Diet partitioning, habitat preferences and behavioral interactions between juvenile yellow perch and round goby in nearshore areas of Lake Erie

Journal of Great Lakes Research 37 (2011) 101–110

Contents lists available at ScienceDirect

Journal of Great Lakes Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j g l r

Diet partitioning, habitat preferences and behavioral interactions between juvenile yellow perch and round goby in nearshore areas of Lake Erie Janelle M. Duncan, Caroline A. Marschner, María J. González ⁎ Department of Zoology, Miami University, Oxford, OH 45056, USA

a r t i c l e

i n f o

Article history: Received 13 March 2010 Accepted 19 October 2010 Available online 7 January 2011 Communicated by Thomas Hrabik Index words: Yellow perch Round goby Diet Habitat Behavior Competition

a b s t r a c t Ecological interactions between native and non-indigeneous species depend on interspecies dietary and habitat overlap and species-specific behavior. In the Great Lakes, the exotic round goby (Apollonia melanostoma) is very abundant in littoral areas used by the native yellow perch (Perca flavencens). We examined yellow perch-round goby interactions using multiple approaches. Field surveys analyzing dietary overlap among three size classes of yellow perch and round goby detected significant overlap only between juvenile perch (b 95 mm TL) and gobies (b60 mm TL). Laboratory experiments using juvenile stages tested for habitat preference differences (open sand, macrophytes and dreissenids) in solitary, intraspecific (2 perch) and interspecific (1 perch, 1 goby) treatments. In macrophyte and dreissenid habitats, we tested for treatment differences in fish behavior (intraspecific vs. interspecific) and yellow perch growth (solitary, intraspecific and interspecific). Round goby consistently preferred complex habitats. Yellow perch showed diurnal preference of complex habitats, but increased nocturnal use of sand in the solitary and interspecific treatments. Activity was greater in dreissenid than macrophyte habitat, but prey attacks showed the opposite trend. Activity and prey attacks were greater in the intraspecific than interspecific treatments. The trend was due to lower prey attacks executed by round goby. In macrophytes, individual yellow perch growth was lower in the intraspecific than in the solitary and interspecific treatments. In dreissenids, intraspecific and interspecific competitors equally decreased yellow perch growth. Our results suggest differences in diet, habitat preference and behavior between juvenile round goby and yellow perch may allow their coexistence in nearshore areas. © 2010 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Introduction Predicting the outcome of biotic interactions among natives and non-indigenous species can be difficult because the interaction strength may depend on their dietary (Bøhn and Amundsen, 2001) and habitat (Li and Moyle, 1999) overlap, which depends on ontogenetic diet and habitat shifts (Huckins et al., 2000). Habitat structure can also differentially influence prey capture rates (Crowder and Cooper, 1982), aggression (Breau and Grant, 2002), and territory size (Sundbaum and Näslund, 1998) of native and non-indigenous species. Here, we examine the interaction between yellow perch, Perca flavenscens and the non-indigenous round goby, Apollonia melanostoma (formerly Neogobius melanostomus; Stepien and Tumeo, 2006). Yellow perch is a common species in northern USA (Carlander, 1997) and an important component of fish communities in the Laurentian Great Lakes (Fielder and Thomas, 2006; Ohio Division of Wildlife (ODW), 2006). Yellow perch undergo ontogenetic diet shifts ⁎ Corresponding author. Tel.: +1 513 529 3189. E-mail addresses: [email protected] (J.M. Duncan), [email protected] (C.A. Marschner), [email protected] (M.J. González).

from zooplanktivory to benthivory when the young-of-year (YOY) reach 30–35 mm (Wu and Culver, 1992) and from benthivory to piscivory at about 150–200 mm (Truemper et al., 2006). Severe fluctuations in yellow perch abundance in the Great Lakes have been linked with the introduction of non-indigenous fish species such as alewife and white perch (Brandt et al., 1987; Parrish and Margraf, 1994). The round goby has successfully spread to all five Great Lakes and many tributaries since first reported in 1990 (Jude et al., 1992), reaching higher abundances in nearshore areas (Bergstrom et al., 2008). Round gobies may reduce recruitment of lake trout and smallmouth bass by consuming eggs and fry (Chotkowski and Marsden, 1999; Steinhart et al., 2004), interfere with spawning of mottled sculpin (Janssen and Jude, 2001) and compete for habitat with native darters (Jude et al., 1995; Balshine et al., 2005) and sculpins (Dubs and Corkum 1996). Round gobies can also reduce the density of benthic invertebrates (Lederer et al., 2008) consumed by benthivorous stages of yellow perch such as amphipods, dipterans and tricopterans (Tyson and Knight, 2001; Truemper et al., 2006). Bottom trawl surveys in Lake Erie indicate the co-occurrence of yellow perch and round goby (Ohio Division of Wildlife, 2006) but their interactions are not well studied. Although the yellow perch

0380-1330/$ – see front matter © 2010 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jglr.2010.11.015

102

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

population in Lake Erie has recovered during the last decade following a drastic abundance decline during the 1990s, yellow perch is not as abundant as during their peak in the late 1980s (Lake Erie Committee Great Lakes Fishery Commission, 2004). In addition, during the last decade, the overall conditions in Lake Erie have substantially changed due to drastic alterations in littoral food web structure due to the successful establishments of zebra and quagga mussels (Dreissena polymorpha, D. burgensis), the amphipod Echinogammarus ischinus, and round goby (Vanderploeg et al., 2002). Finally, ontogenetic diet shifts and the high degree of omnivory showed by yellow perch (Tyson and Knight, 2001) can make yellow perch-round goby interactions complex. For example, round goby has negatively affected several benthic fish consumed by yellow perch such as mottled sculpin and johnny darter (Jude et al., 1995; Janssen and Jude, 2001), but round goby has become a dominant diet item for the piscivorous stages of yellow perch (Truemper et al., 2006). However, relatively little is known about the potential interactions between the benthic stages of round goby and yellow perch. Round gobies and benthic stages of yellow perch have shown strong preferences for rocky substrates (Ray and Corkum, 2001; Janssen and Luebke, 2004) and macrophytes (Weaver et al., 1997; Jude et al., 1992), and observational data has shown them to overlap in both habitats. No study has quantified the degree of dietary overlap between yellow perch and round gobies collected in the same geographic area in the Great Lakes, but published diets of yellow perch (Parrish and Margraf, 1994; Tyson and Knight, 2001; Truemper et al., 2006) and round goby (Jude et al., 1995; Charlebois et al., 1997; French and Jude, 2001) suggest a size-dependent dietary overlap due to the consumption of benthic invertebrates by juvenile stages of both species. However, habitat structure and species-specific behavior may mediate interactions between juvenile yellow perch and round goby. In littoral zones, the abundance and species richness of benthic freshwater macroinvertebrates also increases with habitat complexity (Gilinsky, 1984). Furthermore, interactions among fish species are affected by habitat complexity (Werner, 1986). Fish foraging efficiency typically declines in submerged vegetation (Savino and Stein, 1982; Diehl, 1988) and in dreissenid colonies (Cobb and Watzin, 2002) as habitat structural complexity increases. However, the effect of increased habitat complexity may be species-specific. For example, the number of prey captured by European perch, bream, and roach decreased as habitat complexity increased. European perch were less affected than the other species (Diehl, 1988). Additionally, visually oriented animals, such as yellow perch, may find it more difficult to detect intruders and defend their territories in highly complex habitats. Thus, as habitat complexity increases, fish territory size, time allocated to patrolling, and aggression rate decline (Breau and Grant, 2002; Sundbaum and Näslund, 1998). The outcome of biotic interactions between yellow perch and round goby may depend on species-specific behavioral responses. Previous studies suggest that interference competition can cause intraspecific variability in YOY yellow perch growth (Post et al., 1997), and a negative relationship exists between growth and activity in yellow perch (Rennie et al., 2005). Thus, increases in yellow perch activity level, such as bouts of aggressive behavior in the presence of round goby, may reduce yellow perch growth rate (Westerberg et al., 2004). However, the intensity of behavioral responses of both species may vary between macrophyte and dreissenid habitats due to the greater vertical structural complexity provided by macrophytes. In this study, we compared diet overlap and composition of three size classes of yellow perch and round goby during summer in Hatchery Bay, western Lake Erie. In laboratory experiments, we also tested differences in habitat preference, behavior and the potential effect of round goby on yellow perch individual growth. Interspecific differences in habitat preferences were tested among open sand, macrophyte and dreissenid colonies. We predicted that both species would prefer complex habitats (dreissenids or macrophytes) to open

