By John Lovett
University of Arkansas Agricultural Bureau
Arkansas Agricultural Experiment Station
Fayetteville, Ark – Satellite data used by archaeologists to find traces of ancient ruins hidden under a dense forest canopy can also be used to improve speed and accuracy to measure the amount of carbon retained and released in the forest.
Understanding this carbon cycle is key to climate change research, according to Hamdi Zurqani, an assistant professor of geospatial science at the Arkansas Forest Resource Center and the University of Forestry, Agriculture and Natural Resources at Monticello's Arkansas University. The Center is headquartered in UAM and conducts research and expansion activities through the Arkansas Agricultural Experiment Station and the Cooperative Expansion Services of the University of Arkansas Agricultural Research and Outreach Arms Division.
“The forest is often called the lungs of our planets and for good reason,” Zurukani said. “They store around 80% of the world's terrestrial carbon and play an important role in regulating the global climate.”
To measure the carbon cycle in a forest, calculations of the forests of biomass on the ground are required. Effective, traditional, ground-based methods for estimating biomass on forests are labor-intensive, time-consuming and limited spatial coverage capabilities, Zurqani said.
A recently published study in Ecological Informatics, Zurqani demonstrates how, even in remote areas where accessibility is often a problem, large forest biomass can be quickly and accurately mapped, and information from open access satellites can be integrated into Google Earth Engine for rapid and accurate mapping.
Zurqani's novel approach uses data from NASA's Global Ecosystem Dynamics Investigation Lidar. It is also known as GediLidar. This includes three lasers installed at the International Space Station. This system can accurately measure the height of a three-dimensional forest canopy, vertical canopy structure and surface height. LIDAR stands for “light detection and range” and uses light pulses to measure distance and create 3D models.
Zurqani also used image data from the European Space Agency's Earth Observation Collection of Copernicus Sentinel Satellites (Sentinel-1 and Sentinel-2). Combining 3D images from GEDI and optical images from Sentinel, Zurqani improved the accuracy of biomass estimation.
In this study, four machine learning algorithms were tested to analyze the data. Gradient Tree Boost, Random Forest, Classification and Regression Tree, or Cart, and Vector Machines. Gradient Tree Boost achieved the highest accuracy score and lowest error rate. Random Forest came in second, proving reliable but slightly inaccurate. Cart provided reasonable estimates, but tended to focus on smaller subsets. Zurqani said that the support vector machine algorithm is struggling, and the study emphasizes that not all AI models are equally suitable for estimating ground forest biomass.
Zurqani says the most accurate prediction combines the GEDI Lidar dataset, combining Sentinel-2 optical data, vegetation index, topographic features, and canopy height with the GEDI Lidar dataset, which serves as reference inputs for both training and testing machine learning models, indicating that multi-source data integration is important for reliable biomass mapping.
Why is it important?
Zurqani said accurate forest biomass mapping has real-world implications for better accounting for carbon and improving forest management on a global scale. A more accurate assessment will allow governments and organizations to more accurately track carbon sequestration and emissions from deforestation and inform policy decisions.
The road ahead
The study has been marked dramatically when measuring forest biomass on the ground, but Zurqani said the remaining challenges could include impacts on satellite data. In some regions, there is still no high-resolution rider coverage. He added that future research could explore deeper AI models such as neural networks to further improve predictions.
“One thing is clear,” Zukani said. “As climate change intensifies, these technologies are essential to protecting our forests and planets.”
For more information about the Agricultural Research Department, please visit the Arkansas Agricultural Experiment Station website
/Public release. This material of the Organization of Origin/Author is a point-in-time nature and may be edited for clarity, style and length. Mirage.news does not take any institutional position or aspect, and all views, positions and conclusions expressed here are the views of the authors alone.
