Predator depletion on Cape Cod (USA) has released the herbivorous crab Sesarma reticulatum from predator control leading to the loss of cordgrass from salt marsh creek banks. After more than three decades of die-off, cordgrass is recovering at heavily damaged sites coincident with the invasion of green crabs (Carcinus maenas) into intertidal Sesarma burrows. We hypothesized that Carcinus is dependent on Sesarma burrows for refuge from physical and biotic stress in the salt marsh intertidal and reduces Sesarma functional density and herbivory through consumptive and non-consumptive effects, mediated by both visual and olfactory cues. Our results reveal that in the intertidal zone of New England salt marshes, Carcinus are burrow dependent, Carcinus reduce Sesarma functional density and herbivory in die-off areas and Sesarma exhibit a generic avoidance response to large, predatory crustaceans. These results support recent suggestions that invasive Carcinus are playing a role in the recovery of New England salt marshes and assertions that invasive species can play positive roles outside of their native ranges.
Citation: Coverdale TC, Axelman EE, Brisson CP, Young EW, Altieri AH, Bertness MD (2013) New England Salt Marsh Recovery: Opportunistic Colonization of an Invasive Species and Its Non-Consumptive Effects. PLoS ONE 8(8): e73823. https://doi.org/10.1371/journal.pone.0073823
Editor: Mary O'Connor, University of British Columbia, Canada
Received: April 11, 2013; Accepted: July 25, 2013; Published: August 29, 2013
Copyright: © 2013 Coverdale 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.
Funding: This work was funded by NSF BIO OCE-0927090. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Unchecked human population growth has threatened the persistence of natural ecosystems  by escalating extinctions , ecosystem phase shifts , habitat loss , and species invasions . Species introductions can have negative ecological impacts and, consequently, are often viewed as destructive [6,7]. Recently however, invasive species have been shown to restore lost ecological functions and promote recovery within heavily degraded ecosystems , stimulating debate on the costs and potential benefits of species outside native ranges e.g. [6,9–12], particularly in light of the extent and severity of human impacts on ecosystems.
Ecosystem recovery after anthropogenic disturbance has been documented in terrestrial , freshwater  and marine systems , but full recovery has been observed in only a third of ecological recovery studies . Invasive predators could have a particularly large impact on the recovery of degraded communities if their impact is exerted through both consumptive and non-consumptive effects . Non-consumptive effects have been hypothesized to be a more potent community structuring force than predation alone because a single predator can influence more prey through non-consumptive interactions than it can consume directly, resulting in larger community effects [17,18]. Elucidating recovery mechanisms, including the potential for invasive species to aid in recovery, is essential for informing conservation to improve management success, attain sustainable human ecosystem use, and test ecological theory [19,20].
Overexploitation of predators is one of the greatest threats to coastal marine ecosystems , so the resilience and recovery potential of ecosystems damaged by predator depletion is of considerable conservation and management importance [22,23]. The importance of predators on coral reefs  and kelp forests  is well established, but their role in salt marshes remains contentious [26,27]. Recent die-offs of salt marsh cordgrass across the western Atlantic [28–30], however, illustrate that in the absence of top predators salt marshes can be heavily damaged by herbivory. Such results suggest that, under continued predator depletion, salt marshes worldwide may become vulnerable to consumer-driven die-off .
Herbivore-driven die-offs on Cape Cod (MA), first reported in 2002, result from overgrazing by the native, nocturnal marsh crab Sesarma reticulatum on the low marsh cordgrass Spartina alterniflora [30,31], the foundation species critical for New England marsh growth and the provisioning of ecosystem services. Sesarma are common in New England, but die-off is not found in undisturbed salt marshes with robust predator populations and low Sesarma densities [32–34]. At sites with heavy recreational fishing >50% of marine predators (e.g. striped bass Morone saxatilis, blue crab Callinectes sapidus) have been removed, increasing Sesarma densities by ~400% and triggering cordgrass die-off . On Cape Cod, Sesarma-driven die-off has denuded >95% of creek banks at impacted sites and is prevalent at >90% of marshes regionally . At elevated densities, Sesarma dig communal burrow networks in denuded peat banks. Burrows can displace >65% of peat volume and large burrow complexes can contain >25 Sesarma, which rely on this refuge from predation and desiccation to persist in the marsh intertidal .
Recently, invasive European green crabs (Carcinus maenas) have colonized the intertidal zone of sites with high Sesarma densities. Although Carcinus do not dig burrows, they have been shown to use Sesarma burrows and evict resident crabs. Carcinus are >50X more common in the intertidal zones of die-off than healthy sites, where Sesarma burrow density is >5X greater . These sites lack robust predator populations, have high Sesarma densities and have experienced severe cordgrass die-off over the last ~35 years [32,33]. Recently, sites colonized by Carcinus have experienced cordgrass regrowth, suggesting that Carcinus may act as compensatory predators , restoring predation pressure lost to localized overfishing for recreationally targeted species.
Our previous work suggests that the interaction between Sesarma and Carcinus is largely dictated by a behavioral response of Sesarma to the presence of, but not predation by, Carcinus . In this paper we test the hypothesis that Carcinus opportunistically utilize Sesarma burrows for refuge and non-consumptively reduce Sesarma activity and herbivory through olfactory and visual cues. Specifically, we hypothesized that in the intertidal (1), Carcinus displace Sesarma from their burrows to avoid predation and desiccation, allowing them to remain in the intertidal during low tide (2), Carcinus play a compensatory predation role by reducing Sesarma functional density and herbivory and (3) non-consumptive interactions between Sesarma and Carcinus are mediated by olfactory and/or visual cues.
Why do Carcinus use Sesarma burrows?
To test the hypothesis that Carcinus use Sesarma burrows as a refuge habitat from desiccation and/or predation in the intertidal, we ran a fully factorial tethering experiment crossing burrow and predator exclusion at two heavily burrowed sites. Predator exclusion cages (40 x 40 x 40 cm) and burrow exclusion panels (40 x 40 cm) were constructed of 12 mm galvanized hardware cloth. Cages and burrow exclusion panels were attached to the marsh surface with garden staples to prevent access by burrowing predators and the escape of tethered Carcinus. Carcinus were tethered with 15 cm of 50 lb braided fishing line threaded between the second and third walking legs and attached to the carapace with cyanoacrylic glue. Carapace pieces attached to the tether at the end of the experiment provided evidence of predation, while dead intact Carcinus were evidence of physical stress-induced mortality. Previous tethering experiments revealed that crab behavior and survivorship are unaffected by this tethering method . Predator exclusion cages prevented predation but allowed access to Sesarma burrows, while burrow exclusion panels prevented burrow use by tethered Carcinus and allowed access to predators. Burrow densities at both sites were >115/m2 and tethered Carcinus with access to burrows immediately entered them when deployed. Tethered Carcinus were randomly assigned to one of four treatments: open (burrow access and predator exposure), burrow exclusion (hardware cloth floor preventing burrow access), predator exclusion (cage preventing predator access), and predator and burrow exclusion (n=15/treatment/site). Carcinus mortality was scored after 48 hours and analyzed with a two-factor ANOVA (caged vs. uncaged and burrow access vs. burrow exclusion).
To examine the generality of Carcinus reliance on Sesarma burrows, we surveyed creek bank Carcinus and Sesarma densities at healthy and die-off sites (n=3 sites/site type) in 2011. At each site, three replicate creek banks (10 m long, 1 m wide and 1 m deep) were surveyed for Carcinus and Sesarma. Species-specific abundances (Carcinus or Sesarma) were pooled by site. Species-specific abundance was analyzed with a one-factor ANOVA (healthy vs. die-off sites). To examine how Carcinus abundance varies temporally, sites were surveyed again in 2012. Carcinus abundance per creek bank was aligned rank transformed using ARTool  for nonparametric factorial data analysis and analyzed with ANOVA (site-type, year, and site-type*year).
Does the presence of Carcinus reduce Sesarma functional density and herbivory?
