https://orcid.org/0000-0002-7980-0654
Figures
Abstract
Non-indigenous species are often identified as threats to native species and communities. Yet, the mechanisms that enable many of these invaders to thrive and alter their newly invaded habitats are still not fully understood. This applies to habitats such as widespread sedimentary shorelines characterized by the presence of scattered biogenic clumps of blue mussels (Mytilus edulis) structurally more complex than bare sediments. In Atlantic Canada, some of these shorelines are numerically dominated by native mud crabs (Dyspanopeus sayi) but have been gradually invaded by the European green crab (Carcinus maenas). This study describes between-habitat (mussel clump vs. bare sediment) differences in density and diversity of invertebrates. It also tests the impact of juvenile green crabs in comparison to native mud crabs using two approaches: First, measuring habitat-related differences in these crabs’ feeding rates on a common prey (soft-shell clams, Mya arenaria). Second, measuring their influence on invertebrate communities associated with mussel clumps. The results show that mussel clumps hold higher invertebrate density and diversity than surrounding sedimentary bottoms. In the laboratory, the feeding rates of native mud crabs were dependent on the type of habitat (sand flat > mussel clump), whereas those of green crabs were significantly higher and unrelated to the habitat in which predation occurred. In field experiments, juvenile green crabs were also the only predators that changed community structure in the mussel clump habitat. These results indicate that green crabs can cause a significant impact on native species and communities. Moreover, they suggest that the ability of this species to overcome the refuge provided by complex biogenic habitats for prey may represent an unexplored mechanism to explain this invader’s expansion here and elsewhere.
Citation: Malyshev AV, Tummon Flynn P, Cox R, Duarte C, Quijón PA (2020) Community disruption in small biogenic habitats: A coastal invader overcomes habitat complexity to alter community structure. PLoS ONE 15(10): e0241116. https://doi.org/10.1371/journal.pone.0241116
Editor: Maura (Gee) Geraldine Chapman, University of Sydney, AUSTRALIA
Received: January 29, 2020; Accepted: October 9, 2020; Published: October 26, 2020
Copyright: © 2020 Malyshev et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study was funded by the Natural Sciences and Engineering Research Council, Canada (NSERC) through a Discovery Grant (5261) to PAQ.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The spread of marine non-indigenous species has been a recurrent topic in the ecological literature [1–4]. Nonetheless, the mechanisms mediating their success remain poorly understood and their impacts are still far from predictable [5–7]. Most studies addressing the influence of non-indigenous predators have focused on consumption rates, prey preferences [8, 9] and intra-guild interactions [10, 11]. As a result, the ultimate impact of invaders on species assembly and diversity has often received less attention than the study of more direct effects on individual (often commercially valuable) species [12–14]. However, a growing number of studies have shown that non-indigenous species have community-wide and food web effects that should not be disregarded, particularly when these invaders are predators [4, 15, 16]. In habitats such as marine soft-bottoms, epibenthic predators can have an unparalleled impact in regulating community structure and function [17–20]. However, when these relatively smooth and featureless sediments are associated with structurally complex biogenic habitats such as eelgrass, mussel or oyster beds [21], the impact of predators becomes less clear [22–25].
Clumps of mussels create three-dimensional substrate of hierarchical, irregularly shaped patches [21, 26–28] and represent an interesting system to explore the impact of new predators in the context of habitat-related interactions. The structural and functional characteristics of mussels can influence patterns of biodiversity and play a role in structuring soft-bottom communities. Their physical presence creates spatial refuges from predation and environmental stress, provides substrate for epifauna, stabilizes sediments, and alters water flow [25, 29–30]. They are also able to selectively filter suspended particles from the water column, including larvae of benthic species [31] and deposit organic material within the bed [32]. Previous studies comparing invertebrate communities inside and outside of mussel beds have reported different species composition, including greater diversity and/or abundance, in communities associated with mussel patches in soft-bottoms [33–38] and hard-bottoms [39–43].
Mussel clumps and beds are ubiquitous features of coastal habitats around the world and are a prominent feature of Atlantic Canadian sedimentary shorelines. In the Atlantic region, these patchy biogenic habitats are numerically dominated by small native mud crabs (Dyspanopeus sayi) but have also been colonized by juvenile stages of the non-indigenous European green crab (Carcinus maenas) [44–46]. Changes in predatory guild composition and abundance following the arrival of non-indigenous species can have cascading effects on communities [4, 15]. Green crabs, in particular, have been negatively associated with native crustacean and bivalve populations [46–49]. They have also been found to be responsible for habitat disruption, including local-scale changes (e.g., invertebrate density in feeding pits; [50]) and much larger-scale disruptions of biogenic habitats, for example the loss of large eelgrass beds in Nova Scotia and Newfoundland [51, 52] and formerly extensive soft-bottom mussel beds in the Gulf of Maine [53–57].
The first goal of this study was to document the contribution of blue mussels (Mytilus edulis) to the local-scale abundance and diversity of intertidal invertebrate communities in soft-bottom habitats in Atlantic Canada. We hypothesized that the biogenic structural complexity introduced by mussels alters community diversity and/or abundance relative to that on bare sediments. The second goal of this study was to assess the impact of juvenile green crabs in comparison to the impact of the similar-sized native mud crab. We hypothesized that crab predation on an individual species is altered in habitats with more complexity (mussel clumps vs. bare sand) and that native and non-indigenous crab predators differ in their impacts on infaunal density and community composition in complex habitats. We tested these hypotheses by measuring habitat-mediated feeding rates on a common type of prey (juvenile soft-shell clams, Mya arenaria) in lab enclosure experiments, and then assessing predator influence on community composition and abundance in mussel clumps through field enclosure experiments.
Material and methods
Study area and predator densities
Three intertidal sites with extensive sandy flats and abundant mussel clumps were sampled in the Hillsborough estuarine system, southern Prince Edward Island (Fig 1): Primrose Point (46’12’35”N, 6311’01”W), Meadow Bank (46’11’19”N, 63’14’22”) and Stewart Cove (46’13’23”N, 63’07’09”W). The three areas are located within 5 km from each other within the distribution range of green crabs [44, 45] and mud crabs [46]. These three sites are characterized by similar tide ranges and intertidal zones with gentle slopes associated with eelgrass beds and prominent salt marsh formations. In-situ densities of mud crabs in the past decade (S1 Fig) have consistently been higher in mussel clumps (~1.0–5.5 crabs 625 cm-2; hereafter 1.0–5.5 crabs quadrat-1) than on nearby sandy sediments (~0–0.3 crabs quadrat-1) (see field survey methodology below). Those measurements (2006, 2007 and 2014) established a gradual decline in the average number of mud crabs over time, a trend also reported for the southern Gulf of St. Lawrence [46] using time-series data of mud crab abundance measured with beach seine samples from multiple locations. In contrast, abundances of non-indigenous green crabs in the study area and the region have gradually increased since their establishment [46], although they are highly variable and strongly site-dependent (see Pickering et al. 2017 [9] and Poirier et al. 2017a [45] for data on temporal changes in density of green crabs).
