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.