• Entrar
    Ver item 
    •   Página inicial
    • BIBLIOTECA DIGITAL: Teses & Dissertações
    • 40001016015P0 Programa de Pós-Graduação em Engenharia Florestal
    • Teses
    • Ver item
    •   Página inicial
    • BIBLIOTECA DIGITAL: Teses & Dissertações
    • 40001016015P0 Programa de Pós-Graduação em Engenharia Florestal
    • Teses
    • Ver item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Correlações e estimadores de biomassa acima do solo em florestas no sul do Brasil

    Thumbnail
    Visualizar/Abrir
    R - T - JONATHAN WILLIAM TRAUTENMULLER.pdf (2.633Mb)
    Data
    2019
    Autor
    Trautenmüller, Jonathan William, 1988-
    Metadata
    Mostrar registro completo
    Resumo
    Resumo: Um método para gerar estimativas de biomassa é o emprego de equações matemáticas, que podem ser ajustadas por diferentes técnicas. O presente trabalho teve como objetivo geral testar, analisar e comparar métodos de ajuste dos estimadores de biomassa total acima do solo e seus componentes quanto ao desempenho e a consistência biológica de suas estimativas, bem como o desmembramento das correlações simples entre as variáveis dendrométricas e alométricas, em florestas nativas no Brasil. Os dados foram provenientes de oito áreas localizadas no Paraná e Rio Grande do Sul, onde foram coletadas informações de diâmetro a 1,3 m do solo (dap), altura total, altura do fuste, altura da copa, biomassa dos componentes fuste, galhos, folhas e biomassa total acima do solo. Funções nãolineares foram ajustadas de forma independente (ENGLS) e simultânea (WNSUR), tomando-se o dap e a altura total como variáveis independentes nos modelos de regressão. Para o ajuste das equações de razão foi utilizado o volume do cilindro com base no diâmetro quadrático médio (??????) como variável independente. Para correção da heteroscedasticidade dos estimadores por razão os dados foram estratificados em dois níveis, o primeiro por classes de dap e o segundo pelo coeficiente angular do estimador por razão entre a biomassa total acima do solo e o ??????. Para agrupar as árvores no segundo estrato a posteriori foi aplicada a análise discriminante. A correlação de Spearman foi utilizada devido a não normalidade dos dados e a análise de trilha foi aplicada para desmembrar as correlações simples nos efeitos diretos e indiretos entre as variáveis. As equações de regressão, ajustadas de forma simultânea (WNSUR), apresentaram maior eficiência e consistência biológica quando comparados com o ajuste independente. O ajuste simultâneo apresenta melhores propriedades estatísticas e biológicas para gerar estimadores de biomassa, porque em seu ajuste considera as correlações contemporâneas entre os componentes de biomassa. Os estimadores por razão propiciaram a aditividade de forma natural dos componentes de biomassa e são promissores para melhorar a precisão das estimativas. Na comparação de desempenho, os estimadores por razão foram melhores que os estimadores por regressão nas classes de diâmetro acima de 20 cm. A análise de trilha evidenciou ser interessante que os componentes de biomassa possam ser integrados nos modelos de regressão como variáveis endógenas. Assim, a análise de trilha indicou que pelo menos um componente de biomassa deve ser incluído como variável independente na modelagem da biomassa por meio de regressão. Palavras-chave: Estimadores por regressão. Estimadores por razão. Análise de trilha. Consistência biológica. Estimadores eficientes.
     
    Abstract: One method to generate biomass estimates is the use of mathematical equations, which can be adjusted by different techniques. The aim of the present work was to test, analyze and compare methods of adjustment of the total biomass estimators aboveground and its components as to the performance and biological consistency of their estimates, as well as the dismemberment of the simple correlations between the dendrometric and allometric variables, in natural forests in Brazil. Data were collected from eight areas located in Paraná and Rio Grande do Sul, diameter at 1.30 m aboveground (dbh), total height, stem height, crown height, stem biomass, branches, leaves and total biomass above ground. Nonlinear functions were adjusted independently (ENGLS) and simultaneous (WNSUR), taking the dap and total height as independent variables in the regression models. For the adjustment of the ratio equations, the cylinder volume based on the mean square diameter (??????) was used as the independent variable. For correction of the heteroscedasticity of the estimators by ratio the data were stratified into two levels, the first by dbh classes and the second by the ratio estimator between the total biomass aboveground and ??????. The discriminant analysis was applied to group the trees in the second stratum a posteriori. Spearman's correlation was used due to non-normality of the data and the track analysis was applied to dismember the simple correlations in the direct and indirect effects among the variables. The regression equations, adjusted simultaneously (WNSUR), presented higher efficiency and biological consistency when compared with the independent fit. Simultaneous adjustment presents better statistical and biological properties to generate biomass estimators, because in its adjustment it considers the contemporary correlations between the biomass components. Ratio estimators have provided the natural additivity of the biomass components and are promising to improve the accuracy of estimates. In the performance comparison, the ratio estimators were better than the regression estimators in the diameter classes over 20 cm. The path analysis showed that it is interesting that the biomass components can be integrated into the regression models as endogenous variables. Thus, path analysis indicated that at least one biomass component should be included as an independent variable in the modeling of biomass by means of regression. Key-words: Regression estimators. Ratio estimators. Path analysis. Biological consistency. Efficient estimators.
     
    URI
    https://hdl.handle.net/1884/60280
    Collections
    • Teses [387]

    DSpace software copyright © 2002-2022  LYRASIS
    Entre em contato | Deixe sua opinião
    Theme by 
    Atmire NV
     

     

    Navegar

    Todo o repositórioComunidades e ColeçõesPor data do documentoAutoresTítulosAssuntosTipoEsta coleçãoPor data do documentoAutoresTítulosAssuntosTipo

    Minha conta

    EntrarCadastro

    Estatística

    Ver as estatísticas de uso

    DSpace software copyright © 2002-2022  LYRASIS
    Entre em contato | Deixe sua opinião
    Theme by 
    Atmire NV