Recent Additions | Contact Us Search: All EPA This Area

• You are here: EPA Home
• Great Lakes > Invasive Species > The Effect of Zebra Mussels on Cycling and Potential Bioavailability of PCBs: Section 6

Invasive Species

The Effect of Zebra Mussels on Cycling and Potential Bioavailability of PCBs: Case Study of Saginaw Bay
FINAL COPY

SECTION 6

6.1 CONCLUSIONS and Recommendations   

6.1.1   Conclusions

A screening-level multi-stressor aquatic ecosystem modeling framework was developed, incorporating numerous processes, interactions, and feedbacks present in the Saginaw Bay ecosystem. The model included hydrology, hydraulic transport, nutrient dynamics, multi-class phytoplankton growth and succession, zebra mussel growth and dynamics, particulate matter processing and cycling by mussels, ∑PCB loading, fate and transport of ∑PCBs, and bioaccumulation of ∑PCBs in the lower food chain. The model simulation evaluates the ∑PCB concentration in water column, sediments, and in participating organisms such as phytoplankton, zooplankton, and zebra mussels in Saginaw Bay. The results were field tested with the published data on ∑PCB concentration both in zebra mussels and water column. The match of the predictions with the available data showed that the model demonstrated a reasonably accurate representation of physical reality. Although necessary data for model calibration was lacking, the exercise was tremendously useful in identifying data deficiencies and theoretical gaps. Also, this dynamic modeling approach highlights the important feedback interactions and major impacts on PCB cycling and bioaccumulation arising from the zebra mussel invasion.

It is envisaged that the integrated screening-level model comprising of europhication and fate and transport of ∑PCBs is a preliminary tool to assess the ecosystem level impacts of zebra mussel stressor in Saginaw Bay. Better understanding of the cycling processes and the factors governing the accumulation and distribution of contaminants within zebra mussels are critical to analyzing the mussel-mediated trophic transfer. The inclusion of effects of zebra mussel stressor in cycling and bioaccumulation of contaminants is important for research and management of Great Lakes water resources.

The following conclusions can be drawn based on the model results:

• The invasion of zebra mussels has led to a re-direction of the energy and nutrients 
from the pelagic food chain to the benthic food chain through enhanced removal of particles from the water column to the sediments. This additional particle flux has resulted in an increase in ∑PCB flux to the sediments. By this, a portion of the contaminant inventory has been transferred to the sediments. Assuming a constant load of ∑PCBs to the system, the ∑PCB concentrations in the sediments were higher than those in the absence of zebra mussels.
 
• The most differential buildup of ∑PCBs in sediments was in the near shore area of the bay due to high mussel densities and location in the path of Saginaw River inflow, which is the major source of ∑PCBs to the Bay.
   
• The introduction of zebra mussels did not lead to an increased biomagnification of  ∑PCBs in lower pelagic food chain with two trophic levels; phytoplankton, herbivorous zooplankton, and carnivorous zooplankton. The shift in bioaccumulation pattern in the presence of mussels depends on various ecosystem interactions and feedbacks in the system. A key consideration in assessing the impacts of mussels on bioaccumulation was the filtering activities of mussels on different types of suspended particles. Zebra mussel filtration on blue greens algae and rejection on herbivorous zooplankton did result in incremental ∑PCB bioaccumulation in pelagic lower food chain.
   
• The high ∑PCB level in zebra mussels in segments where ∑PCB concentrations were high showed that zebra mussels closely track the environmental conditions. The highest ∑PCB concentration in zebra mussels was found near the mouth of the Saginaw River (i.e., in segment 1). This highlights the use of mussels as biomonitors of contaminants.
   
• The ∑PCB concentrations in water column and zebra mussels were influenced by the circulation pattern of the Saginaw River plume, which has resulted in a reduction in concentration for the segments away from the Saginaw River mouth.
   
• The total water column ∑PCB concentration was directly proportional to Log K_ow, whereas dissolved ∑PCB concentration was inversely proportional to Log K_ow.
   
• ∑PCB body burden of zebra mussels is a function of Log K_ow since efficiency of chemical uptake, chemical dietary assimilation efficiency, and the elimination rates are function of Log K_ow. ∑PCB concentration in zebra mussels was directly proportional to Log K_ow.
   
• The total water column ∑PCB concentration was directly proportional to external ∑PCB loadings and was not strongly dependent on zebra mussel densities, although a small decline in water column concentration occurred in presence of mussels.
   
• ∑PCB body burden of zebra mussels was directly proportional to external ∑PCB loadings and inversely proportional to mussel densities.
   
• In response to external phosphorus loadings, it was found that the total water and sediment ∑PCB concentration were directly proportional, whereas dissolved water and ∑PCB concentration in phyptoplankton were inversely proportional.
   
• Responses to changes in external phosphorus loadings, PCB loadings and zebra mussel densities were greater in near shore zone than in the open water.
   
• Sensitivity analysis of the selected model parameters indicated that Log K_ow and lipid content were quite sensitive for ∑PCB concentration in Dreissena. ∑PCB body burden of zebra mussels was directly proportional to lipid content of the mussels. ∑PCB body burden of larger, adult zebra mussels was higher than smaller, juvenile mussels.
   
