SPATIAL DEPENDENCE STUDY OF Eucalyptus grandis DENDROMETRIC CHARACTERISTICS
Titel:
SPATIAL DEPENDENCE STUDY OF Eucalyptus grandis DENDROMETRIC CHARACTERISTICS
Auteur:
José Marcio de Mello João Luís Ferreira Batista Marcelo Silva de Oliveira Paulo Justiniano Ribeiro Júnior
Verschenen in:
Cerne
Paginering:
Jaargang 11 (2005) nr. 2 pagina's 113-126
Jaar:
2005
Inhoud:
The knowledge of spatial continuity structure of dendrometric characteristics is crucial in forest inventory, managementand planning. The inclusion of spatial continuity effect in inventory analysis provides stable and safe results that can be used in forestmanagement and planning. This paper evaluated the structure of spatial continuity of four dendrometric characteristics obtainedby two sampling procedures. Data were collected in 987 hectares of Eucalyptus grandis, located in the south of São Paulo State.Two sampling procedures were used, systematic grading and unencumbered systematic. For each plot from the two samples, thefollowing dendrometric characteristics were measured: volume, quadratic average diameter, basal area and average height of thedominant trees. For each one of these characteristics, anisotropic and isotropic semi variogrammes were built, in order toevaluate the spatial continuity structure. The semi variogrammes were built using the moment estimator method. The mainauthorized functions were adjusted to the experimental semi variogrammes, by The Minimum Square Method. The behavior of thespatial continuity was evaluated through the degree of spatial dependence and of the assigned semi variogrammes for the fourcharacteristics, in the appraised sampling methods, respectively. All appraised characteristics presented spatially structured,independently of the appraised sampling procedure. The continuity structure of the four characteristics was isotropic, i.e., thevariance among pairs of points depends on the separation vector h. Therefore, unidirectional semi variogrammes can be built forall appraised characteristics. The results suggest that, in forest inventory, the spatial component should be considered, i.e., plotsshould not be treated separately.
Uitgever:
Universidade Federal de Lavras - UFLA (provided by DOAJ)