To test the hypothesis that Carcinus reduce Sesarma activity and herbivory we performed a Carcinus addition experiment at Blackfish Creek (Wellfleet, MA), a die-off site with little recovery and few naturally occurring Carcinus (7.3 ± 4.3 crabs/100 m2). We randomly selected 20 plots on creek banks with conspicuous Sesarma herbivory, separated by >4 meters. Ten plots were randomly assigned as Carcinus additions and the others assigned as unmanipulated controls. All plots had high fiddler crab (Uca pugnax) densities, so both Carcinus addition and control plots had high ambient crab activity. Carcinus placed in addition plots were of similar size to the large Carcinus used in avoidance response trials and predation experiments described below. To assess how the presence of Carcinus affects the spatial extent of Sesarma herbivory, we transplanted 3 cores (7.5 cm diameter) of cordgrass into each plot 0, 0.5 and 1.0 m from the center, parallel to the shore. In crab addition plots, a tethered Carcinus was added to the center on a 25 cm tether and provided with an artificial burrow for refuge. We checked all replicates biweekly for Carcinus survival and signs of predation on Sesarma. Carcinus were replaced as necessary throughout the experiment to ensure constant presence of live Carcinus in addition plots. We quantified Sesarma activity by sampling functional Sesarma density 0, 0.5 and 1.0 m from each plot’s center with pitfall traps . Functional densities were measured before and 24 hours after Carcinus addition to test the hypothesis that the presence of Carcinus reduces Sesarma activity. After a month, the number of stems grazed by Sesarma on each cordgrass culm was quantified to test the hypothesis that the presence of Carcinus reduces Sesarma herbivory and that this effect decreases with distance from Carcinus. Sesarma functional density and herbivory were analyzed with 2-factor ANOVAs (treatment x distance).
What cues trigger an avoidance response by Sesarma?
We performed avoidance response trials in field mesocosms to test the hypothesis that non-consumptive effects mediate interactions between Sesarma and Carcinus. Mesocosms had opaque sides and mimicked the submerged intertidal but were flat and lacked burrows to allow quantification of escape time in the absence of refugia. Trials were performed shortly after dusk because Sesarma are nocturnal crabs and leave their burrow complexes at night to forage. Mesocosms were supplied with fresh seawater for each trial to avoid the accumulation of olfactory cues. An arena was established within the mesocosm and its size (17 cm radius) was based on the average distance to the nearest burrow in field plots (9.8±0.5 cm). Sesarma (2.0±0.2 cm carapace width) were placed in the center of the mesocosm under a smaller container to allow time for habituation after which the smaller container was removed and the time for each Sesarma to move outside the arena was recorded. To examine whether the induction of Sesarma avoidance behavior is species-specific, trials were run with three similarly sized large predatory crabs, Carcinus (7.0 cm carapace width), Atlantic rock crab (Cancer irroratus; 9.6 cm carapace width) and blue crab (Callinectes sapidus; 13.2 cm carapace width), as well as two non-predatory crabs commonly found in New England marshes: the horseshoe crab (Limulus polyphemus; 7.8 cm carapace width), and spider crab (Libinia emarginata; 5.8 cm carapace width). To test whether Sesarma avoidance response is size specific, trials were run with small (3.9 cm) and large (7.0 cm) Carcinus. To test the mechanism(s) of avoidance behavior, visual and olfactory cues were isolated in separate trials. For visual trials, a large Carcinus was placed in a clear, sealed glass container visible to Sesarma; for olfactory trials, water with Carcinus effluent was released into the mesocosm prior to the insertion of Sesarma. Avoidance responses were compared against control trials where only Sesarma were placed in the mesocosm. Species-specific, Carcinus size-specific and non-consumptive mechanism trials were analyzed with one-way ANOVAs with escape time as the response variable. Data was pooled by treatment for analysis with Bonferroni corrections used to calculate experiment-wide error for avoidance response trials (α’ = 0.017).
We also tested species- and size-specific predation in field mesocosms. Species were placed within flat bottom circular (radius 9 cm) or rectangular mesocosms (42.5 x 30.2 cm) depending on trial species size, that were filled with fresh seawater, covered with hardware cloth mesh to prevent crabs from escaping, and staked into the marsh overnight. All trials included a Sesarma (1.95±0.03 cm carapace width), and either had no predatory crab, a large Carcinus (6.2±0.1 cm carapace width), small Carcinus (3.7±0.2 cm carapace width), Libinia (4.1±0.3 cm carapace width), Cancer (10.1±0.4 cm carapace width), Callinectes (13.0±0.3 cm carapace width), or Limulus (18.7±1.2 cm carapace width). Predation events were scored the following morning. Species-specific and Carcinus size-specific predation rates were analyzed with one-way ANOVAs with Sesarma mortality as the response variable.
