Northeastern Naturalist 434 M.L. Kuhlmann 22001166 NORTHEASTERN NATURALIST 2V3(o4l). :2433,4 N–4o5. 34 Invasion-related Change in Crayfish Density Affects a Stream Macroinvertebrate Community Mark L. Kuhlmann* Abstract - Orconectes rusticus (Rusty Crayfish) have invaded streams of the upper Susquehanna River catchment, NY, replacing native crayfish and, in some areas, increasing overall crayfish density. Crayfish are important consumers and significant agents of disturbance in aquatic communities, so the introduction and expansion of Rusty Crayfish could affect the invaded community through the change in crayfish species composition, the increase in crayfish density, or some combination of the two. Other macroinvertebrates are prey of, competitors with, or subject to disturbance by crayfish and so are likely to be affected by changes in the crayfish assemblage. I conducted experiments in stream enclosures to determine the effects of crayfish species and density on the macroinvertebrate community. Increasing crayfish density reduced macroinvertebrate density but did not significantly affect macroinvertebrate taxon richness, diversity, or community composition. At moderate densities, the effects of native O. propinquus (Clearwater Crayfish) and invasive Rusty Crayfish did not differ significantly, although experimental power to detect small differences was low. These results suggest that the invasion of Upper Susquehanna River catchment streams by Rusty Crayfish will impact the macroinvertebrate community most strongly where or when Rusty Crayfish achieve high densities. Introduction Biotic invasions are widely recognized as a significant cause of changes in biodiversity and ecosystem function, often with large economic costs (Pejchar and Mooney 2009, Simberloff et al. 2013, Strayer 2012). Understanding the ecology of biotic invasions is important both for practical reasons (e.g., control of invasive species, predicting impacts) and because it can contribute to our understanding of ecological and evolutionary processes (Gurevitch et al. 2011, Sax et al. 2007, Simberloff 2004). Crayfish are frequent and important invaders of freshwater ecosystems, where invasive crayfish often greatly reduce if not completely replace populations of native crayfish (Lodge et al. 2000, 2012). As one of the largest invertebrates in freshwater communities, crayfish, whether native or invasive, have important roles as consumers, prey, and agents of disturbance (Covich et al. 1999, Momot 1995, Nyström 2002), so changes in the composition or abundance of a catchment’s crayfish fauna could have significant community-wide impacts. Although metaanalyses conclude that invasive crayfishes often have stronger effects than native species (James et al. 2015, McCarthy et al. 2006, Twardochleb et al. 2013), very few studies have directly experimentally compared the effects of invasive and native crayfish in the field (but see Lagrue et al. 2014). *Biology Department, Hartwick College, Oneonta, NY 13820; kuhlmannm@hartwick.edu. Manuscript Editor: David Yozzo Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 435 The effects of a crayfish invasion can arise from a change in total crayfish density, the change in the crayfish species assemblage (e.g., from species-specific per-capita effects), or a combination of the two. Differentiating between these mechanisms can lead to greater understanding of how an invasion will affect a community. For example, if per capita effects of the native and invasive species are similar, the invader will cause major changes to the invaded community only at high densities, which is likely only in some places or times (Hansen et al. 2013c). I investigated hypotheses about how crayfish invasions affect macroinvertebrate communities in streams of the upper Susquehanna River catchment, NY, that have been invaded by Orconectes rusticus (Girard) (Rusty Crayfish). Rusty Crayfish is native to the Ohio River catchment but has been widely introduced to both streams and lakes across northern and eastern North America, where it often completely replaces native or previously established crayfish species (Butler and Stein 1985, Daniels 1998, Jansen et al. 2009, Olden et al. 2006, Taylor and Redmer 1996, Wilson et al. 2004). Sometime after 1969, the Rusty Crayfish was introduced to the upper Susquehanna River catchment and has since spread into many tributaries, replacing at least 2 regionally native congeners, Orconectes propinquus (Girard) (Clearwater Crayfish) and Orconectes obscurus (Hagen) (Allegheny crayfish) (Barber 2013, Kuhlmann and Hazelton 2007). I used field sampling data to test the hypothesis that crayfish density increases following invasion by the Rusty Crayfish, and field enclosure experiments to test the hypotheses that both changes in crayfish density and crayfish species composition will affect the macroinvertebrate community. Because the native species being replaced in this system, the Clearwater Crayfish, is closely related to and appears to be ecologically similar to the invading Rusty Crayfish, I predicted that the effects on the macroinvertebrate community of increasing crayfish density would be stronger than the effects of changing crayfish species. Field-site Description The study region of the upper Susquehanna River catchment, centered around Oneonta, NY (42°27'N, 75°04'W), included the Susquehanna River mainstem upstream of Bainbridge, NY, to its headwaters at Otsego Lake and the 2nd–3rd order tributaries in the region (Fig. 1). I conducted field sampling of crayfish in habitats that were common in all streams in the region, could be sampled rapidly with a standardized protocol, and were likely to contain crayfish: shallow (<0.8 m depth) stream reaches with rocky substrate. These habitat criteria include riffles and