DE VINGERAFDRUK VAN BOMEN, EN WAT WE ERMEE KUNNEN BEREIKEN

Jitse
Deketelaere

Beïnvloedbare bossen

Kun je aan een stuk hout zien uit welk bos het afkomstig is? Verrassend genoeg wel. Net zoals mensen een unieke vingerafdruk hebben, draagt ook elke boom een eigen bijna-unieke chemische samenstelling. Die samenstelling wordt beïnvloed door tal van factoren, zoals de boomsoort, de bodem waarop de boom groeit en het klimaat waaraan de boom wordt blootgesteld.

Deze chemische vingerafdruk maakt het mogelijk om de herkomst van hout te achterhalen. Dit kan met behulp van een AI-model. Zo wordt het eenvoudiger om de oorsprong van hout te controleren en illegale houtkap op te sporen en tegen te gaan.

Een boom op uw bord

Terwijl grote bosbranden de krantenkoppen halen deze zomer, verdwijnt er elk jaar miljoenen hectaren aan bos over heel de wereld om verschillende redenen. Al deze ontbossing heeft nefaste gevolgen voor onder andere de biodiversiteit, het klimaat en de gemeenschappen die van deze ecosystemen afhankelijk zijn.

Eén van de hoofdredenen? Het eten op uw bord. De uitbreiding van de landbouw is elk jaar opnieuw de grootste oorzaak van ontbossing. Een deel van die ontbossing gebeurt bovendien illegaal. Het gekapte hout vindt vervolgens zijn weg naar de internationale markt, vaak met behulp van vervalste documenten. En daar is veel geld mee gemoeid.

Een houtmijn

Banken overvallen is niet meer van deze tijd. Tegenwoordig bestaat er een nieuwe criminele goudmijn. De wereldwijde illegale houtkap wordt geschat op zo’n 52 tot wel 157 miljoen Amerikaanse dollar, en is daarmee de meest winstgevende misdaad met betrekking tot natuurlijke hulpbronnen. Tot zo’n 50% van al het hout dat verhandelt wordt wereldwijd, zou frauduleuze documenten bevatten. Als op papier alles volgens de regels lijkt te verlopen, hoe toon je dan aan dat er in werkelijkheid fraude is gepleegd?

Die uitdaging vormde het vertrekpunt van mijn thesis. In samenwerking met de Plantentuin Meise en World Forest ID ging ik op zoek naar de beste en meest begrijpbare manier om de oorsprong van hout op een wetenschappelijk onderbouwde manier te voorspellen.

Hoe zeker is zeker genoeg?

Door een AI-model te trainen, leert het patronen herkennen die typisch zijn voor bepaalde regio's. Wanneer vervolgens een nieuw en onbekend houtstaal wordt geanalyseerd, kan het model op basis van die chemische samenstelling een voorspelling maken van de oorsprong. Die voorspelling kan betrekking hebben op een exacte locatie, maar evengoed op een land of een vooraf afgebakende geografische regio. Absolute zekerheid bestaat daarbij echter nooit.

Net die onzekerheid vormt een belangrijke uitdaging. Stel bijvoorbeeld dat een container hout aankomt in de haven van Antwerpen. Om na te gaan of het hout daadwerkelijk afkomstig is uit de opgegeven regio, worden verschillende stalen genomen en chemisch geanalyseerd. Het AI-model gebruikt deze gegevens om de vermoedelijke herkomst te bepalen.

Maar wat als die voorspelde herkomst niet overeenkomt met de oorsprong die in de documenten vermeld staat? Hoe kan men dan aantonen dat de voorspelling van het model voldoende betrouwbaar is?

Een eenvoudige vergelijking maakt duidelijk dat de (on)zekerheid van resultaten van cruciaal belang kunnen zijn, zeker in juridische context. Stel dat je geflitst wordt voor een snelheidsovertreding en op de boete staat vermeld dat men voor 70% zeker is dat het jouw wagen was die geflitst werd. Zou jij bereid zijn om de boete te betalen? 

Daarom richtte het tweede deel van mijn thesis zich op de vraag hoe de betrouwbaarheid van dergelijke voorspellingen kan worden verhoogd. Met andere woorden: hoe kunnen we een voorspelling met een zekerheid van bijvoorbeeld 70% omzetten naar een betrouwbaarheid van 90%, 95% of een andere vooraf vastgelegde zekerheidsgraad?

