Scale effects of DEM size on SWAT watershed delineation and runoff volume
Martha Otero, investigator
Social and natural scientists acknowledge the importance of scale effects and how relationships and processes vary at different scales. The term scale is used differently across disciplines and it is often used to refer to the magnitude of a study area, to the degree of detail and to the thematic scale or the temporal scale. Scales issues in hydrological modeling for watershed management have been of great concern since hydrological processes occur at a wide range of spatial and temporal scales. Predictions of physical models such as SWAT depend on the quality and accuracy of the input data such as de digital elevation model (DEM). For example, the resolution of the DEM is expected to affect the delineation of the watershed which at the same time will affect the model’s outputs such runoff, sediments and nutrients yields. This paper addresses the issue of scale effects DEMs size on runoff hydrological modeling outputs by using the Soil and Water Assessment Tool (SWAT). The buffering method BOS (buffer-overlay-statistics) was used to estimate the accuracy and displacement of the river network delineation. The runoff volume estimated with both DEMs at the outlet of the watershed was compared with the actual measurement. Results indicate that the DEMs were accurate enough to capture significant topographic variability for watershed and river network delineation. Although the delineation obtained with 90 m DEM was less accurate than the results obtained with 30 m DEM, this mesh size still reflects the real topographic features in the landscape of this watershed. Simulation results over a 10 year period showed that runoff predictions were overestimated during seasonal peak flows but accurate enough on the mean annual runoff volume.
