Effect of Zebra Mussels, Dreissena polymorpha, on Macroinvertebrates in a Lake-outlet Stream.
Zebra mussels, Dreissena polymorpha, typically dominate the benthic fauna in lakes following their introduction (e.g., Stewart and Haynes, 1994). Increased densities of benthic macroinvertebrates, especially deposit feeders, have been found in dense zebra mussel colonies in lakes (Dermott et al., 1993; Stewart and Haynes, 1994; Wisenden and Bailey, 1995). Mussels principally affect macroinvertebrate density and biomass by altering the structure of benthic substrate (Ricciardi et al., 1997; Stewart et al., 1998). Mussel shells increase habitat space by increasing surface area (Mellina and Rasmussen, 1994) and shells also provide a complex architecture that many macroinvertebrates can use as refuge from predation (Stewart et al., 1998). Macroinvertebrates have also been shown to respond positively to the enhanced benthic organic matter (BOM) deposited by mussels in the form of feces and psuedofeces (Stewart et al., 1998).
Zebra mussels are colonizing many inland lakes in North America (Johnson and Carlton, 1996). The outflowing streams of these lakes have a high probability of being colonized as well (Horvath et al., 1996). As observed in lakes, zebra mussels should increase the surface area and complexity of substrata in lake-outlet streams. However, flowing water may prevent the accumulation of mussel feces and pseudofeces, at least in erosional zones. Consequently, the response of stream macroinvertebrate populations to zebra mussels could differ from that observed in more depositional habitats if BOM production is important.
We investigated the effect of zebra mussels on macroinvertebrate abundance and diversity on hard substrate in a lake-outlet stream by manipulating zebra mussels on artificial substrata. We predicted that zebra mussels would increase macroinvertebrate density and family richness because of changes in substrate heterogeneity alone. We separated the biotic and abiotic effects of zebra mussels by using treatments of both live mussels and dead mussel shells at four representative densities ranging from zero to high.
STUDY AREA AND METHODS
This study was conducted in Christiana Creek, a lake-outlet stream in southwestern Michigan, that was colonized by zebra mussels in 1993. Christiana Creek drains two connected lakes, Eagle and Christiana lakes, that were colonized by zebra mussels in 1991. Zebra mussel density in the stream is highest at the lake outlet ([approximately]1600[center dot][m.sup.-2]) and declines with distance downstream to [less than]1 mussel[center dot][m.sup.-2] at 2 km downstream (Horvath et al. 1996). The site of our experiment was 400 m downstream from the lake outlet where the ambient density of mussels was very low (50[center dot][m.sup.-2]) and stable from 1994-1996. The stream has low gradient near the lake outlet (0.1 m/km; determined from topographic maps), but the site could be considered an erosional zone because the streambed consisted mostly of small gravel with depositional sediments occurring only along the stream margins. Channel width was 15 m and mean depth was 86 cm. Discharge during the study was about 2 [m.sup.3][center dot][s.sup.-1] (measured 200 m upstream) and no floods occurred during the study. Water velocity rarely exceeded 0.25 m[multiplied by][s.sup.-1] at the site. The macrophyte Vallisneria americana and the macroalga Chara sp. cover of the stream bottom increased from 0% at the left bank to 100% at the right bank, thus creating a lateral gradient of macrophyte density. Seston concentration in the water column is generally 7 [[micro]gram][center dot][L.sup.-1] (Horvath and Lamberti, in press).
In our experiment, we used. a crossed factorial design with two states of mussels (live or dead) and three density levels [low (10 mussels/rock; 200[center dot][m.sup.-2]), medium (25 mussels/rock; 500[center dot][m.sup.-2]) or high (50 mussels/rock; 1000[center dot][m.sup.-2])], plus a noncrossed control (no attached mussels). Artificial substrata (called rocks) were made by pouring concrete into hemispherical molds. Rocks were dried for 2 wk and then flushed with running water for 24 h to leach any residual compounds. Surface area of rocks averaged 0.05 [+ or -] 0.004 [m.sup.2] (measured using the foil method of Lamberti et al., 1991). Zebra mussels (shell length = 13-15 mm) were collected from Christiana Creek at the outlet of Christiana Lake and kept in aquaria in the laboratory overnight. We estimated the mean surface area of a single mussel shell to be 3.4 [+ or -] 0.08 [cm.sup.2] ([+ or -]1 SE) using the foil method on a subsample of these mussels (n = 40). Half of the mussels were dried at 60 [degrees] C for 24 h, the dried tissue was removed and shell halves were glued together with cyanoacrylate glue. A middle point of the ventral surface of empty shells was glued to rocks with epoxy. Preliminary studies showed that live mussels tended to emigrate from the rocks. Therefore, live mussels were also carefully glued to rocks with epoxy such that the ventral side of only one valve was attached to the rock to allow for normal filtering activity. The epoxy was allowed to set for 1 h while mussels were kept moist. Rocks with live mussels were then kept in aerated coolers overnight and observed the next day to ensure that mussels were still alive (indicated by active filtering), thus ensuring that the epoxy was nontoxic.
