Laser scanning for tree inventories: what studies report
A Ghent study of 95 street and park trees found tripod and walking laser scans measured trunk diameter with a bias under 2 cm against the benchmark. Shrubs, ivy and leaf cover limited the other measures.
Checked against the sources at the bottom of this page on October 16, 2026. Rules, fees and pay change: the source has the last word.
Laser scanning, also called LiDAR, builds a three-dimensional point cloud of a tree from millions of laser pulses. Three publications report on scanning real trees and where it fell short. One compares scanning platforms on city trees in Belgium, one is an Alabama Extension note on urban tree carbon, and one is a Welsh report on woodland timber.
Three platforms compared in Ghent
D'hont, Calders, Bartholomeus, Lau, Terryn, Verhelst and Verbeeck published their study in Urban Forestry and Urban Greening, volume 99, in September 2024. The university record lists the Laboratory of Geo-information Science and Remote Sensing at Wageningen University and Research. The team scanned trees in Ghent, Belgium, from three platforms: terrestrial laser scanning (TLS), where the scanner stands on the ground, mobile laser scanning (MLS), where it is carried through the site, and airborne laser scanning (ALS). TLS and MLS ran with leaves on and with leaves off, and ALS ran leaf-on.
The authors measured trunk diameter at breast height, tree height, crown projection area and crown volume on 95 individual trees. They used the TLS scans as the benchmark. Their findings:
- Trunk diameter from TLS and MLS showed a bias under 2 cm, with a concordance correlation coefficient of about 1.
- With leaves on, shrubs and ivy blocked part of the view.
- Leaf-on MLS also lost the top of large crowns. Crown volume showed a bias of -116 m2, as the record prints it, and a coefficient of 0.85. Tree height showed a bias of -0.38 m and a coefficient of 0.99.
- Crown projected area was affected least, with a bias of 0.49 m2 and a coefficient of 0.99.
The gap to the benchmark grew with tree size and with surrounding clutter such as buildings. For MLS the authors point to weak satellite positioning, traffic and walking routes that were not ideal. They write that the results should help city councils and tree managers pick a platform by purpose, site complexity and budget.
Urban carbon from standing trees
Georgios Arseniou of Auburn University wrote the Alabama Cooperative Extension System publication FOR-2159, dated April 2024. It describes terrestrial laser scanning as a way to estimate the carbon in urban trees without felling them. A tree's above-ground biomass is its dry mass above the ground, and the publication puts about 45 to 50 percent of that mass at carbon.
The method it describes has four parts. The scanner on a tripod captures millions of laser pulses a second. Software fits cylinders to the point cloud and adds up their volumes to give the woody volume. That volume is multiplied by wood density, taken from public databases or from increment cores. The publication then puts carbon at 50 percent of the woody biomass. Its example clouds hold eight million points for a leafless red maple and twenty million for a leaf-on honey locust.
The publication says a trained professional is needed both to run the scanner and to analyse the data, and that the method suits institutions doing large urban forest inventories.
Woodland timber and carbon in Wales
Heddwyn Bye of Woodknowledge Wales and Bangor University wrote a report funded by the Welsh Government's Home-Grown Homes project. The work scanned plots at eight woodlands, from Douglas fir plantation to unmanaged farm woodland with alder and hazel. Eleven plots of 10 m by 10 m were scanned, and the sawlog estimates cover 55 trees.
The report compares scan measurements with measurements by hand. Trunk diameter agreed closely. Height, taper, sweep and lean agreed slightly less well. A sawlog selection routine then estimated the logs each tree could yield, using 2.4 m, 3.6 m and 4.8 m lengths in red and green grades.
The report lists limits of scanning in complex woodland:
- Weather. The executive summary says the scanners cannot be used in rain or in wind over 15 mph. The limitations section puts the wind limit at 10 mph, since tree movement adds noise to the cloud.
- Field work. Several scans are needed to cover a plot, joined with spheres or reflective targets placed in it.
- Processing. The data needs training and time, and the author says an automated processing platform must exist before real-world use.
- Cost. The report says scanning costs more than a conventional inventory.
Related pages cover LiDAR and remote sensing for tree work and tree inventories.
Sources
- D'hont and colleagues, Evaluating airborne, mobile and terrestrial laser scanning for urban tree inventories: A case study in Ghent, Belgium, Urban Forestry and Urban Greening 99 (2024), article 128428, Wageningen University and Research portal record, read October 16, 2026
- Alabama Cooperative Extension System, Quantification of Urban Forest Carbon Stocks Using Terrestrial Laser Scanning Technology, FOR-2159, April 2024 (PDF), read October 16, 2026
- Woodknowledge Wales, Estimating Timber Value and Carbon in Complex Woodlands Using Terrestrial Laser Scanning, Welsh Government funded Home-Grown Homes project (PDF), read October 16, 2026