Lake Water Clarity: State Level
Using Landsat satellite imagery, we have been using procedures explained in the “Step-by-step: Mapping Water Clarity” page to measure water clarity of all lakes 8 hectares (20 acres) or larger in the state of Minnesota. This geographic scale requires the processing and classification of 19 Landsat satellite images. Thus far we have completed seven complete mappings (or censuses) of Minnesota lake water clarity for the nominal years 1975, 1985, 1990, 1995, 2000, 2005, and 2008. The classifications at approximately five-year intervals from 1975 to 2008 provide an unprecedented assessment of lakes in terms of the number of lakes and geographic and temporal extent. We currently are processing imagery to add the years 2010 and 2015 to the database.
We use the term “nominal” to reflect the fact that in most cases clear imagery for the late-summer index period (late July to mid-September) that was used for the mapping was not available for the entire state. Imagery from the previous or subsequent year(s) was used for areas lacking clear imagery in the nominal year. For example, in the 1990 census, we used imagery acquired between 1990 and 1992. Secchi depth (SD) measurements collected within a few days (usually + 3 days) of each Landsat image were obtained from the Citizen Lake Monitoring Program (CLMP), which relies on citizen volunteers who measure SD on ~900-1100 Minnesota lakes (the exact number has varied over the years) multiple times during the open-water season. CLMP is managed by the Minnesota Pollution Control Agency, which maintains the SD database.
After the models (described on previous pages) were applied, the resulting R2 values generally were in the range 0.80-0.90. Click on the map to the left to see a PDF with the seven classifications from 1975 to 2008. This work was published (Olmanson et al. 2008) in the leading remote sensing journal, Remote Sensing of Environment. Persons interested in seeing more detailed information, including temporal trends in clarity for individual lakes should go to the Lake Browser.
We have analyzed the data for temporal and geographic patterns and trends, and relationships to land use and other factors that may cause changes in lake quality. Key results (Olmanson et al., 2013) from that analysis are shown in the three figures below. From 1985 to 2005, mean water clarity at the state level remained relatively stable; 4.6% of lakes had increased clarity and 6.2% decreased. However, there are strong geographic patterns with lower clarity in the south and higher clarity in the north. Deeper lakes tend to have higher clarity and are more stable than shallow lakes and agricultural and urban land use are associated with lower clarity.
Olmanson, L. G., M. E. Bauer and P. L. Brezonik. 2008. Development and analysis of a 20-year Landsat water clarity census of Minnesota’s 10,000 lakes. Remote Sens. Environ. 112: 4086-4097.
Olmanson, L. G., P. L. Brezonik, and M. E. Bauer. 2013. Geospatial and temporal analysis of a 20-year record of Landsat-based water clarity in Minnesota’s 10,000 lakes. J. Amer. Water Resour. Assoc. 50(3): 748-761.