sand, but that yellow perch would prefer macrophytes to dreissenids, while round goby would prefer dreissenids to macrophytes. Behavioral differences between intraspecific and interspecific treatments were tested in dreissenid and macrophyte habitats. Within habitats, we predicted fish to be less active, less aggressive and closer together, and have lower prey attack rates in the interspecific than in the intraspecific treatment based on round gobies' stronger association with substrate. Among habitats, it is expected that fish would be less active, less aggressive and closer together in the macrophyte than in the dreissenid habitat due to the reduction of visual encounters provided by macrophytes' vertical structure. Finally, based on previous feeding experiments (González and Burkart, 2004; Duncan, unpublished data), we anticipated greater prey attack rates in the macrophyte than in the dreissenid habitat. Lastly, the individual growth of yellow perch was compared under solitary conditions, in the presence of a conspecific and in the presence of a round goby in both habitats. We expected greater yellow perch growth in macrophytes than in the dreissenid habitat, and that yellow perch growth would be greatest under solitary conditions. Lower yellow perch growth in the presence of round goby than a conspecific would indicate a negative ecological effect of round goby on yellow perch populations. Methods Diet composition and overlap in the field Samples for gut content analysis were collected from June 14 to July 2, 2002 and July 30 to August 16, 2002 in the western basin of Lake Erie near Hatchery Bay, South Bass Island, Ohio. We established three 10-m transects oriented perpendicular to shore, parallel to each other and 10 m apart, in two sites, representative of shallow nearshore habitats in of the western basin. The Peach Point (PP) site was steeply inclined, ranging between 1 and 3 m in depth. The substrate was rocky and widely interspersed with dreissenid colonies and macrophytes. The Perry's Monument (PM) site was less steep, ranging between 2 and 3 m in depth with a mainly sandy substrate and scattered rocks. Macrophytes were dominant at the PM site, with dreissenid colonies interspersed in the deeper, rockier areas. We collected fish only at PP in June, and at both sites in August. Fish were collected using cast nets, seines, electrofishing and hook-and-line. In general, smaller fish were collected with cast nets and seines, while larger fish were collected using electrofishing and hook-and-line. Fish were sacrificed and frozen for later analysis. In the laboratory, fish were thawed, total length (TL) was measured to the nearest millimeter, and wet mass was measured to the nearest 0.1 g. Fish stomachs were removed and stomach content volume was determined using a graduated cylinder. Prey items were counted under a dissecting microscope and identified to order and their percent contribution to total volume was calculated (Jude et al., 1995). Fish collected in PP and PM in August showed similar trends in diet composition; therefore we calculated mean diet composition (% of gut volume) for June and August. Yellow perch and round gobies undergo ontogenetic diet shifts (Wu and Culver, 1992; French and Jude, 2001); therefore, we divided the data set into three size classes based on discriminant analysis of gut contents using SAS software. Maximum possible discrimination between three size classes of fish was used to create objectively defined categories based on similarity of diet. For each size class of yellow perch and round goby, we calculated an average proportion of each prey category in the diet. We used the Schoener index (SI) based on prey abundance to compare dietary overlap (Schoener, 1970):  n  α = 1
0:5 ∑ jPxi −Pyi j i=1

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

where α is the SI value, Pxi is the proportion of the ith prey item in fish class x, and Pyi is the proportion of the ith prey item in fish class y. The SI values range between 0 (no overlap) and 1 (complete overlap), where an index of 0.6 or greater suggests some degree of competition for food between species (Wallace, 1981). For both species, we calculated SI values among size classes (nine interspecific comparisons) for both sampling dates. Site was not considered as a variable for the SI comparisons because fish was not collected at PP in June, nor were we able to collect individuals from all size classes from both sampling sites in either June or August. The SI yields only a single number for each comparison, so no standard deviation can be determined. This makes the robustness of the statistic difficult to determine and was of particular concern in this study, where sample sizes were small in comparison to other studies (Parrish and Margraf, 1994; Tyson and Knight, 2001). Therefore, to determine the precision of our dietary overlap results, bootstrapping was employed to construct confidence intervals for each SI value (Bennewitz et al., 2002). In this analysis, we used sampling with replacement to construct two bootstrap samples with the same number of observations as the original data, and a SI value was computed from that particular realization of the sampling. This process was repeated 5000 times for each comparison, and a confidence interval was constructed by removing the largest and smallest 2.5% of the bootstrapped SI produced. Laboratory experiments All fish were maintained in a flow-through system (22.1 °C±0.72 SD) using lake water and held at a constant photoperiod (12 h light/12 h dark) for 2 weeks prior to our experiments. During the acclimation period, fish were fed a mixture of amphipods (G. fasciatus) and trout mix ad libidum. Dreissenid and macrophyte densities used in all experiments were within the range observed in Hatchery Bay, Lake Erie (González and Burkart, 2004). Fish used in the laboratory experiments did not express reproductive characteristics and could not be sexed. Habitat preference experiments We conducted laboratory experiments to test for differences in habitat preference between yellow perch and round goby during both day and night periods. Two experiments tested habitat preference between low complexity (open sand) and high complexity habitats (macrophytes or dreissenids), and each experiment was conducted twice (two 24-h trials). A third experiment tested habitat preference between macrophytes and dreissenids with four 24-h trials. In each experiment, fish were assigned to one of four treatments: (i) one yellow perch (YP), (ii) one round goby (RG), (iii) two yellow perch (2YP), and (iv) one yellow perch and one round goby (YP + RG). Each treatment had three to five replicates per trial depending on availability of aquaria. A preliminary experiment showed a higher mortality in marked juvenile yellow perch. Thus, in the 2YP treatment we paired yellow perch that slightly differed in size; thus, we could recognize each individual yellow perch in the videotapes. In each experiment, we used 75-L aquaria filled with 63-μm filtered lake water and constant aeration. Aquarium sides and front were covered with black plastic sheeting. The aquarium backs were covered with white plastic sheeting, which provided contrast to monitor fish. The aquarium bottoms were filled with 0.5 kg of sieved and rinsed play sand. Two types of habitat were placed in each side of the aquarium, with each habitat making up to 50% of the total area; aquaria had either 15 live macrophytes (Vallisenaria americana) attached to galvanized nuts, six titles encrusted with live dreissenid (D. polymorpha and D. bugensis), or open sand. We used V. americana because it is the dominant macrophyte species in Hatchery Bay, Lake Erie (Stuckey and Moore, 1995) and field surveys in Hatchery Bay indicate that both yellow perch and round goby use V. americana beds