Rocks were transported to Christiana Creek in coolers and placed on the streambed on 14 September 1996 in a randomized block design. Blocks (n = 7) were spaced 1 m apart and arranged perpendicular to flow. A block design was used because of lateral gradients in macrophyte cover, depth and velocity across the 15-m-wide channel. The seven treatments were randomly assigned within the blocks and rocks were set 0.5 m apart within each block. The experiment was run for 28 d; rocks were checked every 3-4 d by snorkeling to ensure that live mussels were still attached and actively filtering. Rocks were collected on 12 October 1996 by placing a net (100-[[micro]meter] mesh) over the rock, bringing the rock to the surface within the net, placing the rock and any net contents into a plastic bag and transporting the sample to the laboratory. Macroinvertebrates, including newly recruited zebra mussels, were removed from the rock using a brush, forceps and a spray of 70% ethanol. All macroinvertebrates were preserved in 95% ethanol, identified to family when possible and counted. Zebra mussels originally glued to the rocks were recounted.
Statistical analysis. - Factors in our experiment were mussel state and density. At the beginning of the experiment, mussels added 170 [cm.sup.2] (34%), 85 [cm.sup.2] (17%) and 34 [cm.sup.2] (7%) to the surface area of the rock at high, medium and low densities, respectively. Some of the originally glued zebra mussels were lost during the experiment. The final number of glued mussels was significantly different among the density treatments ([F.sub.2,36] = 89.76, P 0.001; 2-way ANOVA), but did not differ between the live and dead mussel treatments ([F.sub.1,36] = 0.56, P = 0.46), indicating that mussel treatments were maintained during the experiment. At the end of the experiment, mussels added a mean area ([+ or -]SE) of 154.2[+ or -] 3.1 [cm.sup.2] ([approximately]31%), 80.4 - 1.8 [cm.sup.2] ([approximately]16%) and 31.6 - 1.5 [cm.sup.2] ([approximately]6%) to the surface area of the rock at high, medium and low densities, respectively. Differences in macroinvertebrate density, family richness and Simpson's diversity index (in Begon et al., 1990) among treatments were tested separately with a blocked two-way ANOVA. Taxa were only identified to family because all collected macroinvertebrates could not be identified to genus or species. Control rocks were not analyzed with 2-way ANOVA because they created an unbalanced design for the live-dead treatment. Nonsignificant (P [greater than] 0.05) Bartlett's test for homogeneity of variances and normal probability plots of residuals indicated that the data met the assumptions of an ANOVA, thus no data transformations were used. ANOVA was significant, therefore, treatments were compared pairwise with Tukey's HSD test. We recalculated macroinvertebrate density after including the additional surface area provided by attached mussels and analyzed density using ANCOVA with surface area as the covariate. Analyses were done using SYSTAT version 5.0 (SPSS, Inc., Chicago Ill.).
Twenty-five families of macroinvertebrates were found on the rocks. Hydridae (Hydra spp.) accounted for 64% of all macroinvertebrates collected, Trichoptera (caddisflies) represented 14% and Gastropoda (snails) 7% (Table 1). Abundance of most macroinvertebrates increased with higher zebra mussel abundance (density treatments), with the exception of Heptageniidae, Hydrachnida and rare taxa (Table 1).
A significant block effect was detected in the ANOVA only for Simpson's diversity (P 0.001). The live-dead treatment did not significantly affect total macroinvertebrate abundance ([F.sub.1,35] = 1.486, P = 0.231), Simpson's diversity ([F.sub.1,35] = 0.651, P = 0.806) or family richness ([F.sub.1,35] = 0.489, P = 0.489). No significant interactions with density were found. Therefore, we pooled live and dead treatments and tested for density differences using a one-way ANOVA with the control treatment included. Total macroinvertebrate abundance differed significantly among density treatments ([F.sub.,344] = 10.7, P [less than] 0.001; [ILLUSTRATION FOR FIGURE 1 OMITTED]) and was significantly greater at the high density than all other treatments (Tukey's HSD test). Simpson's diversity was not significantly different among densities ([F.sub.3,44] = 1.774, P = 0.185; [ILLUSTRATION FOR FIGURE 2 OMITTED]), but family richness differed significantly among densities ([F.sub.3,44] = 8.625, P [less than] 0.001; [ILLUSTRATION FOR FIGURE 2 OMITTED]). More taxa were found at higher mussel densities. The ANCOVA also indicated a significant effect of zebra mussel density on total macroinvertebrate abundance ([F.sub.3,44] = 6.341, P = 0.001), with the high-density treatment different from all other treatments.
