Indice de sitio e relação entre altura dominante e teores nutricionais das aciculas em povoamentos de Pinus elliottii var. elliottii no Segundo Planalto Paranaense
Abstract
O objetivo do presente trabalho de pesquisa foi de comparar modelos usados para relacionamento entre alturas dominantes e idades, para determinar a melhor equação que represente as curvas de Índice de Sitio para plantações de Pinus elliottii do Segundo Planalto do Estado do Paraná; relacionar teores de nutrientes das acículas que melhor se correlacionam com o crescimento em altura. Foram coletadas amostras de análise de tronco e acículas de treze locais diferentes, distribuídos em todas as classes de sítio com idades entre 16 e 18 anos. As estatísticas e a análise de resíduos mostraram que o modelo de Prodan foi o melhor entre os testados, para relação entre altura dominante e idade e por isso foi usado para construir Curvas de Índice de Sítio. A estabilidade das curvas também foi testada, provando haver confiabilidade no uso das mesmas. Os efeitos dos teores de nutrientes das acículas sob o crescimento em altura foram estudados através de análise de regressão. A variável dependente foi a altura dominante aos 15 anos de idade. As variáveis independentes foram obtidas através de análise de teores de nutrientes das acículas, sendo consideradas apenas as variáveis que apresentaram Significância do teste "F" a P=0,95 de probabilidade. Estas variáveis foram submetidas a dois métodos de seleção: Análise dos Componentes Principais (ACP) e "BACKWARD". A ACP selecionou as variáveis: teores de nitrogênio, magnésio, cobre, potássio e zinco foliares. O método "BACKWARD" selecionou as seguintes variáveis: teores de magnésio, potássio, zinco e cobre foliares. De acordo com as estatísticas os teores de magnésio e zinco foram os fatores mais limitantes ao crescimento em altura. The objectives of this research work were: to compare used models for the relationship between dominant heights and ages, and to chose the better equation to represent site index curves for Pinus elliotti plantations in the second plateau of the Parana State; to relate the nutrient contents of the needles with height growth. The data came from places,covering all old. Statistics and was the best one used to construct stem analysis and needles from 13 different site classes. Ages ranged from 16 to 18 years residual analysis showed that the Prodan model to relate dominant height over age, thus being site index curves. The estability of the curves was also tested, prooving there be confiability in using such curves. The effects of nutrient contents of needles on height growth were studied through regression analysis. The dependent variable was dominant height at 15 years of age. The independent variables were obtained through analysis of nutrient content of the needles, being considered only those nutrients variables which presented significant "F" test for P=0,95 of probability. These variables were submited to 2 methods of variables selection: Principal Component Analysis (PCA) and BACKWARD. The PCA selected the following variables: nitrogen content, magnesium. copper, potash and zinc. The BACKWARD method selected the following variables: magnesium. potash, zinc and copper in decreasing order. According to statistic analysis, magnesium and zinc content were the most limiting factors to the height growth. RESUMO
O objetivo do presente trabalha de pesquisa foi de comparar modelos usados para relacionamento entre alturas dominantes e idades, para determinar a melhor equação que represente as curvas de índice de Sítio para plantações de Pinus elliottii do Segundo Planalto do Estado do Paraná; relacionar teores de nutrientes das acículas que melhor se correlacionam com o crescimento em altura. Foram coletadas amostras de análise de tronco e acículas de treze locais diferentes, distribuídos em todas as classes de sítio com idades entre 16 e 18 anos. As estatísticas e a análise de resíduos mostraram que o modelo de Prodan foi o melhor entre os testados, para relação entre altura dominante e idade e por isso foi usado para construir Curvas de índice de Sítio. A estabilidade das curvas também foi testada, provando haver confiabilidade no uso das mesmas. Os efeitos dos teores de nutrientes das acículas sob o ;crescinento em altura foram estudados através de análise de regressão. A variável dependente foi a altura dominante aos 15 anos de idade. As variáveis independentes foram obtidas através de análise de teores de nutrientes das acículas, sendo consideradas apenas as variáveis que apresentaram significancia do teste "F" a P=0,95 de probabi1 idade. Estas variáveis foram submetidas a dois métodos de seleção : Análise dos Componentes Principais (ACP) e "BACKWARD". A ACP selecionou as variáveis: teores de nitrogênio, magnesio, cobre, potássio e zinco foliares. O método "BACKWARD" selecionou as seguintes variáveis: teores de magnesio, potássio, zinco e cobre foliares. De acordo com as estatísticas os teores de magnesio e zinco foram os fatores mais limitantes ao crescimento em a 1 tura. SUMMARY
The objectives o-f this research work were: to compare used models for the relationship between dominant heights and ages, and to chose the better equation to represent site index curves for Pinus elliottii plantations in the second plateau of the Parana State; to relate the nutrient contents of the needles with height growth. The data came from stem analysis and needles from 13 different P1 aces,covering all site classes. Ages ranged from 16 to 18 years old. Statistics and residual analysis showed that the Prodan model was the best one to relate dominant height over age,- thus being used to construct site index curves. The estábility of the curves was also tested, prooving there be confiability in using such curves . The effects of nutrient contents of needles on height growth were^studied through regression analysis. The dependent variable was dominant height at 15 years of age. The independent variables were obtained through analysis of nutrient content of the needles, being considered only those nutrients variables which presented significant "F" test for P=0,95 of probability. These variables were submited to S methods of variables selection: Principal Component Analysis (PCA) and BACKWARD. The PCA selected the following variables: nitrogen content, magnesium, copper, potash and zinc. The BACKWARD method selected the following variables: magnesium, potash, zinc and copper in decreasing order. According to statistic analysis, magnesium and zinc content were the most limiting factors to the height growth.
Collections
- Teses & Dissertações [8689]