103

as habitat (Marschner personal observation). Lighting was overhead full spectrum fluorescent lighting (12 h light/12 h dark) and water temperature was 23.6 °C ± 1.13 SD. We had a limited availability of round gobies b60 mm; therefore, few round goby slightly larger than 60 (size range: 55–67 mm) were used in these experiments. Regression analyses between the proportion of time spent by round goby in a given habitat and round goby size in both round goby treatment (RG and RG + YP) showed no significant trends (r2 = 0.007–0.30; p ≥ 0.12). Fish were weighed and measured (yellow perch: 55.7 mm ± 5.3 SD; round goby: 63.7 mm ± 4.4 SD) before their placement into the experimental aquaria. No differences in length or weight were found among yellow perch (ANOVA, length: F2,98 = 1.34, p = 0.27; weight: F2,98 = 1.00, p = 0.37) or round goby (ANOVA length: F2,64 = 2.34, p = 0.11; weight: F2,64 = 1.40, p = 0.26, SAS Institute, 2003) for each experiment. Fish were fed 24 h before experimentation and no food was supplied during these experiments. In each trial, fish were released in the center of the aquarium and allowed to acclimate for 2 h before the experiment was initiated. The location of each individual was recorded every 2 h for a 24-h period (12 observations/fish/aquarium) using a video camera. Nighttime observations were made using a 25-W red bulb. For statistical analyses, we first calculated the average proportion of time each fish species spent in a given habitat at each observation period and then calculated an overall average proportion of time spent at a given habitat during the day (07:00–17:00) and night time (19:00–05:00). A constant was added to each proportion due to zeros in the data set. Since these observations are serially correlated, independence was achieved by computing the ratio between time spent in each habitat (i.e., time spent in dreissenids/time spent in sand), which was then log transformed (Aebischer et al., 1993). For each experiment, MANOVA (SAS Institute, 2003) was used to test for treatment differences in habitat preferences of yellow perch (alone vs. intraspecific vs. interspecific) and round goby (alone vs. interspecific) during the light and dark periods. A focal yellow perch from the two yellow perch treatment was randomly selected to use in treatment comparisons. Significant differences in habitat use were identified when the Wilk's λ value was less than or equal to 0.05. Behavioral observations We conducted two experiments to estimate behavioral differences between intraspecific (two yellow perch) and interspecific (one yellow perch and one round goby) treatments in two habitats (dreissenid and macrophyte). For each experiment, five consecutive 1-day trials in 20-L aquaria with two replicates per treatment (total of 10 replicates/treatment) were conducted in each habitat. We used 20 L aquaria in the behavioral observations because the scale allowed for better visual observations than 75-L aquaria during video playback. The macrophyte habitat consisted 36 plants, anchored with six galvanized nuts, evenly spaced within each aquarium. The dreissenid habitat consisted of three dreissenid-encrusted tiles evenly distributed along the center of each aquarium. Rulers were placed along the bottom, front and side of each aquarium to aid in distance calculations. Fish were randomly assigned into one of two treatments. Different fish were used for each trial and each fish was used only once during the experiment. All experimental fish had food withheld 48 h prior to experimentation. Fish were placed into experimental aquaria overnight (12 h) to acclimate; trials started the following morning. Each experimental aquarium was filmed for 30 min immediately after a food addition (30 amphipods/tank) by a video camera from above and then moved randomly among aquaria until all tanks had been filmed for that day. We randomly selected all yellow perch used in the intraspecific treatment, but for identification purposes smaller individuals were labeled YP1 and larger individuals as YP2, though the differences in

104

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

sizes between individuals were not statistically significant. Mean temperature during the experiments was 20.8 °C ± 1.1 SD. Our videotapes were analyzed to measure the following parameters: (1) the activity level of each individual, (2) the amount of time engaged in aggressive interactions, (3) the distance between individuals, and (4) the number of prey attacks. We recorded activity levels for each individual within a treatment every minute as either swimming or stationary. Aggressive interactions were counted as all behavioral observations that involved swimming toward another individual and forcing it to retreat. These interactions were scored using a scale of increasingly aggressive responses: approach, chase, and bite (Dubs and Corkum, 1996). However, all aggressive responses were compiled for analyses due to the limited number of interactions in some categories. Distance (mm) on a video monitor (ca. one quarter of the real distance) was measured from one fish to the next fish every minute, and data were log transformed to meet normality assumptions. The number of prey attacks included all attempted attacks on prey, not just those that resulted in consumption, because visual obstruction by habitat sometimes made it impossible to know whether an amphipod was swallowed. Inspection of the aquaria after each trial indicated that round gobies did not consume mussels during the experiments. The overall mean sizes of yellow perch used in the experiment were 63.5 ± 0.6SD (dreissenid habitat) and 63.1 ± 1.9SD (macrophyte habitat). The overall mean sizes of round goby used in the experiment were 63.5 ± 2.8SD (dreissenid habitat) and 62.2 ± 4.0SD (macrophyte habitat). Similar to the habitat preference experiment, we conducted a regression analysis between round goby behavioral observations and round goby mean size. We detected a significant trend only between the activity level and round goby size in the macrophyte treatment (r2 = 0.56; p ≤ 0.05). This trend was caused by the high activity level of the largest round goby (67.8 mm). This fish was considered as an outlier and it was not included in our statistical analyses of all behavioral parameters. Elimination of this round goby decreased the overall mean size to 60.2 ± 3.2SD in the macrophyte habitat. Lengths for yellow perch and round gobies did not differ between habitats (F1,78 = 1.87, p = 0.18), treatments (F1,78 = 0.33, p = 0.57) or species (F1,78 = 0.007, p = 0.98). Weights for yellow perch and round gobies also did not differ between habitats (F1,78 = 1.00, p = 0.32) or treatments (F1,78 = 0.31, p = 0.58). However, weights differed between species because round gobies were slightly heavier (2.97 g ± 0.6SD) than yellow perch (2.36 g ± 0.5SD; F1,79 = 18.2, p = 0.001) and fish were slightly heavier in the interspecific (2.7 g ± 0.6SD) than in the intraspecific (2.4 g ± 0.5SD) treatment (F1,78 = 7.44 p = 0.01). For all behavioral parameters we used a two-way ANOVA (SAS Institute, 2003) to test for differences between treatments (intra vs. interspecific) and habitats (dreissenid vs. macrophyte). We were interested on the effect of round goby presence on yellow perch attack rate, thus we conducted an additional two-way ANOVA focus only on yellow perch prey attacks. Growth We evaluated the effect of round goby on yellow perch growth rate in two 14-day laboratory experiments using dreissenid or macrophyte habitats. Experiments in each habitat were conducted separately due to space limitations in the laboratory. Experimental setup was similar to the habitat preference experiments (75-L aquaria). Water temperature was 21.8 °C± 1.6 SD. Macrophyte habitat consisted of 30 live plants attached to galvanized nuts randomly placed in each aquarium. Dreissenid habitat consisted of 12 live dreissenid-encrusted tiles randomly placed in two rows into each aquarium. For both experiments, fish were acclimated for 24 h in their experimental aquaria before initiating the experiments. We had three treatments with eight replicates per treatment: (i) one yellow perch (8 fish), (ii) two yellow