shallow runs but exclude pools, stream reaches with unconsolidated sediments, and deeper portions of the larger rivers. Experiments were conducted in Charlotte Creek, a 2nd–3rd order tributary of the Susquehanna River, adjacent to Hartwick College’s Pine Lake Environmental Campus (PLEC), West Davenport, NY (Delaware County: 42°27'N, 74°55'W). Northeastern Naturalist 436 M.L. Kuhlmann 2016 Vol. 23, No. 4 Methods Crayfish sampling To examine changes in crayfish density associated with the invasion of O. rusticus, I used a subset of data from periodic sampling of streams in the upper Susquehanna River catchment during the summers (June–August) of 1999–2015 (Kuhlmann and Hazelton 2007; M. Kuhlmann, unpubl. data). This data set includes 44 sites sampled in 1–12 years. From these, I identified those sites that were sampled in multiple years during a transition from mostly native crayfish to mostly invasive crayfish by selecting sites (n = 6) that met the following criteria: initial relative abundance of Rusty Crayfish < 50%, final relative abundance of Rusty Crayfish > 50%, and >3 sample years. Sampling procedures are described in detail in Kuhlmann and Hazelton (2007). Briefly, at each site, I sampled 3 points along each of 3–4 cross-stream transects at 10-m intervals upstream of a randomly selected starting point. Samples were haphazardly placed but stratified to include both the edge and center of the stream to a maximum water depth of 0.8 m (in most streams, summer water levels were much below this across the complete stream width). I sampled crayfish using 2 methods: Figure 1. Map of the streams of the Upper Susquehanna River catchment, showing the crayfish sampling sites (x) named in Figure 2 and the location of the enclosure experiments (Pine Lake Environmental Campus). Inset shows the location of the study area in New York. Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 437 quantitative kicknetting (when current was sufficient) and quadrat sampling (in slow-moving, shallow water). The modified kicknet (Hauer and Resh 2006), 1-m wide x 1-m tall with a ~0.5-m bag made of 3-mm mesh, was positioned across the current and held against the substrate with 2 poles by 1 person. A 2nd person turned over all rocks and stirred up the substrate in a 0.6-m2 trapezoidal sampling area upstream of the net, allowing the current to carry disturbed crayfish into the net. During quadrat sampling, I carefully searched a 1-m2 area by hand, capturing the crayfish in aquarium nets. For both methods, I also recorded any crayfish observed escaping from the sample area. In the field, I identified all captured crayfish to species based on external morphological features (R. Daniels, New York State Museum, Albany, NY, unpubl. data; Peckarsky et al. 1990). For density and relative abundance calculations, I divided escaped crayfish (seen but not captured to determine species) proportionally among species captured in the sample. Because the effectiveness of both of these methods depends heavily on sampling conditions, especially depth, current speed, and the rugosity of the substrate, the reported densities should be treated as semi-quantitative estimates of true d ensities. For each site, I examined the relationship between total crayfish density and the proportion of crayfish captured that were non-native Rusty Crayfish both graphically and with a Spearman non-parametric correlation, using yearly samples as replicates. Crayfish-density experiment I conducted a crayfish density experiment during the summers (June–August) of 2007 and 2010 in enclosures placed in Charlotte Creek. Each 0.75-m2 enclosure (0.75 m wide x 1 m long x 0.3 m tall) consisted of a clear acrylic plastic bottom and sides, a front and back of 1.27-cm diamond-mesh plastic netting to allow water flow, and a removable 0.64-cm-mesh plastic-netting top. One week prior to the start of an experimental run, I placed 6 enclosures into areas of relatively high flow in the stream. Each enclosure was placed on, and anchored to, 3 concrete blocks, either full-sized (19.5 cm high) or partition (9 cm high) blocks, selected so that the enclosure’s top was approximately at the water’s surface. I added a single layer of gravel- and flat cobble-sized rocks from the nearby stream to each enclosure, including at least 4 larger rocks to serve as crayfish shelters. The enclosures remained in the stream for a week to allow macroinvertebrate colonization by stream drift. I cleaned the mesh front and back panels twice daily while the enclosures were in the stream to maintain flow through the interiors. After the 1-week colonization period, I recorded stream current velocity inside and in the vicinity of the enclosures and water depth inside each enclosure. I took a single macroinvertebrate sample (Hess sampler, 0.086 m2) from the center of each enclosure and 3 reference samples of the natural stream community from nearby areas with similar depth and flow to conditions in the enclosures. All macroinvertebrate samples were preserved in 70% ethanol for later processin g. For each run of trials, I randomly assigned 2 enclosures to each of 3 crayfish density treatments: no crayfish, low crayfish density (3 Rusty Crayfish/enclosure = Northeastern Naturalist 438 M.L. Kuhlmann 2016 Vol. 23, No. 4 4 crayfish/m2) or high crayfish density (8 Rusty Crayfish/enclosure = 10.6 crayfish/ m2). These experimental densities fall within the range of crayfish densities measured in streams within the region (range: 0–19.3 crayfish/m2, median: 4.3 crayfish/ m2; M. Kuhlmann, unpubl. data). I did 2 runs of trials (n = 6 enclosures per run) each summer for a total of n = 8 replicates of each experimental density. Just prior to each run of trials, I collected Rusty Crayfish from Charlotte Creek in the vicinity of the enclosures where