Calibração de um modelo hidrológico de escala diária para bacias hidrográficas do semiárido cearense

Abstract

Planning and managing water resources depends heavily on the availability of hydrometeorological data. However, in Brazil, the absence of data or errors in the existing series is common, especially when working with river-flow data. In Ceará, there is a great availability of pluviometric data, however, there are few fluviometric stations with data series long enough for hydrological modeling. In this context, rainfall-runoff modeling is an alternative to the lack of data. It can be used to fill the erros in the flow series of the fluviometric stations. In this Bachelor thesis, the Soil Moisture Accounting Procedure (SMAP) for daily scale proposed by Lopes, Braga and Conejo (1982) was used. We chose to use the Multiobjective evolutionary algorithm MOPSO (Multiobjective Particle Swarm Optimization) for the automatic calibration of the model parameters. The model was calibrated and validated for the river basins of Arneiroz (36020000), Icó (36290000), Iguatu (36160000) and Senator Pompeu (36470000), which are located near reservoirs of high relevance for the Ceará's water supply. The best calibrations were chosen according to the best Nash-Sutcliffe efficiency coefficient of the simulated series. In all 4 basins, the NSE of the series obtained were above 0.70, indicating the high quality and consistency of the calibration parameters. Although the calibration parameters perform poorly in some performance indices and validation, they can still be used in the absence of observed data in the basin.