Modelagem da intermitência e do aporte hídrico a reservatórios no semiárido brasileiro: o caso da bacia do rio umbuzeiro, Ceará

• Ano de defesa: 2021
• Autor(a) principal: Lima, Gabriela Domingos
• Orientador(a): Araujo, José Carlos de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Recurso hídrico; Modelagem hidrológica; Escoamento; Modelo WASA; Water resource; Hydrological modeling; River discharge; WASA model
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/58893

The semiarid region in Brazil is known for recurrent dries that result in a hydrologic vulnerable area. This way, the management of water resources is more and more required, as well as the need to identify and understand the availability of water to attend its increasing demands. The lack of hydrological data in semiarid basins creates concern that can compromise the effective management of water resources and the evaluation of the water availability. For this reason, hydrological modeling is a solid ally to supply the lack of data and assist water management. In addition, researches across the globe indicate an increasing tendency of intermittency in rivers of different spatial scales. In this perspective, this research was conducted to classify through measurements and hydrological modeling the intermittency and water supply to the reservoirs of the Brazilian Semi-Arid area. The model used in this research, WASA, has a strong physical basis and has been used in several dry regions of the globe. The basin under study is the Umbuzeiro River, located in the city of Aiuaba, Ceará, it was evaluated in three nested basins: the Benguê (933 km²), the Aroeira (800 km²) and the Aiuaba Experimental Basin (BEA, 12 km²). Simulations were made for eleven parameterization sets, changing the saturated hydraulic conductivity of the soil on the slopes and the riverbed; the hydraulic conductivity of the crystalline basement; the depth of the riverbed; and the model rainfall adjustment parameter. The results illustrate that; among the eleven parameterization sets, none was successful in representing the two variables, (intermittency and water inflow to the reservoirs) in the three basin scales simultaneously. The largest basins (Benguê and Aroeira) were better represented than the small-scale basin. The simulation of the Experimental Basin of Aiuaba (smaller scale) obtained a notable improvement after the application of the hydraulic conductivity of the soils according to in-situ measurement. That indicates that the model has a strong physical base. The WASA model was able to represent well the water supply (and, consequently, the dynamics of stored volumes) to the Benguê Reservoir (18 hm³, NSE > 0.9) through the application of the saturated hydraulic conductivity measured in the field, without the need for a rainfall parameter of adjustment. In the Boqueirão reservoir (0.06 hm³), even the best performance was unsatisfactory (NSE = 0.38), although this result was obtained with the application of measured data. These observations indicate that the WASA model responds best in basins with a larger spatial scale. Therefore, it is recommended to carry out a wide calibration process of the WASA model, using multivariate optimization techniques. Given the complexity of the model and the relatively long computation time, such program should be done in a highly efficient computational environment. It is hoped, therefore, that the WASA model will be able to represent both the water supply and the intermittency of rivers for different scales. This will make it possible to prepare studies of global changes.