Why do Carcinus use Sesarma burrows?
Carcinus had higher mortality when exposed to predation and/or restricted from burrows (predation effect, F1,4 = 13.70, P < 0.05; burrow effect, F1,4 = 52.07, P < 0.01; Figure 1). By restricting burrow access, exposure to physical stress alone led to higher mortality than exposure to predation, but both treatments resulted in higher mortality rates than the predator exclusion with burrow access treatment. All Carcinus that lacked burrow access had clear signs of desiccation mortality, while mortality events in replicates without cages left broken, predated carapaces. Carcinus exposed to predation without burrows experienced the highest mortality, but the interaction between exposure to predation and burrow access was not significant (F1,4 = 3.78, P = 0.12, Figure 1).
Tethered Carcinus with access to burrows to avoid desiccation and in cages to avoid predation experienced the highest survivorship, while those exposed to both stressors experienced significant mortality. These results underscore the role of Sesarma burrows as refuges from desiccation, which transform inhospitable die-off banks into benign intertidal habitats capable of sustaining large, burrow-dwelling Carcinus populations.
Recovering marshes have higher burrow densities and wider burrow complexes than healthy sites  and we found that Carcinus were >50X more common at burrowed, die-off sites than healthy sites with few Sesarma burrows (F1,4 = 7.73, P < 0.05). Sesarma density was also higher at die-off sites (F1,4 = 13.23, P = 0.02), but Sesarma and Carcinus were never found in the same burrow (Figure 2). Over two years, Carcinus abundance was again higher at die-off than healthy sites (F1,1 = 90.678, P < 0.05) and increased between 2011 and 2012 (F1,1 = 4.54, P < 0.05). There was also an interaction between site-type and year (F1,1 = 5.89, P < 0.05). Carcinus abundance increased from 2011 to 2012 at die-off sites but remained zero at healthy sites (Figure 3).
Note the order of magnitude difference in crab densities between site types. Carcinus outnumbered Sesarma at both sites, but were only common at sites with high Sesarma densities and consequently many burrow complexes and expansive die-off (bottom). Carcinus and Sesarma were never found in the same burrow and no evidence of predation was ever observed, suggesting Sesarma may exhibit a strong avoidance response to the presence of Carcinus.
Note not only the magnitude difference in Carcinus densities between site types but also the difference in abundance trends across years. At healthy sites, Carcinus remains low over both 2011 and 2012. At die-off sites, however, Carcinus increases from 2011 to 2012.
Do Carcinus reduce Sesarma functional density and herbivory?
Before Carcinus addition, functional Sesarma densities were similar at all distances in all plots (all F1,42 < 4.25, all P > 0.05). Forty eight hours after Carcinus addition, Sesarma functional density decreased in pitfall traps 0 m (F1,42 = 4.53, P < 0.05) and 0.5 m (F1,42 = 4.98, P < 0.05) from the tethered Carcinus with >3X decrease in Sesarma density at all distances (Figure 4B). Carcinus addition reduced Sesarma grazing over the duration of the experiment, an effect that decreased with distance (0 m: F1,41 = 10.06, P < 0.0068; 0.5 m: F1,42 = 0.03, P = 0.87; 1.0 m: F1,42 = 2.39, P = 0.35; Figure 4A) and was significant only in the center of experimental plots. There was no evidence of predation on Sesarma during the course of the experiment.
(A) Sesarma grazing was reduced by the presence of a single, tethered Carcinus at 0m, and (B) Sesarma density was reduced at 0 and 0.5m but there was no evidence of predation, which is commonly seen in healthy marshes. This suggests that a single, large Carcinus can reduce Sesarma functional density and herbivory without directly consuming Sesarma (* denotes significant difference at P < 0.05).
What cues trigger an avoidance response by Sesarma?
