Estimating the effective clearance rate and refiltration by zebra mussels, Dreissena polymorpha, in a stratified reservoir

Abstract

Summary1. The grazing impact of zebra mussels, Dreissena polymorpha Pallas, is often evaluated by applying the individual filtration rate measured in the laboratory to the field abundance and then by comparing the total volume of water filtered with the whole lake volume. Since this approach overlooks refiltration, it overestimates the grazing impact of zebra mussels. To deal with this problem, the present authors developed an in situ method for collecting faeces and pseudofaeces to measure the actual volume of water that is cleared of suspended particles by Dreissena in a unit time under a given set of environmental conditions. This is termed the effective clearance rate (ECR).2. The experiment was conducted in Hargus Lake, OH, U.S.A., a small thermally stratified reservoir, to test the effects of spatial aggregation, mussel density and the concentration of particulate inorganic matter (PIM) on the effective clearance rate of Dreissena.3. Over 40 measurements, the ECR values ranged from 15.3 to 68.6 mL ind−−1 h−−1. Much of the variation can be explained by colony form, mussel density and seston concentration. The effects of these variables were all statistically significant. The average ECR for isolated individuals was higher than that for those in clumps (40.4 versus 32.8 mL ind−−1 h−−1), which is attributed to increased refiltration in the cores of the clumps. The ECR decreased with increased zebra mussel density because of intensified competition for food particles within the group. The ECR increased with increased PIM concentration in the lake water, which may be interpreted as a result of enhanced water mixing which ultimately caused increases in both sediment resuspension and particle delivery to the mussels.4. Taking the filtration rates for a 20‐mm mussel to be between 116 and 234 mL ind−−1 h−−1, based on data from the literature, the clumped mussels under the present experimental conditions would have a refiltration ratio between 3.4 and 6.9.5. The present authors developed an areal clearance model which predicts that seston removal by the Dreissena population is limited by the particle delivery from the ambient water to the mussel bed and will reach a maximum value beyond which no further increase will occur with increased population density.6. It is concluded that the direct grazing impact of zebra mussel on phytoplankton in thermally stratified lakes is much less effective than predicted from simple filtration rate estimation.