Pushing the Switch: Functional Responses and Prey Switching by Invasive Lionfish May Mediate Their Ecological Impact

Pushing the switch: functional responses and prey switching by invasive lionfish may mediate their ecological impact

McCard, M., South, J., Cuthbert, R. N., Dickey, J. W. E., McCard, N., & Dick, J. T. A. (2021). Pushing the switch: functional responses and prey switching by invasive lionfish may mediate their ecological impact. Biological Invasions. https://doi.org/10.1007/s10530-021-02487-7

Published in: Biological Invasions

Document Version: Publisher's PDF, also known as Version of record

Queen's University Belfast - Research Portal: Link to publication record in Queen's University Belfast Research Portal

Publisher rights Copyright 2021 the authors. This is an open access article published under a Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited.

General rights Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights.

Take down policy The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the Research Portal that you believe breaches copyright or violates any law, please contact [email protected].

Download date:28. Sep. 2021 Biol Invasions

https://doi.org/10.1007/s10530-021-02487-7 (0123456789().,-volV)( 0123456789().,-volV)

ORIGINAL PAPER

Pushing the switch: functional responses and prey switching by invasive lionﬁsh may mediate their ecological impact

Monica McCard . Josie South . Ross N. Cuthbert . James W. E. Dickey . Nathan McCard . Jaimie T. A. Dick

Received: 18 February 2020 / Accepted: 23 January 2021 Ó The Author(s) 2021

Abstract Biodiversity is declining on a global scale notorious and widespread marine invader. Functional and the spread of invasive alien species (IAS) is a responses and prey switching propensities were quan- major driver, particularly through predatory impacts. tified towards three representative prey species: Thus, effective means of assessing and predicting the Artemia salina, Palaemonetes varians, and Gammarus consequences of IAS predation on native prey popu- oceanicus. Lionfish exhibited potentially destabilising lation stability remains a vital goal for conservation. Type II FRs towards individual prey species, owing to Here, we applied two classic ecological concepts, high consumption rates at low prey densities, whilst consumer functional response (FR) and prey switch- FR magnitudes differed among prey species. Func- ing, to predict and understand the ecological impacts tional response attack rates (a) were highest, and of juveniles of the lionfish (Pterois volitans), a handling times (h) lowest, towards A. salina, followed by P. varians and then G. oceanicus. Maximum feeding rates (1/h) and functional response ratios Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/ (FRR; a/h) also followed this impact gradient for the s10530-021-02487-7.

M. McCard (&) Á J. South Á R. N. Cuthbert Á R. N. Cuthbert J. W. E. Dickey Á J. T. A. Dick GEOMAR Helmholtz Centre for Ocean Research Kiel, Institute for Global Food Security, School of Biological Du¨sternbrooker Weg 20, 24105 Kiel, Germany Sciences, Queen’s University Belfast, 19 Chlorine Gardens, Belfast, Northern Ireland BT9 5DL, UK J. W. E. Dickey e-mail: [email protected] Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB), 12587 Berlin, Germany J. South Centre for Invasion Biology, South African Institute for N. McCard Aquatic Biodiversity (SAIAB), Makhanda 6140, South Environmental Sciences, School of Science and Africa Technology, Bournemouth University, Fern Barrow, Poole, Dorset BH12 5BB, UK J. South Á R. N. Cuthbert DSI/NRF Research Chair in Inland Fisheries and Freshwater Ecology Laboratory, South African Institute for Aquatic Biodiversity (SAIAB), Makhanda 6140, South Africa

123 M. McCard et al. three prey species. Lionfish, however, displayed a actions and allocate limited resources, we need potentially population stabilising prey switching enhanced predictive capacity in invasion ecology, propensity (i.e. frequency-dependent predation) when especially to forecast potential ecological impacts multiple prey species were presented simultaneously, (Dick et al. 2017b; Cuthbert et al. 2019a, b). where disproportionately less of rare prey, and more of Predicting invasion patterns and processes could abundant prey, were consumed. Whilst FR and FRR help elucidate future risks posed by IAS towards magnitudes indicate marked per capita lionfish preda- native species (Ricciardi 2003; Simberloff 2011; tory impacts towards prey species, a strong prey 2013; Dick et al. 2014, 2017a). In turn, prediction switching propensity may reduce in-field impacts by methods must also be robust when facing combina- offering low density prey refuge in biodiverse com- tions of abiotic and biotic environmental drivers or munities. Our results thus corroborate field patterns ‘‘context-dependencies’’ (Ricciardi et al. 2003; documenting variable impacts of lionfish, with prey Alexander et al. 2014). Invasive species generally extirpations less likely in diverse communities owing demonstrate a higher resource use efficiency as to frequency-dependent predation. compared to native analogues, and this has been linked to their higher ecological impact (Morrison and Keywords Frequency-dependent predation Á Hay 2011; Dick et al. 2014, 2017a, b; Cuthbert et al. Interaction strength Á Invader impact prediction Á 2019b). Functional responses (FRs: resource con- Marine biology Á Pterois volitans sumption as a function of resource density; Holling 1959) have been used effectively to assess and predict the ecological impacts of invaders (see Dick et al. 2014; Iacarella et al. 2015; South et al. 2017; Cuthbert Introduction et al. 2019b). Three types of FR have been described. A Type I response is characterised by a lack of The introduction, establishment, and spread of inva- handling time, where consumption rate increases sive alien species (IAS) presents major threats to linearly as resource density rises (Jeschke et al. biodiversity and economies worldwide (Simberloff 2014; Hoxha et al. 2018). Hyperbolic Type II et al. 2013; Dick et al. 2017a, b; IPBES 2019; responses are characterised by high proportional Haubrock et al. 2020), and the rate at which invaders consumption at low resource densities followed by are arriving remains high owing to increasing geo- an asymptote, and are widely regarded as destabilising graphical connectivity (Seebens et al. 2018). Invasive on resource populations, such as prey species that alien species spread has been facilitated by globalised cannot escape predation when relatively rare (Mur- transport networks (Zieritz et al. 2016), and their doch and Oaten 1975; Hassell 1978). Conversely, establishment in new areas is being further aided by Type III responses, with a sigmoidal curve, are changing climates and anthropogenic alterations of deemed to be more stabilising towards prey popula- ecosystems (Didham et al. 2007; Muhlfeld et al. 2014). tions, with refuge occurring for prey at low prey Marine coastal systems have become the most heavily densities (Colton 1987; Hassell 1978). The fitting of invaded regions due to increases in aquaculture, global FR models allow parameters of interest to be estimated shipping activities and connectivity, such as the Suez that relate to predator foraging behaviour (Rogers Canal creating a corridor for invasions into the 1972; Jeschke et al. 2002). In that context, the attack Mediterranean (Briski et al. 2015; Stuer-Lauridsen rate is the scaling coefficient that corresponds to the et al. 2018). Impacts of IAS can be far-reaching, and initial slope of the curve (Hassell and May 1973), and affect ecosystems, biodiversity, the economy, food therefore predators that consume more prey at low security, and human, animal and plant health (Lowry densities should have a higher attack rate. The et al. 2013; Cuthbert et al. 2021). With numerous handling time is a second parameter of interest for examples of failed attempts at IAS eradication and impact prediction, whereby predators with a lower control (Courchamp et al. 1999; Rayner et al. 2015), handling time will reciprocally exhibit a higher consensus is being reached that management focus maximum feeding rate (FR curve height) towards should lie with prevention, rather than eradication prey (Jeschke et al. 2002). (Piria et al. 2017). However, in order to prioritise 123 Pushing the switch: functional responses

