O impacto do uso de ponderadores de dados agrupados na geoestatistica

AUTOR(ES)
DATA DE PUBLICAÇÃO

2003

RESUMO

This dissertation presents some techniques on calculating a weighted up set, ITom clustered data, that are associated to them. These data, when applíed to the sample set, can indicate statistical values that are closer to the true ones than the previous. This operation is called declustering. There are four described methods that intend to provide this weighted up set, through the analysis of the advantages and disadvantages of each method. These methods are: Area-of-Influence Polygon, Cell Declustering, Kriging ofthe Mean and Richmod Method. The ways that some statistical parameters should be modified so that the considered values may be used in these calculations, are also described. In order to analyze the influence of the weighted up set , there are seven known populations being studied . The populations taking part in this study are the following: declus.dat, data set used for the development of the cell declustering method development; the data set used in Isaaks (1989), nominated Walker Lake, and tive populations specificalIy simulated for this work, by using sequential Gaussian simulation. After obtaining the weighted variogram, a study about the impact in the Kriging and the Gaussian simulation where was done at the location of the studied evento From each method there are made conclusions, which lead to abandoning the CelI Declustering Method and the Influence Polygon, because their results are not consistent for the most studied populations. The Richmod Method deserves to be studied in order that his declustering power can be reduced. The most reliable method presented in this dissertation is the Kriging ofthe Mean.

ASSUNTO(S)

geoestatistica

ACESSO AO ARTIGO

Documentos Relacionados