The insert shows the location of Meadows Bank (MB), Primrose Point (PP), and Stewart Cove (SC). The photograph illustrates mussel clumps scattered within sedimentary bottoms in the field sites.
Field survey
To characterize the local composition and density of invertebrates associated with mussel clumps (i.e., discrete patches of blue mussels and shells; Fig 1) and surrounding sandy sediments, a one-time survey was conducted to collect samples from one of the sites referred to above (Stewart Cove) during low spring tides of June and July 2007. Eight mussel clumps, approximately 20 cm in diameter, were randomly selected and sampled using 25x25 cm stainless steel quadrats (625 cm2) inserted ~5 cm into the substrate. Clumped mussels, shells, sediments, and debris were collected using spatulas and then washed through 2 mm sieves. The choice of sieve size was based on feasibility for processing and handling. While we acknowledge a risk of underestimating the diversity of small species, a large and diverse number of invertebrates were collected in the samples. Each sample of a mussel clump was paired with a sample collected from bare sediments (those lacking mussel clumps or shells), from the immediate vicinity (0.5–1.0 m away). The material retained in the sieves was preserved in 70% ethanol, sorted and identified to the species level using standard identification keys for the region [58, 59].
Laboratory enclosure experiment: Habitat-mediated crab feeding rates
Based on the results of the field survey (see Results), juvenile soft-shell clams (Mya arenaria) were selected as prey to compare feeding rates of native and non-indigenous crabs as it was one of the most abundant species in both mussel clumps and bare sediments. This species is a component of the diet of mud crabs and green crabs [60] and juvenile clams (6–10 mm Shell Length, SL) are known to be consumed by both species in the study area. Previous experiments have also used this species to compare crab feeding rates. Soft-shell clams were collected manually from nearby sediments in Stewart Cove. Adult mud crabs (25–35 mm CW) were compared to juvenile green crabs of a similar size range because predator (crab) size can influence feeding trial outcomes [9, 61]. Juvenile green crabs were considered the logical counterpart of native mud crabs because they exploit the same prey items and habitats as consistent occupants of intertidal mussel, oyster and algae clumps (P. Quijón Pers. Obs.; [62]). All the crabs for experiments were collected from Belvedere Pond, located approximately 2 km NE from Stewart Cove, within the same estuarine system. Only unharmed (without missing claws or legs) male crabs of each species were used to avoid potential biases associated with sex and condition [63]. The crabs were starved for 48 h prior to their use in any experiment to avoid potential biases associated with crab’s prior diet [64].
Twenty juvenile soft-shell clams were placed into each of ten 19.8 L glass tanks containing seawater (~25 ppt; 18°C) and a bottom 3 cm layer of cleaned sandy sediments from the study area. The clams were allowed to acclimate and bury themselves in the sand for approximately 1 h before one individual mud crab or green crab was introduced. Trials were run once for 24 h with the top and sides of each tank covered with dark paper to avoid crab escape and to reduce light exposure and potential visual distractions [65]. In addition to trials with bare sediments, additional trials (mussel clump mimic treatment) were conducted in identical tanks in which a 3 cm layer of mussel shells (considered representative of the thickness of mussel layers in the field) was added on top of the sediment to resemble the physical structure of a mussel clump. While shells cannot fully mimic live mussels [see 56, 57, 66] they were collected from the same field site, had a similar size range (~15–35 mm SL), and have been used in the past as habitat mimics in other studies [46]. At the end of each trial the predator was removed from the tank and the number of remaining soft-shell clams was counted to estimate percent mortality of prey (or % of clams eaten). Experiments were initiated and terminated at about the same time of the day (late morning) A few trials in which crabs showed signs of molting were not considered for any analyses and conducted again with different crabs. To avoid potential learning after repeated trials, each crab was used only once [65].
Field enclosure experiment: Crab effects on mussel communities
Mussel clumps of approximately the same size (~20 cm diameter) were randomly selected from the mid-low intertidal of Stewart Cove to conduct a 1-wk field enclosure experiment in 2008. These discrete clumps were 1–2 m apart and embedded in bare unstructured sandy sediments (Fig 1). Mussel clumps located at least 10 m apart were identified (tagged with straws) and randomly assigned to the following treatments: green crab enclosure (GC), mud crab enclosure (MC), cage with no crabs added (control cage; CC) and un-manipulated control (clump with no cage or crabs added; NC). The experiment included a total of eight replicates per treatment. Three mud crabs or three green crabs were placed in each of the enclosure cages (MC and GC treatments, respectively). These crabs were collected from the same location and had the same size range as those used in the laboratory experiments. Their density within the cages was approximately half the highest mean density observed in the field surveys conducted in 2006, 2007 and 2014 (up to a mean of 5.5 crabs clump-1; see details in S1 Fig) and was consistent with repeated observations of mussel clumps in the study area and other areas nearby during those years. Cages were circular, (35 cm in diameter and 30 cm high) bucket-shaped cylinders built with ~0.5 cm wire mesh around a metal rod frame. The cages were inserted upside-down (open bottom, covered top) approximately 5 cm into the sediment and anchored with long plastic pegs. To avoid unnecessary disturbance of sediments and invertebrates associated with the mussel clumps, only visible mud crabs or green crabs were manually removed from the clumps before cage placement.
