Effect of Spatial Distribution of Precipitation Data on Temporal and Spactial Uncertainty of Swat Output

Accurate modeling of water quality, water availability, and transport of pollutants at the watershed scale requires an accurate representation of the precipitation data. For this reason, the ability of hydrologic models to predict accurate outputs depends to a great extent on how well the rainfall data is distributed. In general, rainfall intensities can vary in space and time, particularly in convective events. A number of schemes are available to account for temporal and spatial uncertainties of precipitation data. The simplest method is the arithmetic mean, which assumes the rainfall is uniformly distributed over the watershed. The Thiessen polygon method is an improvement over the arithmetic approach, by assigning the record from the closest rain gauge to the unstamped location. The Centroid method is another popular method. The centroid method uses the rain gauge nearest to the centroid of each subbasin. The objectives of this project are (1) to assess the impacts of using different interpolation schemes for incorporating spatially variable precipitation data into the Soil and Water Assessment Tool (SWAT) and (2) assess the impacts of using three levels of subbasin delineation on the streamflow