Whilst the utility of FRs in predictions of ecological Hackerott et al. 2017), and thus may benefit from impact towards single prey species have been recur- investigation with respect to per capita and prey rently displayed (see before), there has hitherto been switching propensities under key context-dependen- limited consideration for impacts towards more cies such as prey community diversity. diverse prey communities, which represents a key Juvenile lionfish are a particularly understudied context-dependency (Cuthbert et al. 2018, 2019a; IAS demographic, leading to a gap in the literature Joyce et al. 2019). Prey switching (i.e. frequency- compared to that of adults (South et al. 2017) and that dependent predation) could enhance impact predic- may contribute to the impact variability mentioned tions from comparative FRs, and is characterised by above. Adults ontogenetically switch to piscivorous disproportionately reduced consumption where prey feeding, which has implications for fisheries manage- are rare and disproportionately high consumption ment in invaded areas (Morris and Akins 2009); where prey are abundant (Murdoch 1969). A form of however, juveniles are critical due to their feeding on frequency-dependent predation, prey switching by planktonic species such as larval fish and crustaceans predators thus has the potential to stabilise prey (Cure et al. 2012). This could lead to competition populations by providing rare prey with refuge from between juvenile lionfish and native fish species for predation (Hughes and Croy 1993). Classical work prey, posing another threat, while further enhancing found prey switching behaviours in predatory fish that lionfish impacts. There is a distinct lack of quantifi- were disproportionately influenced by the abundance cation of juvenile lionfish ecological impact in marine of spatially-partitioned benthic and surface prey ecosystems, due to their ability to avoid detection and within aquatic environments (Murdoch et al. 1975). capture at small sizes. Juvenile lionfish, below 15 cm On the other hand, prey populations may be desta- are notably absent in the trophic ecology literature, bilised (i.e. extirpated) when switching is not evi- despite the well recorded size-dependent differences denced and predation is consistent towards one prey in available gut content data (for size-dependent species irrespective of environmental abundance trophic ecology at 15 cm and above, see Dahl et al. (Cuthbert et al. 2018). That is because consumption 2014; Munõz et al. 2011; Mizrahi et al. 2017; Dahl remains elevated towards a particular prey type, even et al. 2017). Regardless, even within these size classes, when that prey is rare in the environment, threatening smaller lionfish diet was composed of 62% inverte- population persistence. Thus, incorporating quantita- brates in Bacalar, Mexico (Dahl et al. 2017) and tive assessments to determine evidence for prey crustacean prey volume in lionfish gut contents drove switching, or lack thereof, for existing and emerging the highest percentage difference between small and IAS may provide important insights for the prediction large lionfish diets on hard bottomed reefs in the USA of ecological impacts, and in turn inform their (Munõz et al. 2011). management and mitigation. Given their fast growth rate, cryptic nature (Darling A notorious marine IAS is Pterois volitans, the red et al. 2011), undetectable chemical scent (Lo¨nnstedt lionfish (hereafter, lionfish), which has rapidly and McCormick 2013) and size-dependent refuge invaded the Caribbean Sea and western Atlantic from mechanical removal (Barbour et al. 2010), there Ocean (Albins and Hixon 2011). Lionfish are a is a potential that juvenile lionfish are exerting commercially-available aquarium fish species glob- undocumented ecological impact upon prey species ally, and as a result, were the first marine invaders to in their invasive ranges. However, observed lionfish establish an invasive population resulting from the impacts also seem to vary widely and there is debate as aquarium trade (Betancur et al. 2011). More recently, to their true ecological effects (Albins and Hixon an incipient Mediterranean invasion from Lessepsian 2011; Ingeman et al. 2017; Hackerott et al. 2017). migrants via the Suez Canal has occurred, with six Therefore, this presents a critical need to investigate countries in the Mediterranean basin now invaded by the predatory impact, specifically of juvenile lionfish, P. volitans and related species P. miles (Andradi- towards a variety of prey types, with a focus on Brown 2019). Their ecological impacts on prey invertebrates such as crustaceans (Chagaris et al. species can be severe, but these impacts are charac- 2017). In the present study, we thus quantify predatory terised by variability that thus far has been unex- impacts of juvenile lionfish in simple and marginally plained (Albins and Hixon 2011; Ingeman et al. 2017; more complex experimental prey communities, asking 123 M. McCard et al. whether their ecological impacts are mediated by per (Dahl et al. 2017). Palaemonentes varians represents capita effects and frequency-dependent predation palaemonid shrimp, species of which are abundant where multiple prey species occur. We thus first across the lionfish invasive range and represented determine and compare the FRs and associated widely in lionfish diets (Layman and Allgeier 2012; parameters of lionfish towards three prey species, Layman et al. 2014). In addition, their native range Artemia salinas, Palaemonetes varians and Gam- includes the North-East Atlantic and West Mediter- marus oceanicus, when individual prey species are annean (Falciai 2001) which may be subject to lionfish presented at different densities, using the classic attack invasion commensurate with range expansion under rate (a) and handling time (h), as well as the new thermal change. Gammarus oceanicus represents functional response ratio (a/h). Secondly, we examine benthic crustacean prey species, such as amphipods whether lionfish exhibit prey switching behaviour and isopods, both found in lionfish diets in their when presented with multiple prey species simultane- invasive range (Morris and Akins 2009; Ortiz et al. ously at different prey ratios, i.e. frequency-dependent 2015). Further, Gammarus oceanicus is native across predation. the North Atlantic, and whilst lionfish have not established further north than Cape Hatteras, they have been detected as far as New York. Therefore, it is Materials and methods not unlikely that with warming scenarios their invasive range will overlap with that of G. oceanicus and Animal collection and maintenance similar species (Meister et al. 2005; Grieve et al. 2016; Pinsky et al. 2020). However, we note that these prey Experiments were undertaken at Queen’s Marine do not currently overlap with lionfish distributions. Laboratory (QML), Portaferry, Northern Ireland, All prey species were purchased from Seahorse between February and September 2018. Juvenile Aquarium, Dublin, and maintained under identical lionfish (n = 14; mean total length mm ± SE: conditions to the predators in separate holding tanks 7.87 ± 1.29 cm) from Grosvenor Tropicals, Lisburn, (W: 600 9 L: 800 9 H:700, 10 L). Intraspecific prey size were kept in a holding tank (W: 12.600 9 L: was standardised throughout all trials (total length 6000 9 H:1800, 220 L) with external filtration contain- mm ± SE: A. salina 6.2 ± 0.8 mm; P. varians ing UV- and sand-filtered Strangford Lough water. 10.9 ± 0.7 mm and G. oceanicus 10.9 ± 0.8 mm). Predators used were of a similar size as the mass of Pilot trials were carried out to determine appropriate predators relative to prey can influence predatory prey densities for use during the experiments such that impact (Holling 1964). The water was changed daily the FR type and asymptote could be determined (see and maintained at 25.0 (± 1.0 °C) using an aquarium Alexander et al. 2012). We complied with all neces- heater under a natural light regime. Lionfish were sary ethical protocols sought from the School of maintained daily ad libitum on frozen anchovy to Biological Sciences ethics committee, Queen’s avoid predator learning behaviour to the focal exper- University Belfast. imental representative prey species. Feeding experiments were conducted within a glass tank (W: Functional responses 1300 9 L: 1800 9 H: 1200, 45 L) maintained at 25.0 (± 1.0 °C). All fish were allowed to acclimate to Each prey species was, separately, supplied at 16 experimental arenas for 30 min prior to densities per 45 L tank (2, 4, 6, 8, 12, 16, 20, 25, 30, 35, experimentation. 40, 45, 50, 55, 60, 70; n = 7 per prey species per Prey species used for the experiments were brine density; 0.04–1.56 ind. L-1) in a randomised pattern shrimp (Artemia salina), dwarf white shrimp (Palae- temporally to eliminate time as a confounding factor. monetes varians) and a marine gammarid (Gammarus Functional response experiments were initiated oceanicus). Prey species were chosen to be taxonom- through the addition of the allotted prey density to ically relevant but also available easily in high experimental tanks containing an individual lionfish. quantities (See Chagaris et al. 2017 for dietary Lionfish were then allowed to feed for 3 h before importance of crustaceans for lionfish). In this case, being removed for enumeration of prey consumed. In A. salina represents small pelagic crustacean prey total, 14 lionfish were used, with no lionfish used more 123 Pushing the switch: functional responses than once at any prey density and prey species to avoid consumed, we therefore fit Rogers’ random predator pseudoreplication. Reuse of individuals was, however, equation to model FRs (Rogers 1972): essential due to limited numbers of fish available (see N ¼ N ðÞð1 À exðÞ aðÞ N h À T 1Þ Alexander et al. 2014). All individual lionfish had e 0 e three days of recovery time between experiments. where Ne is the amount of prey consumed, N0 is initial Controls consisted of one replicate of each prey type prey density, a is the attack rate, h is the handling time across all densities in the absence of lionfish predators. and T is the total time available. We note that, at low prey densities, lionfish often consumed all available Prey switching prey rapidly within the 3 h experimental period, and therefore our estimations of the attack rate parameter Owing to similarities in body size and consumption (i.e. FR initial slope) are likely conservative among rates by lionfish in FR trials (see later), P. varians and prey types. Nonetheless, the random predator equation G. oceanicus were selected as prey types for the prey is robust to total prey depletion when estimating attack switching experiment. Palaemonetes varians and G. rates and handing times (Cuthbert et al. 2020). Further, oceanicus were supplied to the lionfish at nine in comparative FR analyses, it is the similarities and different prey ratios per 45 L tank (45:5, 40:10, differences that are important rather than absolute 35:15, 30:20, 25:25, 20:30, 15:35, 10:40, 5:45; n =7 values. Functional responses were then non-paramet- per prey ratio; 1.1 ind. L-1). Lionfish were then rically bootstrapped (n = 2000) to produce 95% allowed to feed for 1 h and all prey were replaced as confidence intervals using initial maximum likelihood they were consumed to maintain nominal prey species estimates of a and h. The indicator variable approach ratios throughout each replicate. Due to the logistics of outlined in Juliano (2001) was used to determine the experiment, prey were replaced manually by the differences between FR parameters (a and h) across authors. Although this process of replacing prey may prey types. The handling time parameter was then have resulted in some physical disturbance and used to determine maximum feeding rates (1/h)of increased visibility of added prey, we believe such lionfish across prey types. Furthermore, the functional effects would have been minor and well-balanced response ratio (FRR; a/h) was calculated for each prey between prey types. type to amalgamate information from these two FR parameters (Cuthbert et al. 2019b; South et al. 2019). Statistical analyses The FRR metric allows for comparative insights through synthesis of both FR parameters, resolving All statistical analyses were undertaken in R (R Core issues of which should be selected when determining Development Team 2018). Logistic regression was ecological impacts by practitioners. used to deduce FR types based on analyses of In the prey switching experiment, overall prey proportional prey consumption across prey densities, consumption was analysed using a generalised linear with ‘prey density’ included as a continuous variable mixed model (Bates et al. 2015). Errors were assumed (Pritchard et al. 2017). Here, a significantly negative to be Poisson distributed given that prey were first-order term is indicative of a Type II FR (Juliano continually replaced, and were found not to be 2001). A significantly positive first order term fol- overdispersed through analysis of residual deviance. lowed by a significant negative second order term We incorporated ‘prey species’ (2 levels) and ‘ratio’ (9 would indicate a Type III FR (Juliano 2001). To levels) as fixed effects. To account for non-indepen- further illustrate the direction and shape of propor- dence of data, each prey pair was included as a random tional consumption of prey at different prey densities, effect in the model. In other words, because both prey a locally weighted scatterplot smoothing, which had a types were presented simultaneously within the same smoothing factor of 6/10, was fit to the FR data for the experimental ‘unit’, with two separate prey mortality three prey types (Pritchard et al. 2017). This smooth- measures from each replicate, we captured this ing indicated a Type II functional response given variation as a random intercept for each experimental consumption rates decreased with increasing densi- unit, with random slopes for each prey type [i.e. ties. As prey were not replaced as they were (1 ? prey|unit)].