For the duration of the experiment, the cages were checked daily during low tide for potential stranding and accumulation of eelgrass (Zostera marina), rockweed (Fucus serratus) or Irish moss (Chondrus crispus) or build-up of fine sediments (i.e., to visually assess for potential cage-artefacts; [67]). Eelgrass or seaweed stranding were observed on the cage sides in only a few cases and were removed immediately. There was no visual evidence of sediment accumulation within or outside the cages. At the end of the experiment, cages were carefully opened and the mussel clumps examined until the three crabs that had been introduced were found, recorded and removed. Mud crabs were released on site whereas green crabs were retained and returned to the laboratory (none of the green crabs or mud crabs escaped the cages and no additional crabs were found in any of the cages). After crab removal, a quadrat was inserted immediately around the mussel clump, and samples were collected and processed. Blue mussels were in general much larger (> 2cm SL) and less abundant than other invertebrates and therefore were not considered plausible prey for the small predators used in this experiment. In addition, blue mussels were not considered part of the infaunal communities described below.
Data analyses
Invertebrate species richness and density recorded in bare sediment and mussel clumps in the field were compared using paired t-tests. Density comparisons included total number of invertebrates and the density of polychaetes, bivalves, and four of the most abundant species. Data on species composition and density were also analyzed to assess community similarity between habitats using the Bray-Curtis index and the program Plymouth Routines In Multivariate Ecological Research (PRIMER 6). Data were fourth-root transformed to proportionally reduce the influence of the most abundant species. Visual differences among treatments were inspected using a non-metric Multi-Dimensional Scaling (MDS) plot. Analysis of Similarity (ANOSIM) was then used to assess whether or not the differences between habitats were significant [68]. The SIMPER routine was also applied to identify the species that contributed the most to the dissimilarity between habitats.
Comparisons of the number of soft-shell clams eaten in the laboratory (percent mortality) by either mud crabs or green crabs in bare sediments or mussel clump habitats were analyzed with a two-way ANOVA. Independent factors included the predator (green crab vs. mud crab) and the habitat (mussel clump vs. sand flat); both factors were considered fixed.
For the field enclosure experiment, species richness and densities (total and density of bivalves, polychaetes, and the four most abundant species) recorded in the four treatments (GC, MC, CC and NC) were compared using one-way ANOVAs. In cases where the ANOVAs detected significant differences, Holm-Sidak a posteriori tests were used to identify individual pairwise differences between treatments. Assumptions of normality (Shapiro-Wilk test) and equal variance were checked in all instances. Due to violations to the first assumption, a log(n+1) transformation was applied to the data except in two cases: species richness (raw data) and density of H. filiformis (sq-root transformation). These analyses were conducted using Minitab® routines. Untransformed data on species composition and density were also analyzed to assess community similarity among treatments using PRIMER 6 routines (MDS, ANOSIM and SIMPER) as described for the field survey.
Permits, animal care and ethics
Permit to access field sites in Prince Edward Island were obtained from the Department of Fisheries and Oceans Canada–Gulf Division (seasonal permit to collect organisms for scientific purposes). All applicable institutional (ACC-UPEI) and national guidelines for the care and use of animals were followed. This article does not contain any studies with human participants.
Results
Field survey
Species composition and average density are summarized in S1 Table. The average number of species was three times higher in sediments associated with mussel clumps than in those associated with surrounding bare sands (7.75 and 2.5 species quadrat-1, respectively; paired t-test p = 0.001; Fig 2; Table 1). Total densities, total density of bivalves and polychaetes, and densities of soft-shell clams (Mya arenaria), gem clams (Gemma gemma) and the spionid polychaete Heteromastus filiformis, were on average between 3 and 7 times higher in the mussel clumps than in surrounding sandy sediments (all differences were significant, paired-t test p≤0.036; Fig 2; Table 1). In contrast, the polychaete Glycera dibranchiata was found at similar average densities (1.88 and 1.75 individuals quadrat-1 in mussel clumps and sand flats, respectively; paired t-test p = 0.857; Fig 2; Table 1).
Different letters above the bars stand for significant differences between habitats (paired t-test p<0.05).
N = 8 and DF = 7 in all the analyses. Significant p-values are in bold.
The nMDS plot (Fig 3) identified a strong difference in composition and abundance of communities associated with mussel clumps and bare sediments. The nMDS plot was also a good representation of the variation among samples (2D stress value = 0.08) and showed that samples associated with mussel clumps segregated apart from samples associated with bare sediments. ANOSIM tests confirmed that between-habitat differences were significant (Global R = 0.588, p = 0.002). The average level of dissimilarity between habitats was 65.8% and the most important species driving the dissimilarity between habitats were the bivalves Gemma gemma and Mya arenaria and the isopod Jaera marina (See detailed results in S2 Table).
Lines encircling samples apart from all the other samples are based on significant differences detected by ANOSIM (p<0.05).
Laboratory enclosure: Habitat-mediated crab feeding rates
Predator identity and habitat each had a significant influence on the percent mortality of juvenile soft-shell clams (two-way ANOVA p<0.001 and p = 0.025, respectively), whereas their interaction was not significant (two-way ANOVA p = 0.230). Green crabs consumed significantly more juvenile soft-shell clams than mud crabs, eating ~60% and ~38% more clams than mud crabs in mussel clumps and sand flats, respectively (Fig 4). When comparing feeding rates in relation to habitat, mud crabs consumed substantially fewer soft-shell clams in mussel clumps than in the sand flat (average clam percent mortality was ~40% and 60%, respectively). In contrast, juvenile green crabs consumed only marginally different amounts of soft-shell clams: average mortality of ~82% and 88% in mussel clumps and sand flats, respectively; Fig 4).
The experiments were conducted in tanks with mimics of mussel clumps and sandy flats. Different letters above the bars stand for significant differences between predators (two-way ANOVA p<0.05). Between-habitat differences were also found but are not included for clarity.
Field enclosure: Crab effects on mussel clump communities
Species composition and average density are summarized in S3 Table. The average number of species in the mussel clumps was significantly lower in green crab enclosures (~6.6 species quadrat-1) than in any other treatment (mud crab enclosure, caged and uncaged controls; one-way ANOVAs p = 0.009; Fig 5, Table 2). In general, similar results were obtained when quantifying total densities and the density of bivalves (one-way ANOVA p<0.001 and p = 0.003, respectively). The total density of polychaetes was significantly different only when comparing green crab and mud crab enclosures (one-way ANOVA p<0.001; Fig 5). Regarding the density of numerically dominant species (Fig 6), mussel clumps exposed to green crab predation generally exhibited significantly lower averages of soft-shell clams and Nereis succinea (one-way ANOVA p<0.001 and p = 0.003, respectively; Table 2). However, the density of small gem clams (Gemma gemma) and Heteromastus filiformis did not differ among treatments (p = 0.083 and p = 0.089, respectively; Fig 6; Table 2).
