Nonpoint Source Pollution Model

LGA - a nonpoint source pollution model.
Building off a model originally completed by the University of Albany for the South Shore Estuary on Long Island, the LGA worked with the Department of State (DOS) - Division of Coastal Resources and applied a similar model to the Lake George Watershed. Environmental Protection Funds provided under title 11 were secured to contract with the State University of New York College of Environmental Science and Forestry (SUNY ESF) and the Institute for the Application of Geospatial Technology (IAGT) to create datasets for the model. The LGA then created a Nonpoint Source Pollution Model (NPSPM); we used a combination of image processing software and GIS to create a static model to show potential pollution hot spots in the Lake George Watershed, depicted at right.


  • Working with the DOS, IAGT and Lake George Watershed Coalition (LGWC), we started building a Lake George model in 2000.
  • A Landsat 7 satellite image from May of 2001 was captured and processed to derive a land cover classification. IAGT ran the classification based on locations identified by LGA throughout the watershed (Oct - 2002).
  • Streams, lakes, ponds, wetlands and the Lake George shoreline were digitized and added to the database (Jan - 2003).
  • The Lake George Watershed Coalition contracted with SUNY ESF to digitize the soil maps for the areas of the watershed located in both Warren, Washington and Essex County. The soils data were received, corrected and hydrologic soil values were added in April of 2004.
  • An accuracy assessment completed by IAGT showed the land classification to be rather low (43%). The model was run regardless, but it was decided that an improved classification was needed. Results from the first run are shown above.
  • The model shows areas with high runoff potential. Due to to the scale of the model we will not be able to identify an exact problem, but rather the model shows us locations that should garner closer scrutiny.
  • Using recommendations from IAGT in the accuracy assessment, a second classification was completed using different techniques and refining the data. This was completed in 2005 with better results (66%).
  • To improve the model itself, we have looked into utilizing LiDAR elevation data. The LiDAR data will improve the current USGS dataset tenfold and give us much better results. We have attempted to secure funding for this dataset, but to date we have been unsuccessful in obtaining funding.
  • We have also considered creating a build out analysis for the entire basin and using those results in the NPSP model to project future runoff potential with different build out scenarios (e.g. 50%, 75% or 100%). With these models, we can try to monitor and project the basin’s development, helping to ensure that Lake George remains the “Queen of American Lakes."
Search Engine Optimization Services and Website Design by Mannix Marketing, Inc.