EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Structured machine learning modeling to support conservation of deep‐sea benthic biodiversity

Gustavo Fonseca iD, Danilo C Vieira iD, Juliane C Carneiro, Renato S Carreira, Milena Ceccopieri, Thais N Corbisier, Adriana Galindo Dalto, Alberto G Figueiredo, Fabiane Gallucci, Paula Gheller, Simone Brito de Jesus, Helena Passeri Lavrado und 12 weitere

Conservation Biology · 2026-08 · 40 (4)

DOI: 10.1111/cobi.70255 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Conservation Biology, online erschienen am 10.03.2026. Verfasst von Gustavo Fonseca und 23 weiteren Autor:innen. Beteiligte Einrichtung: Instituto do Mar, UNIFESP, Santos, Brazil. Hauptthemen laut MEDLINE: Biodiversity, Conservation of Natural Resources, Machine Learning, Invertebrates. Die Arbeit nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Conselho Nacional de Desenvolvimento Científico e Tecnológico-CNPQ, Agência Nacional do Petróleo, Gás Natural e Biocombustíveis.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Biodiversity monitoring programs need to deliver accurate, timely, and actionable predictions. To establish a predictive monitoring program for deep‐sea benthos of the Santos Basin, Brazil, we developed a two‐stage structured model that allowed comparison of biodiversity predictions obtained from environmental simulations (2M‐Sim). We also modeled the environmental variables as a function of spatial and temporal variables and compared this model's predictions with predictions obtained from real environmental data (2M). We built unstructured models (1M) as references to evaluate whether the proposed structured approach was reliable. We expected no significant differences between 1M and 2M or between 2M and 2M‐Sim. Data were obtained from 100 stations at depths of 25–2400 m during two surveys (2019 and 2021). In our model framework, we used 12 benthic macro‐ and meiofaunal variables, 44 sediment and water column environmental variables, and four spatial and temporal variables. We applied a random forest algorithm to the structured and unstructured models. All comparisons were performed with 20% of the dataset set aside for validation. The average accuracy was 72%, 69%, and 68% for the 1M, 2M, and 2M‐Sim models, respectively. Accuracies of 2M ranged from 38% to 84% and were generally higher for macrofauna. The observed accuracy loss from 1M to 2M (3%) and from 2M to 2M‐Sim (1%) was not significant for any biodiversity variable. The 2M model identified 30 significant environmental variables; bottom water parameters and sedimentary phytopigment and carbonate concentrations were the best predictors. Our approach supports biodiversity conservation by optimizing data needs and future sampling and by guiding data‐driven management decisions for benthic biodiversity.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

BiodiversityConservation of Natural ResourcesMachine LearningInvertebratesAquatic OrganismsEnvironmental MonitoringBrazilAnimalsPredictive Learning Models

Stichwörter der Autor:innen

benthosbentosbosque aleatoriocontinental marginenvironmental essential variablesmargen continentalmodelo estructuralmonitoreomonitoringrandom foreststructure modelingvariables ambientales significativas大陆边缘海底生物环境关键变量

Autor:innen und Institutionen

  • Instituto do Mar, UNIFESP, Santos, Brazil.
  • Geology and Geophysics Department, Fluminense Federal University, Niteroi, Brazil.
  • Departamento de Química, PUC-Rio, Rio de Janeiro, Brazil.
  • Instituto Oceanográfico, USP, São Paulo, Brazil.
  • Departamento de Biologia Marinha, IB, UFRJ, Rio de Janeiro, Brazil.
  • PETROBRAS Research Center, Centro de Pesquisas Leopoldo Américo Miguez de Mello (CENPES), Rio de Janeiro, Brazil.

ORCID-Profile: Gustavo Fonseca, Danilo C Vieira, Luciana Erika Yaginuma

Förderung

  • Conselho Nacional de Desenvolvimento Científico e Tecnológico-CNPQ
  • Agência Nacional do Petróleo, Gás Natural e Biocombustíveis

Zitieren

APA

Fonseca, G., Vieira, D. C., Carneiro, J. C., Carreira, R. S., Ceccopieri, M., Corbisier, T. N., Dalto, A. G., Figueiredo, A. G., Gallucci, F., Gheller, P., de Jesus, S. B., Lavrado, H. P., Lazzari, L., Marcon, E. H., Moreira, D. L., de Moura, R. B., Pape, E., Santarosa, A. C. A., Dos Santos Filho, J. R., …, & Watanabe, W. (2026). Structured machine learning modeling to support conservation of deep‐sea benthic biodiversity. Conservation Biology, 40(4). https://doi.org/10.1111/cobi.70255

Vancouver

Fonseca G, Vieira DC, Carneiro JC, Carreira RS, Ceccopieri M, Corbisier TN, et al. Structured machine learning modeling to support conservation of deep‐sea benthic biodiversity. Conservation Biology. 2026;40(4). doi:10.1111/cobi.70255

BibTeX
@article{Fonseca2026,
  author = {Fonseca, Gustavo and Vieira, Danilo C and Carneiro, Juliane C and Carreira, Renato S and Ceccopieri, Milena and Corbisier, Thais N and Dalto, Adriana Galindo and Figueiredo, Alberto G and Gallucci, Fabiane and Gheller, Paula and de Jesus, Simone Brito and Lavrado, Helena Passeri and Lazzari, Letícia and Marcon, Eduardo Hilzendeger and Moreira, Daniel Leite and de Moura, Rafael Bendayan and Pape, Ellen and Santarosa, Ana Cláudia Aoki and Dos Santos Filho, João Regis and de Mello E Sousa, Silvia Helena and Vicente, Thaisa Marques and Yaginuma, Luciana Erika and Yamashita, Cinthia and Watanabe, Wandrey},
  title = {Structured machine learning modeling to support conservation of deep‐sea benthic biodiversity},
  journal = {Conservation Biology},
  year = {2026},
  volume = {40},
  number = {4},
  doi = {10.1111/cobi.70255}
}
Kostenlos starten →