Een nieuwe manier van voorspellen

Ik ging aan de slag met een algoritme genaamd Conformal prediction. Deze methode werd als een extra laag rond het AI-model geplaatst en veranderde de manier waarop de resultaten werden geïnterpreteerd.

In plaats van één mogelijk antwoord, genereert een model met Conformal Prediction een verzameling van mogelijke antwoorden. Het grote voordeel hiervan is dat de gebruiker zelf kan bepalen hoe zeker het resultaat moet zijn. Naarmate een hogere betrouwbaarheid wordt gevraagd, zal het model een ruimer voorspellingsgebied voorstellen. In plaats van één specifieke locatie kan het bijvoorbeeld meerdere regio's of landen aanduiden. Tegelijkertijd blijven de resultaten statisch onderbouwd, en kunnen dus als bewijs dienen in de rechtbank.

Wetenschap als wapen tegen de illegale ontbossing

Deze bruikbare tool is niet enkel handig voor juridische instanties. Ook voor bedrijven kan ze een waardevol hulpmiddel zijn. Volgens de Europese ontbossingswetgeving moeten handelaars namelijk zelf kunnen aantonen dat het hout waarmee ze werken geen onaanvaardbaar risico op illegale herkomst inhoudt. In de praktijk gebeurt deze risicobeoordeling vaak op basis van documentatie, waarvan de juistheid vaak moeilijk te controleren is. De wetenschappelijke aanpak uit deze thesis kan een objectiever en betrouwbaarder alternatief zijn. 

Rubber dat uit een boom gehaald wordt.

Mijn onderzoek focuste zich nu enkel op hout. Er bestaat echter de mogelijkheid om deze methodologie uit te breiden naar andere plantaardige producten, zoals cacao, koffie of rubber. Van deze producten wordt de chemische samenstelling namelijk ook beïnvloed door de geografische herkomst. Hierdoor ontstaat een brede waaier aan mogelijke toepassingen binnen de strijd tegen fraude en illegale handel.

Daarnaast zet men sterk in om de voorspellingsmodellen nog te verbeteren. Dit zou bijvoorbeeld kunnen door de chemische data te combineren met satellietbeelden. Men kijkt op deze beelden of de voorspelde oorsprong van het hout al dan niet realistisch is. De strijd tegen de illegale houtkap is dus volop aan de gang, en ook het wetenschappelijk onderzoek dat deze strijd ondersteunt, blijft zich voortdurend verder ontwikkelen.

Bossen beschermen met betrouwbare voorspellingen

Wat begon met een stukje hout en een vraag naar zijn herkomst, groeide uit tot een zoektocht naar betrouwbare en wetenschappelijk onderbouwde antwoorden. Want in de strijd tegen de illegale houtkap is niet alleen de vraag waar komt dit hout vandaan? belangrijk, maar ook hoe zeker kunnen we daarvan zijn?

