Aplicação do modelo SWAT para simulação de cenários de uso da terra na bacia hidrográfica do rio Jucu

• Autor(a) principal: Correia, Tamires Partélli
• Data de Publicação: 2016
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal do Espírito Santo (riUfes)
• Texto Completo: http://repositorio.ufes.br/handle/10/7688
• topic: Jucu, Rio (ES); Modelagem hidrológica; Modelo SWAT; Hydrological modeling; Jucu River; Recursos Florestais e Engenharia Florestal; 630

The Jucu river watershed, along with the Santa Maria da Vitoria river basin are responsible by 50% of the water supply in the Espírito Santo state, and the production of 11% of the electricity in the state. The Jucu river watershed has important remaining forests for biodiversity preservation. Due to the importance of this area for the conservation and maintenance of water sources, it is necessary to use tools that aiming the optimizing of land use and understanding of behavior of the water dynamics in the watershed. Among the techniques used to represent the hydrological processes, stands out modeling. Among the several hydrological models, it emphasizes the SWAT model (Soil and Water Assessment Tool), a physical model and semi-distributed. This study had the main goal to simulate scenarios of use and occupation of land, using the SWAT model in the Jucu river watershed, located in the Espírito Santo state. Were defined 6 scenarios to evaluate the minimum, average and maximum flow, as follows: 20% of native forest area in the basin (F20) (scenario 1); scenario 2: 36% of native forest area in the basin (current use of land) (F36); scenario 3: 50% of native forest area in the basin (F50); scenario 4: 100% of native forest area in the basin (F100); scenario 5: 100% of the basin area occupied with eucalyptus cultivation (E100); scenario 6: 100% of the basin area occupied by pastures (P100). The model was calibrated and validated for daily and monthly scale, with statistical indices: Nash efficiency coefficient and Sutcliffe, Nash efficiency coefficient and modified Sutcliffe, determination coefficient and PBIAS index, which showed satisfactory performance in the calibration and validation phase of the model. All the scenarios demonstrated a sensitivity of the model to land use changes in the basin. For the study of minimum reference flows (Q7,10 and Q90), the highest values were found for the scenarios 4 (F100), 3 (F50) 5 (E100), 2 (F36) 1 (F20) and 6 (P100), respectively, in descending order. The study of minimum flows is extremely important to assess how important the forests to keep the water within the system, especially in dry periods. In relation to the current scenario (F36), it was observed that there was an increase in the average flow in the scenarios 6 (P100) and 1 (F20), and decrease of the average flow in the scenarios 4 (F100), 3 (F50) and 5 (E100). Regarding maximum flows, the highest values occurred in the scenarios 6 (P100), 5 (E100), 1 (F20), 2 (F36) 3 (F50) and 4 (F100), respectively, also in descending order. The study allowed to show the importance of the forests to maintain a higher water courses regularity, associated with higher minimum flows and lower maximum flows in scenarios with more forest, having become evident that only the assessment of the average flow is not sufficient for the analysis of water behavior in the basin.