GC: green crab enclosure (filled bars); MC: mud crab enclosure; CC: clump in a control cage; NC: no cage or clump without a cage. Different letters above the bars stand for significant differences among treatments based on Holm-Sidak post-hoc comparisons (p<0.05).
Other details as in Fig 5.
N = 8 and DF = 3,28 in all the analyses. For simplicity, only Mean Squares (MS), F- and p-values are presented. Significant p-values are in bold. Holm-Sidak post-hoc comparisons are illustrated in Figs 5 and 6.
The nMDS plot (Fig 7) identified a strong effect of green crabs on community structure and a fairly good representation of the variation among samples (2D stress value = 0.19). The samples associated with green crab enclosures formed a distinct cluster separated from the samples from all other treatments (mud crab enclosures, and caged and un-caged controls). ANOSIM tests confirmed that the green crab enclosure was the only treatment significantly different from others (Global R = 0.169, p = 0.003). P-values for pair-wise comparisons between green crab enclosure and the other treatments were all significant (p = 0.001–0.006), whereas p-values for pair-wise comparisons among the remaining three treatments were all non-significant (p = 0.322–0.714). The average level of dissimilarity between green crab enclosures and the three other treatments ranged between 68.3 and 72.30%. The most important species driving the dissimilarity between habitats were the polychaete Heteromastus filiformis and the bivalves Gemma gemma and Mya arenaria (See S4 Table for details).
GC (filled symbols): green crab enclosures; MC: mud crab enclosures; CC: control cage; NC: no cage or clump with no cage. Lines encircling green crab enclosure samples apart from all the other samples are based on significant differences detected by ANOSIM (p<0.05).
Discussion
Green crabs can be effective predators in unstructured sediments such as sandy and muddy tidal flats [47, 50, 69]. The results gathered here suggest that they can also be effective predators in habitats that typically reduce or preclude the effects of epibenthic predation [24, 70]. Juvenile green crabs were more effective predators than native mud crabs, and they demonstrated an ability to catch and consume prey in both unstructured and structured habitats. These results reinforce the potential for green crabs to have negative community-wide effects [16]. Moreover, they suggest that the ability of this species to overcome the refuge provided by complex biogenic habitats could represent a fairly unexplored mechanism to explain this invader’s expansion here and elsewhere. If this argument is applicable to other non-indigenous predators, it may well represent an additional facilitation mechanism for successful invasions [2, 7], more specifically for invasions by predators. This argument fits in the broader examination of the relative effects of native and non-indigenous predators on prey [71, 72]. A review of the literature found that in comparison to their native counterparts, non-indigenous predators were generally more detrimental to terrestrial and avian vertebrates [71] and prey associated with terrestrial or freshwater habitats, but not always with marine habitats [72]. Our study of invertebrate predators and prey adds an interesting case to that theoretical debate.
Field survey and habitat differences
In comparison to surrounding unstructured sediments, mussel clumps held remarkably high densities and diversity of species. These results support the hypothesis that mussel clumps represent islands of biogenic habitat whose structural complexity increases diversity relative to bare sediment. This is in line with what has been observed in coastal shorelines in Europe [34–36], but not necessarily in eastern North America, where increases in abundance but not in diversity have been reported [33, 37, 38, 56, 57, 73]. It is unclear why diversity results differ between the study area and Maine, USA. Such differences may be the result of different green crab invasion histories [48] or the use of different gear for the sampling of infauna: various quadrats sizes and different sieve mesh sizes. Regardless, these results highlight the importance of biogenic habitats to the full shoreline ecosystem [40, 73]. Understanding the influence of those habitats on the impact of native and non-indigenous predators is relevant, as demonstrated by our laboratory experiments: The analysis of crab feeding rates on a single prey type suggests that native and non-indigenous crabs do not have the same impact and that this impact depends on habitat (see also [92]). Added habitat complexity (mussel shells mimicking biogenic clumps) effectively diminished mud crab feeding rates on soft-shell clams (measured here as clam percent mortality). However, it did little to prevent consistently high feeding rates by juvenile green crabs. Native predators fed on soft-shell clams, but when placed in a more complex habitat, they followed a pattern previously observed elsewhere in e.g., blue crabs consuming scallops and oysters [23, 74]. Although our mussel clump mimics did not fully resemble a clump of live mussels, these mimics reduced clam mortality caused by mud crabs by as much as 40%. This was likely because of a reduction in predator-prey encounter rates resulting from the added difficulty crabs faced finding and handling clams [22, 74].
The clear influence of mussel clump mimics on mud crab feeding rates supports the hypothesis that predation levels are altered (reduced) in habitats with more complexity. Ultimately, habitat complexity may have reduced prey profitability (sensu Norberg 1977 and Stephens and Krebs 1986; [75, 76]) for the native predator, although we acknowledge that adding a layer of mussel shells on top of sediments also changed the overall depth of the habitat available to the clams to seek refuge. In contrast, juvenile green crabs consumed significantly more clams than mud crabs did, and they did so regardless of habitat complexity and depth. It is unclear what enables juvenile green crabs to be successful predators in sharply different habitats. However, from prior observations of green crabs digging for prey around roots and shoots of eelgrass (Zostera marina) [77], it can be speculated that this ability is related to an unusually broad repertoire of movements allowing them to seek prey in relatively complex habitats. Prior studies have shown that green crabs possess quick learning skills and adaptability to handle new types of prey and to gather food (bivalves) using novel techniques [78, 79]. For prey such as the soft-shell clam, which typically gains refuge from predation within structured habitats such as seaweeds and shellfish clumps and beds [40], the evidence presented here is also concerning from a practical point of view. The arrival of predators well-suited to effectively exploit structured as well as unstructured habitats may entail a new, and potentially substantial, source of mortality for multiple prey. Species of commercial importance, such as the soft-shell clam, make the outcome of these interactions potentially detrimental for coastal growers and harvesters [9, 12, 74] and the industries they support.
