R-LASH: implementação e aplicação do modelo hidrológico em linguagem R

• Ano de defesa: 2023
• Autor(a) principal: Moura, Maíra Martim de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Pelotas; Programa de Pós-Graduação em Recursos Hídricos; UFPel; Brasil
• Palavras-chave em Português: Módulo ArcLASH; Algoritmo multiobjetivo; Calibração automática; Gestão de recursos hídricos; ArcLASH module; Multiobjective algorithm; Automatic calibration; Water resources management; ENGENHARIAS; ENGENHARIA SANITARIA; RECURSOS HIDRICOS
• Link de acesso: http://guaiaca.ufpel.edu.br/xmlui/handle/prefix/12930

Growing industrialization and urbanization, associated with climate change, have impacted the space-time distribution of water resources and favored the increase in the unpredictability of hydrological processes. Aiming at a better understanding of the functioning of the water balance and the dynamics of the numerous processes that control the movement of water and their probable impacts on its quantity, several hydrological models have been developed. One of the models that has stood out in several works carried out in Brazil is the Lavras Simulation of Hydrology (LASH). The great advantage of LASH is the reduced number of parameters for representing hydrological processes. In addition, the model allows the use of less detailed databases, which favors its use, especially in developing countries. Despite having a small number of parameters to be optimized and having already performed well in studies located in different regions of Brazil, LASH still needs improvements related to its processing, mainly regarding calibration strategies. The main objective of this work was to implement the LASH hydrological model in an open programming environment, using the R programming language, as well as to evaluate its implementation, efficiency, and usability for three watersheds. The study area comprised the watersheds of the Piratini River (BHRP), the Fragata River (BHAF), and the Pelotas River (BHAP), located in the south of Rio Grande do Sul. In the proposed version, the LASH model, now called R-LASH, had its hydrological routine adapted to the R language and, in addition to the SCE-UA mono-objective calibration algorithm, three multi-objective algorithms (DREAM, MOPSO, and NSGA II) were added to the model routine. Based on the proposal, it was possible to expand the functionalities of the LASH model, mainly by standardizing the input files, performed in the ArcLASH module, and by the possibility of using multiobjective algorithms for model calibration. The calibration of the R-LASH for the watersheds analyzed resulted in adequate sets of model parameters, which allowed the simulation of the physical processes in a satisfactory way.