• The sensitivity of model output to external loads indicated that it is important to account for different nutrient loading on ∑PCB dynamics because nutrients impact the phytoplankton biomass, which in turn controls the ∑PCB dynamics in the system.
   
• This is not a predictive model. This model is not intended to predict, but is mainly an heuristic exercise to evaluate the fate and transport of ∑PCBs with the introduction of mussels. This work helps in understanding the changes that occurred in the network of processes with the introduction of zebra mussels in the system, which is essential in elucidating the factors important for determining the fate and transport of contaminants in aquatic systems. This modeling effort will help guide our understanding of bioaccumulation of contaminants in fish, which is a concern for human and ecosystem health.   

6.1.2   Recommendations 

• The combination of parameters for the bioaccumulation model is based on limited data available in literature. The screening level concentration predictions in various compartments of the model may not be unreasonable even with the present formulation of model based on limited data, but it did not allow parameter calibration. Formal calibration of model results with the required field data is needed.
   
• The cycling of contaminants to benthic feeders and the associated food chain accumulation as a result of mussel activities and enhanced deposition should be evaluated. Field data should be collected on contaminant concentration in benthic invertebrates, which feed on zebra mussel feces and pseudofeces. The coupling between the pelagic and benthic food chain might lead to funneling of contaminants into the higher trophic levels via benthic pathway.
   
• Field measurements on ∑PCB concentration in the lower food web such as in phytoplankton and zooplankton species are required in order to compare the model predictions.
   
• Additional studies on chemical accumulation and bioprocessing of different types of solids are required to account for their relevant effects on body burden of mussels and their transport in the sediments. Specifically, the processing of ∑PCBs associated with different types of solids need special attention to verify the model results.
   
• In order to assess their potential as indicators, it is necessary to monitor the concentration of ∑PCBs in waters and mussels at various times of the year and over a spectrum of contamination. 
   
• The fraction of an individual’s total mass comprised of lipid can vary among individuals over more than an order of magnitude, so a detailed study on lipid contents of species as a function of season, reproduction, and food availability is required. Additional studies on change in lipid contents of mussel with growth and reproduction should be carried out. The change in lipid weight as a function of time should be considered in the model formulation.
   
• For the present study, a regression of chemical assimilation efficiency depending on K_ow for one age group of mussels was used. The extension of this regression for other age classes may be questionable. So more studies on toxicokinetics factors of different age classes would improve the resolution of model results.
   
• The development of zebra mussel bioenergetics and bioaccumulation model was based on empirical and laboratory data. The validation of those parameters to field data should be ascertained.
   
• The model predictions were based on the actual time series of forcing functions for the 1991 calibration year. A sensitivity analysis by settling ∑PCB and phosphorus loads with ¡ 50% values of 1991, while keeping all other parameters at the base values of 1991, showed the importance of these forcing functions. So, in order to investigate the differences between and among individual years in actual contaminant conditions in the system, a complete data set on forcing functions of each year is required.  

6.1.3 Model Limitations 

A number of limitations in present model formulation that may impact the ∑PCB dynamics in the system are the following:  

• The filtration of zebra mussels on different types of particles has affected the ∑PCB accumulation in the pelagic lower food chain. A pivotal question that remains to be answered is the impact of zebra mussel on bioaccumulation in top predator fish like walleye, where concentration are not only impacted by pelagic food chain but also by benthic food chain. To answer this, a complete food web model comprising of both benthic and pelagic interactions is required. Since the introduction of mussels has resulted in elevated levels of ∑PCBs in the sediment and has increased the bioaccumulation potential in the benthic food web. Also, the feeding and burrowing activities of benthic feeders can play an important role for cycling the sediment contaminants. This may result in an increase in the bioaccumulation in the pelagic food web due to linkages of benthic species with the pelagic food web. 
   
• The developed model is capable of predicting the ∑PCB body burden of zebra mussels based on ∑PCB concentration in the water and food of the mussels. However, the model is unable to differentiate the food value for assimilation of ∑PCBs of different phytoplankton groups and zooplankton.
   
• The importance of lipid content suggests the possibility that the ∑PCB concentration in zebra mussels is directly proportional to lipid contents of the mussels. It is reported in the literature that lipid content of mussels varies with season and size, at present the model does not account for it.
   
• There is not yet a good understanding of zebra mussel food processing capabilities on different types of phytoplankton groups. Since the lipid content of phytoplankton groups differ so the inter-specific differences in phytoplankton lipid content can influence the chemical accumulation.
   
• Since the actual time series of forcing functions were available only for the 1991 calibration year, the model could not be used to investigate the differences between and among individual years. 
   
• By removing the phytoplankton from water column, zebra mussels remove the food source of microscopic zooplankton, which in turn are the food for larval and juvenile fishes and other-plankton feeding forage fish. The depletion of food sources of forage fish impacts the sport and commercial fisheries. The developed framework does not include the changes mediated by zebra mussel invasion in higher trophic levels.

 

Local Navigation

• Great Lakes Home
• Basic Information
• Interagency Task Force
• Interested Parties
• US Agencies
• States
• Canada
• Tribal Nations
• Others
• Federal Programs
• Legacy Act
• Policies & Strategies
• Monitoring & Indicators
• Ecosystems
• Toxics Reduction & Pollution Prevention
• Funding

• EPA Home
• Privacy and Security Notice
• Contact Us