Sediment Transport Model Application using Basins SWAT for the Cayuga Creek Watershed, Niagara Falls, NY, with Specific Focus on the Impact of Land Use Change

Joseph J. Gould and Kim N. Irvine


To assist state and local watershed management with decision making regarding water quality issues in the Cayuga Creek watershed, a modeling approach was taken using the Environmental Protection Agency’s (EPA) Soil Water Assessment Tool (SWAT). To calibrate and validate the model, flow and sediment data were collected at an upstream location characterized as an agricultural region and a downstream location with a mixed urban, residential, and agricultural land use. Flow and sediment data collected for 2008 were used to calibrate the model for both sites. Flow and sediment data collected for 2007 were used to validate the model for downstream site. Due to the lack of flow data for the upstream site the model was validated primarily using sediment concentration data. Statistical analysis of the model results show that more data collection is needed to fully calibrate the model. However preliminary results provide some insight into land use impacts. Land use within the watershed was classified and mapped using airphotos, ArcGIS9.2, and the Anderson Level 2 system for conditions in 2005 and 1958. There was an increase in urban and brush areas between 1958 and 2005 and a decrease in agricultural land use. Agricultural land use has become more fragmented since 1958. Even by 1958 there was a relatively small amount of forest land. SWAT model results show that monthly mean TSS concentrations in the summer through fall for the 2005 land use might be in the 10-75 mg/L range, with concentrations being slightly higher at the downstream site than the upstream site. Using the same rainfall time series, the model results also show that sediment concentrations are slightly lower in 2005 than 1958. This probably is related to land use change, specifically the decrease in agricultural land and increase in brush land. This initial model set up has helped identify information gaps that will need to be addressed in order to more fully implement the model as a management tool.

Permanent link: