Modelagem hidrodinâmica uni e bidimensional aplicada ao rompimento da Barragem do Fundão em Mariana/MG

• Ano de defesa: 2019
• Autor(a) principal: Brandão, Mariana Ribeiro de Lima
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio de Janeiro; Brasil; Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia; Programa de Pós-Graduação em Engenharia Civil; UFRJ
• Palavras-chave em Português: Estudo de ruptura de barragem; Barragem de rejeitos; HEC-RAS; RiverFlow2D; CNPQ::ENGENHARIAS::ENGENHARIA CIVIL
• Link de acesso: http://hdl.handle.net/11422/13674

Dams are man-made structures that offer numerous benefits to mankind, however, in the event of a dam break, the uncontrolled wave released from the reservoir significantly increases river discharge, exposing the downstream area to flooding. Given the potential damage, hypothetical rupture studies of tailings dams are of utmost importance regarding the safety of the population living in the downstream valley, however the methodological uncertainties inherent in these studies require delimitation. DNPM Ordinance no . 70,389 assumes that the mining dam safety plan includes the emergency action plan in specific cases, but does not specify the methodology to be employed. In this context, a methodological approach was developed to deal with assumptions and procedures associated with the development of tailings dam rupture studies through uni and bidimensional hydrodynamic modelling supported by computational tools taking as reference the case of the Fundão dam failure in Mariana/MG, which was characterized by remote sensing techniques and field visit. The results showed that the DEM stream burning is important to estimate the wave arrival time and the maximum flood depth. Regarding the flow type, newtonian or nonnewtonian, it was observed that there was little interference to the maximum flood depth. In contrast, scenarios that propagated newtonian fluids represented better the wave arrival time than models that propagated non-newtonian fluids, even though both scenarios delayed the wave arrival time in comparison to the observed one. In addition, the dam breach opening time had no significant influence on the wave arrival time or the peak flow rate of the hydrograph.