Please use this identifier to cite or link to this item:
Title: Métodos para análise de configurações pontuais em redes lineares
Other Titles: Method for analysis of point configurations in linear networks
Authors: Scalon, João Domingos
Scalon, João Domingos
Oliveira, Marcelo Silva de
Oliveira, Deive Ciro de
Keywords: Redes lineares
Processos pontuais
Manejo florestal
Matas ciliares
Estatística espacial
Linear network
Pontual processes
Forest management
Riparian forests
Spatial statistics
Issue Date: 16-Jul-2018
Publisher: Universidade Federal de Lavras
Citation: SILVA, V. F. da. Métodos para análise de configurações pontuais em redes lineares. 2018. 49 p. Dissertação (Mestrado em Estatística e Experimentação Agropecuária)-Universidade Federal de Lavras, Lavras, 2018.
Abstract: In order to understand patterns using puntual processes in a scenario different from the usual one where we have a spatial area of study most often bivariate, that is, the observations scattered over a surface where the observed points can be located anywhere in that area , being able to be georeferenced by latitude and longitude, aiming to present and describe the methodology under study when the points are positioned on a section of line, that when united is called a linear network. Knowing that Brazil has the largest rainforest in the world and in it is also situated the largest hydrographic basin in the world. It will be used as a linear network a river, which cuts off the Fazenda Canário located in the state of Acre, which will be playing the role of linear network and the georeferenced event observed along the banks of this river will be the Heura Crepitans with properties above the established slaughtering measure. It is proposed to analyze how they are distributed in this type of network space scenario. From the analysis of first-order effects, using linear kernel smoothing, where it was shown to be better than the planar kernel which in turn underestimated the intesity rates. Using K-linear function for second-order effects, the results show no clustering present along the river, effective compared to the planar approach that verified the clustering patterns, thus the method is effective for the study and important for the analysis in a forest river census of the Amazon Forest, defining its patterns.
Appears in Collections:Estatística e Experimentação Agropecuária - Mestrado (Dissertações)

Files in This Item:
File Description SizeFormat 
DISSERTAÇÃO_Métodos para análise de configurações pontuais em redes lineares.pdf1,81 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.