Proposição de método multivariado de estratificação vertical em floresta natural

• Ano de defesa: 2011
• Autor(a) principal: AMARAL, Rogério Carlos Costa do
• Orientador(a): FERREIRA, Rinaldo Luiz Caraciolo
• Banca de defesa: STOSIC, Borko, STOSIC, Tatijana, ALVES JÚNIOR, Francisco Tarcísio
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal Rural de Pernambuco
• Programa de Pós-Graduação: Programa de Pós-Graduação em Biometria e Estatística Aplicada
• Departamento: Departamento de Estatística e Informática
• País: Brasil
• Palavras-chave em Português: Análise de agrupamento; Análise discriminante; Estratificação florestal
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA
• Link de acesso: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/5302

The objective of this research was to propose a new methodology for vertical stratification of natural forests, based on total tree height, from the application of multivariate techniques. The methodology was compared to others methods used in forest stratification. The structural data used in this work came from a research by Costa Junior (2006) of an Atlantic forest fragment known as ‘Mata das Caldeiras’, with 38.56 ha, located in Catende, PE. The measures used were the total tree height, diameter of breast height (DBH) and basal area of 766 trees with circumference at breast height (CBH) greater than or equal to 15 cm. The proposed methodology of multivariate vertical stratification resulting from the combination of principal component analysis, cluster and discriminant analysis was compared to the methods of Vega (1966), IUFRO, Longhi (1980), Sousa and Leite (1993), Guedes-Bruni et al. (2006) ,IUFRO-Lorey and Sanquetta (1995 ). The efficiency of the method used was considered when there was a decrease in the coefficient of variation (CV) within the stratum after the use of discriminant analysis. It was observed significant gains in accuracy of CV for the multivariate method proposed in relation to the others when the discriminant analysis was applied. However, the difficulties of vertical stratification are clear because the gains in CV vary widely depending on the methodology used. The new methodology can be used for vertical stratification of natural forests and presents as main advantage the removal of the researcher’s subjectivity.