Abordagem bayesiana para modelos dinâmicos de biomassa

• Ano de defesa: 2018
• Autor(a) principal: XAVIER, Érika Fialho Morais
• Orientador(a): ANDRADE, Humber Agrelli de
• Banca de defesa: DUARTE NETO, Paulo José, STOSIC, Tatijana, SILVA, Antonio Samuel Alves da, OLINDA, Ricardo Alves de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal Rural de Pernambuco
• Programa de Pós-Graduação: Programa de Pós-Graduação em Biometria e Estatística Aplicada
• Departamento: Departamento de Estatística e Informática
• País: Brasil
• Palavras-chave em Português: Análise bayesiana; Modelo de produção; Biomassa; Espadarte; Albacora bandolim
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA
• Link de acesso: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7237

Tuna and tuna-like species represent a great attraction for companies and fishers in general, due to their meat quality, their high commercial value and their extensive distribution in the oceans. Appropriate management is fundamental to the survival of some species, and to the sustainability of fi sheries. In this way, knowledge about abundance estimates and population biomass variations is paramount for sheries management. Dynamic models of biomass are essential techniques for calculating the estimates. This study proposed the use of Bayesian analysis for the models of Fox, Schaefer and Pella-Tomlinson. Initially, the objective was to evaluate the accuracy and precision of the information acquired about the parameters of the models. For this, catch and catch per unit effort data of the South Atlantic sword sh (Xiphias gladius) were used. It was concluded that there was no progress on precision and accuracy, taking as true "values" for the parameters the estimates of the last report of The International Commission for the Conservation of Atlantic Tunas (ICCAT). Sequentially, the objective was to evaluate the dynamic model of the biomass of Pella and Tomlinson, considering the variation and the fixation of the shape parameter in a scalar value. In this case, were used catch and catch per unit effort data of bigeye tuna (Thunnus obesus) of Atlantic. It was concluded that the model of Pella and Tomlinson with estimation of the shape parameter through of a prior distribution is more susceptible to in influences of non informative data and with discrepant points, and produces estimates more imprecise than when the form parameter is set as a scalar value, in the models of Schaefer and Fox.