123 M. McCard et al.

Chesson’s selectivity index, assuming prey replace- Functional responses ment, was used to infer prey preferences for P. varians and G. oceanicus across nominal ratios (Chesson First order terms were significantly negative, indicat- 1978). Selection towards a particular prey types was ing Type II FRs by lionfish towards all prey species determined by: (Table 1; Fig. 1). Lionfish displayed the highest attack Xm ÀÁ rate towards A. salina, intermediate rates upon P. ai ¼ ðÞri=ni = rj=nj ð2Þ varians and lowest towards G. oceanicus (Table 1). j¼1 Attack rates towards A. salina were significantly higher compared to both P. varians (z = 13.56, where a is Chesson’s selectivity index for prey type i, i p \ 0.001) and G. oceanicus (z = 43.69, p \ 0.001). n is the number of prey type i available at the start of i In turn, P. varians had significantly greater attack rates the experiment, r is the number of prey type i i than G. oceanicus (z = 3.96, p \ 0.001). Lionfish consumed, m the number of prey types, r is the j handling times were shortest towards A. salina, number of prey type j consumed and n the number of j intermediate towards P. varians and longest towards prey type j available at the start of the experiment. In a G. oceanicus (Table 1). Handling times were signif- two-prey system, values of a range between 0 – 1, i icantly shorter towards A. salina than P. varians with 0.5 indicating null preference and those closer to (z = 13.93, p \ 0.001) and G. oceanicus (z = 12.72, 1 indicating increasing preference, whilst values p \ 0.001), whilst the latter two species were not closer to 0 indicate avoidance. Chesson’s indices were significantly different (z = 1.48, p = 0.14). Accord- transformed to reduce extremes (0 s, 1 s): ingly, maximum feeding rates were highest towards A. at ¼ ðÞaiðÞþn À 1 0:5 =n ð3Þ salina, whilst feeding rates towards the other two prey species were more similar (Table 1; Fig. 2). Moreover, where at is the transformed output and n is the sample FRRs were highest towards A. salina, with P. varians size (Smithson and Verkuilen 2006). Beta regression intermediate and G. oceanicus lowest, illustrating the was then used to compare between observed trans- power of FRR to give better resolution of differences formed Chesson’s indices with those expected under in per capita effects (Table 1). null preference (0.5) (Cribari-Neto and Zeileis 2010). Here, ‘observed/expected index’ (2 levels) and ‘ratio’ Prey switching (9 levels), for each reciprocated prey type were included as explanatory variables. We followed a Significantly more P. varians were consumed than G. backward stepwise deletion process in all models that oceanicus overall (v2 = 11.59, df = 1, p \ 0.001; omitted non-significant terms and interactions (Craw- Fig. 3), and prey consumption was significantly ley 2007). Likelihood ratio tests were performed to greater where a given prey species was available at derive the overall significance of effects. Where a higher proportions (v2 = 681.29, df = 8, p \ 0.001). factor was significant at the 95% confidence level, However, there was a significant ‘prey species 9 ra- Tukey tests were employed post hoc for multiple tio’ interaction (v2 = 44.36, df = 8, p \ 0.001). This pairwise comparisons (Lenth 2016). reflected greater dissimilarities in overall consumption in favour of P. varians when matched with G. oceanicus under higher proportions. Results Lionfish exhibited a strong prey switching propensity between P. varians and G. oceanicus (Fig. 3). For Across control groups for all prey species, survival both prey species, this was reflected by a significant exceeded 99% in the absence of lionfish Therefore, all interaction between ‘observed/expected index’ and mortality was assumed to be due to predation in the FR ‘proportion’, signalling that observed Chesson’s and switching experiments, and we observed preda- indices were variable as proportional availability tion events frequently. changed for the prey species (v2 = 146.64, df = 8, p \ 0.001). For both prey species, selectivity was significantly lower than expected under null preference when present at low proportions (0.1, 0.2, 0.3, 123 Pushing the switch: functional responses

Table 1 First order terms, functional response (FR) types, rounded FR parameter estimates (a, h and 1/h) with associated p values, alongside functional response ratio (FRR: a/h) estimates for all prey species treatments Prey species First order term, p FR Type Attack rate (a), p Handling time (h), p Maximum feeding rate (1/h) FRR (a/h)

A. salina - 0.11 \ 0.001 II 13.595 \ 0.001 0.017 \ 0.001 56.97 774.47 P. varians - 0.07 \ 0.001 II 8.884 \ 0.001 0.025 \ 0.001 38.97 346.24 G. oceanicus - 0.06 \ 0.001 II 5.418 \ 0.001 0.027 \ 0.001 36.94 200.17

