Illegal logging is a serious problem in many of the world's forests, damaging ecosystems, undermining sustainable management practices, and fuelling other forms of crime. It has been estimated that between 8% and 29% of all internationally traded timber has been harvested illegally, which would make the illegal trade in wood the world's third largest form of trans-border crime. This problem is particularly acute in tropical regions, for example the proportion of internationally traded timber from Central Africa thought to have been taken illegally is thought to be between 50% and 90%. Many timber importing countries have legislation which require importers to show that timber has been sourced legally, with data on both the origin and species of wood required, but this is often difficult to prove, and current international trade systems are thought to be rife with abuse. The new European Union Regulation on Deforestation Free Products, which is due to come into effect on 30 December 2025, mandates that timber importers demonstrate that timber has been sourced legally, and in a way that does not contribute to deforestation, and also requires that exact geolocations are provided for each piece of timber. However, independently verifying both the species of tree from which wood originates and the location from which it was obtained is likely to prove challenging.
A variety of methods have been used to try to verify the origins of timber, including the anatomy of the wood, genetic identification, near-infrared spectroscopy, mass spectroscopy, and stable isotope analysis. These methods have proven effective in identifying wood to the species level, but much less effective at determining its point of origin, leaving tracing efforts reliant on external documents, tags, or scans, all methods which are known to be vulnerable to fraudulent practices. Methods such as genetic analyses, stable isotope ratios and multi-element analysis, have failed to achieve the level of accuracy needed to track tropical timbers effectively, at least in part because of a lack of variation between samples.
In a paper published in the journal Communications: Earth & Environment on 15 October 2025, a team of scientists led by Laura Boeschoten of the Forest Ecology and Forest Management Group at Wageningen University & Research, and the Department of Ecology, Evolution and Environmental Biology at Columbia University, and Barbara Rocha Venancio Meyer-Sand also of the Forest Ecology and Forest Management Group, and of the Forest and Nature Conservation Policy Group at Wageningen University & Research, describe a multi-variant method which combines genetic analysis, stable isotope ratio analysis, and multi element analysis to trace the origin of Azobé, or Red Ironwood, Lophira alata, timber from locations in Cameroon, Gabon, and the Republic of the Congo.
Boeschoten and Meyer-Sand et al. reasoned that these different methods can present complementary information when trying to determine the origin of wood, and note that similar methods have been used to trace a range of agricultural commodities, wood from archaeological sites, and timber from temperate forests. Central Africa presents an ideal test area for the use of these methodologies in a tropical context. Levels of illegal logging are known to be high, there is an absence of the type of geographical barriers which lead to genetic isolation, and limited variations in climate, topography, bedrock, or soil type, all of which tend to hamper single-method tracing, but which leave open the possibility of multi-technique methodologies providing sufficient data to allow for accurate tracing of wood samples. To this end, Boeschoten and Meyer-Sand et al. combined genetic, isotopic, and elemental tracing methods, to evaluate if, in concert these could provide a reliable method to trace timber from Central Africa.
For trees, a genetic landscape can often be determined through the distribution of genetic traits within a population, which in turn is determined by reproductive, demographic and historical biogeographic factors, such as seed and pollen dispersal, topographic barriers like rivers and mountain ranges, and evolutionary history, including glaciation cycles during the Pleistocene that led to forest refugia. Genetic methods have been used to determine the origin of timber on a regional scale, for example to differentiate between West African and West African timber, and has sometimes proven useful at identifying the country of origin for wood, with a few cases of the method being used to differentiate timbers at shorter ranges.
Stable isotope ratios represent the proportion of isotopes such as oxygen¹⁸, hydrogen² (deuterium), carbon¹³, nitrogen¹⁵, and sulphur³⁴ within the total samples of their respective elements from a sample. While these isotopes have the same chemical properties as the more common isotopes of the same elements, they have different masses, and therefore are subject to sorting by some biological, geological, and environmental processes. This method has been used extensively in timber tracing from other regions, and has proven effective as a way of differentiating timbers from locations very distant from one-another, with a higher spacial definition achieved in areas where local isotopic variations tend to be larger, such as in mountain ranges. This method is often combined with element analysis. In Central Africa, some studies have found this to be an effective way to tell timber from different locations apart, but in other areas little difference has been seen between remote locations within the same country.
