Essay Beyond Competition: Incorporating Positive Interactions between Species to Predict Ecosystem Invasibility

Fabio Bulleri*, John F. Bruno, Lisandro Benedetti-Cecchi

ne of the many unintended to become established. Predictions spread of exotic species. For example, consequences of global from this model are, however, based our expectations of how species Ocommerce has been the on the assumption that natural richness and resource availability affect translocation of countless plants and communities are largely structured by invasibility can be dramatically altered animals to new regions, continents, competitive interactions and that the when positive effects of extant species and oceans [1,2]. Such “exotic” species effects of native species on invaders are (including natives and established have colonized nearly every habitat predominantly negative. exotics) on exotic invaders are taken on Earth, and modern ecosystems There is, however, growing into account. are now made up largely of species evidence that facilitation (positive The potential role of facilitation originating from geographically species interactions—see Glossary) in modifying the diversity–invasibility distinct regions [3–5]. Most exotic plays an equally important role in relationship can be illustrated by species have negligible or no negative shaping communities and ecosystems constructing a series of simple effects, but a small handful have had [16–20]. One species can facilitate models that relate species richness to substantial impacts on native species another by ameliorating stressful invasibility under different scenarios and ecosystem processes [3,6]. For abiotic conditions or by providing of community assembly. Two basic example, the introduction of the refuges from natural enemies assumptions of the models are: (1) Nile perch (Lates niloticus) into Lake such as predators. Nonetheless, that the relationship between species Victoria has not only caused the positive species interactions are richness and resource availability is extinction of two-thirds of the endemic rarely incorporated into conceptual negative [25,26] and (2) that the fish fauna, but has changed the entire ecological theories that describe the probability that the native assemblage food web of the lake by reducing the complex dynamics of species invasions includes facilitators is positively grazing by phytoplanktivores [7,8]. [19,21]. Facilitation has been included correlated with extant species richness Given the sizable ecological and in invasion scenarios to describe the [13,15,19,27,28]. Since invasion economic costs of species invasions case of extant exotic species enhancing success will be greater when an exotic [9], understanding the environmental the colonization of new exotics (e.g., species does not have to compete factors that regulate them has become invasional meltdown [22]). Yet a large with residents for resources, any a major goal for basic and applied body of evidence from terrestrial and factor causing a temporary increase ecologists. One major research theme marine habitats indicates that native in resource availability will increase a is the investigation of the relationship species also commonly facilitate community’s vulnerability to invasion between native species richness (the exotic colonizers through a variety of number of local native species) and mechanisms. For example, shading the ability of exotic species to colonize by the native shrub, Atriplex vesicaria, Citation: Bulleri F, Bruno JF, Benedetti-Cecchi L (2008) Beyond competition: Incorporating positive and thrive in new habitats (termed fosters the establishment of the exotic interactions between species to predict ecosystem community “invasibility”) [10,11]. A succulent, Orbea variegata, in South invasibility. PLoS Biol 6(6): e162. doi:10.1371/journal. longstanding concept in ecology is that Australia [23], while native sessile pbio.0060162 habitats with high levels of diversity invertebrates protect the introduced Copyright: © 2008 Bulleri et al. This is an open-access are difficult to invade (the biotic oyster, Crassostrea gigas, from predation article distributed under the terms of the Creative Commons Attribution License, which permits resistance hypothesis—see Glossary) on the rocky shores of Western Canada unrestricted use, distribution, and reproduction in [11–15]. This is because, in theory, a [24]. any medium, provided the original author and source more diverse assemblage of plants or are credited. animals can utilize resources more fully Including Facilitation in Resource- Abbreviations: HEER, high efficiency in exploiting than a less diverse community, thus resources Based Invasion Theory increasing the intensity of competition Fabio Bulleri and Lisandro Benedetti-Cecchi are in and making it harder for new species Incorporating facilitation into the Department of Biology, Università di Pisa, Pisa, ecological theories that can be applied Italy. John F. Bruno is in the Department of Marine to species invasions could advance Sciences, University of North Carolina, Chapel Hill, North Carolina, United States of America. our understanding of the processes The Essay section contains opinion pieces on topics underlying the colonization and * To whom correspondence should be addressed. of broad interest to a general medical audience. E-mail: [email protected]