perch (n= 16 fish), and (iii) one yellow perch and one round goby (n =16 fish). In both experiments, fish were fed 60 amphipods (G. fasciatus) per day for the first 7 days. The daily ration was increased to 120 amphipods per day for the last 7 days to account for increased growth during week 1. A preliminary experiment showed that these food rations were lower than those needed to achieve maximum growth rates (2–3% of their body weight per day; Duncan, unpublished data). Since fish growth is a good index of overall fitness (Werner, 1986), fish were measured (TL, mm) and weighed (g) on the initial and final days of the experiment. We tested whether yellow perch growth (fish growth = finalmm or g − initialmm or g) differed among treatments using a one-way ANOVA and used a Tukey's honestly significant difference (HSD) test to compare treatment means (SAS Institute, 2003). A preliminary experiment showed a higher mortality in marked juvenile yellow perch. Therefore, because we could not identify individual yellow perch in the intraspecific treatment, we averaged change in length and weight of both individuals and weighted these means to control for the unequal proportional size of the total number of individuals across treatments. Yellow perch (ANOVA, length: F5,62 = 0.84, p = 0.53, mean: 60.5 mm ± 2.8SD; weight: F5,62 =1.45, p=0.22, mean: 2.23 g±0.46SD) and round gobies (t-test length: t0.05,8 = −1.12, p = 0.30, mean: 60.2 mm ± 1.6SD; weight: t0.05,8 =−0.67, p=0.52, mean: 2.6 g±0.3SD) initial lengths and weights did not differ between experiments. We did not compare growth between fish species because their intrinsic growth may differ. Results Diet composition and overlap in the field Using discriminant analysis on the diet composition of 99 yellow perch and 135 round gobies, we grouped fish into three size classes for yellow perch (b95 mm, n = 29; 95–150 mm, n = 38; and N150 mm TL, n = 32)) and round gobies (b60 mm, n = 48; 60–105 mm, n = 40; and N105 mm TL, 47). Mean diet overlap varied over time among perch and goby size classes (Fig. 1). During both sampling periods, the largest diet overlap occurred between smallest size classes of yellow perch and round gobies. However, the overlap index was only N0.6 in August (Fig. 1b). A relatively high but not biologically significant overlap was observed between medium sized yellow perch and round gobies in August (0.52, Fig. 1d). Large yellow perch displayed very little overlap with any round goby size class (Fig. 1e,f). Small yellow perch showed a more diverse diet in June than in August. Zooplankton, amphipods, ephemeropterans, and odonates represented 70% of the diet for yellow perch in June, and relatively few dipterans were consumed (Fig. 2a). In August, small yellow perch mostly consumed zooplankton and dipterans (Fig. 2b). Medium-sized yellow perch consumed primarily benthic invertebrates, but the relative proportion of different invertebrate groups varied with time. In June, 50% of consumption was comprised of ephemeropterans, odonates, and fish (Fig. 2a), while amphipods and crayfish represented 18% of the diet. In August, medium yellow perch consumed mostly amphipods, ephemeropterans and trichopterans (Fig. 2b). The diet of large yellow perch was mainly composed of ephemeropterans and fish in June (Fig. 2a), and trichopterans, crayfish, and fish in August (Fig. 2b). Similar to small yellow perch, small round gobies consumed zooplankton, amphipods, and dipterans, but ephemeropterans, tricopterans and dreissenids only represented a very small proportion of the diet (5–7%). In both months, the proportion of dreissenids in goby diets increased with each successive size class (Fig. 2). The diet compositions of medium and large round gobies were similar between June and August. Amphipods, dipterans and dreissenids represented 59–66% of the diet of medium gobies. Large round gobies consumed mainly dreissenids (55–57%), but gastropods, amphipods, ephemeropterans, odonates and tricopterans were also present in the diet (Fig. 2b).

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

105

Fig. 1. Schoener index values (± 95% confidence interval) for interspecific comparisons among yellow perch size classes, (a, b) b 95 mm (c, d) 95–150 mm (e, f) N 150 mm, and all round goby size classes (b60, 60–105 and N 105 mm) in June and August.

The high degree of diet similarity between small yellow perch and round goby in August, as revealed by the Schoener index (Fig. 1b), resulted mostly from a high level of consumption of dipterans and zooplankton by both species (Fig. 2b). Small and medium size classes of both fish also consumed amphipods and ephemeropterans. Fish, crayfish and odonata were only consumed by medium and large yellow perch, while dreissenids and snails where only found in round goby stomachs (Fig. 2). Habitat preference experiments As we predicted, yellow perch preferred more complex habitats (dreissenids or macrophytes) to sand, but in some treatments light levels affected yellow perch habitat preferences (Fig. 3). When habitat was composed of dreissenid colonies and sand, yellow perch strongly

preferred dreissenids in the presence of a conspecific or round goby during both day and night periods (Fig. 3a; MANOVA: F1,59 = 2.47, p = 0.12). However, yellow perch alone preferred the dreissenid habitat during the day and the sand habitat at night (Fig. 3a; MANOVA: fish × light level: F4,59 = 8.80, p b 0.001). When presented with the choice of macrophytes or sand habitat, yellow perch with a conspecific strongly preferred macrophytes over sand (Fig. 3b; MANOVA: F4,59 = 3.75, p = 0.01), but the habitat preference of yellow perch alone and the interspecific treatments varied with light levels (Fig. 3b; MANOVA: F4,59 = 4.35, p = 0.04). In both treatments (dreissenid vs. sand, and macrophyte vs. sand), yellow perch increased their use of the sand habitat during the night (Fig. 3b; MANOVA: fish × light level: F4,59 = 4.87, p = 0.002). We observed no effect of yellow perch or light levels on habitat preferences of round gobies. Round gobies strongly preferred

106

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

Small

Medium

Large

1.0

a

between yellow perch and round goby but only in the macrophyte habitat (Table 1; Fig. 4d).