Multi-element analysis works by measuring the proportions of elements such as magnesium, calcium, and lanthanum in wood samples. These elements are not vital for the survival of trees, but are taken up with ground water during transpiration, then deposited into the tissues of the plant. The proportion of each of these elements present reflects the chemistry of the soil and underlying bedrock. This method can be quite efficient at differentiating the origin of timber over small distances, but is less accurate over larger areas. It has been combined with stable isotope analysis to determine the origin of some timbers from Eastern Europe.
Azobé timber is widely traded in the Congo Basin due to its dense, durable, and decay-resistant wood, which can be used for structures such as roads and jetties. The species is listed as Vulnerable on the International Union for the Conservation of Nature's Red List of Threatened Species, due to heavy harvesting of the trees, destruction of the forests where they grow, and a slow growth and regeneration rate. Boeschoten and Meyer-Sand et al. sampled 234 Azobé trees at 23 locations in Cameroon, Gabon, and the Republic of Congo, with the two closest sites being 15 km apart, and the two most remote separated by a distance of over 1000 km. Samples were tested for chloroplast genome-wide single-nucleotide polymorphisms (individual gene variations from chloroplasts, which do not undergo genetic recombination each generation, but are inherited from the mother Plant in the same way as mitochondria), the proportions of the isotopes oxygen¹⁸, hydrogen², and sulphur³⁴, and the presence of 41 elements, including trace elements known to be used by plants as well as 16 rare earth elements.
Overview of the three methods for timber tracing in this study and their drivers of geographic variation. Possible drivers of geographic variation shown as examples are: (1) For single-nucleotide polymorphisms (SNPs): former forest refugia during cold Pleistocene glaciation cycles with refugia in dark and forest cover in light green. (2) For stable isotopes: the oxygen isotope signal in rainwater, with higher rainwater the proportion of oxygen¹⁸ in lighter blue. (3) For multi-element analysis: soil clay content ranging from high clay % in green, mid-levels of clay % in yellow and low clay % in pink. Boeschoten & Meyer-Sand et al. (2025).
The primary focus of the study was to address origin identification ('Where did this timber come from?'), with origin verification ('Does this timber come from the stated place of origin?') being a secondary question. Origin identification is considered to be a crucial step in the creation of forensic methods, as it gives insight into variation across different points of origin, and is therefore the most widely used metric in studies; making this essential to compare the value of one particular study to those carried out by other research groups. However, Boeschoten and Meyer-Sand et al. note that origin verification is the more common goal of forensic studies, so both aims were included in the study.
The genetic analysis identified three main genetic clusters, as well as recovering previously identified genetic splits, including one which divides the Lower Guinean Region, and another that follows the Cameroonian Volcanic Line. The largest of these three clusters incorporated ten sites in West Cameroon, to NorthWest Congo and Central-East Gabon, with two sites in West Cameroon forming a distinct sub-cluster. The second cluster was located in West Gabon, and was the most genetically distinct, potentially representing an unknown cryptic species of Lophira. The third group was located in the Northern Congo, and again represents a distinctive, and previously unknown genetic population, although in this case probably not distinctive enough to merit being identified as a separate species.
Despite this apparent distinctiveness, genetic analysis of individual samples was only able to place them at the correct site of origin 46.2% of the time, with 62.2% of trees identified to a location within 100 km of where they grew and 85.6% to within 300 km. All trees were placed within 500 km of their growing sites.
Stable isotope ratios were found to vary a great deal, with the variation between trees at the same site often as great as that between trees at sites hundreds of kilometres away. Some patterns could be observed, particularly with regard to hydrogen², but even when all isotopes examined were included, it was possible to identify the site of origin for a sample only 40.7% of the time, and to place a tree within 100 km of its location only 49.8% of the time. Trees were frequently placed more than 1000 km from their actual location.
Levels of trace elements varied from between 0.001 g per kg for ytterbium and 4 g per kg for potassium. There were clear differences between sites, and trees could be placed at the correct site 73.4% of the time, and within 100 km 81.3% of the time. The elements tungsten, barium, molybdenum, potassium, and chromium were most useful in determining sample origins. Trees could be placed accurately at two sites in Cameroon 100% of the time, although some trees from other sites were still misplaced by more than 1000 km.
