Science & Technology (Commonwealth Union) – Published in the journal Earth System Science Data, the completed 3D tree census dataset is giving researchers a clearer picture of how much biomass — or plant material — forests hold. This is a crucial step toward accurately estimating the total amount of carbon stored across entire forest ecosystems.
By supplying precise ground-based measurements, the project is also strengthening satellite-based forest monitoring systems used to assess how forests are reacting to climate change. Satellite missions can rely on the study’s reference data to fine-tune their algorithms and enhance the accuracy of global biomass maps.
The UCL-led initiative, known as ForestScan, forms part of a broader international collaboration called GEO-TREES. Together, the teams are developing a worldwide network of carefully selected forest sites where scientists can conduct highly detailed measurements of trees and forest structure, with a particular focus on biomass and the carbon it stores. These locations are referred to as Forest Biomass Reference Measurement Sites (FBRMS).
For the study, researchers concentrated on three tropical rainforest regions: Paracou in northern French Guiana, Lopé National Park in central Gabon, and Kabili-Sepilok Forest Reserve in northeast Malaysia.
Using laser scanners mounted on aircraft and drones, the team surveyed representative plots across these areas covering nearly 550 hectares (around two square miles), collecting data on more than 200,000 individual trees.
In collaboration with local scientists and researchers in each country, the team gathered highly detailed information by manually measuring, tagging, and conducting ground-based scans of around 7,000 trees across smaller sections within the study plots.
Lead author Dr Cecilia Chavana-Bryant (UCL Geography) indicated that specific data on forest biomass is vital for understanding how forests capture and store carbon, and how they react to climate change.
She further indicated that by integrating cutting-edge 3D scanning technologies across three continents, we have assembled one of the most comprehensive datasets ever produced for tropical forests — creating an important reference point for satellite missions and Earth observation systems.
Dr Cecilia Chavana-Bryant also pointed out that the research establishes a strong foundation for more dependable global forest monitoring, supporting scientists, policymakers, and conservationists in making well-informed decisions to safeguard forests and address climate change at both local and international levels.
At each site, the scientists made use of laser scanners to produce 3D models of each tree within the chosen plots.
On top of that, as a component of a tree census, local researchers took into account all the trees, the circumference for every tree trunk was gaged manually with the utilization of a tape measure, and the species of everyone taken note of.
“This was a tremendous undertaking, and it would not have been possible without the support of our collaborators in each region. It has resulted in the most detailed dataset ever collected for these forest areas, helping us better understand and characterise forests globally. By using a diverse set of tools in ForestScan, we have achieved exceptional accuracy in our data and gained valuable insights into the real-world challenges of collecting field measurements at this scale” explained Dr Chavana-Bryant.
Precisely calculating forest biomass is becoming ever more vital, given the crucial role forests play in absorbing carbon dioxide, a major greenhouse gas, from the atmosphere. Trees and other vegetation take in carbon as they grow, storing it within their trunks, branches and tissues. In fact, roughly half of a tree’s living mass consists of carbon captured from the air.
Determining the exact volume of forest biomass — and therefore how much carbon is locked away — is especially significant for the expanding carbon offset market. In this system, organisations invest in planting new forests or safeguarding endangered woodland areas to compensate for their own carbon emissions.
Moreover, reliable biomass data enables scientists to assess how much carbon is released during events such as wildfires or illegal logging. A significant share of the doubts surrounding carbon offset schemes arises from limited high-quality ground data, which is needed to train the deep learning and AI models increasingly relied upon to estimate forest carbon stocks.
