Modelagem e cenários hidrológicos em bacia representativa da Mata Atlântica no Sul do Brasil

Abstract

Human actions modify the environment and increase the degradation of water resources, which are limited and fundamental to the maintenance of life. Thus, it is vital that these resources are quantified, so they can be evaluated and protected, in search of self-sustainability. The application of a hydrological model allows the simulation of different scenarios, becoming a tool for planning and management for the respective committee of the basin, where alternative uses can be simulated beforehand and measures for the preservation of the same. Hydrological processes, such as climatic variables, change the water regime and the hydrological response of the basin. This study presents the results of the simulations with the SWAT (Soil and Water Assessment Tool) model in a small watershed in Southern Brazil (latitude 29°38'37.49" and longitude 53°48'21.99"), representative of the Atlantic Forest biome. This area was monitored by two sequential stations, each one with one rain gauge and one stage gauge, having a contribution area of 4.5 km² and 12 km ² respectively. The altitudes in the basin range from 316 m to 431 m and vegetation is predominantly composed of 55% of native forest and 39% of native pasture. The simulated period was from June 2008 to December 2009, corresponding to the period of monitoring. The temperature ranged from -2.2°C to 39.2°C. The annual rainfall average ranged between 2005 mm and 2250 mm. The quality of the results was characterized by the Nash-Sutcliffe efficiency index (NSE) and by the coefficient of determination (R²). The model was evaluated in a monthly and daily time step. At the monthly time step, the values obtained for NSE, in the calibration phase, were 0.80 and 0.60, respectively for the two sections. The values obtained for the R² were 0.81 and 0.61 in the same sections. In the daily time step, values obtained for NSE, in the calibration phase, was 0.36, for both sections. For R², these values were 0.38 and 0.37 in the respective sections. These results show that the fit was good for monthly values, but not for the daily. Due to the short period of monitoring, the validation of the model results was realized with the observed data from the middle section, with an area of 4.5 km². The values obtained for the NSE in the validation phase were 0.75 and 0.37 for the monthly and daily time step respectively, and for R², 0.75 and 0.37 for the monthly and daily values, which validate the fit. These results are in agreement with the values found by Paiva and Paiva (2006), who evaluated the SWAT in a downstream section, in the same basin, with an area of 18 km². These results show that the SWAT model fits well in this region with monthly time step. Probably, the low values of the adjustment for the daily time step and the main problems of using the model are due to the size of the basin, the short period of observed data and the need for more information on the physical characterization of the soil. Future climatic scenarios A2 (reflects the high emission of greenhouse gases and is considered pessimist) and B2 (reflects the low emission of greenhouse gases and is considered optimist) show that the maximum flows reach more extreme values in the future, the average temperature, maximum and minimum showed a variation of about 1° C. The pessimistic scenario considered A2 showed more extreme values so calculated as to the input data of water flow.