Data for "Watershed and Streambank Erosion Modeling in a Coldwater Stream Using the GWLF-E Model"
Sediment pollution is a major cause of stream degradation throughout the United States. Quantifying this pollution and understanding erosion risk is important for decision makers, but also problematic. Data at the catchment scale is often limited by availability, cost, representativeness, and reliability, making it unrealistic to assess erosion risk over a large area. While many models estimate overland flow related soil loss, streambank erosion is often overlooked and can be the dominant source of sediment in a catchment. We used the Enhanced Generalized Watershed Loading Functions (GWLF-E) model and MapShed plugin for MapWindow GIS to simulate the water budget, field erosion from the landscape, and streambank erosion in 20 subbasins along a gradient of agricultural to urban land cover from 1997-2015 in the Indian Mill Creek watershed of Michigan, USA. Annual water budget results suggest the creek is primarily groundwater fed, but that a per-subbasin average of 6% to 15% of precipitation becomes runoff. Field erosion contributed a per-subbasin average of 0.5 to 2.5 Mg ha-1 yr-1 of sediment, while streambank erosion accounted for 0.2% to 50.1% of the subbasins’ total sediment yields. Average lateral erosion rate of streambanks in subbasins ranged from 0.04 to 7.37 cm yr-1, with 4 subbasins exceeding 1.0 cm yr-1. Urban areas had more streambank erosion than agricultural segments due the runoff from impervious surfaces. Our findings suggest that models that simulate both field and streambank erosion can provide valuable data to prioritize restoration programs that reduce sediment loading.