Allometric models of biomass and essential oils of Aniba rosaeodora Ducke in a tropical middle terrace soil forest of Masisea, Ucayali, Peru
Carregando...
Arquivos
Fontes externas
Fontes externas
Data
Orientador
Coorientador
Pós-graduação
Curso de graduação
Título da Revista
ISSN da Revista
Título de Volume
Editor
Tipo
Artigo
Direito de acesso
Arquivos
Fontes externas
Fontes externas
Resumo
In this study we developed an allometric model to predict the total biomass of the tropical tree of economic value Aniba rosaeodora to manage the sustainable use of natural populations through the use of branches and leaves, in the same way that allows us to evaluate the potential for its domestication and develop plantations. For this, 15 trees were sampled with an average diameter at breast height (DBH) of 9 cm (σ = 7.5 cm, CV = 44 %), height (Th) of 17.6 m (σ = 5.5 m, CV = 31 %), an average total biomass of 277.7 kg (σ = 203.9 kg, CV = 73 %), resulting in an average of 4.5 l of essential oil (σ = 3.3 l, CV = 75 %), and with an oil yield of 1.51 % (σ = 0.34 %, CV = 22.3 %). Thus, 29 allometric models were selected to estimate biomass of the stem (Cb), biomass of green branches (Pb), biomass of secondary branches (Sb), leaf biomass (Lb), total tree biomass (Tb) and essential oil, from variables of easy measurement (diameter) and strong correlation (Spearman's Rho 0.63–0.99; P < 0.05). From the tests, the models with the best correlation coefficients, R2 and R2aj were selected to estimate the biomass and essential oil content of each tree. Thus, the equations whose predictor variable is the D50 was the best fit, where Tb=(-1.73388+0.835102D50)2, ρ = 0.96; R2aj = 0.962 and Oil=exp(-6.4554+2.54862ln(D50)), ρ = 0.82; R2aj = 0.930, but by an indirect method using total tree biomass (Tb) a better fit can be obtained for essential oil, Oil=1/(-0.0793641+82.2948/Tb), ρ = 0.84; R2aj = 0.998.
Descrição
Palavras-chave
Allometric model, Amazon, Biomass, Carbon, Rosewood
Idioma
Inglês
Citação
Trees, Forests and People, v. 16.




