Simulação de eventos extremos de cheia de curto-prazo usando precipitação estimada por radar

• Ano de defesa: 2018
• Autor(a) principal: Queiroz, Klewerton dos Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso embargado
• Idioma: por
• Instituição de defesa: Universidade Federal de Alagoas; Brasil; Programa de Pós-Graduação em Recursos Hídricos e Saneamento; UFAL
• Palavras-chave em Português: Inundações – Impactos socioeconômicos; Modelo hidrológico; Radar meteorológico; Precipitação observada; Floods; Hydrological model; Radar; CNPQ::ENGENHARIAS::ENGENHARIA SANITARIA::RECURSOS HIDRICOS
• Link de acesso: http://www.repositorio.ufal.br/handle/riufal/5014

Flooding events caused by river overflows are a natural phenomenon that may cause severe socio-economic losses and environmental impacts due to human occupation along of natural floodplain areas. Such absence in urban plain regarding risk áreas occupation makes the riverine population vulnerable to flood disasters, such as the one occurred in 2010 in the city of Rio Largo, Brazil. In this sense, the hydrological modeling comes as an unevaluable tool for the planning and management of water resources, aiding the hydrological forecast conditions. Accuracy in precipitation data is critical for predicting floods, which depending on the density of the rainfall network may not be appropriate for an adequate characterization of the spatial and temporal distribution of precipitation across the watershed. Alternatively, estimates of precipitation by a weather radar may offer gain in represent the actual variability of precipitation at different scales in space and time. Therefore, the present study has the objective of simulating extreme short-term flood events through a distributed hydrological model in a tropical/semi-arid climate basin, which presents a historical of flood events. Observed and estimated radar hourly precipitation data were used as input to the hydrological model. The hydrological model was calibrated and validated for four extreme rainfall events (observed and estimated by radar), obtaining a good representation between the estimated and observed discharge data. The NS coefficient obtained using the observed rainfall was 0.92 with the manual calibration and 0.94 with the automatic calibration, getting better performance using the Muskingum-Cunge model, compared to the Inercial model as a method of flow propagation. The radar precipitation obtained better results in the flow estimates when a constant correction factor of 1.9 was applied in comparison with the correction factor varying with the distance. The results obtained shown that the meteorological radar comes as an alternative to precipitation estimation as a way to overcome the limitation of the rainfall network or to improve the information of the existing network, providing subsidies to the managers in the planning of water resources.