Zoo Gas

Growth

0.8 Am Eph

0.6

Dip

Dre

Cra

0.4

Tri

Proportion of gut volume

Odo

0.2

Fish

Oth

0.0 1.0

0.8

b Am

Zoo

Discussion

0.6

Eph Cra

0.4

Dre Dip

Tri

0.2

Fish

Oth

0.0 YP

The effects of interspecific and intraspecific competitors on yellow perch growth varied between habitats. As we predicted in the dreissenid habitat, solitary yellow perch showed a greater increase in length (F2, 23 = 14.38, p b 0.001; Fig. 5a) and weight (F2, 23 = 18.44, p b 0.001; Fig. 5b) than in both treatments with competitors. However, yellow perch length and weight changes were similar in the intraspecific and interspecific treatments, and perch lost weight in both of these treatments. In the macrophyte habitat, yellow perch grew significantly less in the intraspecific treatment than in the solitary and interspecific treatments (F2, 23 = 16.47, p b 0.001; Fig. 5c). Furthermore, yellow perch lost weight only in the intraspecific treatment (F2, 23 = 7.99, p = 0.003; Fig. 5d).

RG

YP

RG

YP

RG

Species Fig. 2. Mean diet composition (% of gut volume) of three size classes of yellow perch (YP) and round goby (RG) in (a) June and (b) August. Am = Amphipoda; Eph = Ephemeroptera sp.; Cra = Crayfish; Dre = Dreissena spp.; Dip = Diptera; Gas = Gastropoda; Odo = Odonata; Oth = Others; Tri = Tricoptera; Zoo = zooplankton.

dreissenids (Fig. 3a; MANOVA: F4,59 = 15.49, p b 0.001) and macrophytes (Fig. 3b; MANOVA: F1,59 = 4.35, p = 0.04) to sand in both solitary and interspecific treatments during day and night periods. When given a choice between two complex habitats, yellow perch preferred the macrophytes in all treatments, and round goby exhibited a strong preference for dreissenids (MANOVA: F4, 59 = 15.64, p b 0.001; Fig. 3c). Habitat preference for both species was similar during light or dark periods (MANOVA: F1, 59 = 0.06, p = 0.81) and there were no habitat × light level interactions (MANOVA: F4, 59 = 2.31, p = 0.07).

Behavioral observations Activity levels were significantly higher in the dreissenid than in the macrophyte habitat, and fish were significantly more active in the intraspecific than in the interspecific treatment (Table 1; Fig. 4a). We detected no significant differences in aggression levels (Table 1; Fig. 4b). Overall, the number of prey attacks was significantly greater in the macrophyte than in the dreissenid habitat (Table 1; Fig. 4c). The intraspecific treatment had a significantly greater number of prey attacks than the interspecific treatment. However, when we excluded the round goby attacks from the analysis, the rate of prey attacks by yellow perch were similar between treatments (F1,38 = 0.0026; p = 0.95), indicating that the treatment differences were mainly due to the lower number of prey attacks executed by round goby (Table 1; Fig. 4b). Distance between fish was significantly higher in dreissenid than in the macrophytes; but we detected a significant habitat × treatment interaction in the distance between fish. The distance between both yellow perch was significantly greater than the distance

Our results showed weak negative interactions between yellow perch and round goby, which may allow coexistence of these two species in benthic habitats. Our dietary analyses indicated a low dietary overlap of yellow perch and round goby; significant overlap was detected only in one comparison (perch b95 mm TL and gobies b60 mm TL in August). Overall, juvenile yellow perch showed a strong preference for macrophytes and dreissenid habitats over sand, which was not influenced by the presence of a conspecific or round goby. Yellow perch was more active and had higher prey attack rates in the present of a conspecific or round goby. Furthermore, round goby exhibited lower activity and prey attack rate relative to yellow perch, particularly in the macrophytes. Finally, round goby presence significantly decreased yellow perch growth only in the dreissenid habitat, but the effect was comparable to that caused by the presence of a conspecific. Diet composition and overlap in the field Our dietary analyses agree with previous studies that documented zooplankton, dipterans and amphipods as important dietary items for yellow perch (Parrish and Margraf, 1994; Tyson and Knight, 2001; Truemper et al., 2006) and round goby (Jude et al., 1995; Charlebois et al., 1997; French and Jude, 2001). We observed size-dependent dietary overlap, with small size classes showing greatest overlap. However, yellow perch consumed a higher proportion of ephemeropterans, trichopterans, fish and crayfish as fish increased in size, while round gobies consumed a higher proportion of dreissenids and gastropods. Previous studies also report an increase in dreissenid consumption with increased goby size (Jude et al., 1995; Charlebois et al., 1997; French and Jude, 2001). The higher consumption of dipterans by both fish species in our study during August may be related to emergence patterns (Parrish and Margraf, 1994). There is some evidence that round gobies may cause changes in benthic invertebrate communities by decreasing the abundance of Dreissena (Lederer et al., 2008), amphipods (González and Burkart, 2004) isopods and tricopterans (Lederer et al., 2008). However, Truemper et al. (2006) suggest that a generalist feeding strategy has allowed yellow perch to adjust to changes in benthic invertebrate communities in Lake Michigan from 1984 to 2002, including those caused by round goby invasion. The wide array of prey items consumed by yellow perch and limited overlap in our study confirm a similar trend in nearshore habitats in Western Lake Erie. Habitat preference experiments As we predicted, both species showed a preference for macrophyte or dreissenid habitat over sand. However, for yellow perch night

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

Light

107

Dark

100

a

80

Dreissenid Sand

60 40 20

Proportional Habitat Use

0 100

b Macrophyte Sand

80 60 40 20 0 100

c Dreissenid Macrophyte

80 60 40 20 0 1 YP 2 YP YP RG RG (Alone) (Focal) (Inter) (Inter) (Alone)

1 YP 2 YP YP RG RG (Alone) (Focal) (Inter) (Inter) (Alone)

Treatment Fig. 3. Mean proportional habitat use during light and dark periods in the (a) dreissenid and sand, (b) macrophyte and sand, and (c) dreissenid and macrophyte experiments.

habitat preference was mediated by competitor presence and habitat structure. Yellow perch increased the use of the sand habitat during night trials when alone or in the interspecific treatment in the

Table 1 Two-way ANOVAs testing for the effect of habitat (dreissenid vs. macrophyte) and treatment (intraspecific vs. interspecific) on the following behavioral parameters: time of activity (minutes), aggressive interactions (minutes), rate of prey attacks (# attacks/ 30 min and distance (mm) including round goby. Significant differences (Habitat and Interaction df = 3,78; Treatment df = 1,78; α = 0.05) are highlighted in bold text. Behavior

