, 390, 111444. https://doi.org/10.1016/j.agrformet.2026.111444
- Study type: Observational field study, a retrospective analysis of an existing hourly laser-scan dataset; peer-reviewed journal article (Agricultural and Forest Meteorology, open access).
- Sample size: 17 dead branches, one per tree (14 living trees, 3 dead trees); 65 hourly scans; 933 observations.
- Exposure: Wood equilibrium moisture content, modelled from air temperature and relative humidity at a nearby research station, not measured in the branches.
- Comparison group: None; branches were compared by species (7 Scots pine, 6 Norway spruce, 1 silver birch, 3 on dead spruce trees).
- Follow-up: 3.5 days of hourly scans in June 2020, selected for light wind and little rain, at Hyytiรคlรค, Finland.
- Funding / conflicts of interest: German Research Foundation, Research Council of Finland and the Tulanet postdoctoral programme; the authors declare no competing interests.
- Data availability: Code and higher-resolution point clouds on the heiDATA repository (doi 10.11588/DATA/HLA7AF); selected scans in the LiPheStream dataset.
- Preregistration: Not reported.
- Main limitation: Dead branches were picked by eye from the point clouds with no ground-truth health data; one birch branch and one 3.5-day window limit how far the pattern generalizes.
FAQ
Do dead branches need a living tree to move?
Dead branches did not need a living tree to move in these scans. Three of the 17 branches sat on trees that were dead outright, and they behaved like branches on living pine and spruce. That fits the idea that the wood’s response to moisture is what moves them, though three branches is a small number.
Why do dead branches lag behind the humidity?
Dead branches lag behind the humidity because water moves slowly through wood. The branches trailed modelled wood moisture by 3.5 hours on average, with individual lags between 1 and 5.5 hours. The delay fits water diffusing slowly through wood, though the team could not say how strongly each factor contributed.
Why would a birch branch move the opposite way to pine and spruce?
A birch branch may move the opposite way to pine and spruce because tension wood grows on top of a broadleaf branch rather than underneath. The single birch branch in the scans rose by day and sank by night, the reverse of the conifers and what the researchers expected. One branch cannot speak for a species, so that result is a hint rather than a rule.
Could laser scanners help spot unhealthy trees?
Laser scanners could help spot unhealthy trees only in principle so far. The Heidelberg team suggests that the movement of dead branches, which earlier work found differs from living ones, might one day help flag branches that have died, and the paper calls its work a proof of concept. These scans included no living branches and did not test tree health directly.
Cite This Page
