Mapeamento de áreas inundáveis através de sensoriamento remoto: análise de incertezas

• Ano de defesa: 2023
• Autor(a) principal: Abib, Viviane Caroline Oliveira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Engenharia Ambiental; UFSM; Programa de Pós-Graduação em Engenharia Ambiental; Centro de Tecnologia
• Palavras-chave em Português: Inundação; Sensoriamento remoto; Planejamento urbano; Mapeamento de áreas inundáveis; Inundation; Remote sensing; Urban planning; Mapping of floodable areas; CNPQ::ENGENHARIAS
• Link de acesso: http://repositorio.ufsm.br/handle/1/30273

The mapping of floodable areas is a non-structural measure, which acts as a control measure, assisting the decision-making process of State entities, which have the legislative obligation to carry out urban planning aimed at preventing disasters and calamities, such as floods. This study aimed to analyze the uncertainties of a flood mapping made from remote sensing, using satellite images (ALOS and SRTM), compared to the mapping of a flood event, in a sub-basin located within the basin hydrographic analysis of the Piabanha River, made with field topography data. The two digital elevation models (DEM) were filtered in the CloudCompare software, generating two new DEMs for analysis. The DEMs were subjected to the modeling process in the HE CRAS software, following the same parameters used in the original reference modeling. The results showed that the mapping of flood areas, made using remote sensing, can be applied in places where there is no mapping of the flood area, in an accessible way and can be carried out from any location. The modeling must be done with the filtered DEMs, which obtained higher levels of combination with the reference mapping and that a margin of error is added around the modeling, especially in stretches where the watercourse has sinuous curves, since the Modeling does not precisely delimit these areas. The data shows that approximately 40% of the modeling done by remote sensing would occupy a zone with a high probability of flooding and the remainder can be delimited as a medium and low probability zone, at the planner's discretion. Finally, the objective of the study was concluded, identifying the uncertainties of flood mapping with remote sensing and suggesting points of attention for the application of the method in other analyses. In this way, one of the main urban problems in Brazilian municipalities, floods, can have a source of supporting data, so that prevention measures can be taken, without the need for excessive professional and financial resources. Furthermore, researchers and universities can remotely map the basins and make them publicly available, so that the committees responsible for the river basins or public administrators can use them during the management of municipalities, helping urban planning that prevents the impacts of the flooding process.