the species is allopatric. From among those specimens, I then selected for use in the trials crayfish with 2 intact chelae and carapace length (CL) >25 mm, the smallest size unable to escape through the mesh panels of the enclosures (M. Kuhlmann, unpubl. data). Crayfish were divided into size- and sex ratio-matched groups and randomly assigned to enclosures of the appropriate treatment. Crayfish were left in the enclosures for a ~3 week experimental period, after which I again measured current velocity and water depth and collected a macroinvertebrate sample from the front of each enclosure and 3 reference samples from the nearby stream. I removed all crayfish from each enclosure and compared the number recaptured to the starting number to determine the efficacy of the experimental treatment. After the first experimental period of a summer, the enclosures were emptied of rocks, cleaned, and repositioned if necessary before repeating the procedures above for the next run of trials. I processed macroinvertebrate samples in the laboratory under magnification and identified and enumerated common and diverse insect orders (Coleoptera, Ephemeroptera, Diptera, Trichoptera, Plecoptera) to family and other taxa (e.g., bivalves, isopods) to higher taxonomic levels. For each sample, I calculated the density and proportion of each taxonomic group, the total macroinvertebrate density, the number of taxa (taxon richness), and “taxon” diversity using the Shannon diversity index (H'; Southwood and Henderson 2000). I also grouped taxa into functional feeding groups, following Merritt and Cummins (2006) and Voshell (2002), and computed the density of each group. For analysis, I included the reference samples taken from the nearby stream as an experimental treatment level (“outside”) to compare the enclosures’ macroinvertebrate communities with the stream’s. I did separate analyses for the initial (start) and final (end) samples. I examined the effects of the experimental treatment on the measures of the macroinvertebrate community using fully factorial multivariate analysis of variance (MANOVA), with crayfish density and run as factors. Dependent variables were transformed when necessary (based on examination of residual plots) to minimize heteroscedasticity. When significant multivariate treatment effects were found, I conducted univariate tests of individual variables (ANOVA) followed by post-hoc means comparisons (Tukey HSD) where appropriate (Abdi and Williams 2010). To further examine the effect of crayfish treatments on macroinvertebrate com - munity composition, I used principal component analysis (PCA). PCA simplifies multivariate relationships by generating new variables (components) that are combinations of original variables and show common patterns of variation (Gotelli and Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 439 Ellison 2004). I transformed counts of each taxon into standardized Z-scores and used the Z-scores as variables in the PCA; taxa that correlate most strongly with the new components are most important in determining observed patterns. To test for differences in community composition, I compared values of the first 2 components among treatments using analysis of variance (ANOVA). Crayfish-species experiment The procedures and analyses for the crayfish species experiment were the same as the crayfish density experiment with the following changes. In each run of trials, I randomly assigned 2 enclosures each to 1 of 3 experimental treatments: no crayfish, invasive crayfish (4 Rusty Crayfish per enclosure), or native crayfish (4 Clearwater Crayfish per enclosure). I ran 2 sets of trials during the summer of 2012 for a total of n = 4 replicates per treatment; planned additional replicates were precluded by the destruction of most of the enclosures during a strong spate. I collected Clearwater Crayfish at sites on upper Charlotte Creek and Butternut Creek, both Susquehanna River tributaries, and on the upper West Branch Delaware River; the Clearwater Crayfish was sympatric with the Rusty Crayfish and Cambarus bartonii (Fabricius) (Appalachian Brook Crayfish) at these locations. Prior to use in the experiments, I housed Clearwater Crayfish in single-species groups in 76- and 57-L aquaria in the laboratory for up to 4 weeks. Because there were only 2 runs of trials of this experiment and preliminary analyses indicated that it did not explain a significant amount of variance, I did not include run as an analysis factor. For dependent variables with a statistically significant univariate treatment effect, I also tested 3 sub-hypotheses with preplanned contrasts. First, I tested whether enclosures differed from the nearby stream (outside vs. no crayfish, native crayfish, and invasive crayfish). Second, I tested the hypothesis that crayfish affect the macroinvertebrate community (no crayfish vs. native crayfish and invasive crayfish). Third, I tested the hypothesis that Clearwater and Rusty Crayfish have different effects on the macroinvertebrate community (native crayfish vs. invasive crayfish). Results Crayfish sampling Crayfish density tended to be higher when Rusty Crayfish were present in higher proportion at 5 of the 6 sites (Fig. 2), although the relationship was only statistically significant at 2 of the sites. At Schenevus Creek, in contrast, the pattern was distinctly non-linear, with higher crayfish densities at both low and high relative abundances of Rusty Crayfish compared to intermediate proportion s (Fig. 2b). Crayfish-density experiment Stream current velocity in the enclosures ranged from 0.09 m/s to 0.59 m/s (mean = 0.26 m/s) and generally declined over a summer. Current velocity inside the enclosures was reduced 18–35% (mean = 25%) compared to the adjacent stream. Two of the low crayfish and 5 of the high crayfish enclosures lost ≥1 crayfish during Northeastern Naturalist 440 M.L. Kuhlmann 2016 Vol. 23, No. 4 the experimental period (low: 0–1 crayfish lost, median = 0 lost; high: 0–3 crayfish lost, median = 1 lost), but the ranges of final crayfish densities did not overlap among treatments. Figure 2. Relationship between invasion status (proportion O. rusticus) and crayfish density at 6 sites (panels) in the upper Susquehanna River catchment. Points are labeled by the sample year. Results of Spearman’s rank correlations for each site are inset. Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 441 Table 1. Multivariate and univariate analyses of macroinvertebrate density, taxon richness, and diversity and feeding functional group densities from the crayfish-density experiment. Source: Trt = experimental treatment; levels include the 3 experimental crayfish densities in the enclosures and the outside reference samples from the nearby creek. a. MANOVA: Source Pillai’s trace F Hypothesis d.f. Error d.f. P Trt 1.45 3.39 15 54 less than 0.005 Run 2.25 10.85 15 54 less than 0.005 Trt x Run 2.02 1.50 45 100 0.048 b. Univariate ANOVAs: Dependent variable Source M.S. F d.f. P Macroinvertebrate density Trt 3,531,436.57 4.38 3 0.016 Run 7,988,900.84 9.91 3 less than 0.005 Trt x Run 1,064,478.85 1.32 9 0.288 Error 806,575.44 20 Taxon richness Trt 14.87 3.31 3 0.041 Run 42.92 9.56 3 less than 0.005 Trt x Run 8.82 1.96 9 0.100 Error 4.49 20 Diversity (Shannon H') Trt 0.44 13.07 3 less than 0.005 Run 0.84 25.25 3 less than 0.005 Trt x Run 0.12 3.44 9 0.010 Error 0.03 20 Shredder density A Trt 1.09 1.90 3 0.162 Run 1.41 2.45 3 0.093 Trt x Run 0.22 0.39 9 0.928 Error 0.57 20 Collector density A Trt 0.28 4.64 3 0.013 Run 0.84 13.71 3 less than 0.005 Trt x Run 0.07 1.11 9 0.402 Error 0.06 20 Scraper density A Trt 0.99 7.95 3 0.001 Run 0.68 5.43 3 0.007 Trt x Run 0.24 1.89 9 0.113 Error 0.13 20 Predator density A Trt 0.24 1.79 3 0.181 Run 0.89 6.53 3 0.003 Trt x Run 0.25 1.81 9 0.129 Error 0.14 20 Omnivore density A Trt 1.46 10.68 3 less than 0.005 Run 3.31 24.23 3 less than 0.005 Trt x Run 0.64 4.69 9 0.002 Error 0.14 20 ALog transformed (Y'= Log10 [Y + 1]) Northeastern Naturalist 442 M.L. Kuhlmann 2016 Vol. 23, No. 4 At the start of the experiment, the multivariate effect of the crayfish density treatments on the macroinvertebrate community variables was not statistically significant (Pillai’s Trace: F24, 24 = 1.01, P = 0.476), indicating no strong differences in the macroinvertebrate community among enclosures or between the enclosures and the stream; I did not conduct univariate tests on individual va riables. At the end of the experiment, the multivariate effect of the crayfish-density treatment on measures of the macroinvertebrate community was statistically significant (Table 1a). The experimental treatment had a statistically significant effect on macroinvertebrate density, taxon richness, taxon diversity, and the densities of the collector, scraper, and omnivore functional groups (Table 1b). However, the post-hoc means comparisons found no significant difference between the enclosure treatments (no crayfish, low density of crayfish, and high density of crayfish) for taxon richness, diversity, scraper density, and omnivore density; thus, there was no crayfish-density effect, only an enclosure effect (Figs. 3, 4). Macroinvertebrate density differed significantly between the no crayfish and high crayfish treatments, and median macroinvertebrate density declined as crayfish density increased (Fig. 3a). Collector density in the high crayfish-density treatment was significantly lower than in all other treatments (Fig 4b). Principal components analyses found significant differences in macroinvertebrate community composition only between the enclosures and the nearby stream, not among crayfish densities. At the start of the experiment, the experimental treatments separated significantly only along PC axis 2 (PC1—ANOVA: F3, 20 = 0.481, P = 0.699; PC2—ANOVA: F3, 20 = 5.209, P = 0.001); however, the only significant differences were between the reference samples and the enclosures. At the end of the experiment, the experimental treatments separated significantly on both of the first 2 PCA axes (PC1—ANOVA: F3, 20 = 4.192, P = 0.019; PC2—ANOVA: F3, 20 = 5.119, P < 0.005), but means comparisons did not find any significant differences between the crayfish-density treatments; again, the reference sa mples were significantly different from the enclosures. Crayfish-species experiment Stream current velocity during the 2012 experiment was much lower than during the crayfish-density experiment. Current velocity in the enclosures ranged from 0.05 m/s to 0.20 m/s (mean = 0.07 m/s) and was 0–50% lower (mean = 21%) in the enclosures compared to the adjacent stream. No crayfish were lost from the enclosures during this experiment. At the start of the experiment, the multivariate effect of crayfish species on the macroinvertebrate community variables was not statistically significant (Pillai’s Trace: F24, 27 = 1.314, P = 0.245), so I did not conduct univariate tests on individual variables. At the end of the experiment, the multivariate effect of the crayfish species treatment on measures of the macroinvertebrate community was statistically significant (Pillai’s Trace: F24, 27 = 2.502, P = 0.011). Univariate tests showed significant treatment effects on diversity, taxon richness, and the densities of the shredder, collector, and omnivore functional groups (Figs. 5, 6). The preplanned contrasts indicated that Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 443 Figure 3. Effects of crayfish density on macroinvertebrate (a) density, (b) taxon richness, and (c) diversity. Data are from Hess samples from n = 8 e nclosures per crayfish treatment and