Bibliografie

[1] Abraham, U. I. (2020). Deforestation, air quality degradation and increased cardiopulmonary diseases. SRMS Journal of Medical Science, 5(02).
[2] Aitchison, J. (1983). Principal component analysis of compositional data. Biometrika, 70(1), 57–65.
[3] Alnuaimi, A. F., & Albaldawi, T. H. (2024). An overview of machine learning classification techniques. In BIO Web of Conferences (Vol. 97, p. 00133). EDP Sciences.
[4] Angelopoulos, A., Bates, S., Malik, J., & Jordan, M. I. (2020). Uncertainty sets for image classifiers using conformal prediction. arXiv preprint arXiv:2009.14193.
[5] Angelopoulos, A. N., & Bates, S. (2023). Conformal prediction: A gentle introduction. Foundations and Trends in Machine Learning, 16(4), 494–591.
[6] Aparicio-Muriana, M. M., Hong, Y., Chilaka, C. A., Quinn, B., Montes-Niño, A. M., Birse, N., & Elliott, C. T. (2025). Mapping the global origins of soybean: a study using ICP-MS and chemometrics. npj Science of Food, 9(1), 263.
[7] Baccini, A., Goetz, S. J., Walker, W., Laporte, N., Sun, M., Sulla-Menashe, D., Hackler, J., Beck, P., Dubayah, R., Friedl, M., et al. (2012). Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nature Climate Change, 2(3), 182–185.
[8] Biau, G., & Scornet, E. (2016). A random forest guided tour. TEST, 25(2), 197–227.
[9] Boeschoten, L. E., Rocha Venancio Meyer-Sand, B., Boom, A., Bouka, G. U. D., Ciliane-Madikou, J. C., Engone Obiang, N. L., Guieshon-Engongoro, M., de Groot, A., Loumeto, J. J., Mbika, D.-M. M., et al. (2025). Combined genetic and chemical methods boost the precision of tracing illegal timber in Central Africa. Communications Earth & Environment, 6(1), 789.
[10] Boeschoten, L. E., Sass-Klaassen, U., Vlam, M., Comans, R. N., Koopmans, G. F., Meyer-Sand, B. R. V., Tassiamba, S. N., Tchamba, M. T., Zanguim, H. T., Zemtsa, P. T., et al. (2022). Clay and soil organic matter drive wood multi-elemental composition of a tropical tree species: Implications for timber tracing. Science of the Total Environment, 849, 157877.
[11] Boeschoten, L. E., Vlam, M., Sass-Klaassen, U., Rocha Venâncio Meyer-Sand, B., Adzkia, U., Bouka, G. D., Ciliane-Madikou, J. C., Engone Obiang, N. L., Guieshon-Engongoro, M., Loumeto, J. J., et al. (2023). A new method for the timber tracing toolbox: applying multi-element analysis to determine wood origin. Environmental Research Letters, 18(5), 054001.
[12] Boner, M., Sommer, T., Erven, C., & Förstel, H. (2007). Stable isotopes as a tool to trace back the origin of wood. In Proceedings of the International Workshop “Fingerprinting Methods for the Identification of Timber Origins” (pp. 3–5).
[13] Booksh, K. S., Celani, C. P., Ralbovsky, N. M., & Smith, J. P. (2025). Assessing classification models of pharmaceuticals with conformal prediction. Journal of Chemometrics, 39(3), e70017.
[14] Bowen, G. J., Wassenaar, L. I., & Hobson, K. A. (2005). Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia, 143(3), 337–348.
[15] Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
[16] Camin, F., Boner, M., Bontempo, L., Fauhl-Hassek, C., Kelly, S. D., Riedl, J., & Rossmann, A. (2017). Stable isotope techniques for verifying the declared geographical origin of food in legal cases. Trends in Food Science & Technology, 61, 176–187.
[17] Cheadle, C., Vawter, M. P., Freed, W. J., & Becker, K. G. (2003). Analysis of microarray data using z score transformation. The Journal of Molecular Diagnostics, 5(2), 73–81.
[18] Craine, J. M., Brookshire, E., Cramer, M. D., Hasselquist, N. J., Koba, K., Marin-Spiotta, E., & Wang, L. (2015). Ecological interpretations of nitrogen isotope ratios of terrestrial plants and soils. Plant and Soil, 396(1), 1–26.
[19] Cresswell, J. C., Sui, Y., Kumar, B., & Vouitsis, N. (2024). Conformal prediction sets improve human decision making.
[20] Cutler, A., Cutler, D. R., & Stevens, J. R. (2012). Random forests. In Ensemble Machine Learning (pp. 157–175). Springer.