Green crab effects on mussel clump communities
During the field experiment conducted here, the density and diversity of species associated with mussel clumps were altered solely in the presence of the non-indigenous predator. Unlike the native mud crab, green crabs reduced overall diversity and abundance by ~30%. The impact of green crabs in this habitat was surprising as the complexity of the mussel clumps was expected to limit crab foraging abilities [23, 92]. However, the results presented here are consistent with the results from the laboratory experiment and suggest that this non-indigenous predator may thrive in unstructured as well as structured habitats. These results also provide strong support to the hypothesis that native and non-indigenous crab predators differ in their impacts on infaunal communities in complex habitats. The literature on soft-bottom communities contains a wealth of examples where population densities or life history traits (e.g., size or growth rates) have changed following the addition of epibenthic predators [17–20, 80, 81]. However, only a handful of studies have documented changes in the number of species in a community. In the Atlantic region, the native rock crab (Cancer irroratus) is one such case [82]. Those authors documented a marked decrease in density and species richness following the enclosure of rock crabs [82]. The changes reported here are not as striking quantitatively, but were measured in habitats (mussel clumps) expected to be less prone to strong predator effects.
The effects of green crabs may not be restricted to those documented here. Gehrels et al. [46] recently showed that adult green crabs also feed on mud crabs, which suggests multiple, perhaps synergistic negative effects (sensu Myers 1989 [83]) on this native species. Similarly, competitive interactions or various indirect effects between crabs inhabiting mussel clumps are plausible and cannot be discounted. Due to logistic limitations, the design of our field experiment did not include a fifth treatment where mud crabs and green crabs were confined together. Interactions between these predators may change their individual effects on infauna, perhaps adding predation pressure (additive or non-additive effects) or, more likely, attenuating their overall effects [47]. The examination of intra-guild interactions was beyond the scope of this study, but would likely shed light on some of the results recorded in the clumps that lack a cage (NC or uncaged treatment). As expected, exposure to surrounding predators caused an “intermediate” effect (i.e., between control and strong green crab effects) on total densities and the densities of polychaetes (including N. succinea and H. filiformis), but not on species richness of bivalve densities. The latter may suggest limitations associated with the use of cages [84, 85], which normally include accumulation of fine sediments or changes in predator or prey behavior [84, 85]. We consider the accumulation of fine sediments highly unlikely given the short duration of the field trial [69], and the daily monitoring of cages showing no evidence of cage artefacts. The potential for behavioral responses to the presence of e.g., adult green crabs [16, 46] or other species should be explored further. Similarly, interactions between infaunal organisms may also have an effect here and should be assessed. Prior studies suggest that at least some of these species are predatory infauna [69, 86], a functional group including carnivorous polychaetes and nemertines (e.g., Lineus viridis in Maine [33]), which likely add further complexity to the trophic role played by the green crab in these habitats.
Implications
Invasions stand among various other pervasive sources of anthropogenic disturbance of coastal habitats [87–89]. While the changes measured in this study occurred at the scale of small patchy habitats, they must be considered in the context of the continued expansion of this non-indigenous species in the Atlantic region [44, 45] and shoreline regions in most continents. Additional studies should address larger-scale questions such as the ongoing alteration of structured habitats (beds of eelgrass, mussels and oysters) by green crabs, which may result in increased fragmentation and ultimately, a gradual change in the entire shoreline ecosystem. The decline of once extensive mussel beds in the Gulf of Maine has been linked to the green crab invasion as well as climate change, and represents a key example of a non-indigenous species’ ability to influence entire food webs [16] by simultaneously consuming native species while disrupting biogenic habitat. Over time, as little as 2–4% of the original live mussel coverage remains in mussel beds increasingly fragmented into mussel clumps [56, 57]. As the slow recovery of similar habitats elsewhere suggests (e.g., oyster beds and eelgrass beds [51, 90]), such a change has long-term implications for ecosystems [91] and harvesting industries.
The further spread of green crabs is expected to promote changes among individual prey (soft-shell clams, blue mussels and various other shellfish species [9, 12]), intra-guild counterparts such as mud crabs or other crustaceans ([11, 46, 92], this study), and habitat-forming species such as mussels, oysters and eelgrass [51, 78, 93]. While assessing the likelihood of future invasions by this and other species [94], their ability to modify communities associated with structured habitats needs to be seriously considered. Establishing or recovering foundation species and ecosystem engineers that create structured habitats is broadly perceived as a positive measure for the ecology and function of coastal ecosystems [92, 95]. However, in light of the evidence presented here, this may not necessarily protect every native species from aggressive invaders like the green crab.
Supporting information
S1 Table. Species composition and average density (+/- Standard Error) of assemblages associated with mussel clumps and sand flats collected during the field survey.
https://doi.org/10.1371/journal.pone.0241116.s002
(DOCX)
S2 Table. Ranking of species contributing the most to the dissimilarity between habitats (up to a minimum of 60% dissimilarity) in the field survey of mussel clumps and sandy flats.
https://doi.org/10.1371/journal.pone.0241116.s003
(DOCX)
S3 Table. Species composition and average density (+/- Standard Error) of assemblages associated with the four treatments in the field enclosure experiment: Green crab enclosure (GC), mud crab enclosure (MC), control cage (CC) and no cage (NC).
https://doi.org/10.1371/journal.pone.0241116.s004
(DOCX)
S4 Table. Ranking of 5 species contributing the most to the dissimilarity between green crab enclosures (GC) and the other three treatments in the field experiment: mud crab enclosure (MC), Control cage (CC) and no cage (NC), up to a minimum of 60% dissimilarity).
Column at the left identifies the comparisons and the average dissimilarity.
https://doi.org/10.1371/journal.pone.0241116.s005
(DOCX)
Acknowledgments
We are grateful of Vanessa Lutz-Collins, John Greenan, Jessica Willis and Lisa Rosenberg (UPEI) for their help in the laboratory and the field. We also thank Diego Quintanilla-Ahumada (UNAB) for reviewing an earlier version of the manuscript and Gabriela Quijón for her help with reference checking. Our gratitude also goes to John A. Commito, Brady A, Quinn and three anonymous reviewers who provided valuable feedback on earlier versions of the manuscript.
References
- 1.
Elton CS (1958) The ecology of invasions by animals and plants. Methuen and Co. / Chapman and Hall, Kluwer Academic Publishers BV, Chicago, 181 p.
- 2. Bax N, Williamson A, Aguero M, Gonzalez E, Geeves W (2003) Marine invasive alien species: a threat to global biodiversity. Mar Policy 27: 313–323.
- 3. Fridley JD, Sax DF (2014) The imbalance of nature: revisiting a Darwinian framework for invasion biology. Global Ecol Biogeogr 23:1157–1166.