Charlotte Creek (Outside, n = 12) at the end of the 3-week experimental treatment period. Letters indicate treatments that are not significantly different by post-hoc means comparisons (Tukey HSD). Box plots show median (bold horizontal line), interquartile range (IQR, boxes), and range (bars) up to 1.5 x IQR past the quartile. Outliers (data points >1.5 x IQR past the quartile) are shown as open circles. Northeastern Naturalist 444 M.L. Kuhlmann 2016 Vol. 23, No. 4 Figure 4. Effects of crayfish density on the density of macroinvertebrate functional feeding groups. Data are from Hess samples from n = 8 enclosures per crayfish treatment and Charlotte Creek (Outside, n = 12) at the end of the 3-week experimental treatment period. Letters indicate treatments that are not significantly different by post-hoc means comparisons (Tukey HSD). Box plots show median (bold horizontal line), interquartile range (IQR, boxes), and range (bars) up to 1.5 x IQR past the quartile. Outliers (open circles) and extreme values (stars) indicate data points >1.5 or 3 x IQR pas t the quartile, respectively. Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 445 Figure 5. Effects of crayfish species on macroinvertebrate (a) density, (b) taxon richness, and (c) diversity. Data are from Hess samples from n = 4 enclosures per crayfish treatment and Charlotte Creek (Outside, n = 6) at the end of the 3-week experimental treatment period. The result of ANOVA of the main effect (Treatment) is shown in the upper left corner of each panel. Any contrasts with significant differences between groups are indicated with lines, connectors, and P values. Contrasts tested the effects of Enclosure (enclosure treatments vs. outside reference samples), Crayfish (no crayfish vs. O. propinquus and O. rusticus) and Species (O. propinquus vs. O. rusticus). Box plots show median (bold horizontal line), interquartile range (IQR, boxes), and range (bars) up to 1.5 x IQR past the quartile. Outliers (open circles) indicate data points >1.5 x IQR past the quartile. Northeastern Naturalist 446 M.L. Kuhlmann 2016 Vol. 23, No. 4 taxon richness and taxon diversity were significantly lower in the enclosures than the nearby stream (Fig. 5 b, c). Collector density was significantly lower in the enclosures compared to the outside (Fig. 6 b). Shredder density was also significantly different between the enclosures and the outside; however, because shredder density was 0 in 2 treatments, the pattern is difficult to interpret (Fig. 6a). Only omnivore density was significantly affected by crayfish (control vs. invasive and native) or crayfish species (invasive vs. native) (Fig. 6e). Principle components analysis found no significant differences among treatments at the start of the experiment (PC1—ANOVA: F3, 14 = 1.187, P = 0.350; PC2—ANOVA: F3, 14 = 0.593, P = 0.630). At the end of the experiment, the experimental treatments separated significantly only on the first PCA axis (PC 1— ANOVA: F3, 14 = 6.077, P = 0.007; PC2—ANOVA: F3, 14 = 2.806, P = 0.078); however, the preplanned contrasts showed that the only significant differences were between the reference samples and the enclosures. Discussion Sampling data from sites in the upper Susquehanna River catchment undergoing the transition from native to invasive crayfishes generally support the hypothesis that overall crayfish density increases as the Rusty Crayfish invades, since, as predicted, at most sites both crayfish density and the relative abundance of Rusty Crayfish increased together over time (Fig. 2). However, these data provide only weak support for the hypothesis. I found sufficient data for analysis from a relatively small number of sites, and some I sampled only a few years, which limits the power of statistical tests. In fact, at one of the sites with the best time-series of data (Schenevus, Schenevus Creek), the pattern did not match the prediction at all. My results are consistent with previous studies showing that crayfish density often but not always increases following an invasion (Hansen et al. 2013b, Lodge et al. 1986, McCarthy et al. 2006, Olsen et al. 1991, Söderbäck 1995, Wilson et al. 2004) or that invasive crayfish densities can but do not always exceed the densities of native crayfishes (Hansen et al. 2013c) In the enclosure experiments, crayfish density but not species affected the macroinvertebrate community. High densities of crayfish caused reduced macroinvertebrate densities, but did not affect diversity or taxon richness (Fig. 3). Effects of crayfish density on community composition appear to be small: one functional Figure 6. (following page). Effects of crayfish species on the density of macroinvertebrate functional feeding groups. Data are from Hess samples from n = 4 enclosures per crayfish treatment and Charlotte Creek (Outside, n = 6) at the end of the 3-week experimental treatment period. The result of ANOVA of the main effect (Treatment) is shown in an upper corner of each panel. Any contrasts with significant differences between groups are indicated with lines, connectors, and P values. Contrasts tested the effects of Enclosure (enclosure treatments vs. outside reference samples), Crayfish (O. propinquus and O. rusticus vs. no crayfish) and Species (O. propinquus vs. O. rusticus). Box plots show median (bold horizontal line), interquartile range (IQR, boxes), and range ( bars). Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 447 group, shredders, decreased in density as crayfish density increased (Fig. 4), but the PCA found no significant differences in overall community composition among the experimental density treatments. At moderate densities, native Clearwater and invasive Rusty Crayfish had similar effects on the macroinvertebrate community (Fig. 5). The only significant Figure 6. (caption on preceding page). Northeastern Naturalist 448 M.L. Kuhlmann 2016 Vol. 23, No. 4 difference between the crayfish-species treatments was in the density of one relatively uncommon functional group, the omnivores; no other measures of the macroinvertebrate community were statistically different. The lack of significant differences suggests that any species-specific differences in the effects of Rusty and Clearwater Crayfish on the macroinvertebrate community are small. These results should be interpreted with some caution since the power of this experiment was low because of the small sample size, which would be exacerbated if difference in per capita effects of the 2 species is small (e.g., compared to the difference between the no crayfish and high crayfish-density treatments in the density experiment). This interpretation is partially supported by comparing the magnitudes of differences between treatment means from the 2 experiments (Table 2). While differences for some variables are smaller for the species experiment than the density experiment, notably macroinvertebrate density, others are not, and the confidence intervals for the species experiment are mostly large. Further investigation of species-specific differences among these species is warranted. The results of my experiments are consistent with comparative field studies showing changes in other North American lake and stream macroinvertebrate communities associated with invasions by Rusty Crayfish (Hansen et al. 2013a, McCarthy et al. 2006, Rosenthal et al. 2006, Wilson et al. 2004). No other studies have experimentally compared the species-specific effects of Rusty and Table 2. Differences between treatment means and their 95% confidence intervals (CI) of crayfish density (density experiment: no crayfish vs. high density of crayfish) and crayfish species (species experiment, O. propinquus vs. O. rusticus) treatments for measures of the macroinvertebrate community in the 2 enclosure experiments. Values are from univariate post hoc means comparisons using Tukey HSD tests (density experiment) and preplanned contrasts (specie s experiment) following MANOVA. 95% CI Difference Lower Upper Experiment Treatments Dependent variable between means bound bound Density No vs. high Macroinvertebrate density 1408.71A 151.86 2665.57 Taxon richness 2.13 -0.84 5.09 Diversity (H') -0.14 -0.40 0.11 Shredder densityB 2.47 -0.70 38.81 Collector densityB 1.45A 0.10 4.37 Scraper densityB 0.41 -0.55 3.37 Predator densityB 1.24 -0.32 6.41 Omnivore densityB 0.62 -0.51 -0.81 Species O. propinquus Macroinvertebrate density 462.21 -249.83 1174.25 vs. O. rusticus Taxon richness 2.25 -1.49 5.99 Diversity (H') 0.13 -0.43 0.68 Shredder densityB 8.33 -0.17 103.71 Collector densityB 0.48 -0.29 2.16 Scraper densityB -0.19 -0.91 6.24 Predator densityB 0.45 -0.94 32.88 Omnivore densityB 25.30 A 3.37 157.49 ATreatment means significantly different at P < 0.05. BBack-transformed from Y' = Log10 (Y + 1). Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 449 Clearwater Crayfish on macroinvertebrates in the field. Laboratory experiments found that Rusty Crayfish had higher consumption rate of snails (Olsen et al. 1991), similar periphyton grazing rates (Luttenton et al. 1998), and only small differences in predation on fish eggs (Ellrott et al. 2007) compared to Clearwater Crayfish. Since the 2 species are closely related (e.g., they can hybridize; Perry et al. 2001), it is not surprising that Clearwater and Rusty Crayfish are ecologically similar and so have similar per capita effects. On the other hand, crayfish (native and non-native) are generally considered to be ecologically important as consumers and ecosystem engineers (Covich et al. 1999, Momot 1995, Nyström 2002), and numerous experiments and field surveys in other systems found changes in the macroinvertebrate community across a gradient in Rusty Crayfish density (Bobeldyk and Lamberti 2008, Charlebois and Lamberti 1996, Hansen et al. 2013a, McCarthy et al. 2006, Nilsson et al. 2012, Stewart et al. 1998), consistent with the effects in this experiment from changes in crayfish density (Fig. 2). Thus, the strongest impacts on macroinvertebrate communities from the Rusty Crayfish’s invasion of streams of the upper Susquehanna River catchment, and possibly other systems, should arise from changes in crayfish density rather than changes in the crayfish species assemblage. The effects on the macroinvertebrate community of changing Rusty Crayfish density in the experimental enclosures, although larger than the species-specific effects, were fairly modest: a decrease in overall macroinvertebrate density, strongest in the collector functional group, but no statistically-detectable changes in diversity or community composition. However, in nature, Rusty Crayfish can achieve much higher densities than used in the experiment—up to ~20/m2 in the upper Susquehanna River catchment (M. Kuhlmann, unpubl. data) compared to a maximum of 10/m2 in the experiment. At higher crayfish densities, the consequences of the invasion are likely to be stronger than those shown in this experiment. In addition, any changes in macroinvertebrate density, even without major changes in community structure, could have indirect effects on components of the community not examined in this study. For example, in lakes, the invasion of Rusty Crayfish leads to changes in the abundance or trophic position of some fish, in part because of competition for macroinvertebrate prey (Kreps et al. 2016, Wilson et al. 2004). Changes in the macroinvertebrate community could also affect leaf-litter processing, with further indirect effects on other parts of the community (Covich et al. 1999, Wallace et al. 1997). Studies of other crayfish invasions, primarily by Pacifastacus leniusculus (Dana) (Signal