Modelagem de biorretenção com uso de modelo concentrado

• Autor(a) principal: Silva, Valquernei Jesus da
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal de Alagoas (UFAL)
• Texto Completo: http://www.repositorio.ufal.br/handle/riufal/6702
• topic: Jardim de Chuva (biorretenção); Método de puls; Solos - Percolação; Rain Garden (Bioretention); Puls method; Soils - Percolation; CNPQ::ENGENHARIAS::ENGENHARIA SANITARIA

The concept of LID (Low Impact Development) recommends the use of compensatory techniques that seek to suppress or mitigate the effect of urbanization through the restoration of the hydrological cycle, one of them being biorretention. In the context of the modeling, there are several options for biorretention, one of them being investigated in this research for quantitative purposes: the Puls method with output flow determined as a function of infiltration / percolation velocity. This study aims to evaluate the behavior of a biorretention device with the use of the Puls method, being the code written in Python language. The methodology involved the characterization of the study area, the description of the constructive aspects and the hydrological monitoring and investigated three conditions for the modeling: infiltration / percolation velocity constant K; infiltration / percolation velocity K as a function of the water level inside the biorretention and 2 constant infiltration / percolation velocity values K1 and K2, one before and after the soil saturation point defined by the local infiltration curve. In the first case, there is a parameter K, in the second case there are two parameters a and b that define the function K x water level, and in the third case there are two parameters K1 and K2. The calibration was performed by trial and error, and validation was also performed, both evaluated by the percentage difference in volume and by the Nash-Sutcliffe (NSE) efficiency coefficient. In the calibration step, the three configurations showed similar performance compared to the NSE, with values close to 0.92. As for the volume difference, the variable K configuration obtained results on average 46% higher than the other two configurations. In the validation stage, the variable K configuration presented most of the events with good results, as well as the highest mean value for the NSE coefficient, 0.41. It was concluded that the model was able to represent well the biorretenção for the three configurations and that the variable K configuration was the most adequate, although it presented a relatively large error in the calibration step.