70 1.0 A. salina A. salina P. varians P. varians G. oceanicus G. oceanicus 60 0.8 50

0.6 40

0.4 30

20 Proportion consumed of prey 0.2 consumed Number of prey

0.0 0 10203040506070 0 0 10203040506070 Initial prey density Initial prey density Fig. 1 Locally-weighted scatterplot smoothing lines fit to the proportion of prey consumed at each prey density for Artemia Fig. 2 Functional response curves for lionfish preying upon salina (blue, solid line), Palaemonetes varians (green, dotted Artemia salina (blue, solid line), Palaemonetes varians (green, line) and Gammarus oceanicus (red, dashed line) dotted line) and Gammarus oceanicus (red, dashed line). Shaded areas are 95% confidence intervals 0.4; all p \ 0.01). Under equal proportions, selectivity promise in forecasting the impacts of invasive species towards prey did not deviate significantly from that (Dick et al. 2017a; Dickey et al. 2018), whether expected (0.5; both prey species p = 0.31). When prey specific predatory effects are mediated by increasing species were presented under higher proportions, prey diversity remains largely unexplored in invasion observed selectivity was significantly higher than science (Cuthbert et al. 2018, 2019a). Here, we expected (0.6, 0.7, 0.8, 0.9; all p \ 0.01). Therefore, demonstrate that lionfish can exert substantial preda- disproportionately less of a given species was con- tion pressure on prey populations, using three repre- sumed when presented under lower proportions, sentative macroinvertebrate species, due to potentially whilst, at higher proportions, disproportionately more destabilising and high magnitude Type II predatory of a given prey species was consumed than expected, FRs. However, we also show that lionfish prey i.e. prey switching occurred (Fig. 3). switching patterns in multiple species communities may somewhat ameliorate this ecological impact by generating low density prey refuge effects, although it Discussion should be caveated that factors such as prey traits also affect lionfish predation (Green and Coˆte´ 2014). Predicting IAS ecological impacts remains a major Assessing prey consumption by lionfish through the challenge in conservation biology. While recent use of FR and prey switching approaches thus serves advances in predictive metrics have shown great

123 M. McCard et al.

1.0 1.0 P. varians experiment conducted here may be subject to similar G. oceanicus conﬁnement effects that can alter the nature of

0.8 0.8 predator–prey interactions, and particularly for in diet

in diet ambush predators such as lionﬁsh. Although, we note that the use of juveniles allowed for a relatively high 0.6 0.6 search volume in aquaria, and we stress that our

P. varians approach was a comparative laboratory-based study G. oceanicus 0.4 0.4 under standardised conditions for all organisms. Nevertheless, FR analyses and associated impact assessment metrics can be powerful indicators of per 0.2 0.2 Proportion of

Proportion of capita impacts and remain useful for comparative purposes (Dick et al. 2014, 2017b; Dickey et al. 2018). 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 Considering prey switching propensities may further Proportion of P. varians or G. oceanicus prey supplied enhance these metrics (Cuthbert et al. 2019a), and demonstrating the propensity of a predator to exhibit Fig. 3 Proportion of Palaemonetes varians (green, circle) or switching may represent a species characteristic that is Gammarus oceanicus (red, square) in the diet of lionfish as a conserved across lab and field. However, switching function of the proportion of each prey species supplied. The solid line indicates the expected values without preference studies could also be advanced further by being between the prey, while the dashed (sigmoidal) line presents a performed in the field, as has recently been proposed hypothetical switching pattern. Means are ± SE (n = 7 per prey and applied for FR studies (e.g. Novak 2010; Novak species per nominal proportion). (Color figure online) et al. 2017). Moreover, our results may have been affected by intraspecific variation given that predators as an important step to improve our knowledge and were reused in experiments; such phenomena could be predictions of the ecological impact and predatory tested using times between captures for individual capacity of this IAS. This is particularly timely as predators in future works (Coblentz and DeLong there is debate as to the reality of the ecological 2020). impacts conferred by lionfish, and emerging observa- Lionfish predation and prey vulnerability to preda- tions of large variation in such impacts (Albins and tion is mediated by prey trait combinations (Green and Hixon 2011; Ingeman et al. 2017; Hackerott et al. Coˆte´ 2014). Lionfish attack rates were significantly 2017). Whether effects found in the present study can highest towards smaller-sized A. salina compared to be extended to other prey types (e.g. fishes) and in larger-bodied P. varians and G. oceanicus, with P. adult lionfish requires further examination. varians intermediate. Active movement in the tank Lionfish displayed Type II FRs towards all three and prey size may thus be contributing factors to representative prey species when each was presented lionfish having higher attack rates on specific prey separately, which might suggest destabilising effects species, as visual fish predators find stationary prey towards prey populations due to high proportional less often than mobile prey (Uiblein et al. 1992). consumption at low prey densities (Murdoch and Furthermore, while nocturnal predators rely on chem- Oaten 1975). However, Type II FRs tend to emerge ical and hydromechanical cues (Wagner and Kro¨ger when predators are unable to switch to alternative 2000; Daghfous et al. 2012), lionfish are more reliant prey, and such switching may drive Type III FRs, on visual cues to maximise hunting success as which are more stabilising towards prey populations juveniles (Arias-Gonza´lez et al. 2011; Black et al. (see Dick et al. 2014). Our quantification of frequency- 2014; South et al. 2017). Lionfish are diurnal hunters dependent predation revealed that lionfish indeed in their invasive range, and therefore visual cues are demonstrate prey switching behaviour, which may important, indicating that reduced consumption of result in low density refugia for prey species. Func- some species supplied may be due to small size which tional responses also tend to experimentally manifest reduces detection. Future work should focus on using as Type IIs, for example, as a result of lack of habitat prey with a variety of functional traits which mediate complexity and arena size effects (Vucic-Pestic et al. vulnerability to predation in more complex settings to 2010). Equally, we caution that the prey switching more realistically represent a reef community. 123 Pushing the switch: functional responses

When feeding on G. oceanicus, lionfish had similar refuge. However, future research should ascertain handling times compared to P. varians, possibly due to whether adult lionfish stages also exhibit a switching both prey being similar in size. Whereas, when feeding propensity between fish prey, as well as whether our on A. salina, a smaller prey, the lionfish had a results hold in field-based conditions that are not significantly shorter handling time, and thus greater subject to any confinement effects. Further, the maximum feeding rate, compared to P.varians and G. potential of some predatory fish to feed on adult and oceanicus. Given that handling time increases con- juvenile lionfish and therefore exert biotic resistance currently with prey size, while attack rate is typically requires examination (Raymond et al. 2014). unimodal at intermediate relative predator–prey sizes The switching experiment shows that juvenile (Brose 2010; McCoy et al. 2011), lionfish likely lionfish may enable frequency-dependent prey consumed fewer G. oceanicus prey given the longer refuges, and this in turn may prevent local extirpations period of time required to subdue and digest such prey of prey species in biodiverse areas. While predation items (Black et al. 2014; Pusack et al. 2016; Davis was, again, significantly higher towards P. varians 2018). Our assimilation of the attack rate and handling overall, both prey species benefited from a low-density time parameters into the FRR (a/h) further demon- refuge, whereby significantly fewer prey were con- strates differential interaction strength between the sumed than expected based on their frequency in the prey species in a manner not immediately deductible environment. Indeed, Hackerott et al. (2017) showed from traditional FR curves alone (Cuthbert et al. no evidence of a negative effect on native species on 2019b; South et al. 2019). Whilst the FRR metric the Belize barrier reef by lionfish, and this may also be cannot distinguish between whether or not the attack explained by prey switching behaviours resulting in a rate or handling time drive differences in FRs, it lack of serious impact. An avoidance of rare prey in allows for a simplified metric for practitioners to more diverse environments may enable prey species discern invader impact. That is because the attack rate persistence in areas of high species richness. Simi- and handling time can often be opposing in their trends larly, the findings of Peake et al. (2018) on lionfish (e.g., one species may simultaneously have higher feeding ecology concluded they are opportunistic attack rates and higher handling times than another), generalists, consuming both vertebrate and inverte- indicating high and low impact simultaneously. Syn- brate species across many trophic guilds. These thesising these parameters thus allows for their predatory effects of lionfish may become more influence to be balanced, and resolves issues of which apparent at different densities depending on the reef parameter should be used to compare ecological diversity and habitat (Green and Coˆte´ 2014). impact. Using this method, we saw the highest FRR Switching propensities found in the present study towards A. salina, followed by P. varians and G. may also have been influenced by differential spatial oceanicus, and thus per capita effects are more clearly occupancies of prey in aquaria, with P. varians discriminated when FRR is applied, given their similar observed to be an active swimmer and G. oceanicus handling times. primarily benthic. In these cases, lionfish may have Previous studies have shown that juvenile lionfish focused their efforts on parts of the water column feed predominantly on small reef fishes and small where the most abundant prey occurred, in turn crustaceans, with a dietary shift where fish prey causing disproportionate consumption. Indeed, prey become more important with increasing lionfish size position in the water column has been found to (Valdez-Moreno et al. 2012; Mizrahi et al. 2017; Dahl facilitate prey switching in guppies, with predatory et al. 2017; Sancho et al. 2018). Therefore, there are fish disproportionately targeting either surface-dwell- implications for a potential bentho-pelagic decou- ing and benthic prey, depending on which was most pling, manifested as the exchange of energy or abundant in the aquaria (Murdoch et al. 1975). nutrients between benthic and pelagic habitats (Grif- Nonetheless, our results provide novel evidence for fiths et al. 2017). The results of the present study thus prey switching in a notorious IAS that may alleviate may impact fisheries management under invasion ecological impact should the trends be representative scenarios, as the presence of frequency-dependent of the wild. We therefore suggest further empirical predation (here, prey switching) suggests that species work to examine prey switching in more realistically rich areas could confer a degree of lionfish predation complex settings. 123 M. McCard et al.