[21] da Silva Araújo, M. G., Amorim, D. J., Almeida, R. F., Mazzi, E., Batista, A. C. G., Souza-Silva, I. M., Brunello, A. T., Nardoto, G. B., Bowen, G. J., & Martinelli, L. A. (2025). Evaluating an isotopic model for tracing the geographic origin of soybeans in Brazil. Food Chemistry, 147586.
[22] DeFries, R. S., Rudel, T., Uriarte, M., & Hansen, M. (2010). Deforestation driven by urban population growth and agricultural trade in the twenty-first century. Nature Geoscience, 3(3), 178–181.
[23] Degen, B., Yanbaev, Y., & Müller, N. A. (2025). Machine learning techniques for continuous genetic assignment of geographic origin of forest trees. PLoS ONE, 20(6), e0324994.
[24] DeNiro, M. J., & Epstein, S. (1979). Relationship between the oxygen isotope ratios of terrestrial plant cellulose, carbon dioxide, and water. Science, 204(4388), 51–53.
[25] Dhillon, G. S., Deligiannidis, G., & Rainforth, T. (2024). On the expected size of conformal prediction sets. In International Conference on Artificial Intelligence and Statistics (pp. 1549–1557). PMLR.
[26] Duncker, P. S., Raulund-Rasmussen, K., Gundersen, P., Katzensteiner, K., De Jong, J., Ravn, H. P., Smith, M., Eckmüllner, O., & Spiecker, H. (2012). How forest management affects ecosystem services, including timber production and economic return: synergies and trade-offs. Ecology and Society, 17(4).
[27] Earthsight. (2023). “Conflict timber” worth over $1bn sold in US since Ukraine invasion. Press release.
[28] Egozcue, J. J., Pawlowsky-Glahn, V., Mateu-Figueras, G., & Barcelo-Vidal, C. (2003). Isometric logratio transformations for compositional data analysis. Mathematical Geology, 35(3), 279–300.
[29] Ehleringer, J. R., Casale, J. F., Lott, M. J., & Ford, V. L. (2000). Tracing the geographical origin of cocaine. Nature, 408(6810), 311–312.
[30] EU. (2023). Commission implementing regulation. Official Journal of the European Union, L 207, 77–82. C/2023/5564.
[31] European Commission. (2026). Regulation on deforestation-free products. https://environment.ec.europa.eu/topics/forests/deforestation/regulatio…. Accessed: 2026-03-16.
[32] Finch, K., Espinoza, E., Jones, F. A., & Cronn, R. (2017). Source identification of Western Oregon Douglas-fir wood cores using mass spectrometry and random forest classification. Applications in Plant Sciences, 5(5), 1600158.
[33] Flores, B. M., Montoya, E., Sakschewski, B., Nascimento, N., Staal, A., Betts, R. A., Levis, C., Lapola, D. M., Esquível-Muelbert, A., Jakovac, C., et al. (2024). Critical transitions in the Amazon forest system. Nature, 626(7999), 555–564.
[34] Food and Agriculture Organization of the United Nations. (2024). The State of the World’s Forests 2024: Forest-Sector Innovations towards a More Sustainable Future. FAO, Rome.
[35] Getirana, A., Libonati, R., & Cataldi, M. (2021). Brazil is in water crisis—it needs a drought plan. Nature, 600(7888), 218–220.
[36] Gislason, P. O., Benediktsson, J. A., & Sveinsson, J. R. (2004). Random forest classification of multisource remote sensing and geographic data. In IGARSS 2004. IEEE International Geoscience and Remote Sensing Symposium (Vol. 2, pp. 1049–1052). IEEE.
[37] Gomes, M. D. G., Costa, F. J. V., Nascentes, C. C., Martinelli, L. A., & Nardoto, G. B. (2025). Multi-elemental analysis for the determination of the geographic origin of tropical timber from the Brazilian Legal Amazon. Forests, 16(8), 1284.
[38] Gomes, V. H., Vieira, I. C., Salomão, R. P., & ter Steege, H. (2019). Amazonian tree species threatened by deforestation and climate change. Nature Climate Change, 9(7), 547–553.
[39] Gori, Y., Deklerck, V., & Camin, F. (2025). Stable isotope ratio analysis to determine the geographical origin of timber: A review. TrAC Trends in Analytical Chemistry, 118448.
[40] Greenacre, M. (2021). Compositional data analysis. Annual Review of Statistics and Its Application, 8(1), 271–299.

