Abundância e detectabilidade das aves comuns do Cerrado de Minas Gerais

• Ano de defesa: 2014
• Autor(a) principal: Souza, Alexander Zaidan de
• Orientador(a): Ribon, Rômulo
• Banca de defesa: Feio, Renato Neves , Peternelli, Luiz Alexandre , Giudice, Gisele Mendes Lessa Del
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Biologia Animal
• Departamento: Biologia e Manejo animal
• País: BR
• Palavras-chave em Português: Ornitologia; Ave - Cerrado (Minas Gerais); Unmarked
• Palavras-chave em Inglês: Ornithology; Bird - Cerrado (Minas Gerais); Unmarked
• Área do conhecimento CNPq: CNPQ::CIENCIAS BIOLOGICAS::ZOOLOGIA
• Link de acesso: http://locus.ufv.br/handle/123456789/2291

This paper evaluates abundance and detectability of the most common species of the Cerrado of Minas Gerais, exploring aspects of population change. For this two traditional methods , index of point count (IPA) and index of frequency lists (IFL) and a more modern method, based on the use of unmarked package of the statistical program R, which is highlighting this type of analysis were used. Sampling was conducted between January and February 2009 , November 2009 , May 2012 , December 2012 and January and February 2013. The methods used for data collection were the MacKinnon lists and point count. Analyses of population change were made based on IFL and IPA . The analyzes of abundance and detectability were obtained in the R program through unmarked package. In this ultimately were built twenty-four models to test the effect of habitat on the abundance of the species studied, and the effect of habitat (pasture, typical cerrado and cerrado) and time of sampling start in detectability of species. In all selected models goodness-of-fit of the Sum of Squares of Errors was applied, the mean value of α > 0.05. The 15 most abundant species were selected, and from the IFL and IPA data and unmarked, generated graphs of variation of abundance (in the entire area and also separated by habitat type) and detectability throughout the day (time linear or squared) function. The model that best explained the abundance in the entire area and also separately by habitat considers the habitat as the most influential variable, and detectability as a linear function. The species showed seasonal variation were : Basileuterus flaveolus, Elaenia cristata, Tangara sayaca and Turdus leucomelas, only E. cristata has published migratory patterns . The selection of the habitat variable says that species abundance is mainly affected according to vegetation structure, and detectability follows the basics of ornithology in which birds are most active during the morning. The results also show migratory movements can be hidden in very common species. So studies with common species are important to detect patterns of variation population still unraveled.