This study shows, for the first time, that lionfish Funding MM is part funded from G & M Williams fund, exhibit prey switching when supplied with multiple Queen’s Marine Laboratory, Portaferry. JS acknowledges funding by the National Research Foundation (NRF)—South prey items. However, their high maximum feeding African Research Chairs Initiative of the Department of Science rate capacity means they have the potential to confer and Innovation (DSI)Technology (Inland Fisheries and high ecological impacts on individual prey species, Freshwater Ecology, Grant No. 110507) and the DSI-NRF even though model prey were used in the present Centre of Excellence for Invasion Biology (CIB). RNC acknowledges funding through a Research Fellowship from study. Lionfish are somewhat philopatric (Tamburello the Alexander von Humboldt Foundation. JWED is funded by and Coˆte´ 2015) and cause patch reef depletion, Inland Fisheries Ireland. severely reducing local prey populations. This could likely be affected by further abiotic contexts, such as Data availability Underlying raw data will be made available temperature and habitat complexity, as well as biotic in the Supplementary Material. factors such as lionfish learning behaviours regarding Compliance with ethical standards food (South et al. 2017; DeRoy et al. 2020a, b). While FR experiments alone are useful in ecological impact Conflict of interest The authors declare that they have no quantifications, when used in tandem with prey conflicts of interest. switching experiments, they can better inform predic- Consent for participate The authors declare that they consent tions by offering insight into interaction strengths to participate. within species-rich communities. This study shows the potential to predict and understand the dynamics of Consent for publication The authors declare that they consent to the manuscript being published. invasion impact by combining these methods, which may help management develop strategies to mitigate Ethics approval Ethical approval for work with lionfish was further impacts of lionfish on native biota (Malpica- granted by the School of Biological Sciences Animal Research Cruz et al. 2016). Effective fisheries management and Ethics Committee, Queen’s University Belfast. protected areas promote biodiverse and functionally Open Access This article is licensed under a Creative Com- diverse reef communities which increases the relative mons Attribution 4.0 International License, which permits use, abundance of a variety of prey species, and potentially sharing, adaptation, distribution and reproduction in any med- enables prey switching. If this is twinned with ium or format, as long as you give appropriate credit to the intensive mechanical removal, the abundance of original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The lionfish may be decreased, thus lowering total impact. images or other third party material in this article are included in We thus predict that lionfish impacts are marked in the article’s Creative Commons licence, unless indicated simple communities, but their effects are alleviated by otherwise in a credit line to the material. If material is not prey switching when multiple resources are available. included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds Additional laboratory and field experiments should be the permitted use, you will need to obtain permission directly considered alongside comparative FR approaches to from the copyright holder. To view a copy of this licence, visit capture effects of diverse prey assemblages and aid http://creativecommons.org/licenses/by/4.0/. future IAS management strategies. Furthermore, FR and prey switching methods could be combined to further inform impact predictions, by simultaneously altering prey density and prey species richness. References

Acknowledgements We extend thanks to Mrs Emma Healey Albins M, Hixon M (2011) Worst case scenario: potential long- for assistance with experimental set-up. term effects of invasive predatory lionﬁsh (Pterois volitans) on Atlantic and Caribbean coral-reef communities. Environ Biol Fishes 96:1151–1157 Authors’ contribution MM and JTAD conceived the study. Alexander ME, Dick JTA, O’Connor NE, Haddaway NR, MM and NM performed the experiments. MM, JS and RNC Farnsworth KD (2012) Functional responses of the inter- conducted statistical analyses. MM wrote the ﬁrst draft and all tidal amphipod Echinogammarus marinus: effects of prey authors contributed to revised versions and gave approval for supply, model selection and habitat complexity. Mar Ecol submission. Prog Ser 468:191–202. https://doi.org/10.3354/meps09978 Alexander ME, Dick JTA, Weyl OLF, Robinson T, Richardson D (2014) Existing and emerging high impact invasive

