Artificial Intelligence Procedures for Tree Taper Estimation within a Complex Vegetation Mosaic in Brazil

Autor(es)

Nunes, Matheus Henrique ^{[1, 2]} ; Gorgens, Eric Bastos ^{[2, 3]}

Afiliação do(s) autor(es)

^[1] Univ Cambridge, Dept Plant Sci, Forest Ecol & Conservat Grp, Cambridge CB2 3EA - England

^[2] Univ Sao Paulo, Coll Agr, Dept Forest Sci, BR-13418900 Piracicaba, SP - Brazil

^[3] Univ Fed Vales do Jequitinhonha e Mucuri, Dept Forestry, BR-39100000 Diamantina, MG - Brazil

• Tipo de documento: Artigo Científico
• Fonte: PLoS One; v. 11, n. 5 MAY 17 2016.
• Citações Web of Science: 5

Resumo

Tree stem form in native tropical forests is very irregular, posing a challenge to establishing taper equations that can accurately predict the diameter at any height along the stem and subsequently merchantable volume. Artificial intelligence approaches can be useful techniques in minimizing estimation errors within complex variations of vegetation. We evaluated the performance of Random Forest (R) regression tree and Artificial Neural Network procedures in modelling stem taper. Diameters and volume outside bark were compared to a traditional taper-based equation across a tropical Brazilian savanna, a seasonal semideciduous forest and a rainforest. Neural network models were found to be more accurate than the traditional taper equation. Random forest showed trends in the residuals from the diameter prediction and provided the least precise and accurate estimations for all forest types. This study provides insights into the superiority of a neural network, which provided advantages regarding the handling of local effects. (AU)

• Processo FAPESP: 12/01044-9 - Determinação da altura relativa e intensidade amostral utilizando geometria analítica em povoamentos inequiâneos, para estimativa de volume e taper individuais
• Beneficiário: Matheus Henrique Nunes
• Modalidade de apoio: Bolsas no Exterior - Estágio de Pesquisa - Mestrado
• Processo FAPESP: 11/19236-9 - Determinação da altura relativa e intensidade amostral utilizando geometria analítica em povoamentos inequiâneos, para estimativa de volume e taper individuais
• Beneficiário: Hilton Thadeu Zarate do Couto
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 10/13723-2 - Determinação da altura relativa e intensidade amostral utilizando geometria analítica em povoamentos inequiâneos, para estimativa de volume e taper individuais.
• Beneficiário: Matheus Henrique Nunes
• Modalidade de apoio: Bolsas no Brasil - Mestrado