[41] Guha, E., Natarajan, S., Möllenhoff, T., Khan, M. E., & Ndiaye, E. (2024). Conformal prediction via regression-as-classification. arXiv preprint arXiv:2404.08168.
[42] Hoaglin, D. C., Iglewicz, B., & Tukey, J. W. (1986). Performance of some resistant rules for outlier labeling. Journal of the American Statistical Association, 81(396), 991–999.
[43] Hoefs, J. (2009). Variations of stable isotope ratios in nature. In Stable Isotope Geochemistry (pp. 93–227). Springer, Berlin, Heidelberg.
[44] INTERPOL. (2016). Uncovering the risks of corruption in the forestry sector. Published 9 December 2016.
[45] Jahren, A. H., Sternberg, L. S. L., et al. (2002). Eocene meridional weather patterns reflected in the oxygen isotopes of Arctic fossil wood. GSA Today, 12(1), 4–9.
[46] James, G., Witten, D., Hastie, T., Tibshirani, R., et al. (2013). An Introduction to Statistical Learning: With Applications in R (Vol. 103). Springer.
[47] Kumar, A., Dhanka, S., Sharma, A., Sharma, A., Nain, M., Kumar, P., Gupta, A. R., Bansal, J., Saxena, N. K., & Pant, R. (2025). A comprehensive review of bias in AI, ML, and DL models: Methods, impacts, and future directions. Archives of Computational Methods in Engineering, 1–31.
[48] Kyere-Boateng, R., & Marek, M. V. (2021). Analysis of the social-ecological causes of deforestation and forest degradation in Ghana: Application of the DPSIR framework. Forests, 12(4).
[49] Köthke, M., Lippe, M., & Elsasser, P. (2023). Comparing the former EUTR and upcoming EUDR: Some implications for private sector and authorities. Forest Policy and Economics, 157, 103079.
[50] Leal Filho, W., Dinis, M. A. P., Canova, M. A., Cataldi, M., da Costa, G. A. S., Enrich-Prast, A., Symeonakis, E., & Brearley, F. Q. (2025). Managing ecosystem services in the Brazilian Amazon: the influence of deforestation and forest degradation in the world’s largest rain forest. Geoscience Letters, 12(1), 24.
[51] Ledoit, O., & Wolf, M. (2004). A well-conditioned estimator for large-dimensional covariance matrices. Journal of Multivariate Analysis, 88(2), 365–411.
[52] Lei, J., G’Sell, M., Rinaldo, A., Tibshirani, R. J., & Wasserman, L. (2018). Distribution-free predictive inference for regression. Journal of the American Statistical Association, 113(523), 1094–1111.
[53] Leung, R. (2021). Empirical observations on the effects of data transformation in machine learning classification of geological domains. arXiv preprint arXiv:2106.05855.
[54] Low, M. C., Schmitz, N., Boeschoten, L. E., Cabezas, J. A., Cramm, M., Haag, V., Koch, G., Meyer-Sand, B. R., Paredes-Villanueva, K., Price, E., et al. (2022). Tracing the world’s timber: the status of scientific verification technologies for species and origin identification. IAWA Journal, 44(1), 63–84.
[55] Lusk, C. H., Wright, I., & Reich, P. B. (2003). Photosynthetic differences contribute to competitive advantage of evergreen angiosperm trees over evergreen conifers in productive habitats. New Phytologist, 160(2), 329–336.
[56] Maes, J., Teller, A., Erhard, M., Conde, S., Vallecillo, R. S., Barredo, C., Paracchini, M.-L., Malak, D. A., Trombetti, M., Vigiak, O., et al. (2021). EU Ecosystem Assessment.
[57] Maor, R., Truszkowski, J., Ablett, F., Jennings, H., Walker, H., Dunn, J., Norman, M., Carrasco, R., Jaime-Arteaga, M., Miles-Bunch, I., et al. (2026). High-resolution soybean tracing for deforestation-free supply chains. Communications Earth & Environment, 7(1), 310.
[58] Marques, L. (2024). Crossing tipping points in the Amazon rainforest: the decisive decade. The Highlander Journal, 3(2), 4–33.
[59] Mavrellis, C. (2017). Transnational crime and the developing world. Published March 27, 2017.
[60] McKinney, C. R., McCrea, J. M., Epstein, S., Allen, H., & Urey, H. C. (1950). Improvements in mass spectrometers for the measurement of small differences in isotope abundance ratios. Review of Scientific Instruments, 21(8), 724–730.

