Impact of Watershed Subdivision and Soil Data Resolution on SWAT Model Calibration and Parameter Uncertainty¹

Abstract

Impact of watershed subdivision and soil data resolution on Soil Water Assessment Tool (SWAT) model calibration and parameter uncertainty is investigated by creating 24 different watershed model configurations for two study areas in northern Indiana. SWAT autocalibration tool is used to calibrate 14 parameters for simulating seven years of daily streamflow records. Calibrated parameter sets are found to be different for all 24 watershed configurations, however in terms of calibrated model output, their effect is minimal. In some cases, autocalibration is followed by manual calibration to correct for low flows, which were underestimated during autocalibration. In addition to normal validation using four years of streamflow data for each calibrated parameter set, cross‐validation (using a calibrated parameter set from one model configuration to validate observations using another configuration) is performed to investigate the effect of different model configurations on streamflow prediction. Results show that streamflow output during cross‐validation is not affected, thus highlighting the non‐unique nature of calibrated parameters in hydrologic modeling. Finally, parameter uncertainty is investigated by extracting good parameter sets during the autocalibration process. Parameter uncertainty analysis suggests that significant parameters show very narrow range of uncertainty across different watershed configurations compared with nonsignificant parameters. Results from recalibration of some configurations using only six significant parameters were comparable to that from calibration using 14 parameters, suggesting that including fewer significant parameters could reduce the uncertainty arising from model parameters, and also expedite the calibration process.

The effect of watershed scale on HEC-HMS calibrated parameters: a case study in the Clear Creek watershed in Iowa, US

“…Figure 4 illustrates that the mean of drainage density tends to decrease as the mean of sub-watershed area increases, while the average of longest flow length within each subwatershed increases. This result is similar to that found by Kumar and Merwade (2009). Further, we find that the function of averaged drainage density and averaged longest flow length versus averaged basin areas follows a power law.…”

“…For example, it has been shown that subwatershed size affects the determination of peak flow magnitude and that the effect changes for different storm types (Ghosh and Hellweger, 2011). Some researchers claim contradictory results regarding the effect of sub-watershed size on determining peak flow magnitudes (Ao et al, 2003;Kumar and Merwade, 2009;Ghosh and Hellweger, 2011). Some have reported that larger sub-watersheds decrease flood peak (Muleta et al, 2007), while others have concluded that the effect can be neglected (Cleveland et al, 2009;Chen et al, 2011).…”

“…The threshold is the minimum upstream drainage area for a channel to originate and can be specified by a percentage of total watershed area (Kumar and Merwade, 2009). In this study, threshold area is specified for river generation, and the values are listed in Table 1.…”

“…These studies include the Soil and Water Assessment Tool (SWAT) model (Tripathi et al, 2006;Muleta et al, 2007;Kumar and Merwade, 2009), MUSIC (Elliott et al, 2009), BTOPMC (Ao et al, 2003), the Storm Water Management Model (SWMM) (Ghosh and Hellweger, 2011), and the Hydrologic Modeling System (HEC-HMS) (Chen et al, 2011;Cleveland et al, 2009). These studies have investigated how watershed aggregation…”

The Curve Number Concept as a Driver for Delineating Hydrological Response Units

“…An obvious approach is to divide the watershed into its natural sub-watersheds, thus preserving the watershed's natural boundaries, flow-paths and channels for realistic water routing [23][24][25][26][27][28][29]. The concepts of critical source area [30][31][32][33][34][35][36], threshold drainage area [37] and aggregated simulation area [38] have also been used to delineate sub-watersheds within semi-distributed models. Moving from semi-to fully-distributed modelling, more detailed concepts are employed, such as grid elements, Representative Elementary Areas (REAs) and Representative Elementary Watershed (REW).…”