Crayfish), show that the relative effects of native and non-native crayfish are highly variable, depending on factors such as habitat and species pairings (Dunoyer et al. 2014; Ercoli et al. 2015a, 2015b; Jackson et al. 2014; Lagrue et al. 2014; Nyström et al. 1999). Two meta-analyses comparing native and invasive crayfish effects on aquatic communities found that, while invaders on average had stronger effects than native species, there was a great deal of variation among studies (James et al. 2015, Twardochleb et al. 2013). Thus, the impact of any particular crayfish invasion may depend both on how similar the niches of the native and Northeastern Naturalist 450 M.L. Kuhlmann 2016 Vol. 23, No. 4 non-native crayfishes are (e.g., relative per capita effects) as well as how much the invasion changes crayfish density. Acknowledgments This project would not have been possible without the field and lab help from numerous summer assistants, including S. Caldwell, R. Oliver, T. Palmer, C. Dresser, A. Nieves, and Z. German. Thanks to Hartwick College’s Pine Lake Environmental Campus for hosting this project and providing logistical support. Partial support for this research came from the Hartwick College Faculty Research Grants Program, the Pine Lake Institute, and the Hartwick College Biology Department. Literature Cited Abdi, H., and L.J. Williams. 2010. Newman-Keuls test and Tukey test. Pp. 1–11, In N. Salkind (Ed.). Encyclopedia of Research Design. Sage, Thousand Oaks, CA. Barber, A. 2013. Distribution of crayfishes in the upper Susquehanna River watershed. M.Sc. Thesis. State University of New York College at Oneonta, Oneonta, NY. 72 pp. Bobeldyk, A.M., and G.A. Lamberti. 2008. A decade after invasion: Evaluating the continuing effects of Rusty Crayfish on a Michigan river. Journal of Great Lakes Research 34:265–275. Butler, M.J., and R.A. Stein. 1985. An analysis of the mechanisms governing species replacements in crayfish. Oecologia 66:168–177. Charlebois, P.C., and G.A. Lamberti. 1996. Invading crayfish in a Michigan stream: Direct and indirect effects on periphyton and macroinvertebrates. Journal of the North American Benthological Society 15:551–563. Covich, A.P., M.A. Palmer, and T.A. Crowl. 1999. The role of benthic invertebrate species in freshwater ecosystems. Bioscience 49:119–127. Daniels, R.A. 1998. Changes in the distribution of stream-dwelling crayfishes in the Schoharie Creek system, eastern New York State. Northeastern Naturalist 5:231–248. Dunoyer, L., L. Dijoux, L. Bollache, and C. Lagrue. 2014. Effects of crayfish on leaf-litter breakdown and shredder prey: Are native and introduced species functionally redundant? Biological Invasions 16:1545–1555. Ellrott, B.J., J.E. Marsden, J.D. Fitzsimons, J.L. Jonas, and R.M. Claramunt. 2007. Effects of temperature and density on consumption of trout eggs by Orconectes propinquus and O. rusticus. Journal of Great Lakes Research 33:7–14. Ercoli, F., T.J. Ruokonen, E. Erkamo, R.I. Jones, and H. Hämäläinen. 2015a. Comparing the effects of introduced Signal Crayfish and native Noble Crayfish on the littoral invertebrate assemblages of boreal lakes. Freshwater Science 34:555–56 3. Ercoli, F., T.J. Ruokonen, S. Koistinen, R.I. Jones, and H. Hämäläinen. 2015b. The introduced Signal Crayfish and native Noble Crayfish have different effects on sublittoral macroinvertebrate assemblages in boreal lakes. Freshwater Biolo gy 60:1688–1698. Gotelli, N.J., and A.M. Ellison. 2004. A Primer of Ecological Statistics. Sinauer, Sunderland, MA. 510 pp. Gurevitch, J., G.A. Fox, G.M. Wardle, Inderjit, and D. Taub. 2011. Emergent insights from the synthesis of conceptual frameworks for biological invasions. Ecology Letters 14:407–418. Hansen, G.J.A., C.L. Hein, B.M. Roth, M.J. Vander Zanden, J.W. Gaeta, A.W. Latzka, and S.R. Carpenter. 2013a. Food-web consequences of long-term invasive crayfish control. Canadian Journal of Fisheries and Aquatic Sciences 70:1109–1122. Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 451 Hansen, G.J.A., A.R. Ives, M.J. Vander Zanden, and S.R. Carpenter. 2013b. Are rapid transitions between invasive and native species caused by alternative stable states, and does it matter? Ecology 94:2207–2219. Hansen, G.J.A., M.J. Vander Zanden, M.J. Blum, M.K. Clayton, E.F. Hain, J. Hauxwell, M. Izzo, M.S. Kornis, P.B. McIntyre, A. Mikulyuk, E. Nilsson, J.D. Olden, M. Papes, and S. Sharma. 2013c. Commonly rare and rarely common: Comparing population abundance of invasive and native aquatic species. PLoS ONE 8:e77415. Hauer, F.R., and V.H. Resh. 2006. Macroinvertebrates. Pp. 435–463, In F.R. Hauer and G.A. Lamberti (Eds.). Methods in Stream Ecology. Academic Press/Elsevier, San Diego, CA. Jackson, M.C., T. Jones, M. Milligan, D. Sheath, J. Taylor, A. Ellis, J. England, and J. Grey. 2014. Niche differentiation among invasive crayfish and their impacts on ecosystem structure and functioning. Freshwater Biology 59:1123–1135. James, J., F.M. Slater, I.P. Vaughan, K.A. Young, and J. Cable. 2015. Comparing the ecological impacts of native and invasive crayfish: Could native species’ translocation do more harm than good? Oecologia 178:309–316. Jansen, W., N. Geard, T. Mosindy, G. Olson, and M. Turner. 2009. Relative abundance and habitat association of three crayfish (Orconectes virilis, O. rusticus, and O. immunis) near an invasion front of O. rusticus, and long-term changes in their distribution in Lake of the Woods, Canada. Aquatic Invasions 4:627–649. Kreps, T.A., E.R. Larson, and D.M. Lodge. 2016. Do invasive Rusty Crayfish (Orconectes rusticus) decouple littoral and pelagic energy flows in lake food webs? Freshwater Science 35:103–113. Kuhlmann, M.L., and P.D. Hazelton. 2007. Invasion of the upper Susquehanna River watershed by Rusty Crayfish, Orconectes rusticus. Northeastern Naturalist 14:507–518. Lagrue, C., T. Podgorniak, A. Lecerf, and L. Bollache. 