Mostrar registro simples

dc.contributor.advisorSanquetta, Carlos Roberto, 1964-pt_BR
dc.contributor.otherUniversidade Federal do Paraná. Setor de Ciências Agrárias. Programa de Pós-Graduação em Engenharia Florestalpt_BR
dc.creatorWendling, Jeferson Luiz Gonçalvespt_BR
dc.date.accessioned2022-11-25T17:18:23Z
dc.date.available2022-11-25T17:18:23Z
dc.date.issued1998pt_BR
dc.identifier.urihttps://hdl.handle.net/1884/25250
dc.descriptionOrientador: Carlos Roberto Sanquettapt_BR
dc.descriptionDissertação (mestrado) - Universidade Federal do Parana, Setor de Ciencias Agrariaspt_BR
dc.descriptionÁrea de concentração: Manejo florestalpt_BR
dc.description.abstractResumo: Descreve-se um estudo desenvolvido com o propósito de ajustar modelos matemáticos que possibilitem estimar a produção de peso comercial (comestível) para creme palmito de Euterpe edulis Mart.. Os dados utilizados para a pesquisa, são provenientes de 142 palmeiras (isoladas ou em pequenos povoamentos) que foram plantadas a céu aberto entre os anos de 1976 a 1995, situadas em propriedades rurais e urbanas localizadas nos municípios de Antonina, Guaraqueçaba e Morretes, no Estado do Paraná. Os dados foram coletados em duas ocasiões: na primeira ocasião (Agosto de 1996) todas as árvores foram selecionadas e identificadas. As seguintes variáveis foram mensuradas: diâmetros (DAP e D60), alturas (total e da estipe) e o comprimento do colmo (porção da estipe que contém o creme de palmito) Adicionalmente, registrou-se informações sobre floração e frutificação, método e época de plantio e o estado fitossanitário de cada árvore. Na segunda etapa (Julho de 1997), todas as variáveis foram remensuradas, e as seguintes observações foram coletadas após a realização do corte da árvore: biomassa total e parcial (folhas, estipe e colmo), peso da cabeça de palmito, peso bruto e comercial (comestível) de creme de palmito e, seus respectivos comprimentos e diâmetros. O procedimento inicial de análise quantitativa dos dados incluiu o desenvolvimento de matrizes de correlação para estudar o relacionamento entre as variáveis (aquelas facilmente mensuráveis e as de diferentes expressões de biomassa) coletadas em campo. Após conhecer o coeficiente de correlação linear para cada par de variáveis, 7 modelos matemáticos para avaliação de biomassa, selecionados na literatura, foram testados. Diferentes expressões de biomassa foram usadas como variáveis dependentes para os ajustes. Como material de aplicação prática, as seguintes características facilmente obtidas foram usadas como variáveis independentes: DAP, D60, Altura Total, Altura da Estipe e Comprimento do Colmo, além da Idade. Dentre as variáveis independentes, selecionou-se aquelas com maior valor para o Coeficiente de Correlação Linear com a variável dependente: o DAP, e a Altura da Estipe. Para todas as variáveis dependentes, os coeficientes de correlação com a Altura da Estipe foram inferiores em relação aos valores verificados para a Altura Total. Por outro lado a Altura de Estipe, pode ser mensurada mais facilmente e com menor possibilidade de erro. O Incremento Corrente Anual, em DAP, D60, Altura Total e Altura da Estipe e Comprimento Colmo para o período 1996-1997, para cada palmeira, foi determinado como a diferença entre os valores observados na primeira e na segunda ocasião, respectivamente. As árvores que por ocasião da segunda mensuração estavam mortas ou cortadas, e aquelas cujos proprietários não permitiram a remensuração, foram desconsideradas. Para a estimativa da produção do Peso de Creme Comestível de palmito, foi desenvolvida em duas etapas de análise quantitativa. Primeiro, obteve-se as estimativas para as variáveis de fácil mensuração (DAP, Altura Total e Altura da Estipe), como função da Idade, mediante o ajuste do modelo de crescimento aos dados mensurados em 1997. Nesta fase foram testados os seguintes modelos: Chapman- Richards, Prodan e Gompertz. Na segunda etapa, obteve-se as estimativas para o Peso de Creme Comestível de palmito a partir das variáveis de fácil mensuração (DAP e Altura da Estipe em função da idade), ajustando-se a equação de biomassa aos dados observados para o peso comestível. Em conformidade com os resultados obtidos, e para o conjunto considerado conclui-se que: O estudo sobre a correlação entre as variáveis dimensionais de uma árvore, é fundamental para predição de variáveis dependentes (de difícil obtenção), através da análise de regressão; A correlação entre as variáveis que exprimem biomassa, em especial as localizadas na parte aérea das árvores apresentaram as correlações mais elevadas entre si; O DAP e a Altura da Estipe, são as variáveis recomendadas para a predição das variáveis de difícil obtenção, em especial o Peso de Creme Comestível de palmito e; O modelo indicado como o mais apropriado para a avaliação de biomassa, em palmeiras da espécie E edulis Mart., foi a equação Y = b . DAP + c . (DAP² . Hestipe). Concluiu-se também, que é possível representar o crescimento e a produção através de modelo matemático como função da idade, sendo os mais apropriados os modelo de Gompertz e de Chapman-Richards; o incremento máximo em DAP, Altura Total e em Altura da estipe, ocorre entre 4 a 7 anos de idade. Recomenda-se, entretanto, o desenvolvimento de pesquisas, com o objetivo de aprimorar o conhecimento das relações biológicas entre as variáveis dimensionais, bem como no sentido de desenvolver modelos matemáticos específicos para árvores individuais e que descrevam o crescimento e a produção de palmeiras.pt_BR
