Comparative Study of Soil and Water Assessment Tool and AnnAGNPS for Prediction of Runoff in a Plain Irrigation District of North China
This study aims to access and compare two distributed hydrological models, the Soil and Water Assessment Tool (SWAT) and the Annualized Agricultural Non-point Source (AnnAGNPS) model. The models are used to calculate runoff in the Fenhe Irrigation District where the catchment area is an agricultural plain region in North China. On the basis of the analysis of the theory and com-ponents of SWAT and AnnAGNPS, we built model databases using GIS and other interrelated observed data. Both models were calibrated for a six-year period (1996-2001) and validated using a five-year period (2002-2006) by comparing simulated and observed monthly and annual runoff at the outlet of the field. In the long-term continuous simulation for the calibration and validation period, both models generally provided good correlation and model efficiency for simulating surface runoff: relative error (RE) of the annual runoff, as well as correlation coefficient (R2) and Nash-Suttcliffe simulation efficiency coefficient (Ens) of the monthly runoff measurements satisfy the evaluation requirements. Comparatively, the results of SWAT simulation show better agreement with measured data, while AnnAGNPS registers comparatively large errors in the years with increasing annual rain-fall or months with rising amount of monthly rainfall and irrigation water. Such conditions increase the overall error. Both models, however, were utilized for runoff simulation of the plain irrigation district in North China.
Runoff Simulation SWAT AnnAGNPS Plain Irrigation District
Wenhui Hu Guangyong Li
College of Water Conservancy and Civil Engineering, China Agriculture University, Beijing, 100083, P. R. China
国际会议
南昌
英文
1101-1107
2010-10-22（万方平台首次上网日期，不代表论文的发表时间）