Abstract
This study assesses the performance of the Soil and Water Assessment Tool (SWAT) in simulating streamflow and sediment for the Song River watershed, with a focus on calibration, validation, and sensitivity analysis. Thirteen parameters were selected for calibration, with eight identified as highly sensitive, reflecting key hydrological processes of the area. The model was calibrated for the period 1974–1995 and validated from 1996 to 2004, with additional testing using field data collected in 2022–2023 through Acoustic Doppler Current Profiler (ADCP) measurements. Key model adjustments, such as the baseflow recession constant (ALPHA_BF) and channel roughness coefficient (CH_N2), were set to 0.05 and 0.04, respectively, to capture the area’s groundwater dynamics and channel characteristics. The calibration results indicated a strong fit, with R2 values of 0.77, NSE of 0.70, and PBIAS of 17.06, demonstrating good agreement between observed and simulated streamflow. Validation showed slightly lower but acceptable performance, with R2 of 0.75 and NSE of 0.68. Further ADCP validation from field data showed R2 values of 0.79 and 0.78 for two monitoring sites, confirming the model’s reliability. Sediment yield simulations at site-2 yielded R2 values of 0.70 and 0.59 for calibration and validation, with NSE values of 0.53 and 0.52, indicating the model’s capability to simulate both streamflow and sediment accurately. These results demonstrate SWAT’s practical utility for water resource management in similar data-limited regions.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call