![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081107061628im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Final Report: Early Signs & Determinants of Biotoxins (Microcystins) in Lakes
EPA Grant Number: R827407Title: Early Signs & Determinants of Biotoxins (Microcystins) in Lakes
Investigators: Sasner, John J. , Haney, James F. , Ikawa, Miyoshi
Institution: University of New Hampshire - Main Campus
EPA Project Officer: Manty, Dale
Project Period: August 16, 1999 through August 15, 2000 (Extended to February 15, 2001)
Project Amount: $149,941
RFA: Futures: Detecting the Early Signals (1999)
Research Category: Ecological Indicators/Assessment/Restoration
Description:
Objective:Microcystins (MC) are potent cellular inhibiting protein-phosphatases, which cause hepatocyte necrosis and hemorrhaging into liver tissues of mammals and other vertebrates, including fish. In response to increasing health-related problems on a global scale, the World Health Organization has established safety guidelines for drinking water at 1.0 µg MC/L. Production of MC most often is associated with phytoplankton blooms of cyanobacteria (a.k.a. blue-green- algae) caused by nutrient enrichment in lakes around the world. In contrast, we recently have found that MC also occurs in relatively clean lakes in New Hampshire with low nutrient concentrations. Some MC-producing cyanobacteria (e.g., Microcystis aeruginosa) have a complex life history that involves overwintering in lake sediments and rising in the water column during warmer summer months. A major hypothesis of this project is that lake sediments can be useful for detecting early signs of MC and MC-producing organisms that may impact water quality. We conducted extensive field sampling and laboratory analyses on 30 lakes in New Hampshire to relate seasonal benthic MC levels to MC concentrations in the water column where biotoxins are most likely to compromise the safe use of lakewater for drinking, aquaculture, and recreational purposes. The project included field limnological parameters (temperature, light, water transparency, dissolved oxygen, conductivity, pH, redox, turbidity, Chlorophyll a), laboratory nutrient analyses (TP, NO3-N, TN), and ELISA estimates of MC levels in benthic sediments, lakewater, and phyto- and zooplankton. All field sampling was performed in triplicate from June 1999 through August 2000, and included at least two sampling dates (spring and summer) for each of the study lakes. Summary/Accomplishments (Outputs/Outcomes):
Microcystins were detected in planktonic and benthic compartments of lakes representing a broad range of trophic conditions, from eutrophic to ultra-oligotrophic mountain lakes. It is particularly noteworthy that nutrient enrichment alone did not account for the high levels of biotoxins (MC) found in certain lakes. MC concentrations in surface lake sediments (0-2 cm depth) were more than an order of magnitude greater (average 340 and maximum 2694 ng MC/L) than those found in the water column (mean 15 and maximum 114 ng MC/L). MC also was found in net phytoplankton (mean 1,798 and maximum 31,472 ng MC/g wet weight) and zooplankton (mean 431 and maximum 8981 ng MC/g wet weight) supporting previous studies in eutrophic and hypereutrophic lakes that showed the passage of MC through the food web.
One goal was to test the hypothesis that MC in the water column could be predicted from benthic MC concentrations. MC levels at the surface (0-2 cm) of sediments taken at deep-water sites (>3 m) were significantly correlated with whole lakewater (WLW) MC, as well as MC levels in the phyto- and zooplankton. Conversely, the MC concentrations at the surface (0-2 cm) of the shallow-water sediments (<3 m) did not predict MC levels in any of the lake compartments in the spring or summer. Summer benthic samples (1999) best predicted the MC content of the phytoplankton in the following spring (2000), accounting for 69 percent of the variation in net phytoplankton MC (ng MC/g wet weight). Benthic MC also was correlated with the net phytoplankton toxicity in the summer (r2 = 0.35), as well as in the spring and in the combined seasons (r2 = 0.37 to 0.38). The strongest relationship between benthic and WLW MC occurred during the summer period, when benthic MC accounted for 28 percent of the MC variation. Spring benthic MC was correlated with summer WLW MC, but was not as predictive (r2 = 0.09) as summer benthic MC. During the spring, deep benthic MC levels were correlated with MC levels in the shallow sediments (r2 = 0.43), to a lesser degree when the seasons were combined (r2 = 0.28), but not significantly in the summer. These findings, combined with the high concentrations of MC found in the sediments, support the hypothesis that benthic MC can be a useful predictor of MC in the water column. Significant linear regressions were found for spring benthic MC levels and the weight-specific MC levels in the net phyto- and zooplankton (r2 = 0.19 and 0.26). Lake concentrations of MC levels in the summer net phyto- and zooplankton (ng MC/L) also were significantly correlated with spring benthic MC (r2 = 0.17 and 0.20, respectively). Again, shallow benthic MC levels in the spring were not a significant predictor of MC levels in any of the water column compartments in the summer.
