We identified how nutrients and exotic zebra mussels interact to promote harmful algal blooms (HABs) in the Great Lakes. Results show the relationship between nutrient loading, herbivore grazing, and HABs is very different in habitats invaded by zebra mussels. The presence of zebra mussels leads to an increase in the biomass of the harmful cyanobacteria Microcystis aeruginosa and the concentration of its microcystin toxin, even in the presence of low amounts of nutrients.

Why We Care
In the 1970s, the Great Lakes led the nation in nutrient control management, contaminant cleanup, ecosystem-based approaches to management, and invasive species control strategies. By the 1980s, the lakes had mostly achieved the phosphorus loading targets called for under the 1972 Great Lakes Water Quality Agreement between the U.S. and Canada and contaminant levels were decreasing or leveling off. However, based on data collected over the past few years, water quality issues have returned.

Over the past 15 years the rate of species invasion in the Great Lakes has accelerated, with substantial impacts on food webs and cycling of nutrients. The most obvious example of these changes resulted from the introduction of zebra mussels in the early 1990s. Zebra mussels have altered energy transfer and nutrient cycling in the lakes and have been identified as a primary cause of the appearance of hazardous algal blooms of the cyanobacteria Microcystis, increased depletion of oxygen (hypoxia), and decreased water clarity with resultant blooms of benthic macrophytes, such as the alga Cladophora.

In addition to the stresses associated with the zebra mussel invasion, the coastal areas of the lakes are being impacted by continuous changes in land use. These issues are now common in several areas, including: Saginaw Bay, Green Bay, and Lake Erie. Also, there has been a massive reduction in Diporeia (a benthic amphipod and critical fish food) in several of the lakes. The strategy developed to manage the lakes by reducing phosphorus loads to set levels of algal chlorophyll did not anticipate or include alteration of key processes in this ecosystem.

What We Did
Our goals were to confirm and expand on prior experimental evidence that an increase in phosphorus loading changes the effect of the zebra mussel Dreissena on M. aeruginosa biomass and to predict the consequences of changes in nutrient loading on harmful phytoplankton abundance in invaded habitats.

Specifically, we sought to:

1. experimentally determine whether the effect of Dreissena on M. aeruginosa changes direction across a gradient of phosphorus loading,
2. identify thresholds in Phosphorus loading at which the Dreissena effect changes direction,
3. understand the mechanisms underlying the complex interaction between Dreissena and M. aeruginosa,
4. determine the degree to which experimental results from inland lakes are relevant to the interaction between Dreissena and M. aeruginosa in the western basin of Lake Erie, and
5. determine the extent to which Dreissena promotion of M. aeruginosa translates into increased levels of cyanobacterial toxin levels in the Great Lakes.

What We Found
The project scientists found a reversal in the typical herbivore-phytoplankton interaction between the zebra mussel (Dreissena polymorpha) and Microcystis aeruginosa, a harmful planktonic cyanobacterium. A pair of large-scale field enclosure (mesocosm) manipulations of mussel density in consecutive years showed that when phosphorus concentrations were very low the effect of Dreissena on the biomass of M. aeruginosa was negative across the full range of sustainable mussel densities (i.e., increased grazing of M. aeruginosa by Dreissena). When the enclosures were fertilized with phosphorus there was a positive effect of Dreissena on M. aeruginosa across the same mussel gradient (i.e., more abundant M. aeruginosa). These contrasting results indicate that D. polymorpha feeds on M. aeruginosa, but that the positive effects of D. polymorpha on M. aeruginosa can be larger than the negative effects of grazing consumption.

Benefits of Our Work
We determined whether the effect of D. polymorpha on M. aeruginosa changes direction across a broad gradient of phosphorus loading, and identified critical nutrient loading thresholds that can be applied in the management of invaded habitats. We achieved a better understanding of the mechanisms underlying herbivore-nutrient-phytoplankton interactions. We directly quantified toxin levels and coupled them to genetic characterizations of Microcystis, furthering our ability to predict when and where toxic blooms are likely to occur in Dreissena-infested habitats. This research helped explain why Microcystis aeruginosa appears to be uniquely responsive to Dreissena invasion relative to other phytoplankton species.

The results provide strong evidence that Dreissena invasions cause an increase in microcystin toxin concentration in lakes with low to moderate nutrients. An increase in M. aeruginosa biomass may negatively impact food webs and public health because microcystins are known to be toxic to aquatic and terrestrial organisms. The generally accepted scenario that low nutrients result in lower toxin amounts is negated. The observation that lower nutrient concentrations in invaded habitats leads to unexpected increases in microcystin toxins suggests that an exclusive management focus on nutrients levels may need to take into account Dreissena invasions.

This project was a joint research effort between Michigan State University and NOAA’s Great Lakes Research Laboratory. The project, in partnership with EPA, was part of the interagency Ecology and Oceanography of Harmful Algal Blooms (ECOHAB) program.