GIS-based Sediment Yield Simulation in a Large River Basin

While the GIS - based distributed watershed model is used to simulate runoff and sediment yield in a micro - scale watershed in the Yellow River Basin, few studies are at meso - scale. The upper watershed of the Luohe River, a tributary of the Yellow River, is selected as the study area to validate the distributed SWAT (soil and Water Assessment Tool) model using GIS technology. Firstly, the basic GIS database integrating DEM, soil, landuse map, weather, and land management data, is established. A two - stage "Brute Force" optimization method is used to calibrate the parameters with the combination of observed monthly flow and sediment from 1992 to 1997. In the process of calibration, the automated digital filter technique is used to separate the direct runoff and base flow. The direct runoff is firstly calibrated, and the base flow, then the total runoff is matched, and the sediment yield was calibrated. Finally, with the calibrated parameters, the model is validated with the observed monthly flow and sediment from 1998 to 1999. Relative error (Re) is within 20%, Determination Coefficient (R(2)) and Nash - Sutcliffe Efficiency (E(ns)) are all equal to or above 0.70 during both calibration and validation periods. The results demonstrate that SWAT model based on GIS could be successfully used to simulate long - term runoff and sediment yield in a meso - scale watershed in the Yellow River Basin.