[61] Mehrtens, H., Bucher, T., & Brinker, T. J. (2023). Pitfalls of conformal predictions for medical image classification. In International Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging (pp. 198–207). Springer.
[62] Meinshausen, N., & Ridgeway, G. (2006). Quantile regression forests. Journal of Machine Learning Research, 7(6).
[63] Meinzer, F. C., Brooks, J., Domec, J.-C., Gartner, B. L., Warren, J., Woodruff, D., Bible, K., & Shaw, D. (2006). Dynamics of water transport and storage in conifers studied with deuterium and heat tracing techniques. Plant, Cell & Environment, 29(1), 105–114.
[64] Mensah, P., Pimenta, A. S., de Melo, R. R., Amponsah, J., Tuo, G., Chakurah, I., Ampadu, S. D., Buckman, I., Nikoi, M., Minkah, E., et al. (2025). The global supply chain of wood products: A literature review. Forests, 16(7), 1036.
[65] Moreira, G. R., Fernandes, E. A. D. N., Bacchi, M. A., Gonzaga, C. L., & de Lima, R. C. (2024). Elemental profile of ipê species to support timber tracing in Amazon rainforest. Trees, Forests and People, 18, 100689.
[66] Mortier, T., Javanmardi, A., Sale, Y., Hüllermeier, E., & Waegeman, W. (2025). Conformal prediction in hierarchical classification with constrained representation complexity. arXiv preprint arXiv:2501.19038.
[67] Mortier, T., Truszkowski, J., Norman, M., Boner, M., Buliga, B., Chater, C., Jennings, H., Saunders, J., Sibley, R., Antonelli, A., et al. (2024). A framework for tracing timber following the Ukraine invasion. Nature Plants, 10(3), 390–401.
[68] Muccio, Z., & Jackson, G. P. (2009). Isotope ratio mass spectrometry. Analyst, 134(2), 213–222.
[69] Nadras, S., Mazlan, R., Hussain, H., & Md Shah, I. (2024). The European Union Deforestation-Free Regulation (EUDR): assessing impacts and strategies for Malaysian and the global oil palm industry. Journal of Sustainability Science and Management, 19(6), 54–74.
[70] Nallapati, R. (2004). Discriminative models for information retrieval. In Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 64–71).
[71] Nellemann, C. (2012). Green carbon, black trade: illegal logging, tax fraud and laundering in the world’s tropical forests.
[72] Osburn, M. R., Dawson, K. S., Fogel, M. L., & Sessions, A. L. (2016). Fractionation of hydrogen isotopes by sulfate- and nitrate-reducing bacteria. Frontiers in Microbiology, 7, 1166.
[73] Pendrill, F., Persson, U. M., Godar, J., & Kastner, T. (2019a). Deforestation displaced: trade in forest-risk commodities and the prospects for a global forest transition. Environmental Research Letters, 14(5), 055003.
[74] Pendrill, F., Persson, U. M., Godar, J., Kastner, T., Moran, D., Schmidt, S., & Wood, R. (2019b). Agricultural and forestry trade drives large share of tropical deforestation emissions. Global Environmental Change, 56, 1–10.
[75] Pereira, N. S., Sial, A. N., Pinheiro, P. B., Freitas, F. L., & Silva, A. M. (2021). Carbon and oxygen stable isotopes of freshwater fish otoliths from the São Francisco River, northeastern Brazil. Anais da Academia Brasileira de Ciências, 93(1), e20191050.
[76] Pourkamali-Anaraki, F. (2025). Aerial image classification in scarce and unconstrained environments via conformal prediction. arXiv preprint arXiv:2504.17655.
[77] Reddington, C., Butt, E., Ridley, D., Artaxo, P., Morgan, W., Coe, H., & Spracklen, D. (2015). Air quality and human health improvements from reductions in deforestation-related fire in Brazil. Nature Geoscience, 8(10), 768–771.
[78] Roden, J., Saurer, M., & Siegwolf, R. T. (2022). Probing tree physiology using the dual-isotope approach. In Stable Isotopes in Tree Rings: Inferring Physiological, Climatic and Environmental Responses (pp. 463–479). Springer.
[79] Romano, Y., Patterson, E., & Candès, E. (2019). Conformalized quantile regression. Advances in Neural Information Processing Systems, 32.
[80] Rousseeuw, P. J., & Croux, C. (1993). Alternatives to the median absolute deviation. Journal of the American Statistical Association, 88(424), 1273–1283.
[81] Sachse, D., Billault, I., Bowen, G. J., Chikaraishi, Y., Dawson, T. E., Feakins, S. J., Freeman, K. H., Magill, C. R., McInerney, F. A., Van Der Meer, M. T., et al. (2012). Molecular paleohydrology: interpreting the hydrogen-isotopic composition of lipid biomarkers from photosynthesizing organisms. Annual Review of Earth and Planetary Sciences, 40(1), 221–249.