Source

F ratio

P value

Time of activity

Habitat Treatment Interaction Habitat Treatment Interaction Habitats Treatment Interaction Habitat Treatment Interaction

7.35 7.23 0.03 0.26 3.64 0.09 6.30 4.71 1.63 3.73 0.47 2.89

0.0003 0.0001 0.64 0.93 0.10 0.26 0.02 0.02 0.51 0.007 0.97 0.01

Aggressive interactions

Rate of prey attacks

Distance

macrophyte-sand experiment. These trends suggest that yellow perch are able to discriminate among competitors. Furthermore, the lower fish activity observed in macrophytes than in the dreissenid habitat during our behavioral observations suggest that lower fish activity level may have triggered the use of open sand. The overall preference of yellow perch for macrophytes agrees with previous field observations (Wu and Culver, 1992; Weaver et al., 1997) and suggests that vertical structure is an important habitat feature for juvenile yellow perch. A previous study suggested that the strong preference for macrophytes by yellow perch may be linked to decreased aggressive interactions among conspecifics because greater vertical structure limits visual detection (Easton and Stamps, 1992). Unfortunately individual aggression levels in the interspecific treatment varied greatly and do not provide support to these previous observations (Fig. 4b). The greater number of prey attacks made by yellow perch utilizing the macrophyte habitat (Fig. 4c) suggests that habitat preference by yellow perch could be linked to increased food availability, (Crowder and Cooper, 1982) in addition to predator avoidance (Savino and Stein, 1982). Field observations show that round gobies prefer cobble habitats to sand regardless of time of day (Ray and Corkum, 2001). Round gobies are well adapted to hiding within crevices (Jude et al., 1992; Charlebois et al., 1997), so it is not surprising that round gobies chose dreissenids over the

108

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

30

20 15 10

6 4

0

0 Dreissenid

Macrophyte

Dreissenid

160

c

Macrophyte

d

Intra Inter

140

30

120

25

Distance (mm)

Number of prey attacks/30 min

8

2

5

35

b

10

Aggression (min)

25

Activity (min)

12

Intra-1YP Intra-2YP Inter-YP Inter-RG

a

20 15 10

100 80 60 40

5

20 0

0 Dreissenid

Macrophyte

Dreissenid

Macrophyte

Habitat Fig. 4. Behavioral observations between intraspecific (open symbols) and interspecific treatments (dark symbols) in macrophyte and dreissenid habitats: (a) mean minutes of activity (±SE), (b) minutes of aggressive interactions (± SE), (c) mean number of prey attacks (± SE) made by individuals, and (d) mean distance (± SE) between individuals.

sand. No previous study had evaluated round gobies' habitat preference towards macrophyte or dreissenid habitats. Although vertical structure may be less important for round gobies than for yellow perch, the preference of macrophytes over sand by round gobies (Fig. 3) suggests that habitat overlap between yellow perch and round gobies may occur in areas with low cobble availability and dominated by macrophytes. Similar to our study, Savino et al. (2007) reported preference for rocky habitat by round goby in the presence or absence of Eurasian ruffe during the day. However, contrary to our study, Savino et al. (2007) found that round gobies decreased use of their preferred rocky habitat during dark conditions. Differences in night habitat preference may be related to speciesspecific interactions between round goby and either ruffe or yellow perch, or differences in experimental conditions between these studies. Our experiment did not evaluate the effect of conspecifics on round goby habitat preference. However, we would expect that the effect of conspecifics may depend on the population size structure. More aggressive large round gobies could induce smaller conspecifics to leave the preferred rocky habitat and move to the less optimal sand habitat (Ray and Corkum, 2001). Furthermore, the presence of large round gobies may increase the habitat overlap between juvenile stages of round goby and yellow perch in the macrophyte habitat if the aggressive behavior of large round gobies towards small conspecifics in the dreissenid habitat causes an increased use of macrophyte beds by juvenile round gobies.

behaviors such as their sit-and-wait foraging strategy (Jude et al., 1995) and increasing movement only when prey are active (Carman et al., 2006). Similarly, round gobies have been shown to be less active compared to ruffe (Savino et al., 2007). Our results also support our predictions of greater activity levels, but lower number of prey attacks, in dreissenids than in macrophytes. These trends were mainly caused by differential behavior of yellow perch between habitats and agree with previous studies suggesting that dreissenids provide better refugia for amphipods relative to macrophytes (González and Burkart, 2004). Contrary to our prediction, yellow perch aggression rate was not greater in the intraspecific than in the interspecific treatment (Table 1, Fig. 4b) mainly due to the high variability in aggression between individuals in the intraspecific treatment in dreissenids. Previous studies have reported contradictory aggression levels among young perch. Westerberg et al. (2004) found clear aggressive behavior among intraspecific juveniles, while Post et al. (1997) found no evidence of aggressive behavior within fish groups. The lower level of aggressive interactions in our interspecific treatments could also be related to the use of juvenile fish in our experiments. Previous studies observed that larger round gobies tend to be more aggressive (Dubs and Corkum, 1996; Savino et al., 2007). Round gobies were also more aggressive towards a conspecific than towards ruffe (Savino et al., 2007) or mottled sculpins (Dubs and Corkum, 1996). In summary, our behavioral observations suggest that ecological interactions between these two species are relatively weak at the juvenile stage.

Behavioral observations Growth Time of activity and prey attack rates within habitats were greater in the intraspecific than in the interspecific treatment, supporting our predictions. This trend may be related in part to specific round goby

As we predicted, yellow perch growth was consistently lower in the presence of a conspecific than in the absence of other individuals, in

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

109

Fig. 5. Change in (a, c) length and (b, d) weight of yellow perch and round goby in the dreissenid (open symbols) and macrophyte (black symbols) habitats. Horizontal lines represent mean values. Different letters denote significant differences at the 0.05 level. Star (*) indicates weighted mean to control for unequal proportion size.

both habitats. However, our results do not support a stronger negative effect of round goby on yellow perch growth, compared to effects of a conspecific. Only in the dreissenid habitat was yellow perch growth similarly reduced in the presence of a round goby and in the presence of a conspecific. The overall lower activity and number of prey attacks displayed by round gobies compared to yellow perch in our experiment (Fig. 4a), as well as the lower level of aggression shown by both species in the interspecific compared to the intraspecific treatment (Fig. 4b), may explain the weak effect of round goby on yellow perch growth. Furthermore, the higher level of activity and lower number of prey attacks exhibited by yellow perch in the dreissenid habitat, relative to the macrophyte habitat, may explain yellow perch's lower growth in the dreissenid habitat in our experiment. Our results contrast with laboratory experiments suggesting that round goby aggressive behavior rather than a greater ability to detect and capture prey allowed round goby to outcompete native sculpins and logperch (Bergstrom and Mensinger, 2009). However, the differences in aggressive behavior between studies may be related to round goby size. The mean size of round gobies used in this experiment was larger (72 mm± 0.1SD) than those in our experiment. The effects of round goby on yellow perch growth in the dreissenid habitat suggest a potential density dependent effect of round goby on yellow perch growth in areas dominated by dreissenids and with high round goby density (Irwin et al., 2009). However, this effect is not particular to round gobies since the presence of a conspecific have a similar effect. Furthermore, diet partitioning under natural conditions and the overall strong preference of yellow perch for macrophytes also suggest that an overall weak effect of round goby on yellow perch growth.