dc.description.abstractAbstract: A study on the fitting of mathematical models for the estimation of commercial weight of heart of palm, produced by Euterpe edulis Martius, is described. The data used in this research were obtained in 142 trees (growing in isolation or in small stands) planted under open growth conditions between 1976 and 1995. All trees were planted in agricultural and urban holdings located in the municipalities of Antonina, Guaraqueçaba and Morretes, state of Paraná, Brasil. The data were collected in two occasions: in the first occasion (August, 1996), all trees were selected and identified. The following variables were then measured: DBH, total and commercial height and length of culm (portion of the stem containing the "heart of palm"), for each tree. Phenology, planting year and method, as well as current health condition were also observed and recorded. In the second occasion (July, 1997), all variables were remeasured, and the following were recorded after all trees were felled: total and partial biomass (leaves, stem and culm), gross and commercial weight of the "hearth of palm" and respective diameters and length. The initial procedure for the quatitative analysis of the data included the development of correlation matrices in order to study the relationships among the variables (between those easily measurable and the different expressions of biomass) as collected in the field. After the coefficient of linear correlation for each pair of variables was determined, seven mathematical models, selected from the literature on biomass evaluation, were tested. Different expressions of biomass were used as dependent variables for all runs. As a mather of practical application, the following easily obtainable characteristics were used as independent variables: DBH, D60, total height, stem height, length of culm and tree age. The independent selected variables were those with the largest value for the coefficient of linear correlation with the dependent variable (DBH and stem height). For all dependent variables, the correlation coefficients with stem height were smaller than with total height. In addition, the stem height can be measured without major difficulties and with smaller errors. The current annual increment (CAI) in DBH, D60, total height, stem height and culm lenght, for the period 1996-1997, was determined as the difference between the values observed in the first and second occasions, respectively. The following trees were not remeasured in the second occasion: dead trees, trees that were cut between the two measuring occasions, and those located in holdings where the owners did not provide permission for felling. The quantitative evaluation of commercial weight of heart of palm was carried out in two steps: in the first one, the estimation was done in respect to the easily measurable variables (DBH, total height and stem height) as a function of age. The following growth models were tested with the data collected in 1997: Chapman-Richards, Prodan and Gompertz. In the second step, the evaluation of the commercial weight of heart of palm as carried out using DBH and stem height (previously estimated in the first step) as independent variables; biomass (commercial weight of heart of palm) was used as dependent variable. From the results it was concluded that: a) the correlation analysis between size variables of the trees is fundamental for the prediction of those variables of difficult mensuration, b) the correlation among the biomass variables, in special among those located in the crown, presented the highest values; c) DBH and stem height are recommended for the prediction of those variables difficult to be obtained, in special the commercial weight of heart of palm; d) the most appropriate biomass model was Y = b . DBH + c . (DBH² . Hstem) ; e) It was also concluded that the representation of growth and yield of palm trees with mathematical models as a function of age is a real possibility; Gompertz and Chapman-Richards were the most adequate models in the present study; f) the maximum CAI for DBH, total and stem height was observed between ages 4 and 7 years. Finally, the development of additional research is recommended; these studies must be carried out in order to improve the knowledge on the biological relations among the dimensional variables, as well in the sense of developing mathematical models capable of describing the growth and yield processes of palm trees.pt_BR
dc.format.extent168f. ; 30cm.pt_BR
dc.format.mimetypeapplication/pdfpt_BR
dc.languagePortuguêspt_BR
dc.subjectÁrvores - Crescimentopt_BR
dc.subjectModelos matemáticospt_BR
dc.subjectDendrometria - Modelos matemáticospt_BR
dc.subjectPalmitopt_BR
dc.subjectTesespt_BR
dc.titleModelos matematicos de crescimento e de produção em biomassa para arvores de Euterpe edulis Mart. plantadas a ceu abertopt_BR
dc.typeDissertaçãopt_BR


Arquivos deste item

Thumbnail

Este item aparece na(s) seguinte(s) coleção(s)

Mostrar registro simples