Diferentes métodos de prognosticar a distribuição diamétrica em povoamentos clonais de eucalipto

• Ano de defesa: 2013
• Autor(a) principal: Oliveira, Onair Mendes de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Ciências Florestais; Centro de Ciências Agrárias e Engenharias; UFES; Programa de Pós-Graduação em Ciências Florestais
• Palavras-chave em Português: Transition matrix; Diameter class distribution models; Matriz de transição; Estrutura diamétrica; Modelos de distribuição diamétrica; Processos de Markov; Eucalipto; Recursos Florestais e Engenharia Florestal; 630
• Link de acesso: http://repositorio.ufes.br/handle/10/5019

This case was aimed to evaluate the efficiency of different methods on the prognosis diameter structure in clonal eucalyptus. Was used 47 plots of fit and 30 plots for validation. Was tested three different models to estimate dominant height (Richards, Logistic and Schumacher), and the Richards model had the best performance. From this model we used the method of the guide curve and the plots were classified into three different site classes (21.5 m, 26.5 m, 31.5 m). The first method was to test matrix Buongiorno and Michie adapted. For both transition matrices were obtained and used in different sites. In general, the transition pattern of trees in different diameter classes was similar between the two sites more productive. It was noticed a significant growth stagnation in individuals larger diameter in the most productive. This may represent the time to intervene in the forest, both in the range of 36-48 months as of 60-72 months. The second method was tested matrix adapted to Peterson. Therefore, we tested different polynomial models for the estimation of transition probabilities. In this case, the model chosen was quadratic. Different stand variables were included in the original model, however only age and number of trees per hectare showed significant influences. From matrices estimated by these equations were designed the diameter distributions. The parameters of the Weibull function obtained in the portions of the adjustment were used to estimate the number of trees per diameter class in each of 30 plots validation database. In all these plots the number of trees per hectare (N / ha) Estimated and observed did not differ significantly by the Kolmogorov-Smirnov test. With respect to projected diameter distributions of the test results demonstrated no significant differences between the observed and estimated projections for the three tested methods. However, analysis of p-value shows that higher ratios of observed values are statistically equal to that observed for the method of Buongiorno and Michie adapted and Weibull function, instead of the method of Peterson adapted. The estimation of mortality, is not efficient in matrix methods. In this case it is suggested to use parallel to a model matrix method to estimate the mortality. Generally, the method of Buongiorno and Michie adapted and Weibull function and obtained similar results were better than the method of Peterson adapted.