- 4. Katsanevakis S, Tempera F, Teixeira H (2016) Mapping the impact of alien species on marine ecosystems: the Mediterranean Sea case study. Divers Distrib 22:694–707.
- 5. Bertness MD, Coverdale TC (2013) An invasive species facilitates the recovery of salt marsh ecosystems on Cape Cod. Ecology 94: 1937–1943. pmid:24279265
- 6. Edelist D, Rilov G, Golani D, Carlton JT, Spanier E. (2013) Restructuring the Sea: profound shifts in the world's most invaded marine ecosystem. Divers Distrib 19:69–77.
- 7. Quijón PA, Tummon Flynn P, Duarte C (2017) Beyond negative perceptions: the role of some marine invasive species as trophic subsidies. Mar Poll Bull 116: 538–539.
- 8. Cohen AN, Carlton JT, Fountain MC (1995) Introduction, dispersal and potential impacts of the green crab Carcinus maenas in San Fransciso Bay, California. Mar Biol 122:225–237.
- 9. Pickering T, Poirier L, Barrett T, McKenna S, Davidson J, Quijón PA (2017) Non-indigenous predators threaten ecosystem engineers: interactive effects of green crab and oyster size on American oyster mortality. Mar Env Res 127:24–31.
- 10. Griffen BD, Byers JE (2006) Intraguild predation reduces redundancy of predator species in multiple predator assemblage. J Animal Ecol 75: 959–966.
- 11. Rossong MA, Williams PJ, Comeau M, Mitchell SC, Apaloo J (2006) Agonistic interactions between the invasive green crab, Carcinus maenas (Linnaeus) and juvenile American lobster, Homarus americanus (Milne Edwards). J Exp Mar Biol Ecol 329:281–288.
- 12. Poirier LA, Symington LA, Davidson J, St-Hilaire S, Quijón PA (2017) Exploring the decline of oyster beds in Atlantic Canada shorelines: Potential effects of crab predation on American oysters (Crassostrea virginica). Helgoland Mar Res 71:13
- 13. Frésard M, Boncoeur J (2006) Controlling the biological invasion of a commercial fishery by a space competitor: a bioeconomic model with reference to the Bay of St-Brieuc scallop fishery. Agricultural and Resource Economics Review 35:78–97.
- 14. Albins MA, Hixon MA (2008) Invasive Indo-Pacific lionfish Pterois volitans reduce recruitment of Atlantic coral-reef fishes. Mar Ecol Prog Ser 367: 233–238.
- 15. Thomsen MS, Byers JE, Schiel DR, Bruno JF, Olden JD, Wernberg T, Silliman BR (2014) Impacts of marine invaders on biodiversity depend on trophic position and functional similarity Mar Ecol Prog Ser 495: 39–47.
- 16. Wong MC, Dowd M (2014) Role of invasive green crabs in the food web of an intertidal sand flat determined from field observations and a dynamic simulation model. Estuaries Coasts 37: 1004–1016.
- 17.
Peterson CH (1979) Predation, competitive exclusion, and diversity in the soft-sediment benthic communities of estuaries and lagoons. In: Livingston RJ (ed) Ecological processes in coastal and marine systems. Plenum, New York, p. 233–264.
- 18. Wilson WH Jr (1991) Competition and predation in marine soft-sediment communities. Annu Rev Ecol Syst 21:221–241.
- 19. Ólafsson EB, Peterson CH, Ambrose WG Jr (1994) Does recruitment limitation structure populations and communities of macroinvertebrates in marine soft sediments: the relative significance of pre- and post-settlement processes. Oceanogr Mar Biol Annu Rev 32:65–109.
- 20.
Byers JE, Grabowski JH (2014) Soft-sediment communities. In: Marine community ecology and conservation. Eds. Bertness MD, Bruno JF, Silliman BR, Stachowicz JJ. Sinauer, p. 227–249.
- 21. Commito JA, Dow WE, Grupe BM (2006) Hierarchical spatial structure in soft-bottom mussel beds. J Exp Mar Biol Ecol 330:27–37.
- 22. Crowder LB, Cooper WE (1982) Habitat structural complexity and the interactions between bluegills and their prey. Ecology 63:1802–1813.
- 23. Hill JM, Weissburg MJ (2013) Habitat complexity and predator size mediate interactions between intraguild blue crab predators and mud crab prey in oyster reefs. Mar Ecol Progr Ser 488:209–219.
- 24. Glaspie CN, Seitz RD (2018) Habitat complexity and benthic predator-prey interactions in Chesapeake Bay. PLoS ONE 13(10): e0205162. pmid:30289889
- 25. Reise K, Buschbaum C, Büttger H, Wegner KM (2017) Invading oysters and native mussels: from hostile takeover to compatible bedfellows. Ecosphere 8(9):e01949.101002/ecs2.1949
- 26. Snover ML, Commito JA (1998) The fractal geometry of Mytilus edulis L. spatial distribution in a soft-bottom system. J Exp Mar Biol Ecol 223: 53–64.
- 27. Crawford TW, Commito JA, Borowik AB (2006) Fractal characterization of Mytilus edulis L. spatial structure in intertidal landscapes using GIS methods. Landscape Ecol 21:1033–1044.
- 28. Commito JA, Commito AE, Platt RV, Grupe BM, Piniak WED, Gownaris NJ, et al (2014) Recruitment facilitation and spatial pattern formation in soft-bottom mussel beds. Ecosphere 5:1–26.
- 29. Albretch A, Reise K (1994) Effects of Fucus vesiculosus covering intertidal mussel beds in the Wadden Sea. Helgoland Mar Res 48:243–256.
- 30. Frechette M, Butman CA, Rockwell Geyer W (1989) The importance of boundary-layer flows in supplying phytoplankton to the benthic suspension feeder, Mytilus edulis L. Limnol Oceanogr 34:19–36.
- 31.
Commito JA, Dankers N (2001) Dynamics of spatial and temporal complexity in European and North American soft-bottom mussel beds. In: Ecological Comparisons of Sedimentary Shores, pp. 39–59. Ed. by Reise K. Springer-Verlag, Heidelberg, Germany.
- 32. Kautsky N, Evans S. (1987) Role of biodeposition by Mytilus edulis in the circulation of matter and nutrients in a Baltic coastal system. Mar Ecol Prog Ser 38:201–212.