The presence of zebra mussel shells increased macroinvertebrate abundance on artificial rocks in Christiana Creek. Macroinvertebrate density was nearly twice as high on high-density treatment rocks than on rocks lacking mussel shells. This positive response was related to physical changes in the 3-dimensional structure of the rock caused by zebra mussels (either live or dead) rather than biotic interactions. We observed no significant differences in macroinvertebrate density between rocks with live or dead zebra mussels, indicating that mussel behavior (e.g., filtration, excretion) did not affect macroinvertebrate colonization in our system.
Similar and more pronounced responses of macroinvertebrate populations to zebra mussels are reported for the Laurentian Great Lakes. ln Lake Erie macroinvertebrate abundance is 2-7 times greater in sediment cores taken under mussel druses than in samples from bare sand (Botts et al., 1996) and 2-5 times higher on tiles with zebra mussels than on tiles without zebra mussels (Stewart et al., 1998). In Lake Ontario total macroinvertebrate abundance is increased 100-900% on cobble and artificial reef substrata following invasion of zebra mussels (Stewart and Haynes, 1994).
In contrast to lake studies mentioned (Botts et al., 1996; Stewart et al., 1998), the response of macroinvertebrate populations to zebra mussels in Christiana Creek did not depend on whether the mussels were live or dead. A possible explanation for this difference is that we used much lower densities of mussels than other studies, although they were 4 to 20 times higher than background densities in Christiana Creek. Organic matter, whether actively or passively deposited, may not accumulate at low mussel densities. Differences in macroinvertebrate community composition between habitats may also contribute to responses. Filter-feeders were the most abundant macroinvertebrate in Christiana Creek, whereas deposit-feeders were more abundant in the lake studies. Filter-feeders may respond differently to BOM than deposit-feeders. Ricciardi et al. (1997) found no significant difference in the density of filter-feeders among live or dead zebra mussels.
Hydrodynamics also differ among aquatic habitats and may determine how macroinvertebrates respond to zebra mussels. As in lakes, dense populations of zebra mussels in rivers can reduce seston concentration through their filter-feeding activity (Caraco et al., 1997). It is likely that feces and pseudofeces produced by mussels in erosional zones (where hard substrates are typically found) will not accumulate as they might in depositional areas. We did not observe accumulations of feces or pseudofeces on the artificial rocks, which may also explain why live zebra mussels had no more effect on benthic macroinvertebrates than dead mussels.
The significant macroinvertebrate response to zebra mussel density observed in Christiana Creek, even after considering the added surface area of attached zebra mussels (ANCOVA analysis), suggests that surface area alone could not explain the positive macroinvertebrate response. It is likely that zebra mussels attached to rocks increased surface heterogeneity (e.g., texture, flow, etc.) compared to rocks with no mussels (see also Mellina and Rasmussen, 1994). The change in the 3-dimensional structure of the rocks could have affected macroinvertebrates in several ways. For example, some invertebrates are negatively phototaxic (e.g., triclads; Thorp and Covich, 1991) and the crevices among zebra mussel shells would provide darker habitats. Crevices also could be used by smaller macroinvertebrates, such as Chironomidae or Hydroptilidae, as refuges from predation (e.g., Charlebois and Lamberti, 1996). Greater microhabitat complexity provided by mussels may explain the increased family richness found on rocks with high densities of mussels, particularly for rare taxa (e.g., Simuliidae and Elmidae). Simpson's diversity did not differ among treatments, probably because Hydra spp. accounted for about 30-60% of the macroinvertebrates, resulting in low evenness overall.
Among aquatic habitats, it appears that benthic macroinvertebrates will respond differently to zebra mussels depending on habitat characteristics, such as native species composition and hydrodynamics. The increase in substrate heterogeneity appears to be more [TABULAR DATA FOR TABLE 1 OMITTED] important in structuring macroinvertebrate populations than biodeposits from live mussels in erosional habitats. Zebra mussels readily colonize lake-outlet streams according to a source (lake) and sink (stream) model (Horvath et al., 1996), but may never attain high densities as in lakes. Our results suggest that mussels have the potential to affect, and even increase, benthic macroinvertebrate communities in these invaded ecosystems, especially if mussels increase substrate heterogeneity, but do not dominate habitat and food resources.