Carcinus and Callinectes were the only species to prey on Sesarma in feeding trials (Figure 5B); there was no Sesarma mortality in Sesarma only trials or in trials with Libinia, Cancer, or Limulus. Carcinus predation on Sesarma was size-specific (F2,27 = 20.52, P < 0.0001), with higher predation rates by large (6.2±0.1 cm) than by small Carcinus (3.7±0.2 cm). Sesarma also exhibited size-specific avoidance to Carcinus (F2,132 = 8.37, P = 0.0004), with large Carcinus eliciting an avoidance response ~2X faster than small Carcinus and Sesarma-only controls (Figure 5A). All Sesarma left the arena after ~11 seconds, mimicking the rapid movement of foraging Sesarma observed in nearby die-off patches. Olfactory and visual cues elicited similar escape responses (F2,134 = 7.93, P = 0.0006; Figure 5A). Avoidance responses were not limited to Carcinus: Sesarma avoided all predatory crabs (Carcinus, Callinectes and Cancer; F5,269 =3.54, P = 0.0041, Figure 5A), but common non-predatory crabs (Limulus and Libinia) did not elicit an avoidance response.
(A) Escape response trials demonstrated that Sesarma flee faster in response to large Carcinus, Carcinus visual and olfactory cues, and other similarly sized predatory decapod crustaceans. Non-predatory crustaceans, small Carcinus, and Sesarma only trials were similar in the amount of time taken to exit the arena. (B) Large Carcinus and Callinectes preyed on Sesarma in overnight feeding trials, but predation was low for small Carcinus, Cancer, Libinia, and Limulus (* denotes significant difference at P < 0.05).
Our results suggest that Carcinus colonize the intertidal at die-off marshes by using Sesarma burrows as refuges from predation and desiccation. At these sites, our results support earlier experiments suggesting that Carcinus displace Sesarma, exposing them to increased thermal stress and predation [33,36]. This displacement increases Sesarma vulnerability to native predators and reduces foraging activity through consumptive and non-consumptive effects, facilitating cordgrass recovery. These results highlight the potential for invasive species to play positive roles outside of their native range , particularly when critical ecological functions have been lost due to human impacts.
Carcinus use of Sesarma burrows
Our data suggest that Carcinus opportunistically invade the intertidal zone of salt marshes on Cape Cod with high Sesarma densities and depleted predator populations and, in the absence of native predators, are becoming numerically dominant predators at die-off marshes . Carcinus are unable to burrow in peat and are reliant on large Sesarma burrow complexes to invade marsh creek banks (Figure 2). Experimental tethering illustrated that Carcinus survival is significantly higher when given access to burrows (Figure 1), suggesting that Sesarma burrows provide Carcinus a refuge from predation and desiccation at low tide. Mud crabs (Panopeus herbstii) and Asian shore crabs (Hemigrapsus sanguineus) were also found in intertidal creek banks and were similar in size to Sesarma. Carcinus has been shown to be a superior competitor over Hemigrapsus  and laboratory feeding trials using Panopeus and Hemigrapsus (Bertness, unpublished data) have shown no evidence of predation on Sesarma, suggesting that Panopeus and Hemigrapsus likely have no impact on the interaction between Carcinus and Sesarma. Panopeus burrows are too small to be invaded by Carcinus and Hemigrapsus is not a burrowing crab, further suggesting these species have no effect on Carcinus and its dependence on Sesarma burrows for persistence. As a result, Carcinus reliance on Sesarma burrows likely explains the high density of Carcinus in the intertidal zone of marshes with severe die-off and high densities of Sesarma and their relative absence from sites without burrows (Figure 2).
Carcinus influence on Sesarma functional density and cordgrass regrowth
Historically, Sesarma densities were controlled by native marine predators such as striped bass (Morone saxatilis), blue crabs (Callinectes sapidus), and smooth dogfish (Mustelus canis). However, decades of recreational fishing have depleted local predator populations within New England salt marshes, releasing Sesarma from top-down control. Carcinus invasion at die-off sites, however, is partially restoring the predation pressure lost to recreational fishing. By inhabiting Sesarma burrow complexes, Carcinus effects on Sesarma are likely greater per capita than those of native predators which are unable to forage both in- and outside of intertidal burrows during low tide.
Our data also suggests that Carcinus reduce Sesarma activity through visual and olfactory cues (Figure 5A). The magnitude of Sesarma response to visual and olfactory cues was similar, and when presented with both stimuli simultaneously, their response was not amplified. These results, coupled with the generic response to predatory crabs exhibited in escape trials with Callinectes and Cancer, suggest that Sesarma are sensitive to visual and olfactory cues from Carcinus despite its relatively recent invasion of the Western Atlantic .
