Water-quality characteristics of Michigan's inland lakes, 2001-10

Abstract

The U.S. Geological Survey and the Michigan Department of Environmental Quality (MDEQ) jointly monitored for selected water-quality constituents and properties of inland lakes during 2001–10 as part of Michigan's Lake Water-Quality Assessment program. During 2001–10, 866 lake basins from 729 inland lakes greater than 25 acres were monitored for baseline water-quality conditions and trophic status. This report summarizes the water-quality characteristics and trophic conditions of the monitored lakes throughout the State; the data include vertical-profile measurements, nutrient measurements at three discrete depths, Secchi-disk transparency (SDT) measurements, and chlorophyll a measurements for the spring and summer, with major ions and other chemical indicators measured during the spring at mid-depth and color during the summer from near-surface samples. In about 75 percent of inland lake deep basins (index stations), trophic characteristics were associated with oligotrophic or mesotrophic conditions; 5 percent or less were categorized as hypereutrophic, and 80 percent of hypereutrophic lakes had a maximum depth of 30 feet or less. Comparison of spring and summer measurements shows that water clarity based on SDT measurements were clearer in the spring than in the summer for 63 percent of lakes. For near-surface measurements made in spring, 97 percent of lakes can be considered phosphorus limited and less than half a percent nitrogen limited; for summer measurements, 96 percent of lakes can be considered phosphorus limited and less than half a percent nitrogen limited. Spatial patterns of major ions, alkalinity, and hardness measured in the spring at mid-depth all showed lower values in the Upper Peninsula of Michigan and a southward increase toward the southern areas of the Lower Peninsula, though the location of increase varied by constituent. A spatial analysis of the data based on U.S. Environmental Protection Agency Level III Ecoregions separated potassium, sulfate, and chloride concentrations fairly well, with a pattern of lower values in northern ecoregions trending toward higher values in southern ecoregions; lower and higher concentrations of magnesium, hardness, calcium, and alkalinity were well separated, but middle-range concentrations in central Michigan ecoregions were mixed. The highest concentrations of chloride and sodium were in the southeastern area of the Lower Peninsula. Lakes with multiple basins showed few statistically significant differences in constituent concentrations at the 95-percent confidence level among combinations of depths between basins. The most statistically significant differences were found for water temperature, with significant differences in somewhat less than half the combinations in the spring and just a few combinations in the summer. The lack of significant differences between major basins of multibasin lakes indicates that monitoring of trophic characteristics in all major basins might not be necessary for the majority of constituents in future sampling programs. Trophic characteristics based on the 2001–10 dataset were compared to trophic characteristics resulting from other Michigan sampling programs, including the volunteer Cooperative Lakes Monitoring Program coordinated by the MDEQ (measurements on 250 lakes in 2011), trophic-state predictions produced by relating existing measurements to remotely sensed data (measurements for about 3,000 lakes), and the National Lakes Assessment (NLA) statistically valid, probability-designed lakes program (measurements for 50 lakes in Michigan and about 1,100 lakes nationally). A higher percentage of oligotrophic lakes resulted when using SDT from the volunteer data and the 2001–10 dataset than when using the predicted measurements from remotely sensed data or the NLA. Comparing trophic characteristics from differently designed programs provides multiple interpretations of lake water-quality status in Michigan lakes. No directional statistically significant difference was found at the 95-percent confidence level among historical nutrients and trophic characteristics when comparing 445 lakes with historical data for 1974–84 with the 2001–10 dataset, though SDT did show statistically significant differences at the 95-percent confidence level. Depending on the primary indicator, 50–66 percent of lakes did not change trophic-status class, 13–23 percent moved towards the oligotrophic end of the TSI scale, and 20–25 percent moved a class towards the eutrophic end of the TSI scale. Increasing percentages of urban-dominant land cover in the drainage areas of lakes had a more positive correlation with chloride concentration than did increased percentages of other land-cover classes; there was also a slight correlation of urban-dominant land cover and calcium concentration. Removing data for lakes in southeastern Lower Michigan, known from previous reports to be higher in chloride, still resulted in a positive relation even though the coefficient of determination (R 2 value) decreased from 0.55 to 0.39. Dominant land-cover drainage areas were not strongly related to nutrients with respect to a linear relation, nor were lake drainage-area sizes.