We also determined if water and sediment chemistry, light penetration, extent of euphotic sediments, benthic Microcystis, and lake morphology were related to MC levels in the lake compartments. Among the nutrients measured, total phosphorus (TP) in the water column in the spring was the best single factor predicting MC levels, accounting for 29, 34, and 31 percent of the variability in summer WLW MC, and the lake concentration of net phyto- and zooplankton MC levels (ng MC/L), respectively. Although TP was significantly correlated with other MC compartments, it was less predictive. Spring TP also was correlated with WLW MC levels in the summer (r2 = 0.25), but less so with the net phyto-, zooplankton, and benthic MC concentrations (r2 = 0.13, 0.11, and 0.01, respectively). Epilimnetic NO3-N was highly correlated with MC in the shallow sediments (r2 = 0.57), as well as with the lake concentrations of phyto- and zooplankton MC (r2 = 0.16 and 0.12, respectively). Other nutrients (TN, sediment TN, and TP) were not significantly correlated with MC content of either the water or the sediments. Acid neutralizing capacity (ANC) was positively correlated with the WLW MC (r2 = 0.12) when seasons were combined. ANC was an even better predictor of the MC concentration in the net phyto- and zooplankton (ng/g wet weight) (r2 = 0.27 and 0.26, respectively), as well as the concentrations of net phytoplankton and zooplankton MC in the lake (ng MC/L) (r2 = 0.30 and 0.28, respectively). Water transparency (Secchi disc depth [SDD]) was negatively correlated with WLW MC levels in both spring and summer seasons (r2 = 0.17 and 0.15, respectively). Although SDD was a significant predictor of MC levels in other lake compartments in the combined seasons, it explained less than 10 percent of their variability. The strongest relationship was between spring SDD and the MC concentration (ng/g wet weight) in the net phytoplankton (r2 = 0.20). The euphotic lake sediments (>1 percent of incident light intensity) were positively correlated with lake water MC levels in the summer (r2 = 0.12) and in the combined seasons (r2 = 0.10), but were not correlated with the MC levels in any other compartment. The density of Microcystis colonies in the sediments was related to MC concentrations in the WLW (r2 = 0.84) as well as in the sediments (r2 =0.27) suggesting Microcystis is an important source of benthic MC.
Lake morphometry also was a determinant of MC toxicity. Generally, the level of MC in various compartments increased with decreasing mean depth of the lake.
Single factor models were improved with multiple regressions, indicating combinations of factors characterize a lake's tendency to develop high MC levels. The prediction of summer WLW MC by spring benthic MC (r2 = 0.18) was increased by adding TP to the model (r2 = 0.43). To a lesser degree, the addition of mean depth as a second independent variable also improved the regression relating spring benthic MC to summer WLW MC (r2 = 0.35). The combined effects of sediment MC, TP, SDD, and relative depth accounted for 53 percent of the variability in summer WLW MC.
Whereas most previous field studies of MC have focused on shallow lakes with eutrophic and hypereutrophic conditions, this project included a broader range of trophic lake types. Notably, using sensitive detection methods, MC levels were found in all lakes examined. Generally, higher concentrations of MC were found in the sediments than in the water column suggesting that the benthos provides a reservoir of potential toxicity in the water column. In lakes where benthic MC becomes redistributed, MC concentrations in the lake sediments can be a useful early signal of MC in the water column, particularly when combined with other variables, such as TP, SDD, and relative depth. We collected benthic samples with a sediment corer that penetrated 20-30 cm into the lake bottom. However, most of the MC resides in the top 2 cm of material. We plan to develop a more efficient and inexpensive sampler that will facilitate the capture of benthic surface materials for MC analyses. This would simplify and allow the incorporation of benthic sampling in water quality monitoring programs. In addition to the potential threat to public health through drinking water, aquaculture, and the recreational use of lakes, there is a need to evaluate the impact of MC on the aquatic food chain, lake productivity, and biodiversity. Results from this project may be especially useful to those responsible for water safety in lakes that have not yet developed high phytoplankton concentrations or obvious blooms of MC-producing cyanobacteria. The tracking of benthic MC also may serve as an early warning signal of future MC problems in eutrophic lakes, where blooms occur infrequently.
Supplemental Keywords:cyanobacteria, Microcystis, microcystins, freshwater biotoxins, hepatotoxins, benthic sampling, ELISA, trophic status of lakes, blue-green-algal blooms, HABS, limnology, lake sediments, indicators, environmental chemistry, northeast United States, EPA Region 1. , Ecosystem Protection/Environmental Exposure & Risk, Water, Scientific Discipline, Waste, RFA, Biology, Nutrients, algal blooms, Chemistry, Hydrology, Contaminated Sediments, marine biotoxins, nutrient kinetics, water quality, aquatic ecosystem, biotoxin risk, toxic cyanobacteria, benthic nutrients, bioindicator development, early warning capabilities, macroalgal populations, benthic algae, contaminated sediment, algal growth, bloom dynamics, environmental monitoring, nutrient transport, liver cancer, nutrient cycling, microbial indicators, nutrient supply
Progress and Final Reports:
2000 Progress Report
Original Abstract