INCORPORATING SHALLOW AQUIFER UNCERTAINTY TO STREAMFLOW PREDICTION IN A FORESTED WETLAND COASTAL PLAIN WATERSHED

Parameterization of shallow aquifer characteristics is crucial when modeling streamflow in Coastal Plain watersheds of the US southeast due to complex surface and subsurface hydrologic interactions and prevalence of wide floodplains that are often covered with dense vegetation. The main objective of this study was to examine the performance of the Soil & Water Assessment Tool (SWAT) watershed-scale streamflow model linked with the Sequential Uncertainty FItting (SUFI-2) uncertainty algorithm. The study site is the Waccamaw River watershed, a low-gradient Coastal Plain watershed in the southeastern United States. Over 90% of the watershed has a shallow water table so model calibration was especially sensitive to four shallow aquifer parameters during the 2003-2007 calibration period: the initial depth of water in the shallow aquifer (SHALLST), the initial groundwater height (GWHT), the specific yield of the shallow aquifer (GW_SPYLD), and the groundwater delay or the time required for water leaving the bottom of the root zone to reach the shallow aquifer (GW_DELAY). These parameters dramatically affected the baseflow and streamflow predictions by adjusting the shallow aquifer water holding capacity. The daily Nash-Sutcliff Efficiency values ranged from 0.83-0.85 while the Mean Squared Error values ranged from 80.17- 248.97. The SUFI algorithm showed that uncertainty bounds of streamflow prediction were related to annual water balances of watershed and shallow aquifer contribution to streamflow can affect the distribution of prediction uncertainty significantly. The uncertainty analysis showed that the shallow aquifer and surface flow was 8.8% and 9.4% of the total water budget, respectively, during a dry year (2007). During a wet year (2006) the shallow aquifer and surface flow was 7.6% and 20.5% of the total water budget, respectively. The shallow aquifer component was highest during January and June in the dry year. During the wet year the shallow aquifer contribution was highest during January and lowest during June. Further investigation of this and other Coastal Plain watersheds is necessary to adequately understand low storage soil impact on the water budget and streamflow, particularly in alluvial riparian floodplains.