2014. An invasive species may be better than none: Invasive Signal and native Noble Crayfish have similar community effects. Freshwater Biology 59:1982–1995. Lodge, D.M., T.K. Kratz, and G.M. Capelli. 1986. Long-term dynamics of three crayfish species in Trout Lake, Wisconsin. Canadian Journal of Fisheries and Aquatic Sciences 43:993–998. Lodge, D.M., C.A. Taylor, D.M. Holdich, and J. Skurdal. 2000. Nonindigenous crayfishes threaten North American freshwater biodiversity: Lessons from Europe. Fisheries 25:7–20. Lodge, D.M., A. Deines, F. Gherardi, D.C.J. Yeo, T. Arcella, A.K. Baldridge, M.A. Barnes, W.L. Chadderton, J.L. Feder, C.A. Gantz, G.W. Howard, C.L. Jerde, B.W. Peters, J.A. Peters, L.W. Sargent, C.R. Turner, M.E. Wittmann, and Y. Zeng. 2012. Global introductions of crayfishes: Evaluating the impact of species invasions on ecosystem services. Annual Review of Ecology, Evolution, and Systematics 43:449–472. Luttenton, M.R., M.J. Horgan, and D.M. Lodge. 1998. Effects of three Orconectes crayfishes on epilithic microalgae: A laboratory experiment. Crustaceana 71:845–855. McCarthy, J.M., C.L. Hein, J.D. Olden, and M.J. Vander Zanden. 2006. Coupling long-term studies with meta-analysis to investigate impacts of non-native crayfish on zoobenthic communities. Freshwater Biology 51:224–235. Merritt, R.W., and K.W. Cummins. 2006. Trophic relationships of macroinvertebrates. Pp. 585–609, In F.R. Hauer and G. Lamberti (Eds.). Methods in Stream Ecology. Academic Press/Elsevier, San Diego, CA. Momot, W.T. 1995. Redefining the role of crayfish in aquatic ecosystems. Reviews in Fisheries Science 3:33–63. Northeastern Naturalist 452 M.L. Kuhlmann 2016 Vol. 23, No. 4 Nilsson, E., C.T. Solomon, K.A. Wilson, T.V. Willis, B. Larget, and M.J. Vander Zanden. 2012. Effects of an invasive crayfish on trophic relationships in north-temperate lake food webs. Freshwater Biology 57:10–23. Nyström, P. 2002. Ecology. Pp. 152–191, In D.M. Holdich (Ed.). Biology of Freshwater Crayfish. Blackwell Science, Oxford, UK. Nyström, P., C. Brönmark, and W. Granéli. 1999. Influence of an exotic and a native crayfish species on a littoral benthic community. Oikos 85:545–553. Olden, J.D., J.M. McCarthy, J.T. Maxted, W.W. Fetzer, and M.J. Vander Zanden. 2006. The rapid spread of Rusty Crayfish (Orconectes rusticus) with observations on native crayfish declines in Wisconsin (USA) over the past 130 years. Biological Invasions 8:1621–1628. Olsen, T.M., D.M. Lodge, G.M. Capelli, and R.J. Houlihan. 1991. Mechanisms of impact of an introduced crayfish (Orconectes rusticus) on littoral congeners, snails, and macrophytes. Canadian Journal of Fisheries and Aquatic Sciences 48:1853–1861. Peckarsky, B.L., P.R. Fraissinet, M.A. Penton, and D.J. Conklin Jr. 1990. Freshwater Macroinvertebrates of Northeastern North America. Cornell University Press, Ithaca, NY. 442 pp. Pejchar, L., and H.A. Mooney. 2009. Invasive species, ecosystem services, and human wellbeing. Trends in Ecology and Evolution 24:497–504. Perry, W.L., J.L. Feder, G. Dwyer, and D.M. Lodge. 2001. Hybrid-zone dynamics and species replacement between Orconectes crayfishes in a northern Wisconsin lake. Evolution 55:1153–1166. Rosenthal, S.K., S.S. Stevens, and D.M. Lodge. 2006. Whole-lake effects of invasive crayfish (Orconectes spp.) and the potential for restoration. Canadian Journal of Fisheries and Aquatic Sciences 63:1276. Sax, D.F., J.J. Stachowicz, J.H. Brown, J.F. Bruno, M.N. Dawson, S.D. Gaines, R.K. Grosberg, A. Hastings, R.D. Holt, M.M. Mayfield, M.I. O’Connor, and W.R. Rice. 2007. Ecological and evolutionary insights from species invasions. Trends in Ecology and Evolution 22:465–471. Simberloff, D. 2004. Community ecology: Is it time to move on? American Naturalist 163:787–797. Simberloff, D., J.L. Martin, P. Genovesi, V. Maris, D.A. Wardle, J. Aronson, F. Courchamp, B. Galil, E. Garcia-Berthou, M. Pascal, P. Pysek, R. Sousa, E. Tabacchi, and M. Vila. 2013. Impacts of biological invasions: what's what and the way forward. Trends in Ecology and Evolution 28:58–66. Söderbäck, B. 1995. Replacement of the native crayfish Astacus astacus by the introduced species Pacifastacus leniusculus in a Swedish lake: Possible causes and mechanisms. Freshwater Biology 33:291–304. Southwood, T.R.E., and P.A. Henderson. 2000. Ecological Methods. Blackwell Science, Oxford, UK. 575 pp. Stewart, T.W., J.G. Miner, and R.L. Lowe. 1998. An experimental analysis of crayfish (Orconectes rusticus) effects on a Dreissena-dominated benthic macroinvertebrate community in western Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences 55:1043–1050. Strayer, D.L. 2012. Eight questions about invasions and ecosystem functioning. Ecology Letters 15:1199–1210. Taylor, C.A., and M. Redmer. 1996. Dispersal of the crayfish Orconectes rusticus in Illinois, with notes on species displacement and habitat preference. Journal of Crustacean Biology 16:547–551. Northeastern Naturalist Vol. 23, No. 4 M.L. Kuhlmann 2016 453 Twardochleb, L.A., J.D. Olden, and E.R. Larson. 2013. A global meta-analysis of the ecological impacts of nonnative crayfish. Freshwater Science 32:136 7–1382. Voshell, J.R., Jr. 2002. A Guide to Common Freshwater Invertebrates of North America. McDonald and Woodward Publishing Co., Blacksburg, VA. 442 pp. Wallace, J.B., S.L. Eggert, J.L. Meyer, and J.R. Webster. 1997. Multiple trophic levels of a forest stream linked to terrestrial litter inputs. Science 277: 102–104. Wilson, K.A., J.J. Magnuson, D.M. Lodge, A.M. Hill, T.K. Kratz, W.L. Perry, and T.V. Willis. 2004. A long-term Rusty Crayfish (Orconectes rusticus) invasion: Dispersal patterns and community change in a north temperate lake. Canadian Journal of Fisheries and Aquatic Sciences 61:2255–2266.