123 Pushing the switch: functional responses

species are characterized by higher functional responses invaded Atlantic coral reefs. Mar Ecol Prog Ser than natives. Biol Let 10:20130946. https://doi.org/10. 467:181–192. https://doi.org/10.3354/meps09942 1098/rsbl.2013.0946 Cuthbert RN, Dickey JWE, McMorrow C, Laverty C, Dick JTA Andradi-Brown DA (2019) Invasive Lionfish (Pterois volitans (2018) Resistance is futile: lack of predator switching and a and P: miles): Distribution, Impact, and Management. preference for native prey predict the success of an invasive Mesophot Coral Ecosyst. https://doi.org/10.1007/978-3- prey species. R Soc Open Sci 5:180–339. https://doi.org/ 319-92735-0_48 10.1098/rsos.180339 Arias-Gonza´lez JE, Gonza´lez-Gańdara C, Luis Cabrera J, Cuthbert RN, Callaghan A, Dick JTA (2019a) Biotic resistance Christensen V (2011) Predicted impact of the invasive from native predators predicts mosquito invasion success lionfish Pterois volitans on the food web of a Caribbean and informs biocontrol strategies. Sci Rep 9:15314. https:// coral reef. Environ Res 111:917–925. https://doi.org/10. doi.org/10.1038/s41598-019-51705-9 1016/j.envres.2011.07.008 Cuthbert RN, Dickey JWE, Coughlan NE, Joyce PWS, Dick Barbour AB, Montgomery ML, Adamson AA, Diaz-Ferguson JTA (2019b) The Functional Response Ratio (FRR): E, Silliman BR (2010) Mangrove use by the invasive advancing comparative metrics for predicting the ecolog- lionfish (Pterois volitans). Mar Ecol Progress Ser ical impacts of invasive alien species. Biol Invas 401:291–294. https:// doi.org/https://doi.org/10.3354/ 21:2543–2547. https://doi.org/10.1007/s10530-019- meps08373. 02002-z Bates D, Maechler M, Bolker BM, Walker SC (2015) Fitting Cuthbert RN, Wasserman RJ, Dalu T, Kaiser H, Weyl OLF, linear mixed-effects models using lme4. J Stat Softw Dick JTA, Sentis A, McCoy MW, Alexander ME (2020) 67:1–48 Influence of intra- and interspecific variation in predator– Betancur-R R, Hines A, Acero PA, Ortı´ G, Wilbur AE, Fresh- prey body size ratios on trophic interaction strengths. Ecol water DW (2011) Reconstructing the lionfish invasion: Evol 10:5946–5962. https://doi.org/10.1002/ece3.6332 insights into Greater Caribbean biogeography. J Biogeogr Cuthbert RN, Pattison Z, Taylor NG, Verbrugge L, Diagne C, 38:1281–1293. https://doi.org/10.1111/j.1365-2699.2011. Ahmed DA, Leroy B, Angulo E, Briski E, Capinha C, 02496.x Catford JA, Dalu T, Essl F, Gozlan RE, Haubrock PJ, Black AN, Weimann SR, Imhoff VE, Richter ML, Itzkowitz M Kourantidou M, Kramer AM, Renault D, Wasserman RJ, (2014) A differential prey response to invasive lionfish, Courchamp F (2021) Global economic costs of aquatic Pterois volitans: prey naivete´ and risk-sensitive courtship. invasive alien species. Sci Tot Environ (in press) J Exp Mar Biol Ecol 460:1–7. https://doi.org/10.1016/j. Daghfous G, Green WW, Zielinski BS, Dubuc R (2012) Che- jembe.2014.06.002 mosensory-induced motor behaviours in fish. Curr Opin Briski E, Gollasch S, David M, Linley RD, Casas-Monroy O, Neurobiol 22:223–230. https://doi.org/10.1016/j.conb. Rajakaruna H, Bailey SA (2015) Combining ballast water 2011.10.009 exchange and treatment to maximize prevention of species Dahl KA, Patterson WF (2014) Habitat-specific density and diet introductions to freshwater ecosystems. Environ Sci of rapidly expanding invasive Red Lionfish, Pterois voli- Technol 49:1 tans, populations in the northern Gulf of Mexico. PLoS Brose U (2010) Body-mass constraints on foraging behaviour ONE 9:e105852. https://doi.org/10.1371/journal.pone. determine population and food-web dynamics. Funct Ecol 0105852 24:28–34. https://doi.org/10.1111/j.1365-2435.2009. Dahl KA, Patterson WF, Robertson A, Ortmann A (2017) DNA 01618.x barcoding significantly improves resolution of invasive Chagaris D, Binion-Rock S, Bogdanoff A, Dahl K, Granneman lionfish diet in the Northern Gulf of Mexico. Biol Invas J, Harris H, Mohan J, Rudd MB, Swenarton MK, Ahrens R, 19:1917–1933. https://doi.org/10.1007/s10530-017-1407- Patterson WF III, Morris JA Jr, Allen M (2017) An 3 ecosystem-based approach to evaluating impacts and Darling E, Green S, O’Leary J, Coˆte´ IM (2011) Indo-Pacific management of invasive lionfish. Fisheries 42:421–431. lionfish are larger and more abundant on invaded reefs: a https://doi.org/10.1080/03632415.2017.1340273 comparison of Kenyan and Bahamian lionfish populations. Chesson J (1978) Measuring preference in selective predation. Biol Invasions 13:2045–2051. https://doi.org/10.1007/ Ecology 59:211–215. https://doi.org/10.2307/1936364 s10530-011-0020-0 Coblentz KE, DeLong JP (2020). Estimating predator functional Davis A (2018) Differential effects of native vs. invasive responses using the times between prey captures. https:// predators on a common Caribbean reef fish. Environ Biol doi.org/10.1101/2020.07.19.208686 Fishes 101:1537–1548. https://doi.org/10.1007/s10641- Courchamp F, Clutton-Brock T, Grenfell B (1999) Inverse 018-0798-z density dependence and the Allee effect. Trends Ecol Evol DeRoy E, Hussey NE, MacIsaac HJ (2020a) Behaviourally- 14:405–410. https://doi.org/10.1016/s0169- mediated learning ability in an invasive marine fish. Biol 5347(99)01683-3 Invasions 22:3357–3369. https://doi.org/10.1007/s10530- Crawley MJ (2007) The R book. John Wiley, Chichester, 020-02329-y pp 27–32 DeRoy E, Scott N, Hussey NE, MacIsaac HJ (2020b) Density Cribari-Neto F, Zeilis A (2010) Beta regression in R. J Stat dependence mediates the ecological impact of an invasive Softw 34:1–24 fish. Divers Distrib 26:867–880. https://doi.org/10.1111/ Cure K, Benkwitt CE, Kindinger TL, Pickering EA, Pusack TJ, ddi.13063 McIlwain JL, Hixon MA (2012) Comparative behaviour of Dick JTA, Alexander ME, Jeschke JM, Ricciardi A, MacIsaac red lionfish Pterois volitans on native Pacific versus HJ, Robinson TB, Kumschick S, Weyl OLF, Dunn AM, 123 M. McCard et al.

