Using a unique combination of old-fashioned fieldwork and sophisticated computer modelling, scientists in Sweden have found a way to trace a single bundle of timber back to the European forest where it came from.
The researchers said the new method, described in a recent article in the journal Nature Plants, could significantly limit the sale of Russian timber, which is banned in the European Union because of the war in Ukraine. However, birch, oak, pine and other types of timber from Russia are still finding European buyers amid growing demand.
Last month, the new approach was used to track down large shipments of illegal Russian timber to Belgium.
The new study examined the chemical composition of 900 wood samples collected from 11 Eastern European countries. The data was fed into a model powered by machine learning, which found patterns that could predict the geographic origin of the samples.
Overall, the model caught 60 percent of the samples that were deliberately labeled with the wrong country of origin. The model could also narrow down the origin of the wood to a radius of about 125 miles, a remarkable achievement on a continent that is about 40 percent forested.
The method is “very, very solid from a technical standpoint,” said Naren Ramakrishnan, a data scientist at Virginia Tech who was not involved in the research.
Led by Victor Deklerck, the study’s lead author, researchers with Preferred by Nature, a Copenhagen-based nonprofit, criss-crossed Europe to collect tree samples using a long, tube-like device that scoops out wood tissue. A tree is not harmed when a specimen is extracted from its trunk, Dr. Deklerck said, because the rest of the organism “snips off” the injured tissue.
The samples were analyzed for the minerals they pulled from the soil, as well as elements, such as nitrogen and carbon, that they absorbed through rainfall.
The result was a “chemical fingerprint” for each tree sample in the study, said Dr. Declerc, who is also the chief scientist at World Forest ID, a nonprofit organization in Washington that fights deforestation.
As precise as these fingerprints were, they represented a piece of knowledge about Europe’s forests. There are 87 billion trees in Sweden alone. Russia, the most forested country in the world, is home to 642 billion trees.
So Jakub Truszkowski, a machine learning expert at the University of Gothenburg, created a spatial model using the samples collected by Preferred by Nature. Powerful computer clusters allowed Dr Truszkowski to extrapolate chemical profiles for vast unsampled areas of forest across Eastern Europe.
The ability of the model to accurately determine the origin of a wood sample varies from country to country. For example, it was able to identify 82 percent of wood originating in Russia (and disguised as originating elsewhere), but only 47 percent of samples coming from neighboring Belarus, which faces the same restrictions as Russia due to its alliance with the Kremlin in the war in Ukraine.
Dr. Truszkowski said the success rate would improve as more wood samples were collected. He also said accurate identifications would be easier under a new European Union law that requires timber producers to be much more detailed in their disclosures about where they cut trees.
“It’s not just the technology that matters here, it’s also the legislation,” Dr. Truszkowski said.
The lumber tests were a kind of proof of concept for the researchers, showing that it was possible for the science to meet real-world needs and do it in a timely manner. The same approach could help track a wide variety of food or agricultural products, including shrimp and palm oil, which are also illegally harvested and trafficked around the world, the researchers said.
“This whole framework can be applied anywhere, in principle,” said Dr. Truszkowski.
More broadly, the study shows how advances in computing will help researchers in all disciplines make sense of vast bodies of data that might otherwise prove unexplored.
“The amount of data is only going to increase, and it’s up to us to make sense of it, whether in a scientific sense or a social sense,” Dr Declerc said.