[82] Sarkar, S., Yousuf, R. B., Wang, L., Mayer, B., Mortier, T., Deklerck, V., Truszkowski, J., Simeone, J. C., Norman, M., Saunders, J., et al. (2025). Chasing the timber trail: Machine learning to reveal harvest location misrepresentation. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 2 (pp. 4796–4805).
[83] Sassen, M., Sheil, D., Giller, K. E., & ter Braak, C. J. (2013). Complex contexts and dynamic drivers: Understanding four decades of forest loss and recovery in an east African protected area. Biological Conservation, 159, 257–268.
[84] Schulz, E., Speekenbrink, M., & Krause, A. (2018). A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions. Journal of Mathematical Psychology, 85, 1–16.
[85] Sharp, Z. (2017). Principles of Stable Isotope Geochemistry.
[86] Siegwolf, R. T., Savard, M. M., Grams, T. E., & Voelker, S. (2022). Impact of increasing CO2, and air pollutants (NOx, SO2, O3) on the stable isotope ratios in tree rings. In Stable Isotopes in Tree Rings: Inferring Physiological, Climatic and Environmental Responses (pp. 675–710). Springer.
[87] Song, X.-P., Hansen, M. C., Potapov, P., Adusei, B., Pickering, J., Adami, M., Lima, A., Zalles, V., Stehman, S. V., Di Bella, C. M., et al. (2021). Massive soybean expansion in South America since 2000 and implications for conservation. Nature Sustainability, 4(9), 784–792.
[88] Sturchio, N. C., Beloso Jr, A., Heraty, L. J., Wheatcraft, S., & Schumer, R. (2014). Isotopic tracing of perchlorate sources in groundwater from Pomona, California. Applied Geochemistry, 43, 80–87.
[89] Tcherkez, G., & Tea, I. (2013). 32S/34S isotope fractionation in plant sulphur metabolism.
[90] Tilman, D., Clark, M., Williams, D. R., Kimmel, K., Polasky, S., & Packer, C. (2017). Future threats to biodiversity and pathways to their prevention. Nature, 546(7656), 73–81.
[91] Tipple, B. J., Berke, M. A., Doman, C. E., Khachaturyan, S., & Ehleringer, J. R. (2013). Leaf-wax n-alkanes record the plant–water environment at leaf flush. Proceedings of the National Academy of Sciences, 110(7), 2659–2664.
[92] Truszkowski, J., Maor, R., Yousuf, R. B., Biswas, S., Chater, C., Gasson, P., McQueen, S., Norman, M., Saunders, J., Simeone, J., et al. (2025). A probabilistic approach to estimating timber harvest location. Ecological Applications, 35(1), e3077.
[93] Van Nguyen, T. T., Heuchenne, C., & Tran, K. P. (2022). Machine learning for compositional data analysis in support of the decision making process. In Machine Learning and Probabilistic Graphical Models for Decision Support Systems (pp. 184–215). CRC Press.
[94] Vovk, V., Gammerman, A., & Shafer, G. (2022). Algorithmic Learning in a Random World (2nd ed.). Springer, Cham.
[95] Wang, H., Liu, H., Lu, B., Ma, M., Chen, J., & Nie, J. (2023). Origin authentication of European and American ash (Fraxinus spp.) based on stable isotope ratio and elemental characteristics combined with chemometrics methods. Chemosensors, 11(10), 536.
[96] Wang, T., & Dobriban, E. (2026). Optimal decision-making based on prediction sets. arXiv preprint arXiv:2602.00989.
[97] Watkinson, C. J., Gasson, P., Rees, G. O., & Boner, M. (2020). The development and use of isoscapes to determine the geographical origin of Quercus spp. in the United States. Forests, 11(8), 862.
[98] Watkinson, C. J., Rees, G. O., Hofem, S., Michely, L., Gasson, P., & Boner, M. (2022). A case study to establish a basis for evaluating geographic origin claims of timber from the Solomon Islands using stable isotope ratio analysis. Frontiers in Forests and Global Change, 4, 645222.
[99] Weisse, M., Goldman, E., & Carter, S. (2024). Tropical forest loss drops steeply in Brazil and Colombia, but high rates persist overall. Global Forest Watch.
[100] Wiedenhoeft, A. C., Simeone, J., Smith, A., Parker-Forney, M., Soares, R., & Fishman, A. (2019). Fraud and misrepresentation in retail forest products exceeds US forensic wood science capacity. PLoS ONE, 14(7), e0219917.
[101] World Forest ID. (2025). Unlocking transparency, exposing material risk: Key findings from timber market study. Insight Report, April 2025.
[102] Zhu, T. (2020). Analysis on the applicability of the random forest. In Journal of Physics: Conference Series (Vol. 1607, p. 012123). IOP Publishing.

Download scriptie (749.13 KB)
Universiteit of Hogeschool
Universiteit Gent
Thesis jaar
2026
Promotor(en) en begeleiders
Willem Waegeman, Victor Deklerck, Sebastian Gabriel Jimenez