- 33. Commito JA, Como S, Grupe BM, Dow WE (2008) Species diversity in the soft-bottom intertidal zone: biogenic structure, sediment, and macrofauna across mussel bed spatial scales. J Exp Mar Biol Ecol 366:70–81.
- 34. Norling P, Kautsky N (2007) Structural and functional effects of Mytilus edulis on diversity of associated species and ecosystem functioning. Mar Ecol Progr Ser 351:163–175.
- 35. Norling P, Kautsky N (2008) Patches of the mussel Mytilus sp. are islands of high biodiversity in subtidal sediment habitats in the Baltic Sea. Aquat Biol 4:75–87.
- 36. Dittman S (1990) Mussel beds–amensalism or amelioration for intertidal fauna? Helgol Mar Res 44:335–352
- 37. Commito JA, Boncavage EM (1989) Suspension feeders and coexisting infauna: an enhancement counterexample. J Exp Mar Biol Ecol 125: 33–42
- 38. Commito JA (1987) Adult–larval interactions: predictions, mussels, and cocoons. Estuar Coastal Shelf Sci 25:599–606.
- 39. Borthagaray AI,Carranza A (2007) Mussels as ecosystem engineers: Their contribution to species richness in a rocky literal community. Acta Oecol 31:243–250.
- 40. Bouma TJ, Olenin S, Reise K, Ysebaert T (2009) Ecosystem engineering and biodiversity in coastal sediments: posing hypotheses. Helgol Mar Res 63:95–106.
- 41. Thiel M, Ulrich N (2002) Hard rock versus soft bottom: the fauna associated with intertidal mussel beds on hard bottoms along the coast of Chile, and considerations on the functional role of mussel beds. Helol Mar Res 56:21–30.
- 42. Seed R (1996) Patterns of biodiversity in the macro-invertebrate fauna associated with mussel patches on rocky shores. J Mar Biol Ass UK 76:203–210.
- 43. Tokeshi M, Romero L (1995). Filling a gap: Dynamics of space occupancy on a mussel-dominated subtropical rocky shore. Mar Ecol Progr Ser 119:167–176.
- 44. Audet D, Davis D, Miron G, Moriyasu M (2003) Geographical expansion of a nonindigenous crab, Carcinus maenas (L.), along the Nova Scotian shore into the southeastern Gulf of St. Lawrence. Canada. J Shellfish Res 22:255–262.
- 45. Poirier LA, Ramsay AP, St-Hilaire S, Quijón PA (2017a) Revisiting recent history: Records of occurrence and expansion of the European green crab across Prince Edward Island, Atlantic Canada. Mar Biodivers Records 10:30.
- 46. Gehrels H, Knysh KM, Boudreau M, Thériault M-H, Courtenay SC, Cox R, et al (2016) Hide and seek: habitat-mediated interactions between European green crabs and native mud crabs in Atlantic Canada. Mar Biol 163:152.
- 47. Gregory G, Quijón PA (2011) The impact of a coastal invasive predator: assessing the role of density and a native counterpart. J Sea Res 66:181–186.
- 48. Rossong M, Barrett T, Quijón PA, Snelgrove PVR, Mackenzie C, Locke A (2012) Regional differences in foraging behaviour and morphology of invasive green crab (Carcinus maenas) populations in Atlantic Canada. Biol Inv 14:659–669.
- 49. Grosholz ED, Ruiz GM, Dean CA, Shirley KA, Maron JL, Connors PG (2000) The impacts of a nonindigenous marine predator in a California Bay. Ecology 81:1206–1220.
- 50. Lutz-Collins V, Cox R, Quijón PA (2016) Habitat disruption by a coastal invader: local community change and recovery in Atlantic Canada sedimentary habitats. Mar Biol 163:178
- 51. Garbary DJ, Miller AG, Williams J, Seymour NR (2014) Drastic decline of an extensive eelgrass bed in Nova Scotia due to the activity of the invasive green crab (Carcinus maenas). Mar Biol 161:3–15.
- 52. Matheson K, McKenzie CH, Gregory RS, Robichaud DA, Bradbury IR, Snelgrove PVR, et al. (2016) Linking eelgrass decline and impacts on associated fish communities to European green crab Carcinus maenas invasion. Mar Ecol Prog Ser 548:31–45.
- 53. Petraitis PS, Dudgeon SR (2015) Variation in recruitment and the establishment of alternative community states. Ecology 96:3186–3196 pmid:26909425
- 54. Sorte CJB, Davidson BE, Franklin MC, Benes KM, Doellman MM, Etter RJ, Hannigan RE, Lubchenco J, Menge BA (2016) Long-term declines in an intertidal foundation species parallel shifts in community composition. Global Change Biol 23:341–352
- 55. Sorte CJB, Jones SJ, Miller LP (2011) Geographic variation in temperature tolerance as an indicator of potential population responses to climate change. J Exp Mar Biol Ecol 400:209–217.
- 56. Commito JA, Jones BR, Jones MA, Winders SE, Como S (2018) What happens when mussels die? Biogenic legacy effects on community structure and ecosystem processes. J Exp Mar Biol Ecol 506: 30–41.
- 57. Commito JA, Jones BR, Jones MA, Winders SE, Como S (2019) After the fall: Legacy effects of biogenic structure on wind-generated ecosystem processes following mussel bed collapse. Diversity 11: 1–20.
- 58. Appy TD, Linkletter LE, Dadswell MJ (1980) A guide to the marine flora and fauna of the Bay of Fundy: Annelida: Polychaeta. Fish Mar Serv Tech Rep 920:124.
- 59. Pocklington P (1989) Polychaetes of Eastern Canada. An illustrated key to polychaetes of Eastern Canada including the Eastern Arctic. Scientific Monograph funded by Ocean Dumping Control Act Research Fund, National Museums of Canada & Department of Fisheries and Oceans.
- 60. Palacios KC, Ferraro SP (2003) Green crab (Carcinus maenas Linnaeus) consumption rates on and prey preferences among four bivalve prey species. J Shellfish Res 22:865–72.
- 61. Sun Y, Wang F, Su X, Liu D, Dong S (2017) Size matters for predator-prey relationships: The case of swimming crab Portunus trituberculatus preying on Manila clam Ruditapes philippinarum. Aquaculture 479:638–644.
