High Resolution Water Body Mapping for SWAT Evaporative Modeling in the Upper Oconee Watershed of Georgia, USA

4 years ago

High Resolution Water Body Mapping for SWAT Evaporative Modeling in the Upper Oconee Watershed of Georgia, USA

Amber R. Ignatius, John W. Jones

Technological improvements in remote sensing and geographic information systems (GIS) have demonstrated the abundance of artificially constructed water bodies across the landscape. While research has shown the ubiquity of small ponds globally, and in the southeastern United States in particular, their cumulative impact in terms of evaporative alteration is less well quantified. The objectives of this study are to examine the hydrologic and evaporative importance of small artificial water bodies in the Upper Oconee watershed in the northern Georgia Piedmont, USA, by mapping their locations and modeling these small reservoirs using the Soil Water Assessment Tool (SWAT). Comparative SWAT models were run with and without the inclusion of small reservoir surface area and volume. The models used meteorological inputs from 1990-2013 to represent years with drought, high precipitation, and moderate precipitation for both the calibration and evaluation periods. Statistical comparison of streamflow indicated that the calibration methodology produced results where the default model simulation without reservoirs fit observed flows more closely than the modified model with small reservoirs included (e.g., Nash-Sutcliffe Efficiency of 0.72 versus 0.64, r2 of 0.73 versus 0.66, and percent bias of 11.4 versus 21.6). In addition, Penman-Monteith, Hargreaves, and Priestly-Taylor evapotranspiration equations were used to estimate actual evaporation from 2,219 small water bodies identified throughout the 1,936.8 km2 watershed. Depending on the evaporation equation used, water bodies evaporated an average of 0.03-0.036 km3/year for the period 2003-2013. Using Penman-Monteith further, if the reservoirs were not considered and average actual evapotranspiration (AET) rates from the rest of the basin were applied, only 0.016 km3 of water would have left the basin as a result of evapotranspiration (ET). This finding suggests construction of small reservoirs increased evaporation by an average of 0.017 km3 per year (approximately 46,500 m3/day). As the construction of small reservoirs continues and high resolution image data used to map these water bodies becomes increasingly available, watershed models that evolve to address the cumulative impacts of small water bodies on evaporation and other hydrologic processes will have greater potential to benefit the water resource management community.

Publisher URL: http://onlinelibrary.wiley.com/resolve/doi

DOI: 10.1002/hyp.11398