Our temporal data also illustrates that Carcinus have remained at low densities at healthy sites and, coincident with recovery, have been increasing at die-off marshes (Figure 3). Therefore, the recent regrowth of cordgrass into formerly denuded creek banks harboring burrow-dwelling Carcinus [36,40] suggests that Carcinus is playing a role in promoting the recovery of salt marshes from die-off through both consumptive and non-consumptive effects. Our Carcinus addition experiment (Figure 4) revealed that a single Carcinus is capable of reducing Sesarma activity and increasing the growth and survivorship of nearby cordgrass. While this effect is limited to <1 m, with Carcinus densities approaching 10 crabs/m3at heavily invaded sites, the consumptive and non-consumptive effects of Carcinus burrow invasion are likely strong enough to drive marsh-wide regrowth. While consumptive effects may be playing a role in the marsh recovery, we have observed few naturally predated Sesarma body parts in the intertidal at die-off sites (Coverdale, personal observation), and none were found in our Carcinus tethering experiment, suggesting that non-consumptive effects may be more prevalent. Similar non-consumptive effects have been shown to produce strong, cascading effects on rocky shores , freshwater lakes  and terrestrial grasslands . By invading burrow complexes, evicting resident Sesarma , and living within Sesarma burrows, Carcinus may also indirectly reduce Sesarma densities by enhancing the effectiveness of depleted native predators.
The restriction of the recent Carcinus colonization of intertidal creek banks to heavily burrowed marshes suggests that Sesarma burrowing facilitates compensatory predation by Carcinus, potentially creating a negative feedback loop whereby elevated Sesarma densities create conditions suitable for predator colonization. By creating a novel intertidal habitat with refuge from predation and desiccation, Sesarma burrows facilitate Carcinus invasion into the intertidal zone of predator-depleted marshes, where Carcinus suppress Sesarma activity and herbivory, promoting cordgrass regrowth and facilitating the recovery of die-off marshes [36,40]. In the absence of burrows at healthy sites, Carcinus are vulnerable to desiccation in the intertidal, suggesting that intertidal Carcinus and Sesarma population fluctuations may be linked in the future. Our results illustrate the severity of human impacts in this system by suggesting that consumptive and non-consumptive top-down control, mediated by an invasive predator, may be facilitating the recovery of heavily degraded Cape Cod salt marshes.
We would like to thank S. Smith and the Cape Cod National Seashore for access to field sites, S. Yin for field support, and Q. He and two anonymous reviewers for comments on the manuscript.
Conceived and designed the experiments: TCC AHA MDB. Performed the experiments: TCC EEA CPB EWY MDB. Analyzed the data: TCC CPB AHA. Wrote the manuscript: TCC CPB.
- 1. Schröter D, Cramer W, Leemans R, Prentice IC, Araújo MB et al. (2005) Ecosystem Service Supply and Vulnerability to Global Change in Europe. Science 310: 1333-1337. doi:https://doi.org/10.1126/science.1115233. PubMed: 16254151.
- 2. Barnosky AD, Matzke N, Tomiya S, Wogan GOU, Swartz B et al. (2011) Has the Earth’s sixth mass extinction already arrived? Nature 471: 51–57. doi:https://doi.org/10.1038/nature09678. PubMed: 21368823.
- 3. Hughes TP (1994) Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral reef. Science 265: 1547-1551. doi:https://doi.org/10.1126/science.265.5178.1547. PubMed: 17801530.
- 4. Vitousek PM (1994) Beyond global warming: ecology and global change. Ecology 75: 1861-1876. doi:https://doi.org/10.2307/1941591.
- 5. Cariton JT, Geller JB (1993) Ecological roulette: the global transport and invasion of nonindigenous marine organisms. Science 261: 78-82. doi:https://doi.org/10.1126/science.261.5117.78. PubMed: 17750551.
- 6. Davis MA, Chew MK, Hobbs RJ, Lugo AE, Ewel JJ et al. (2011) Don’t judge species on their origins. Nature 474: 153-154. doi:https://doi.org/10.1038/474153a. PubMed: 21654782.
