Machine learning-based assessment of leaf-cutting ant infestation in Eucalyptusforest plantations.
CERNE · 2026 · 32
DOI: 10.1590/01047760202632013570 ↗Kurz gefasst
Fachartikel in der Zeitschrift CERNE, 2026. Verfasst von Vitória Fernanda dos Santos und 6 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
Brazilian planted forests play a critical role in global wood and fiber production but face significant productivity challenges from pests and diseases. Leaf-cutting ants (Atta spp. and Acromyrmex spp.) are the main pest in the Brazilian planted forests and cause significant productivity losses every year. Identifying areas most susceptible to colony establishment and growth is crucial for implementing effective management strategies. This study aims to assess the influence of edaphoclimatic and landscape variables on the establishment and expansion of leaf-cutting ant colonies and identify the key factors driving their occurrence.
Results
Based on a decade of monitoring data from 33,000 Eucalyptus stands across five regions, Random Forest models reached accuracies of 83% for predicting initial nests and 78% for predicting large nests.
Conclusion
The machine learning models effectively detected both initial and large nests, revealing that edaphoclimatic and landscape conditions exert varying levels of influence across macro-regions.
Zitieren
APA
Santos, V. F. D., Alvares, C. A., Domingues, M. M., Soliman, E. P., Cegatta, Í. R., Zauza, E. Â. V., & Mafia, R. G. (2026). Machine learning-based assessment of leaf-cutting ant infestation in Eucalyptusforest plantations. CERNE, 32. https://doi.org/10.1590/01047760202632013570
Vancouver
Santos VFD, Alvares CA, Domingues MM, Soliman EP, Cegatta ÍR, Zauza EÂV, et al. Machine learning-based assessment of leaf-cutting ant infestation in Eucalyptusforest plantations. CERNE. 2026;32. doi:10.1590/01047760202632013570
BibTeX
@article{Santos2026,
author = {Santos, Vitória Fernanda dos and Alvares, Clayton Alcarde and Domingues, Maurício Magalhães and Soliman, Everton Pires and Cegatta, Ítalo Ramos and Zauza, Edival Ângelo Valverde and Mafia, Reginaldo Gonçalves},
title = {Machine learning-based assessment of leaf-cutting ant infestation in Eucalyptusforest plantations},
journal = {CERNE},
year = {2026},
volume = {32},
doi = {10.1590/01047760202632013570}
}
EUVIMED