- 62. Thiel M, Dernedde T (1994) Recruitment of shore crabs Carcinus maenas on tidal flats: mussel clumps as an important refuge for juveniles. Helgolander Meeresunters 48:321–332
- 63. Tummon Flynn P, Mellish C, Pickering TR, Quijón PA (2015) Effects of claw autotomy on green crab (Carcinus maenas) feeding rates. J Sea Res 103:113–119.
- 64. Mascaro M, Seed R (2001) Choice of prey size and species in Carcinus maenus (L.) feeding on four bivalves of contrasting shell morphology. Hydrobiologia 449:159–170.
- 65. Cunningham PN, Hughes RN (1984) Learning of predatory skills by shore crabs Carcinus maenas feeding on mussels and dogwhelks. Mar Ecol 16:21–26.
- 66. Commito JA, Rusignuolo BR (2000) Structural complexity in mussel beds: the fractal geometry of surface topography. J Exp Mar Biol Ecol 255:133–152. pmid:11108848
- 67. Quijón PA, Snelgrove PVR (2005a) Predation regulation of sedimentary faunal structure: Potential effects of a fishery-induced switch in predators in a Newfoundland sub-arctic fjord. Oecologia 144: 125–136. pmid:15791429
- 68.
Clarke KR, Gorley RN, Somerfield PJ, Warwick RM (2014) Change in marine communities: an approach to statistical analysis and interpretation,3nd edition. PRIMER-E: Plymouth.
- 69. Quijón PA, Snelgrove PVR (2008) Trophic complexity in marine sediments: New evidence from the Gulf of St. Lawrence. Mar Ecol Prog Ser 371:85–89.
- 70. Ellis JC, Allen KE, Rome MS, Shulman MJ (2012) Choosing among mobile prey species: Why do gulls prefer a rare subtidal crab over a highly abundant intertidal one? J Exp Mar Biol Ecol 416–417:84–91.
- 71. Salo P, Korpimaki E, Banks PB, Nordstrom M, Dickman CR (2007) Alien predators are more dangerous than native predators to prey populations. Proc Royal Soc B, 274, 1237–1243.
- 72. Paolucci EM, MacIsaac HJ, Ricciardi A (2013) Origin matters: Alien consumers inflict greater damage on prey populations than do native consumers. Diversity and Distributions, 19, 988–995.
- 73. Commito J. A., Celano E. A, Celico H. J., Como S., Johnson C. P. (2005). Mussels matter: postlarval dispersal dynamics altered by a spatially complex ecosystem engineer. J Exp Mar Biol Ecol 316: 133–147.
- 74. Hernández Cordero AL, Seitz RD (2014) Structured habitat provides a refuge from blue crab, Callinectes sapidus, predation for the bay scallop, Argopecten irradians concentricus (Say 1822). J Exp Mar Biol Ecol 460:100–108.
- 75. Norberg RA (1977) An ecological theory on foraging time and energetics and choice of optima food-searching method. J Anim Ecol 46:511–529.
- 76.
Stephens DW, Krebs JR (1986) Foraging theory. Princeton University Press, NJ, USA.
- 77. Malyshev A, Quijón PA (2011) Disruption of essential habitat by a coastal invader: New evidence of the effects of green crabs on eelgrass beds. ICES J Mar Sci 68:1852–1856.
- 78. Smallegange IM, van der meer J, Kurvers RHJM (2006) Disentangling interference competition from exploitative competition in a crab–bivalve system using a novel experimental approach. Oikos 113:157–167.
- 79. Quinn BK, Boudreau MR, Hamilton DJ (2012) Inter- and intraspecific interactions among green crabs (Carcinus maenas) and whelks (Nucella lapillus) foraging on blue mussels (Mytilus edulis). J Exp Mar Biol Ecol 412:117–125.
- 80.
Reise K (1995) Tidal flat ecology; an experimental approach to species interactions. Ecological Studies 54. Springer-Verlag, 194 p.
- 81. Thrush SE (1999) Complex role of predators in structuring soft-sediment macrobenthic communities: implications of changes in spatial scale for experimental studies. Aust J Ecol 24:344–354.
- 82. Quijón PA, Snelgrove PVR (2005b) Differential roles of crustacean predators in a subarctic soft-sediment system. Mar Ecol Prog Ser 285: 137–149.
- 83. Myers N (1989) Synergistic interactions and environment. BioScience 39: 506.
- 84. Steele MA (1996) Effects of predators on reef fishes: separating cage artifacts from effects of predation. J Exp Mar Biol Ecol 198: 249–267.
- 85.
Pace ML (2001) Getting it right and wrong: Extrapolations across experimental scales. In: Gardner RH, Kemp M, Kennedy K, Petersen J (Eds). Scaling relations in experimental ecology. Columbia University Press, p. 157–177.
- 86. Ambrose WG Jr (1984) Role of predatory infauna in structuring marine soft bottom communities. Mar Ecol Progr Ser 17:109–115.
- 87. Pauly D, Christensen V, Dalsgaar J, Froese R, Torres F Jr (1998) Fishing down marine food webs. Science 279:860–863. pmid:9452385
- 88. Agardi T (2000) Effects of fisheries on marine ecosystems: a conservationist’s perspective. ICES J Mar Sci 57:761–765.
- 89. Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, et al. (2001) Historical overfishing and the recent collapse of coastal ecosystems. Science 293:629–637. pmid:11474098
- 90. Davis RC, Short FT, Burdick DM (1998) Quantifying the effects of green crab damage to eelgrass transplants. Restoration Ecol 6:297–302.
- 91. Taylor PD, Fahrig L, Henein K, Merriam G (1993) Connectivity is a vital element of landscape structure. Oikos 68:571–573.
- 92. Waser AM, Splinter W, van der Meer J (2015) Indirect effects of invasive species affecting the population structure of an ecosystem engineer. Ecosphere 6:1–12.
- 93. Tummon Flynn P, Lynn KD, Cairns D, Quijón PA (2019) The role of the non-indigenous green crab (Carcinus maenas) in the decline of a unique strain of Irish moss (Chondrus crispus): direct and indirect effects. ICES J Mar Sci. https://doi.org/10.1093/icesjms/fsz130.
- 94. Molnar JL, Gamboa RL, Revenga C, Spalding MD (2008) Assessing the global threat of invasive species to marine biodiversity. Front Ecol Environ 6:485–492.
- 95. Ellison AM, Bank MS, Clinton BD et al. (17 other co-authors) (2005) Loss of foundation species: Consequences for the structure and dynamics of forested ecosystems. Front Ecol Environ 3:479–486.