Acknowledgments. - This study was supported by the US Environmental Protection Agency (CR 820290-01 and CR 820290-02) and by the National Oceanic and Atmospheric Administration through Purdue University (Purdue subgrant No. 643-1500-01 under NOAA Grant No. NA46RG0419-2 to the Illinois-Indiana Sea Grant Program). We thank David Strayer and Pam Silver Botts for valuable comments on the manuscript, Steve Beaty for help with constructing the artificial rocks, Debra Hasfurther for assistance with a pilot study and William Perry and David Lodge for useful discussions leading to this project.
BEGON, M., J. L. HARPER AND C. R. COLINS. 1990. Ecology: individuals, populations, and communities. Blackwell Scientific Publications, Inc. Cambridge, Mass. 945 p.
BOTTS, P., B. SILVER, A. PATTERSON AND D. W. SCHLOESSER. 1996. Zebra mussel effects on benthic invertebrates: physical or biotic? J. N. Am. Benthol. Soc., 15:179-184.
CARACO, N. F., J. J. COLE, P. A. RAYMOND, D. L. STRAYER, M. L. PACE, S. E.G. FINDLAY AND D. T. FISCHER. 1997. Zebra mussel invasion in a large, turbid river: phytoplankton response to increased grazing. Ecology, 78:588-602.
CHARLEBOIS, P. AND G. A. LAMBERTI. 1996. Invading crayfish in a Michigan stream: direct and indirect effects on periphyton and macroinvertebrates. J. N. Am. Benthol. Soc., 15:551-563.
DERMOTT, R., J. MITCHELL, I. MURRAY AND E. FEAR. 1993. Biomass and production of zebra mussels (Dreissena polymorpha) in shallow waters of northeastern Lake Erie, p. 399 - 413. In: T. F. Nalepa and D. W. Schloesser (eds.). Zebra mussels: biology, impacts, and control. Lewis Publishers, Boca Raton, Fl.
HORVATH, T. G., G. A. LAMBERTI, D. M. LODGE AND W. L. PERRY. 1996. Zebra mussel dispersal in lake-stream systems: source-sink dynamics? J. N. Am. Benthol. Soc., 15:564-575.
----- AND G. A. LAMBERTI. 1999. Recruitment and growth of zebra mussels (Dreissena polymorpha) in coupled lake-stream systems. Arch. Hydrobiol., 145:197-217.
JOHNSON, L. E. AND J. T. CARLTON. 1996. Post-establishment spread in large-scale invasions: dispersal mechanisms of the zebra mussel Dreissena polymorpha. Ecology, 77:1686-1690.
LAMBERTI, G. A., S. V. GREGORY, L. R. ASHKENAS, R. C. WILDMAN AND M. S. MOORE. 1991. Stream ecosystem recovery following a catastrophic debris flow. Can. J. Fish. Aquat. Sci., 48:196-208.
MELLINA, E. AND J. B. RASMUSSEN. 1994. Patterns in the distribution and abundance of zebra mussels (Dreissena polymorpha) in rivers and lakes in relation to substrate and other physicochemical factors. Can. J. Fish. Aquat. Sci., 51:1024-1036.
RICCIARDI, A., E G. WHORISKEY AND J. B. RASMUSSEN. 1997. The role of the zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrata. Can. J. Fish. Aquat. Sci., 54:2596-2608.
STEWART, T. W. AND J. M. HAYNES. 1994. Benthic macroinvertebrate communities of southwestern Lake Ontario following invasion of Dreissena. J. Great Lakes Res., 20:479-493.
-----, J. G. Miner and R. L. Lowe. 1998. Quantifying mechanisms for zebra mussel effects on benthic macroinvertebrates: organic matter production and shell-generated habitat. J. N. Am. Benthol. Soc., 17:81-94.
THORP, J. H. AND A. P. COVICH. 1991. Ecology and classification of North American freshwater invertebrates. Academic Press, San Diego, Calif. 911 p.
WISENDEN, P. A. AND R. C. BAILEY. 1995. Development of macroinvertebrate community structure associated with zebra mussel (Dreissena polymorpha) colonization of artificial substrates. Can. J. Zool., 73:1438-1443.
|Printer friendly Cite/link Email Feedback|
|Author:||Horvath, Thomas G.; Martin, Kristine M.; Lamberti, Gary A|
|Publication:||The American Midland Naturalist|
|Date:||Oct 1, 1999|
|Previous Article:||Changes in the Distribution of Freshwater Mussels (Unionidae) in the Upper Susquehanna River Basin, 1955-1965 to 1996-1997.|
|Next Article:||Distributions and Selected Aspects of the Life Histories and Habitat Associations of the Crayfishes Orconectes peruncus (Creaser, 1931) and O....|