KCID journal (한국관개배수논문집)

Korean National Committee on Irrigation and Drainage (한국관개배수위원회)
권호 표지

Effects of Subwatershed Delineation on SWAT Estimation

소유역구분이 SWAT 예측치에 미치는 영향 평가

  • Heo, Seong-Gu (Korean National Committee on Irrigation and Drainage) ;
  • Kim, Gi-Seong (Korean National Committee on Irrigation and Drainage) ;
  • An, Jae-Hun (Korean National Committee on Irrigation and Drainage) ;
  • Im, Gyeong-Jae (Korean National Committee on Irrigation and Drainage) ;
  • Choe, Jung-Dae (Korean National Committee on Irrigation and Drainage)
  • 허성구 (강원대학교 농업생명과학대학) ;
  • 김기성 (강원대학교 농업생명과학대학) ;
  • 안재훈 (농촌진흥청 고령지 농업연구소) ;
  • 임경재 (강원대학교 농업생명과학대학) ;
  • 최중대 (강원대학교 농업생명과학대학)
  • Published : 2006.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The Soil and Water Assessment Tool (SWAT) model has been widely used in hydrology and sediment simulation worldwide. In most cases, the SWAT model is first calibrated with adjustments in model parameters, and then the validation is performed. However, very little study regarding the effects on SWAT estimation of subwatershed delineation was performed. Thus, the SWAT model was applied to the Doam-dam watershed with various threshold values in subwatershed delineation in this study to examine the effects on the number of subwatershed delineated on SWAT estimation. It was found the flow effect of subwatershed delineation is negligible. However there were huge variations in SWAT estimated sediment, T-N, and T-P values with the use of various threshold value in watershed delineation. Sometimes these variations due to watershed delineation are beyond the effects of parameter adjustment in model calibration and validation. The SWAT is a semi-distributed modeling system, thus, the subwatershed characteristics are assumed to be the same for all Hydrologic Response Unit (HRU) within that subwatershed. This assumption leads to variations in the SWAT estimated sediment and nutrient output values. Therefore, it is strongly recommended the SWAT users need to use the HUR specific slope length and slope value in model runs, instead of using the slope and the corresponding slope length of the subawatershed to exclude the effects of the number of subwatershed delineated on the SWAT estimation.

Keywords