Daily Streamflow Simulation in a Small-Scale Farmland Catchment Using Modified SWAT Model

The original SWAT model performs poorly for daily streamflow simulation in small-scale farmland catchments in two ways: (1) the streamflows are usually underestimated for high-flow events while overestimated for low-flow events, and (2) there is an obvious time lag between the simulated and observed peak times. In this study, the SWAT model is modified by recalculating the peak flow rate and peak time based on the physical mechanism. The daily simulation of the modified model is validated with observed streamflows and also compared with the results of the original SWAT model based on a case study of the Pengjiahe Irrigation District, China, during the 1987-1991 flood seasons. Results show that a better match of the peak times was captured between the modified predictions and the measurements. Meanwhile, all peak streamflows agreed well with observations, indicating reasonable ranges of daily streamflow processes. It is also found that the modified model presents more actual and faster water recession rates than the original model. Therefore, it can be concluded that the modifications improved the applicability of SWAT in small-scale farmland catchments.