General implications Round gobies have successfully invaded nearshore benthic habitats in the Laurentian Great Lakes, where they may have detrimental effects on some co-occurring native species. Possible negative interactions include exploitative competition for food, predation on young fish or eggs, habitat displacement, and behavioral interactions. Our field observations provide evidence against the importance of the first two interactions between yellow perch and round goby for the nearshore areas in Western Lake Erie. We observed surprisingly little dietary overlap between the two species despite their co-occurrence. Furthermore, we did not observe consumption of young yellow perch by round gobies. Our laboratory experiments suggest that in addition to the limited dietary overlap, differences in habitat preference and behavior between juvenile benthic stages of the two species may contribute to their coexistence in littoral areas of the Great Lakes. Yellow perch strongly preferred macrophyte habitat. These results reinforce the importance of macrophyte beds as a habitat for juvenile yellow perch (Weaver et al., 1997). However, an important caveat to consider is that our experiments focused on interaction between juvenile stages at a constant density and in the absence of predators, while the actual interactions among size-structured populations of native and non-native species are highly complex. For example, round gobies are potential predators, competitors, and prey for smallmouth bass depending on its life stage (Steinhart et al., 2004). Additionally, adult round gobies have been found to be superior space competitors, which may have deleterious consequences for logperch (Balshine et al., 2005) and mottled sculpins (Janssen and Jude, 2001) requiring sheltered areas for

110

J.M. Duncan et al. / Journal of Great Lakes Research 37 (2011) 101–110

spawning. On the other hand, piscivorous yellow perch consume round goby (Truemper et al., 2006). Future studies should focus on the interactions among different size classes of round gobies on yellow perch and along a fish density gradient to determine the impact on size-structured populations on these native species. Finally, our findings also support recommendations of previous studies to incorporate behavioral observations to (1) understand the mechanisms underlying biotic interactions between native and non-native species (Savino et al., 2007), and (2) help in the development of wellinformed management decisions regarding effects of invasive species on fish communities in the Great Lakes. Acknowledgements We wish to thank J. Hageman, Laboratory Manager, and M. Thomas, Assistant Manager, F.T. Stone Laboratory, for their assistance and use of the facilities. We are grateful for the help of J. A. McAuliffe, G. Gordon, K. Reider, J. Porter, E. Cunningham, J. Valente, S. Andrews, and many student workers at Stone Laboratory. L. Knoll provided laboratory assistance; Dr. Ann Rypstra provided behavioral assistance; and Dr. Michael Hughes for statistical support. Funding was provided by grants to M.J.González from Ohio Sea Grant (R/NIS-7) and Lake Erie Protection Fund (SG 234-04), and a Summer Workshop grant (Department of Zoology, Miami University). References Aebischer, N.J., Robertson, P.A., Kenward, R.E., 1993. Compositional analysis of habitat use from animal radio-tracking data. Ecology 74, 1313–1325. Balshine, S., Verma, A., Chant, V., Theysmeyer, T., 2005. Competitive interactions between round gobies and logperch. J. Great Lakes Res. 31, 68–77. Bennewitz, J., Reinsch, N., Kalm, E., 2002. Improved confidence intervals in quantitative trait loci mapping by permutation bootstrapping. Genetics 160, 1673–1686. Bergstrom, M.A., Evrard, L.M., Mensinger, A.F., 2008. Distribution, abundance, and range of the round goby, Apollina melanostoma, in the Duluth-Superior Harbor and St. Louis River Estuary, 1998-2004. J. Great Lakes Res. 34, 535–543. Bergstrom, M.A., Mensinger, A.F., 2009. Interspecific resource competition between the invasive round goby and three native species: Logperch, slimy sculpin and spoonhead sculpin. Trans. Am. Fish. Soc. 138, 1009–1017. Bøhn, T., Amundsen, P.A., 2001. The competitive edge of an invading specialist. Ecology 82, 2150–2163. Brandt, S.B., Mason, D.M., MacNeill, D.B., Coates, T., Gannon, J.E., 1987. Predation by alewives on larvae of yellow perch in Lake Ontario. Trans. Am. Fish. Soc. 116, 641–645. Breau, C., Grant, J.W.A., 2002. Manipulating territory size via vegetation structure, optimal size of area guarded by the convict cichlid (Pisces, Chichlidae). Can. J. Zool. 80, 376–380. Carlander, K.D., 1997. Handbook of freshwater fishery biology, vol. 3. Iowa State University Press, Ames. Carman, S.M., Janssen, J., Jude, D.J., Berg, M.B., 2006. Diel interactions between prey behavior and feeding in an invasive fish, the round goby, in a North American river. Freshw. Biol. 51, 742–755. Charlebois, P.M., Marsden, J.E., Goettel, R.G., Wolfe, R.K., Jude, D.J., Rudnika, S., 1997. The round goby Neogobius melanostomus (Pallas), a review of European and North American Literature. Illinois-Indiana Sea Grant and Illinois Natural History Survey. INHS Special Publication. No. 20. Chotkowski, M.A., Marsden, J.E., 1999. Round goby and mottled sculpin predation on lake Trout eggs and fry: Field predictions from laboratory experiments. J. Great Lakes Res. 25, 26–35. Cobb, S.E., Watzin, M.C., 2002. Zebra mussel colonies and yellow perch foraging: spatial complexity, refuges, and resource enhancement. J. Great Lakes Res. 28, 256–263. Crowder, L.B., Cooper, W.E., 1982. Habitat structural complexity and the interaction between bluegills and their prey. Ecology 63, 1802–1813. Diehl, S., 1988. Foraging efficiency of three freshwater fish: effects of structural complexity and light. Oikos 53, 207–214. Dubs, D.O.L., Corkum, L.D., 1996. Behavioral interactions between round gobies (Neogobius melanostomus) and mottled sculpins (Cottus bairdi). J. Great Lakes Res. 22, 838–844. Easton, P.K., Stamps, J.A., 1992. The effect of visibility on territory size and shape. Behav. Ecol. 3, 166–172. Fielder, D.G., Thomas, M.V., 2006. Fish population dynamics of Saginaw Bay, Lake Huron, 1998–2004. Michigan Department of Natural Resources, Fisheries Research Report 2083, Ann Arbor. French III, J.R.P., Jude, D.J., 2001. Diets and diet overlap of nonindigenous gobies and small benthic native fishes co-inhabiting the St. Clair River, Michigan. J. Great Lakes Res. 27, 300–311.

