Variação de parâmetros dendrométricos de Pinus taeda L. e a distribuição espacial de atributos do solo por técnicas de mapeamento digital de solos

• Ano de defesa: 2017
• Autor(a) principal: Horst, Taciara Zborowski
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Agronomia; UFSM; Programa de Pós-Graduação em Ciência do Solo; Centro de Ciências Rurais
• Palavras-chave em Português: Solo florestal; Mapeamento digital de solos; Silvicultura; Forest soils; Forestry; Digital soil mapping; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO
• Link de acesso: http://repositorio.ufsm.br/handle/1/14132

The study of the relationship between yield potential and the conditions offered to the development of forest stands is fundamental for the adequate management of the forest, aiming at maximum sustainable yield. Considering the need of using soil spatial information in favor of the forest production system, the overall objective of this dissertation was to evaluate the relationship between dendrometric, topographic and pedological variables and to perform the spatial prediction of the pedological variables correlated with the variation of the dendrometric parameters of Pinus taeda L. The study was conducted in an area of 109 ha of pinus at 29 years, in the city of Campo Belo do Sul, Santa Catarina. Samples were taken from 11 profiles and 126 points and groups were defined for pedologic, topographic and dendrometric variables. Through the analysis of Spearman correlation, variables with higher correlation to the dendrometric parameters of pinus were selected. For each selected variable, prediction functions were built using multiple linear regression and random forest. The best models were used for spatial prediction. Physical and morphological properties of soil prevail over chemical ones in the correlation with dendrometry. Soil depth, thickness of superficial horizon, elevation and vertical distance to channel network were the variables most related to the dendrometric parameters. The best prediction result was obtained with the random forest model, which showed R²= 0.25 and RMSE of 30.23 cm in the prediction of depth of solum, and R²= 0.15 and RMSE=11.75 cm in the prediction of thickness of superficial horizon. The results found in this study confirmed the hypothesis that pinus production is affected by pedologic variables, and that this information can be predicted using digital soil mapping.