Watershed systems provide a wide range of ecosystem services. The spatial patterns in such systems are heavily based upon attributes such as slope and the area per slope length. In terms of defining the watershed system for the development of applied ecological models, it is important to evaluate the extent of variation caused by a change in scale. The present study evaluates the impact of Digital Elevation Models (DEMs) at 10 meter and 30 meter resolutions on non-point runoff predictions for three sub-watersheds in Raritan River basin, New Jersey. These three watershed include: Bound Brook, Lamington and Lawrence Brook. ArcSWAT, an ArcGIS extension of Soil and Water Assessment Tool (SWAT) is utilized to investigate the sensitivity of scale variation in predicting monthly estimates of total ammonia (NH4), total nitrogen (NO2) and sediment transported out of reach. The DEM data was extracted for each sub-watershed from National Elevation Dataset (NED) of U.S Geological Survey. Using land use, slope and soil data for 2012, pollutant loads were calculated for each sub-basin in the watershed over a 10 year simulation period (2012-2022). Monthly average pollutant loads for three watersheds were compared for 10 meter and 30 meter DEMs.
The results show that for most watersheds average monthly run off predictions for sediment, NH4 and NO2 are consistently elevated with 10 meter DEM. The monthly sediment (tons/ha) increased from 217.3 to 238.4, 199.7 to 210.9 and 53.54 to 56.71 for Bound Brook, Lamington and Lawrence Brook respectively. The model shows an increase in NH4 (kg/ha) from 27.14 to 27.89, 9.38 to 10.45 and 8.64 to 9.03 for Bound Brook, Lamington and Lawrence Brook respectively. NO2 (kg/ha) has shown slight increased from 0.011 to 0.012 and 0.143 to 0.136 for Lamington and Lawrence Brook respectively. However, there is a slight decrease in NO2 (kg/ha) from 0.439 to 0.201 for Bound Brook. Spatial patterns of monthly average pollutant loads for each sub-basin in watersheds also show variation in distribution of pollutant runoffs. Selected results show that concentration of NH4, NO2 and sediment have a disperse spatial pattern across hydrological response units in northern and central parts of the Lawrence Brook watershed when using 30 meter DEM compared to 10 meter DEM. Overall statistical and spatial results show that ArcSWAT is sensitive to scale variation and showing efficiency in capturing pollutant runoff data when 10 DEM is used.