Submitted to: IWRA WORLD WATER CONGRESS
Publication Type: Proceedings
Publication Acceptance Date: 7/11/2011
Publication Date: N/A
Citation: Interpretive Summary: Brazil has one of the world’s largest fresh water reserves however it is not distributed uniformly throughout the country. A lack of management and pollution reduce the amount of quality water available. The intensively cultivated crop lands in Southern Brazil have impacted the area’s water balance and are a source of environmental contamination. A computer-based simulation was used to evaluate hydrologic changes in the Arroio Lino watershed due to agricultural activities. The Soil and Water Assessment Tool watershed scale model did not capture the daily surface runoff peaks well and was unable to represent the short duration, high intensity storms that frequently occur in this watershed. Runoff that was underestimated during low flow periods may be due to uncertainty in estimating runoff from daily rainfall. Limited measured data and short calibration and validation periods may have also impacted the simulation results. The model was able to predict reliable estimates for evapotranspiration and identified that baseflow is important than surface runoff within the study area.
Technical Abstract: In Southern Brazil, native forests have been replaced by tobacco crops. These changes are negatively impacting the region’s water balance and resource quality. Understanding hydrologic processes is essential to accurately depicting water and contaminant transport dynamics. The Soil and Water Assessment Tool (SWAT) model was used to evaluate hydrologic processes for the Arroio Lino watershed, located in Southern Brazil. The observed streamflow at the watershed outlet was used for model streamflow sensitivity analysis, calibration and validation. Latin Hypercube (LH) and One-factor-At-a-Time (OAT) sensitivity analyses were performed on 27 input variables. Model calibration was performed with a Shuffled Complex Evolution Algorithm (SCE-UA). Time series plots and standard statistical measures were used to verify model predictions. The most sensitive parameters for runoff were curve number (CN2), soil evaporation compensation factor (ESCO), and baseflow alpha factor (ALPHA_BF). The predicted monthly streamflow matched well with the observed values, with a Nash–Sutcliffe coefficient of 0.87 and 0.76 for calibration and validation, respectively. Daily simulations were less accurate than the monthly predictions. Results indicate that the SWAT model is a promising tool to evaluate small watershed hydrology in subtropical areas for longer time periods. This model will be used for climate and land use change analyses and the impact of various management scenarios on stream water quality.