- 7. Schlaepfer MA, Sax DF, Olden JD (2012) Toward a More Balanced View of Non-Native Species. Conserv Biol 26: 1156-1158. doi:https://doi.org/10.1111/j.1523-1739.2012.01948.x. PubMed: 23082954.
- 8. Schlaepfer MA, Sax DF, Olden JD (2011) The potential conservation value of non-native species. Conserv Biol 25: 428-437. doi:https://doi.org/10.1111/j.1523-1739.2010.01646.x. PubMed: 21342267.
- 9. Nuñez MA, Simberloff D (2005) Invasive species and the cultural keystone species concept. Ecol Soc 10.
- 10. Carroll SP (2011) Conciliation biology: the eco-evolutionary management of permanently invaded biotic systems. Evol Appl 4: 184-199. doi:https://doi.org/10.1111/j.1752-4571.2010.00180.x.
- 11. Simberloff D, Genovesi P, Pyšek P, Campbell K (2011) Recognizing conservation success. Science 332: 419. doi:https://doi.org/10.1126/science.332.6028.419-a. PubMed: 21512020.
- 12. Vitule JRS, Freire CA, Vazquez DP, Nuñez MA, Simberloff D (2012) Revisiting the potential conservation value of non-native species. Conserv Biol 26: 1153-1155. doi:https://doi.org/10.1111/j.1523-1739.2012.01950.x. PubMed: 23083005.
- 13. Broadbent EN, Zarin DJ, Asner GP, Peña-Claros M, Cooper A et al. (2006) Recovery of forest structure and spectral properties after selective logging in lowland Bolivia. Ecol Appl 16: 1148-1163. doi:https://doi.org/10.1890/1051-0761(2006)016[1148:ROFSAS]2.0.CO;2. PubMed: 16827009.
- 14. Schindler DW (1974) Eutrophication and Recovery in Experimental Lakes: Implications for Lake Management. Science 184: 897-899. doi:https://doi.org/10.1126/science.184.4139.897. PubMed: 17782381.
- 15. Worm B, Barbier EB, Beaumont N, Duffy JE, Folke C et al. (2006) Impacts of Biodiversity Loss on Ocean Ecosystem Services. Science 314: 787-790. doi:https://doi.org/10.1126/science.1132294. PubMed: 17082450.
- 16. Jones HP, Schmitz OJ (2009) Rapid recovery of damaged ecosystems. PLOS ONE 4: e5653. doi:https://doi.org/10.1371/journal.pone.0005653. PubMed: 19471645.
- 17. Trussell GC, Ewanchuk PJ, Bertness MD (2002) Field evidence of trait-mediated indirect interactions in a rocky intertidal food web. Ecol Lett 5: 241–245. doi:https://doi.org/10.1046/j.1461-0248.2002.00304.x.
- 18. Peckarsky BL, Abrams PA, Bolnick DI, Dill LM, Grabowski JH et al. (2008) Revisiting the classics: considering nonconsumptive effects in textbook examples of predator-prey interactions. Ecology 89: 2416-2425. doi:https://doi.org/10.1890/07-1131.1. PubMed: 18831163.
- 19. Folke C, Carpenter S, Walker B, Scheffer M, Elmqvist T et al. (2004) Regime shifts, resilience, and biodiversity in ecosystem management. Annu Rev Ecol Evol Syst 35: 557-581. doi:https://doi.org/10.1146/annurev.ecolsys.35.021103.105711.
- 20. Carpenter SR, Folke C, Scheffer M, Westley F (2009) Resilience: Accounting for the Noncomputable. Ecol Soc 14: 13.
- 21. Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW et al. (2001) Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629-638. doi:https://doi.org/10.1126/science.1059199. PubMed: 11474098.
- 22. Walker BH (1995) Conserving biological diversity through ecosystem resilience. Conserv Biol 9: 1-7. doi:https://doi.org/10.1046/j.1523-1739.1995.09010001.x.
- 23. Biggs R, Carpenter SR, Brock WA (2009) Turning back from the brink: Detecting an impending regime shift in time to avert it. P Natl. A Sci. U.S.A. 106: 826-831. doi:https://doi.org/10.1073/pnas.0811729106.
- 24. Mumby PJ, Dahlgren CP, Harborne AR, Kappel CV, Micheli F et al. (2004) Fishing, trophic cascades, and the process of grazing on coral reefs. Science 311: 98-101. PubMed: 16400152.
