Evaluation of SWAT manual calibration and input parameter sensitivity in the Little River Watershed.
Authors:
Feyereisen, G.W. Strickland, T.C. Bosch, D.D. Sullivan, D.G. USDA, ARS
Source:
Transactions of the ASABE. 2007 May-June, v. 50, no. 3, p. 843-855.
NALT Subjects:
agricultural watersheds simulation models statistical analysis Soil and Water Assessment Tool model calibration water quantity stormwater water flow Georgia
Other Subjects:
sensitivity analysis
Issue Date:
May-2007
Abstract:
The watershed-scale effects of agricultural conservation practices are not well understood. A baseline calibration and an input parameter sensitivity analysis were conducted for simulation of watershed-scale hydrology in the Little River Experimental Watershed (LREW) in the Coastal Plain near Tifton, Georgia. The Soil and Water Assessment Tool (SWAT) was manually calibrated to simulate the hydrologic budget components measured for the 16.9 km 2 subwatershed K of the LREW from 1995 to 2004. A local sensitivity analysis was performed on 16 input variables. The sum of squares of the differences between observed and simulated annual averages for baseflow, stormflow, evapotranspiration, and deep percolation was 19 mm 2 ; average annual precipitation was 1136 mm. The monthly Nash-Sutcliffe model efficiency (NSE) for total water yield (TWYLD) was 0.79 for the ten-year period. Daily NSE for TWYLD was 0.42. The monthly NSE for three years with above-average rainfall was 0.89, while monthly NSE was 0.59 for seven years with below annual average rainfall, indicating that SWAT's predictive capabilities are less well-suited for drier conditions. Monthly average TWYLD for the high-flow winter to early spring season was underpredicted, while the low-flow late summer to autumn TWYLD was overpredicted. Results were negatively influenced when seasonal tropical storms occurred during a dry year. The most sensitive parameters for TWYLD were curve number for crop land (CN2(crop)), soil available water content (SOL_AWC), and soil evaporation compensation factor (ESCO). The most sensitive parameters for stormflow were CN2(crop), curve number for forested land (CN2(forest)), soil bulk density (SOL_BD), and SOL_AWC. The most sensitive parameters for baseflow were CN2(crop), CN2(forest), ESCO, and SOL_AWC. Identification of the sensitive SWAT parameters in the LREW provides modelers in the Coastal Plain physiographic region with focus for SWAT calibration.