Hatcher MJ, Paterson RA, Farnsworth KD, Richardson Essl F, Golivets M, Kirichenko N, Kourantidou M, Leroy DM (2014) Advancing impact prediction and hypothesis B, Renault D, Taylor N, Verbrugge L, Courchamp F testing in invasion ecology using a comparative functional (2020). Economic costs of invasive alien species across response approach. Biol Invas 16:735–753. https://doi.org/ Europe. NeoBiota (in press) 10.1007/s10530-013-0550-8 Holling CS (1959) Some characteristics of simple types of Dick JTA, Alexander ME, Ricciardi A, Laverty C, Downey PO, predation and parasitism. Can Entomol 91:385–398. Xu M, Jeschke JM, Saul WC, Hill MP, Wasserman RJ, https://doi.org/10.4039/Ent91385-7 Barrios-O’Neill D, Weyl OLF, Shaw RH (2017a) Func- Holling CS (1964) The analysis of complex population pro- tional responses can unify invasion ecology. Biol Invas cesses. Can Entomol 96:335–347. https://doi.org/10.4039/ 19:1667–1672. https://doi.org/10.1007/s10530-016-1355- Ent96335-1 3 Hoxha T, Crookes S, Lejeusne C, Dick JTA, Chang X, Dick JTA, Laverty C, Lennon JJ, Barrios-O’Neill D, Mensink Bouchemousse S, Cuthbert RN, MacIsaac HJ (2018) PJ, Britton R, Me´doc V, Boets P, Alexander ME, Taylor Comparative feeding rates of native and invasive ascidians. NG, Dunn AM, Hatcher MJ, Rosewarne PJ, Crookes S, Mar Pollut Bull 135:1067–1071. https://doi.org/10.1016/j. MacIsaac HJ, Xu M, Ricciardi A, Wasserman RJ, Ellender marpolbul.2018.08.039 BR, Weyl OLF, Lucy FE, Banks PB, Dodd JA, MacNeil C, Hughes RN, Croy MI (1993) An experimental analysis of fre- Penk MR, Aldridge DC, Caffrey JM (2017b) Invader quency-dependent predation (switching) in the 15-spined Relative Impact Potential: a new metric to understand and stickleback, Spinachia spinachia. J Anim Ecol predict the ecological impacts of existing, emerging and 62:341–352. https://doi.org/10.2307/5365 future invasive alien species. J Appl Ecol 54:1259–1267. Iacarella JC, Dick JTA, Alexander ME, Ricciardi A (2015) https://doi.org/10.1111/1365-2664.12849 Ecological impacts of invasive alien species along tem- Dickey JWE, Cuthbert RN, Rea M, Laverty C, Crane K, South J, perature gradients: testing the role of environmental Briski E, Chang X, Coughlan NE, MacIsaac HJ, Ricciardi matching. Ecol Appl 25:706–716. https://doi.org/10.1890/ A, Riddell GE, Xu M, Dick JTA (2018) Assessing the 14-0545.1 relative potential ecological impacts and invasion risks of Ingeman, K. E., Albins, M. A., Benkwitt, C. E., Green, J., emerging and future invasive alien species. NeoBiota Kindinger, T. L., Tuttle, L. J. and Hixon, M. A. (2017). 40:1–24. https://doi.org/10.3897/neobiota.40.28519 Resolving differences in observed impacts of invasive Didham RK, Tylianakis JM, Gemmeli NJ, Rand TA, Ewers RM lionfish and clarifying advice to managers. Peer Journal 5, (2007) Interactive effects of habitat modification and spe- 34–55. https:/doi.org/https://doi.org/10.7287/peerj. cies invasion on native species decline. Trends Ecol Evol preprints.3455v1 22:489–496. https://doi.org/10.1016/j.tree.2007.07.001 IPBES (2019). Summary for policymakers of the global Falciai L (2001) Occurrence of Palaemonetes varians (Leach, assessment report on biodiversity and ecosystem services 1814) (Decapoda, Palaemonidae) in a brackish pond in of the Intergovernmental Science-Policy Platform on Algeria. Crustaceana 74:697–701 Biodiversity and Ecosystem Services, 86–101. Green SJ, Coˆte´ IM (2014) Trait-based diet selection: prey Jeschke JM, Kopp M, Tollrian R (2002) Predator functional behaviour and morphology predict vulnerability to preda- responses: discriminating between handling and digesting tion in reef fish communities. J Anim Ecol 83:1451–1460. prey. Ecol Monogr 72:95–112. https://doi.org/10.1890/ https://doi.org/10.1111/1365-2656.12250 0012-9615(2002)072[0095:PFRDBH]2.0.CO;2 Grieve BD, Curchitser EN, Rykaczewski RR (2016) Range Jeschke JM, Bacher S, Blackburn TM, Dick JTA, Essl F, Evans expansion of the invasive lionfish in the Northwest Atlantic T, Gaertner M, Hulme PE, Kuhn I, Mrugala A, Pergl J, with climate change. Mar Ecol Prog Ser 546:225–237. Pysek P, Rabitsch W, Ricciardi A, Richardson DM, Sendek https://doi.org/10.3354/meps11638 A, Vila M, Winter M, Kumschick S (2014) Defining the Griffiths JR, Kadin M, Nascimento FJA, Tamelander T, To¨rn- impact of non-native species. Conserv Biol 28:1180–1194. roos A, Bonaglia S, Bonsdorff E, Bru¨chert V, Ga˚rdmark A, https://doi.org/10.1111/cobi.12299 Ja¨rnstro¨m M, Kotta J, Lindegren M, Nordstro¨m M, Norkko Joyce PWS, Dickey JWE, Cuthbert RN, Dick JTA, Kregting L A, Olsson J, Weigel B, Zˇ ydelis R, Blenckner T, Niiranen S, (2019) Using functional responses and prey switching to Winder M (2017) The importance of benthic-pelagic cou- quantify invasion success of the Pacific oyster, Crassostrea pling for marine ecosystem functioning in a changing gigas. Mar Environ Res 145:66–72. https://doi.org/10. world. Glob Change Biol 23:2179–2196. https://doi.org/ 1016/j.marenvres.2019.02.010 10.1111/gcb.13642 Juliano SA (2001) Nonlinear curve fitting predation and func- Hackerott S, Valdivia A, Cox CE, Silbiger NJ, Bruno JF (2017) tional response curves. In: Scheiner S, Gurevitch J (eds) Invasive lionfish had no measurable effect on prey fish Design and analysis of ecological experiments. Oxford community structure across the Belizean Barrier Reef. Peer University Press, Oxford, pp 178–196 J 5:3270. https:/doi.org/https://doi.org/10.7717/peerj.3270 Layman CA, Allgeier JE (2012) Characterizing trophic ecology Hassell MP (1978) The dynamics of arthropod predator-prey of generalist consumers: a case study of the invasive systems. Princeton University Press, Princeton, NJ lionfish in The Bahamas. Mar Ecol Prog Ser 448:131–141. Hassell MP, May RM (1973) Stability in insect host-parasite https://doi.org/10.3354/meps09511 models. J Anim Ecol 42:693-726. https:/doi.org/https:// Layman CA, Jud ZR, Nichols P (2014) Lionfish alter benthic doi.org/10.2307/3133 invertebrate assemblages in patch habitats of a subtropical Haubrock PJ, Turbelin AJ, Cuthbert RN, Novoa A, Angulo E, estuary. Mar Biol 161:2179–2182 Ballesteros-Mejia L, Bodey TW, Capinha C, Diagne C, 123 Pushing the switch: functional responses

Lenth R (2016) Least-squares means: the R Package lsmeans. Novak M, Wolf C, Coblentz KE, Shepard ID (2017) Quantify- J Stat Softw 69:1–33. https://doi.org/https://doi.org/10. ing predator dependence in the functional response of 18637/jss.v069.i01 generalist predators. Ecol Lett 20:761–769. https://doi.org/ Lo¨nnstedt OM, McCormick MI (2013) Ultimate predators: 10.1111/ele.12777 lionfish have evolved to circumvent prey risk assessment Ortiz M, Rodriguez-Zaragoza F, Hermosillo-Nun˜ez B, JordańF abilities. PLoS ONE 8:757–781. https://doi.org/10.1371/ (2015) Control strategy scenarios for the alien lionfish journal.pone.0075781 Pterois volitans in Chinchorro Bank (Mexican Caribbean): Lowry E, Rollinson EJ, Laybourn AJ, Scott TE, Aiello-Lam- Based on semi-quantitative loop analysis. PLoS ONE mens ME, Gray SM, Mickley J, Gurevitch J (2013) Bio- 10:e0130261. https://doi.org/10.1371/journal.pone. logical invasions: a field synopsis, systematic review, and 0130261 database of the literature. Ecol Evol 3:1835–1835. https:// Peake J, Bogdanoff A, Layman C, Castillo B, Reale-Munroe K, doi.org/10.1002/ece3.431 Chapman J, Dahl K, Patterson W III, Eddy C, Ellis R, Malpica-Cruz L, Chaves LCT, Coˆte´ IM (2016) Managing Faletti M, Higgs N, Johnston M, Munõz R, Sandel V, marine invasive species through public participation: Villasenor-Derbez J, Morris J (2018) Feeding ecology of Lionfish derbies as a case study. Mar Policy 74:158–164. invasive lionfish (Pterois volitans and Pterois miles) in the https://doi.org/10.1016/j.marpol.2016.09.027 temperate and tropical western Atlantic. Biol Invas McCoy MW, Bolker BM, Warkentin KM, Vonesh JR (2011) 20:2567–2597. https://doi.org/10.1007/s10530-018-1720- Predicting Predation through Prey Ontogeny Using Size- 5 Dependent Functional Response Models. Am Nat Pinsky ML, Selden RL, Kitchel ZJ (2020) Climate-driven shifts 177:752–766. https://doi.org/10.1086/659950 in marine species ranges: Scaling from organisms to Meister HS, Wyanski DM, Loefer JK, Ross SW, Quattrini AM, communities. Ann Rev Mar Sci 12:153–179. https://doi. Sulak KJ (2005) Further evidence for the invasion and org/10.1146/annurev-marine-010419-010916 establishment of Pterois volitans (Teleostei: Scorpaenidae) Piria M, Copp GH, Dick JTA, Duplic´ A, Groom Q, Jelic´ D, Lucy along the Atlantic Coast of the United States. Southeast Nat FE, Roy HE, Sarat E, Simonovic´ P, Tomljanovic´ T, Tri- 4:193–206 carico E, Weinlander M, Ada´mek Z (2017) Tackling Mizrahi M, Chapman JK, Gough CLA, Humber F (2017) invasive alien species in Europe II: threats and opportuni- Management implications of the influence of biological ties until 2020. Manag Biol Invas 8:273–286. https://doi. variability of invasive lionfish diet in Belize. Aquat Invas org/10.3391/mbi.2017.8.3.02 8:61–70. https://doi.org/10.3391/mbi.2017.8.1.06 Pritchard DW, Paterson RA, Bovy HC, Barrios-O’Neill D Morris JA, Akins JL (2009) Feeding ecology of invasive lionfish (2017) Frair: an R package for fitting and comparing con- (Pterois volitans) in the Bahamian archipelago. Environ sumer functional responses. Methods Ecol Evol Biol Fishes 86:389–398. https://doi.org/10.1007/s10641- 8:1528–1534. https://doi.org/10.1111/2041-210X.12784 009-9538-8 Pusack T, Benkwitt CE, Cure K, Kindinger TL (2016) Invasive Morrison WE, Hay ME (2011) Herbivore Preference for Native Red Lionfish (Pterois volitans) grow faster in the Atlantic vs. Exotic Plants: Generalist Herbivores from Multiple Ocean than in their native Pacific range. Environ Biol Continents Prefer Exotic Plants That Are Evolutionarily Fishes 99:571–579. https://doi.org/10.1007/s10641-016- Naı¨ve. PLoS ONE 6:72–77. https://doi.org/10.1371/ 0499-4 journal.pone.0017227 R Core Development Team (2018) R: a language and environ- Muhlfeld CC, Kovach RP, Jones LA, Al-Chokhachy R, Boyer ment for statistical computing. R Foundation for Statistical MC, Leary RF, Lowe WH, Luikart G, Allendorf FW (2014) Computing, Vienna, Austria Invasive hybridization in a threatened species is acceler- Raymond WW, Albins MA, Pusack T (2014) Competitive ated by climate change. Nat Clim Chang 4:620–624. interactions for shelter between invasive Pacific red lion- https://doi.org/10.1038/nclimate2252 fish and native Nassau grouper. Environ Biol Fishes Munõz RC, Currin CA, Whitfield PE (2011) Diet of invasive 98:57–65. https://doi.org/10.1007/s10641-014-0236-9 lionfish on hard bottom reefs of the Southeast USA: Rayner MJ, Gaskin CP, Fitzgerald NB, Baird KA, Berg MM, insights from stomach contents and stable isotopes. Mar Boyle D, Joyce L, Landers TJ, Loh GG, Maturin S, Perri- Ecol Prog Ser 432:181–193. https://doi.org/10.3354/ men L, Scofield RP, Simm J, Southey I, Taylor GA, Ten- meps09154 nyson AJD, Robertson BC, Young M, Walle R, Ismar SMH Murdoch WW (1969) Switching in general predators: experi- (2015) Using miniaturised radio telemetry to discover the ments on predator specificity and stability of prey popu- breeding grounds of the endangered New Zealand storm lations. Ecol Monogr 39:335–354. https://doi.org/10.2307/ petrel Fregetta maoriana. IBIS 157:754–766. https://doi. 1942352 org/10.1111/ibi.12287 Murdoch WW, Oaten A (1975) Predation and population sta- Ricciardi A (2003) Predicting the impacts of an introduced bility. Adv Ecol Res 9:1–131. https://doi.org/10.1016/ species from its invasion history: an empirical approach S0065-2504(08)60288-3 applied to zebra mussel invasions. Freshw Biol Murdoch WW, Avery A, Smyth MEB (1975) Switching in 48:972–981. https://doi.org/10.1046/j.1365-2427.2003. predatory fish. Ecology 56:1094–1105. https://doi.org/10. 01071.x 2307/1936149 Rogers D (1972) Random search and insect population models. Novak M (2010) Estimating interaction strengths in nature: J Anim Ecol 41:369–383. https://doi.org/10.2307/3474 experimental support for an observational approach. Sancho G, Kingsley-Smith PR, Morris JA, Toline CA, McDo- Ecology 91:2394–2405. https://doi.org/10.1890/09-0275.1 nough V, Doty SM (2018) Invasive Lionfish (Pterois 123 M. McCard et al.

