Índices de dispersão e geoestatística: uma análise sobre dependência espacial e fatores relativos à restrita distribuição de Paepalanthus chiquitensis Herzog nos campos rupestres

• Ano de defesa: 2021
• Autor(a) principal: Alves, Gabriella de Freitas
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Agronomia
• Palavras-chave em Português: Sempre-viva; Campo rupestre; Teoria da Dispersão Lexiana; Dependência espacial; Distribuição agregada; Paepalanthus chiquitensis; Índice de Morisita; Semivariogramas; Krigagem; CNPQ::CIENCIAS AGRARIAS; Agronomia
• Link de acesso: https://repositorio.ufu.br/handle/123456789/31511; http://doi.org/10.14393/ufu.te.2021.75

For sustainable management and conservation plans of Paepalanthus chiquitensis Herzog (Eriocaulaceae), given its ornamental potential, studies that analyzes its spatial distribution is imperative. The objective of this thesis was to characterize the spatial distribution of P. chiquitensis, one of the species known as “everlasting”, not only in relation to its blunt aggregation, but also as to the factors that lead to its restricted occurrence in rupestrian fields of the Cerrado. For this purpose, counting (NI) and georeferencing of individuals of the species was carried out in 272 sample units in an area of rupestrian field, located around the municipality of Tapira-MG. In chapter I, a critical and temporal analysis of the dispersion indices was made and its effectiveness in determining the spatial distribution of the species was tested. In this analysis it was possible to identify that the genesis of the indices is anchored in the Lexian Dispersion Theory, especially in the statistical series of Bernoullian, Poisson and Lexis, which allowed the association of probability distributions with the patterns. The revised indices showed a very close relationship with each other and made little progress in relation to their precursors. All indicated an aggregated distribution of individuals of P. chiquitensis, but did not allow an analysis of the spatial dimension (scale) in which the aggregation occurs, as well as related factors. Another limitation of the indices was not considering the possible existence of spatial dependence between the observations, which is capable of generating important hypotheses for the investigation of factors related to spatial structure. The incorporation of dependence in the analysis represents one of the great advances in ecological research and for that, one of the methods is geostatistics. In chapter II, geostatistical methods associated with factor analysis identified the factors related to their restricted distribution. For this, in addition to the number of individuals, in 60 geo-referenced sample points were obtained data on altitude, humidity, and other physical-chemical properties of the soil. By geostatistical techniques, semivariograms and kriging interpolation maps were obtained. With the semivariograms the aggregation of the specie was estimated in an area with average radius of 50.89 m. The interpolation maps pointed out that the aggregation is closely related to more humid soils and with higher clay contents, with difficult establishment in soils with sandy texture. This information will be useful for management programs, not only of the species, but of other representatives of the Eriocaulaceae family.