Gilinsky, E., 1984. The role of fish predation and spatial heterogeneity in determining benthic community structure. Ecology 65, 455–468. González, M.J., Burkart, G.A., 2004. Effects of food type, habitat, and fish predation on the relative abundance of two amphipod species. Gammarus fasciatus and Echinogammarus ischnus. J. Great Lakes Res. 30, 100–113. Huckins, C.J.F., Osenberg, C.W., Mittlebach, G.G., 2000. Species introductions and their ecological consequences: an example with congeneric sunfish. Ecol. Appl. 10, 612–625. Irwin, B.J., Rudstam, L.G., Jackson, J.R., VanDeValk, A.J., Forney, J.L., Fitzgerald, D.G., 2009. Depensatory mortality, density-dependent growth, and delayed compensation: disentangling the interplay of mortality, growth, and density during early life stages of yellow perch. Trans. Am. Fish. Soc. 138, 99–110. Janssen, J., Jude, D.J., 2001. Recruitment failure of mottled sculpin Cottus bairdi in southern Lake Michigan induced by the newly introduced round goby, Neogobius melanostomus. J. Great Lakes Res. 27, 319–328. Janssen, J., Luebke, M., 2004. Preference for rocky habitat by age-0 yellow perch and alewives. J. Great Lakes Res. 30, 93–99. Jude, D.J., Janssen, J., Crawford, G., 1995. Ecology, distribution and impact of the newly introduced round tubenose gobies on the biota of the St. Clair and Detroit rivers. In: Munawar, M., Edsall, T., Leach, J. (Eds.), In the Lake Huron ecosystem: ecology, fisheries and management. S.P.B., Academic Publishing, The Netherlands, pp. 447–460. Jude, D.J., Reider, R.H., Smith, G.R., 1992. Establishment of gobiidae in the Great Lakes basin. Can. J. Fish. Aquat. Sci. 49, 416–421. Lake Erie Committee Great Lakes Fishery Commission, 2004. Coordinated percid management strategy. 33 pp. Lederer, A.M., Janssen, J., Reed, T., Wolf, A., 2008. Impacts of the introduced round goby (Apollonia melanostoma) on dreissenids (Dreissena polymorpha and Dreissena bugensis) and on macroinvertebrate community between 2003 and 2006 in the littoral zone of Green Bay, Lake Michigan. J. Great Lakes Res. 34, 690–697. Li, H.W., Moyle, P.B., 1999. Management of introduced fishes, In: Kohler, C.C., Hulbert, W.A. (Eds.), Inland fisheries management in North America, 2nd edition. American Fisheries Society, Bethesda, Maryland, pp. 345–374. Ohio Division of Wildlife (ODW), 2006. Ohio's Lake Erie Fisheries 2005. Annual status report. Federal Aid in Fish Restoration Project F-69-P. Ohio Department of Natural Resources, Division of Wildlife, Lake Erie Fisheries Units, Fairport and Sandusky. 110 pp. Parrish, D.L., Margraf, F.J., 1994. Spatial and temporal patterns of food use by white perch and yellow perch in Lake Erie. J. Freshw. Ecol. 9, 29–35. Post, J.R., Johannes, M.R.S., McQueen, D.J., 1997. Evidence of density-dependent cohort splitting in age-0 yellow perch (Perca flavescens), potential behavioral mechanisms and population-level consequences. Can. J. Fish. Aquat. Sci. 54, 867–875. Ray, W.J., Corkum, L.D., 2001. Habitat and site affinity of the round goby. J. Great Lakes Res. 27, 329–334. Rennie, M.D., Collins, N.C., Shuter, B.J., Rajotte, J.W., Couture, P., 2005. A comparison of methods for estimating activity costs of wild fish populations: more active fish observed to grow slower. Can. J. Fish. Aquat. Sci. 62, 767–780. SAS Institute, 2003. SAS/STAT Users Guide, 9.1 Edition. SAS Institute Inc, Cary, NC. Savino, J.F., Riley, S.C., Holuszko, M.J., 2007. Activity, aggression, and habitat use of ruffe (Gymnocephalus cernuus) and round goby (Apollonia melanostoma) under laboratory conditions. J. Great Lakes Res. 33, 326–334. Savino, J.F., Stein, R.A., 1982. Predator-prey interaction between largemouth bass and bluegills as influenced by simulated submersed vegetation. Trans. Am. Fish. Soc. 111, 255–266. Schoener, T.W., 1970. Nonsynchronous spatial overlap of lizards in patchy habitats. Ecology 51, 408–418. Steinhart, G.B., Marschall, E.A., Stein, R.A., 2004. Round goby predation on smallmouth bass offspring in nests during simulated catch-and-release angling. Trans. Am. Fish. Soc. 133, 121–131. Stuckey, R.L., Moore, D.L., 1995. Return and increase in abundance of aquatic flowering plants in Put-In-Bay Harbor, Lake Erie, Ohio. Ohio J. Sci. 95, 262–266. Stepien, C.A., Tumeo, M.A., 2006. Invasion genetics of Ponto-Caspian gobies in the Great Lakes: a ‘cryptic’ species, absence of founder effects, and comparative risk analysis. Biol. Invasions 8, 61–78. Sundbaum, K., Näslund, I., 1998. Effects of woody debris on the growth and behaviour of brown trout in experimental stream channels. Can. J. Zool. 76, 56–61. Truemper, H.A., Lauer, T.E., McComish, T.S., Edgell, R.A., 2006. Response of yellow perch diet to a changing forage base in southern Lake Michigan. J. Great Lakes Res. 32, 806–816. Tyson, J.T., Knight, R.L., 2001. Response of yellow perch to changes in the benthic invertebrate community of western Lake Erie. Trans. Am. Fish. Soc. 130, 766–782. Vanderploeg, H.A., Nalepa, T.F., Jude, D.J., Mills, E.L., Holeck, K.T., Liebig, J.R., Grigorovich, I.A., Ojaveer, H., 2002. Dispersal and emerging ecological impacts of Ponto-Caspian species in the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 59, 1209–1228. Wallace, R.K., 1981. An assessment of diet-overlap indexes. Trans. Am. Fish. Soc. 110, 72–76. Weaver, M.J., Magnunson, J.J., Clayton, M.K., 1997. Distribution of littoral fishes in structural complex macrophytes. Can. J. Fish. Aquat. Sci. 54, 2277–2289. Werner, E.E., 1986. Species interactions in freshwater fish communities. In: Diamond, Case, T.J. (Eds.), Community ecology. Harper and Row, New York, New York, USA, pp. 344–358. Westerberg, M., Staffan, F., Magnhagen, C., 2004. Influence of predation risk on individual competitive ability and growth in Eurasian perch (Perca fluviatilis). Anim. Behav. 67, 273–279. Wu, L., Culver, D.A., 1992. Ontogenetic diet shifts in Lake Erie age-0 yellow perch (Perca flavescens): a size-related response to zooplankton density. Can. J. Fish. Aquat. Sci. 49, 1932–1937.