volitans/miles) feeding ecology in Biscayne National Park, invasive crayfish. NeoBiota 52:9–24. https://doi.org/10. Florida, USA. Biol Invas 20:2343–2361. https://doi.org/10. 3897/neobiota.52.39245 1007/s10530-018-1705-4 Stuer-Lauridsen F, Drillet G, Hansen FT, Saunders J (2018) Seebens H, Blackburn TM, Dyer EE, Genovesi P, Hulme PE, Same Risk Area: An area-based approach for the man- Jeschke JM, Pagad S, Pysˇek P, van Kleunen M, Winter M, agement of bio-invasion risks from ships’ ballast water. Ansong M, Arianoutsou M, Bacher S, Blasius B, Brock- Mar Policy 97:147–155. https://doi.org/10.1016/j.marpol. erhoff EG, Brundu G, Capinha C, Causton CE, Celesti- 2018.05.009 Grapow L, Dawson W, Dullinger S, Economo EP, Fuentes Tamburello N, Coˆte´ IM (2015) Movement ecology of Indo- N, Gueńard B, Ja¨ger H, Kartesz J, Kenis M, Ku¨hn I, Len- Pacific lionfish on Caribbean coral reefs and its implica- zner B, Liebhold AM, Mosena A, Moser D, Nentwig W, tions for invasion dynamics. Biol Invas 17:1639–1653. Nishino M, Pearman D, Pergl J, Rabitsch W, Rojas-San- https://doi.org/10.1007/s10530-014-0822-y doval J, Roques A, Rorke S, Rossinelli S, Roy HE, Scalera Uiblein F, Eberstaller J, Packl M, Winkler H (1992) Effects of R, Schindler S, Sˇtajerova´ K, Tokarska-Guzik B, Walker K, differential prey mobility on the foraging behaviour of a Ward DF, Yamanaka T, Essl F (2018) Global rise in cyprinid fish, Vimba elongata. Ethol Ecol Evol 4:293–297 emerging alien species results from increased accessibility Valdez-Moreno M, Quintal-Lizama C, Go´mez-Lozano R, Gar- of new source pools. Proc Natl Acad Sci 111:2263–2274. cıá-Rivas MC (2012) Monitoring an alien invasion: DNA https://doi.org/10.1073/pnas.1719429115 barcoding and the identification of lionfish and their prey Simberloff D (2011) How common are invasion-induced on coral reefs of the Mexican Caribbean. PLoS ONE ecosystem impacts? Biol Invas 13:1255–1268. https://doi. 7:36–66. https://doi.org/10.1371/journal.pone.0036636 org/10.1007/s10530-011-9956-3 Vucic-Pestic O, Birkhofer K, Rall BC, Scheu S, Brose U (2010) Simberloff D, Martin JL, Genovesi P, Maris V, Wardle DA, Habitat structure and prey aggregation determine the Aronson J, Courchame F, Galil B, Garcia-Berthou E, functional response in a soil predator–prey interaction. Pascal M, Pysek P, Sousa R, Tabacchi E, Vila M (2013) Pedobiologia 53:307–312. https://doi.org/10.1016/j. Impacts of biological invasions: what’s what and the way pedobi.2010.02.003 forward. Trends Ecol Evol 28:58–66. https://doi.org/10. Wagner HJ, Kro¨ger RHH (2000) Effects of long-term spectral 1016/j.tree.2012.07.013 deprivation on the morphological organization of the outer Smithson M, Verkuilen J (2006) A better lemon squeezer: retina of the blue acara (Aequidens pulcher). Philos Trans Maximum-likelihood regression with beta-distributed R Soc Lond Ser B-Biol Sci 355:1249–1252. https://doi.org/ dependent variables. Psychol Methods 11:54–71. https:// 10.1098/rstb.2000.0677 doi.org/10.1037/1082-989X.11.1.54 Zieritz A, Lopes-Lima M, Bogan AE, Sousa R, Walton S, Rahim South J, Dick JTA, McCard M, Barrios-O’Neill D, Anton A KA, Wilson JJ, Ng PY, Froufe E, McGowan S (2016) (2017) Predicting predatory impact of juvenile invasive Factors driving changes in freshwater mussel (Bivalvia, lionfish (Pterois volitans) on a crustacean prey using unionida) diversity and distribution in Peninsular Malay- functional response analysis: effects of temperature, habi- sia. Sci Total Environ 571:1069–1078. https://doi.org/10. tat complexity and light regimes. Environ Biol Fishes 1016/j.scitotenv.2016.07.098 100:1155–1165. https://doi.org/10.1007/s10641-017- 0633-y Publisher’s Note Springer Nature remains neutral with South J, McCard M, Khosa D, Mofu L, Madzivanzira TC, Dick regard to jurisdictional claims in published maps and JTA, Weyl OLF (2019) The Effect of prey identity and institutional affiliations. substrate type on the functional response of a globally

123