Open Journal Systems

MODELAGEM DA RELAÇÃO HIPSOMÉTRICA PARA BRACATINGAIS DA REGIÃO METROPOLITANA DE CURITIBA-PR

Alexandra Consuelo de Plácido e Silva Bartoszeck, Sebastião do Amaral Machado, Afonso Figueiredo Filho, Edílson Batista de Oliveira

Resumo




Este trabalho teve como objetivo testar vários modelos tradicionais e outros genéricos para o ajuste da relação hipsométrica para povoamentos de bracatinga (Mimosa scabrella Benth) da Região Metropolitana de Curitiba, Paraná. Os dados foram obtidos em 20 parcelas permanentes, cujos diâmetros e alturas foram medidos aproximadamente aos 4, 5, 6 e 7 anos de idade e de 124 parcelas temporárias, cobrindo idades de 3 a 18 anos. A existência de dados em vários sítios, idades e densidades permitiu realizar ao todo 751 análises envolvendo a combinação de 23 modelos com três classes de sítios (índices de sítio: 16,8; 13,5 e 10,2 m), idades de 4 a 7 anos para as parcelas permanentes e 3 a 18 para as temporárias, e quatro tratamentos de densidade inicial (testemunha, 2000, 4000 e 8000 árvores/hectare), sendo as temporárias sem tratamento algum. O Índice de Ajuste de Schlaegel (IA), o erro padrão da estimativa em porcentagem (Syx%), seguida da análise de resíduos foram os testes usados para selecionar o melhor modelo. Como se obteve um grande número de combinações e como nenhum dos modelos foi sempre o melhor para todas as combinações, procedeu-se a um ordenamento dos mesmos (ranking) usando as estatísticas (IA e Syx%) para verificar qual dos modelos que, de forma genérica apresentasse a melhor performance. Para as parcelas temporárias e a combinação destas com as testemunhas das parcelas permanentes, os modelos genéricos desenvolvidos por regressão passo a passo tiveram melhor performance. Estes modelos envolveram características do povoamento, tais como, altura dominante (hdom) e diâmetro médio quadrático (dg).

MODELLING HEIGHT-DIAMETER RELATIONSHIP FOR Mimosa scabrella STANDS IN THE CURITIBA METROPOLITAN REGION

Abstract



This research was developed in order to test several traditional end generic models to fit the height-diameter relationship for stands of bracatinga (Mimosa scabrella Benth.) in the Curitiba Metropolitan Region. The data came from 20 permanent sample plots (325 m2 each), whose diameter and heights were measured approximately at ages 4, 5, 6 and 7 years and from 124 temporary sample plots covering ages from 3 to 18 years. The availability of data in several sites, ages and densities made possible to accomplish 751 analysis involving the combination of these mathematical models with three site classes (site indexes: 16.8; 13.5 and 10.2 m), ages from 4 to 7 years for the permanent plots and from 3 to 18 years for the temporary ones, and four treatments (control, 2000, 4000 and 8000 trees/hectare), being the temporary ones without any treatment. The Schlaegel fitting index (I.A.), the standard error of estimate in percent (syx%), followed by analysis of residuals were used to select the models. The existence of a large number of combinations, and given that none of the models was always the best for all combinations, led to the ranking based on I.A. and syx% to identify the best performance. Thus, for the permanent plot data, the Curtis model was the best one. For the temporary plots data, the generic models developed by stepwise procedure provided the best performance. These generic models involved characteristics of the stands, such as, dominant height and quadratic mean DBH.


Palavras-chave


bracatinga (Mimosa scabrella); relação hipsométrica; modelos tradicionais; modelos genéricos Key-words-: “bracatinga” (Mimosa scabrella Benth.); height-diameter relationship; traditional models; generic models.

Texto completo:

PDF


DOI: http://dx.doi.org/10.5380/rf.v32i2.2285