- 25. Estes JA, Tinker MT, Williams TM, Doak DF (1998) Killer whale predation on sea otters linking oceanic and nearshore ecosystems. Science 282: 473-476. doi:https://doi.org/10.1126/science.282.5388.473. PubMed: 9774274.
- 26. Odum EP (1971) Fundamentals of Ecology. Philadelphia: Saunders.
- 27. Bertness MD, Silliman BR (2008) Consumer control of salt marshes driven by human disturbance. Conserv Biol 22: 618-623. doi:https://doi.org/10.1111/j.1523-1739.2008.00962.x. PubMed: 18577090.
- 28. Silliman BR, van de Koppel J, Bertness MD, Stanton LE, Mendelssohn IA (2005) Drought, snails, and large-scale die-off of southern US salt marshes. Science 310: 1803-1806. doi:https://doi.org/10.1126/science.1118229. PubMed: 16357258.
- 29. Jefferies RL, Jano AP, Abraham KF (2006) A biotic agent promotes large-scale catastrophic change in the coastal marshes of Hudson Bay. J Ecol 94: 234-242. doi:https://doi.org/10.1111/j.1365-2745.2005.01086.x.
- 30. Holdredge C, Bertness MD, Altieri AH (2009) Role of crab herbivory in die-off of New England salt marshes. Conserv Biol 23: 672-679. doi:https://doi.org/10.1111/j.1523-1739.2008.01137.x. PubMed: 19183205.
- 31. Smith SM (2009) Multi-decadal changes in salt marshes of Cape Cod, MA: photographic analyses of vegetation loss, species shifts, and geomorphic change. Northeast Nat 16: 183-208. doi:https://doi.org/10.1656/045.016.0203.
- 32. Altieri AH, Bertness MD, Coverdale TC, Herrmann NC, Angelini C (2012) A trophic cascade triggers collapse of a salt marsh ecosystem with intensive recreational fishing. Ecology 93: 1402–1410. doi:https://doi.org/10.1890/11-1314.1. PubMed: 22834380.
- 33. Coverdale TC, Herrmann NC, Altieri AH, Bertness MD (2013) Latent impacts: the role of historical human impacts in coastal habitat loss. Front Ecol Environ 11: 69-74. doi:https://doi.org/10.1890/120130.
- 34. Coverdale TC, Bertness MD, Altieri AH (2013) Regional ontogeny of New England salt marsh die-off. Conserv Biol (. (2013)) doi:https://doi.org/10.1111/cobi.12052. PubMed: 23566036.
- 35. Coverdale TC, Altieri AH, Bertness MD (2012) Belowground herbivory increases vulnerability of New England salt marshes to die-off. Ecology 93: 2085-2094. doi:https://doi.org/10.1890/12-0010.1. PubMed: 23094380.
- 36. Bertness MD, Coverdale TC (2013) Fear of invasive species can trigger recovery of degraded ecosystems. Ecology (in press).
- 37. Wobbrock JO, Findlater L, Gergle D, Higgins JJ (2011) The Aligned Rank Transform for nonparametric factorial analyses using only ANOVA procedures. New York: ACM Press. pp. 143-146.
- 38. Macdonald JA, Roudez R, Glover T, Weis JS (2007) The invasive green crab and Japanese shore crab: behavioral interactions with a native crab species, the blue crab. Biol Invas 9: 837-848. doi:https://doi.org/10.1007/s10530-006-9085-6.
- 39. Say T (1817) An account of the Crustacea of the United States. J Phila: Acad Nat Sci. 1: 57.
- 40. Altieri AH, Bertness MD, Coverdale TC, Axelman EE, Hermann NC et al. (2013) Facilitation drives the resilience of salt marshes and rapid reversal of die-off. Ecology (in press).
- 41. Peacor SD, Werner EE (1997) Trait-mediated indirect interactions in a simple aquatic food web. Ecology 78(4): 1146-1156. doi:https://doi.org/10.1890/0012-9658(1997)078[1146:TMIIIA]2.0.CO;2.
- 42. Schmitz OJ (1998) Direct and indirect effects of predation and predation risk in old-field interaction webs. Am Nat 151: 327–342. doi:https://doi.org/10.1086/286122. PubMed: 18811324.