A experiÃncia tem mostrado que buscas de Ãtimos baseadas em apenas um objetivo, por mais cuidadosas que sejam, nÃo conseguem determinar uma soluÃÃo que modele satisfatoriamente um dado fenÃmeno. O uso de abordagens multiobjetivo pode ainda ser justificado pela natureza dos problemas reais, a qual requer a utilizaÃÃo de mÃltiplos objetivos, muitas vezes conflitantes. Um conceito muito utilizado neste contexto à o de dominÃncia de Pareto, a qual possibilita comparar soluÃÃes usando mÃltiplos objetivos e explorar diferentes caracterÃsticas dos dados observados. Este trabalho tem como foco a aplicaÃÃo de algoritmos evolucionÃrios baseados no acasalamento de abelhas em sua versÃo uni- (Honey-Bee Mating Optimization - HBMO) e multiobjetivo (Multiobjective Honey-Bee Mating Optimization - MOHBMO) na minimizaÃÃo de funÃÃes-teste, calibraÃÃo de modelos hidrolÃgicos e otimizaÃÃo da operaÃÃo de sistemas de reservatÃrios. A versÃo uniobjetivo à aquela proposta por Haddad et al. (2006), enquanto a multiobjetivo à uma proposiÃÃo do presente trabalho. Como algoritmos de referÃncia da performance foram utilizados: PSO (Particle Swarm Optimization), sua versÃo multiobjetivo MOPSO (Multiobjective Particle Swarm), SCEM (Shuffled Complex Evolucion Metropolis) e sua versÃo multiobjetivo MOSCEM (Multiobjective Shuffled Complex Evolucion Metropolis). AplicaÃÃes teÃricas foram realizadas pela minimizaÃÃo de problemas compostos por funÃÃes matemÃticas encontradas na literatura. AplicaÃÃes reais de calibraÃÃo de modelos hidrolÃgicos e operaÃÃes de sistemas de reservatÃrios tiveram como estudo de caso a calibraÃÃo dos modelos HYMOD e SMAP a nÃvel diÃrio para 15 estaÃÃes fluviomÃtricas localizadas nos estados do Cearà e Piauà e a otimizaÃÃo da operaÃÃo do sistema de reservatÃrios que compÃem o sistema de abastecimento de Ãgua da RegiÃo Metropolitana de Fortaleza, respectivamente. / Experience suggests that any single-objective search, no matter how carefully chosen, is not able to identify a solution capable of satisfactorily model a phenomenum of interest. Use of a multiobjective approach can yet be justified by the nature of real world problems, which in general involve multiobjectives, most of the time conflicting objectives. An approach very often used in multicriteria optimization is the concept of Pareto dominance, which allows us to compare different solutions by using different objectives and to explore different characteristics of the observed data. This dissertation employs evolutionary algorithms inspired on honey-bee mating for single (Honey-Bee Mating Optimization - HBMO) e multiobjectives (Multiobjective Honey-Bee Mating Optimization - MOHBMO) in the minimization of test functions and calibration of watershed models. The singleobjective version is the one introduced by Haddad et al. (2006), while its multiobjective version is proposed by the present work. As reference of their performance, the following algorithms were used: PSO (Particle Swarm Optimization), and its multiobjective version MOPSO (Multiobjective Particle Swarm), SCEM (Shuffled Complex Evolucion Metropolis) and its multiobjective version MOSCEM (Multiobjective Shuffled Complex Evolucion Metropolis). Well known theoretical functions were used to test the proposed algorithms. Real world applications on hydrologic model calibration employing HYMOD and SMAP models, with daily time steps, were carried out for 15 streamflow gauge stations located in the states of Cearà e PiauÃ. Besides, an optimization study for the operation of the reservoir system that supplies water for the Metropolitan Region of Fortaleza was also executed.

This dissertation focused on developing a numerical model of spatial and temporal changes in bed morphology of ridge and slough features in wetlands with respect to hydrology and sediment transport when a sudden change in hydrologic condition occurs. The specific objectives of this research were: (1) developing a two-dimensional hydrology model to simulate the spatial distribution of flow depth and velocity over time when a pulsed flow condition is applied, (2) developing a process-based numerical model of sediment transport coupled with flow depth and velocity in wetland ecosystems, and (3) use the developed model to explore how sediment transport may affect the changes in bed elevation of ridge and slough landscape patterns observed in wetlands when a conditional pulsed flow was applied. The results revealed the areas within deep sloughs where flow velocities and directions change continuously. This caused enhanced mixing areas within the deep slough. These mixing areas may have had the potential to affect processes such as sediment redistribution and nutrient transport. The simulation results of solute/sediment transport model also supported the existence of areas within the domain where the mixing processes happened. These areas may have caused that nutrients and suspended particles stay longer time rather than entraining toward downstream and exiting the system. The results of bed simulation have shown very small magnitude of change in bed elevation inside deep slough and no changes on the ridge portion of the study area, when a conditional pulsed flow is applied. These findings may suggest that implementing pulsed flow condition did not increase suspended sediment concentration, which results in insignificant changes in bed morphology of a ridge and slough landscape. Therefore sediment transport may not play an important role in wetland bed morphology and ridge and slough stability. Results from the model development and numerical simulations from this research will provide an improved understanding of how wetland features such as ridge may have formed and degraded by changes in water management that resulted from increasing human activity in wetlands such as The Florida Everglades, over the past decades.