Confusion charts for the identification of wood samples to their respective origin. Identification was based on (A) single-nucleotide polymorphisms (SNP), (B) three stable isotope ratios (ISO), and (C) multi-element analysis of 44 trace elements (EL) and D all three combined. Mean identification accuracy across all sites is indicated in the bottom left. Each site has a unique colour, shown in the inner circle and in the legend. Colours in the outer circle of each symbol indicate to which (other) site(s) the trees of that site were assigned. Primary tropical forest extent from Global Forest Watch is indicated in grey. Boeschoten & Meyer-Sand et al. (2025). Combining any two of these methods proved to be more accurate than any method on its own, confirming that using a combination of methods was a valid approach. Combinations which included the multi-element analysis were particularly efficient, with a combination of multi-element analysis and single-nucleotide polymorphic genetic analysis placing trees at the correct site 79.8% of the time, while a combination of multi-element analysis and stable isotope analysis achieved this 77.3% of the time. When all three methods were combined, trees were placed at the right site 86.9% of the time, within 50 km of that site 91.0% of the time, and within 100 km of that site 94.5% of the time. When all three methods were used in combination, no tree was placed more than 500 km from its site of origin. Individual sites varied in the method which showed the most accuracy, with some sites being identifiable using only a single method. Notably, however, some sites which did not achieve high levels of accuracy with any single method achieved 100% accuracy with a combination of the three.
The combination of methods used by Boeschoten and Meyer-Sand et al. compares well to previously tried methods when attempting to identify the point of origin of samples. However, the question most commonly asked by forensic scientists working with timber-trade regulators is not 'Where did this timber come from?', but 'Does this timber come from where it is supposed to come from?'. In order to assess this, 41 samples were removed from the dataset, which was then recalibrated without them. Each sample was then tested against two claims, firstly that it came from the site that it genuinely came from, and secondly that it came from a randomly chosen alternative site. In this second scenario, all trees from the same site as the tree being tested were excluded, to simulate a tree from a genuinely unknown source, which would not be present in the database.
In the scenario where samples were tested against the correct site of origin, the test identified the sample correctly with an accuracy of 87.8%, whereas for the second test, testing a sample against an incorrect point of origin, the correct answer was produced 95.4% of the time. A purely random test would have identified samples from the correct site only 7.7% of the time, whereas it would have excluded samples from the wrong site 91.7% of the time. Thus, while the second test appeared more accurate, it was not a notable improvement on random chance.
Origin fraud is a major problem for the international timber industry, and no testing method has proven to be completely reliable in all settings. Regions such as Central Africa, where there are few geographical barriers and little environmental variation, are particularly problematic. By using three different methods together, Boeschoten and Meyer-Sand et al. were able to reach a level of tracing accuracy better than had previously been obtained for the region, with 94.5% of wood samples assigned within 100 km of their origin and 91% within 50 km. When the method was used to test timber against claimed points of origin, it correctly confirmed true claims 87.8% of the time and rejected false claims 95.4% of the time.
The 800 m-long Blauwe Loper Bridge in the Netherlands is Europe's longest cycle bridge. It was constructed using Azobé wood from Forest Stewardship Council certified sources. The durable nature of the wood means that it is predicted that it will not need to be replaced for 80 years. About 450 tonnes of Azobé wood was used in the construction of the bridge. Fair & Precious. Combining the three methods provided a greater level of accuracy than using any single method. This is because each variable is driven by a different set of conditions, resulting in a higher level of differentiation, even in a relatively homogeneous environment. The technique was particularly good at placing samples within 50 km of their point of origin compared to other methods, which corresponds roughly to the scale at which soil properties vary. The inclusion of genetic data considerably improved the resolution achieved, even though the lack of geographical barriers resulted in only a gradual genetic change between the sites.
The method compared well to previous studies which have only used a single technique to try to identify timber from the region, and Boeschoten and Meyer-Sand et al. anticipate that methods using multiple different forms of analysis will prove to be useful in areas where environmental gradients are low, but high resolution testing is required. Legal requirements around sourcing timber sustainably are likely to tighten in consumer countries in the future, which will require more reliable methods of testing claims about the origin of timber, as well as for commodities such as palm oil, soy, rubber, beef and cocoa. The methods chosen to verify the origin of such commodities will likely vary, but for items such as timber from the Congo Basin, where fraud is considered to be a significant problem, testing methods will need to be demonstrably reliable with a high degree of accuracy.
See also...
.png)
.jpg)
.jpg)
.jpg)
.jpg)