PLoS Biology | www.plosbiology.org 1136 June 2008 | Volume 6 | Issue 6 | e162 When HEER traits are not represented in the pool of natives, each new species added (with a constant number of individuals) will use a similar amount of resources, resulting in a linear decay with increasing species richness (R1). For example, Hooper and Vitousek [25] found a linear relationship between plant functional group richness and resource use (nitrogen, phosphorus) when nitrogen-fixers were excluded. When HEER traits are uniformly distributed across the native species pool, a small subset of natives can almost completely deplete resources (R2). The same scenario can take place when HEER traits are not distributed uniformly across native species, but are an exclusive characteristic of common or early successional species. Resources are quickly exploited at low diversity in this scenario. The opposite situation—that is, an almost complete use of resources at high levels of species richness (R3)— occurs when HEER traits are possessed by rare or late successional native species. There are several examples in the ecological literature of both patterns of distribution of HEER traits among species [25,29]. In contrast to resource depletion modalities, the shape of the curves describing the probability of including facilitators in the native assemblage as a function of species richness is yet to be determined empirically. Nonetheless, different curves can be drawn for heuristic purposes according to the distribution of the relevant “facilitating traits,” which doi:10.1371/journal.pbio.0060162.g001 ascribe facilitating effects to native species, and to assembly rules (F Figure 1. The Interplay between Resource Availability and Facilitation Regulates Invasibility curves; Figure 1B). If these traits are According to both the distribution of HEER and facilitating traits within the native species pool and realistic assembly rules, different curves can be identified to describe how (A) the availability uniformly distributed across the pool of resources and (B) the probability of exotics species being facilitated by natives vary as functions of natives, the probability of including of native species richness (see text for details). Invasibility varies with native species diversity facilitators will follow a linear increase as a net outcome of two processes: resource depletion (red line) and facilitation (blue line). A combination of R and F curves (C–H) determines the mismatch (upward arrows) between at increasing levels of species richness predictions of invasibility as a plain function of resource availability (dotted line) and predictions (F1). For example, on Mediterranean that account for facilitation (solid black line). rocky reefs, both encrusting and turf- forming algae facilitate the anchoring of stolons of the exotic alga, Caulerpa (the fluctuating resource hypothesis— richness, can be identified according racemosa, by providing a more complex see Glossary) [26]. Near-complete to (1) the occurrence of functional substratum than bare rock [30,31]. exploitation of resources can, however, traits (see Glossary) that confer high Greatest invasion success occurs, occur in both species-rich and species- efficiency in exploiting resources therefore, at low (encrusting corallines poor assemblages, so that invasibility (HEER); (2) the distribution of HEER only) to intermediate (encrusting is not necessarily related to species traits across the pool of native species; corallines plus algal turfs) levels of richness [26]. and (3) realistic assembly rules (i.e., native species/functional richness. In fact, different modalities of the frequency with which a species Conversely, when facilitating traits resource depletion (R curves; Figure occurs or its order of appearance/ are not uniformly distributed across 1A), as a function of native species disappearance in disturbed habitats). native species and are not possessed by

PLoS Biology | www.plosbiology.org 1137 June 2008 | Volume 6 | Issue 6 | e162 dominant species or early colonizers Glossary in disturbed habitats, the probability The biotic resistance hypothesis, formulated by Elton in 1958 [12], predicts that of including facilitators will be greater introduced species often fail to invade communities because strong biotic interactions at high species richness (F2). There is with native species hinder their establishment and spread. It builds on the assumption some empirical evidence for this case that more diverse resident communities, generating more biomass and using resources as well. For example, the establishment more completely, would resist the establishment of invaders. Such effects could be due of weeds in California coastal prairies is either to complementarity in the use of resources among species or to the enhanced enhanced by the native nitrogen-fixing probability of including highly competitive species (strong resource users) at high shrub, [32]; this shrub Lupinus arboreus diversity levels (identity or sampling effect; [13–15]). is not a dominant component of native Facilitation is a biotic interaction in which at least one of the species involved communities and is more likely to be benefits from the presence of the other(s), and neither is negatively affected. Facilitation part of the native pool when species includes interactions between co-evolved, mutually obligate organisms as well as richness is high. facultative interactions between species that are not evolutionarily linked. The presence Finally, when facilitating traits are of one species can facilitate another directly, by improving environmental conditions possessed by few dominant or habitat- (e.g., reducing stress due to physical and/or chemical conditions), or indirectly, by forming species, the probability of lessening consumer and/or competition pressure. facilitation will be sustained across the The fluctuating resource hypothesis, developed by Davis et al. [26], predicts that entire range of species richness (F3). pronounced fluctuations in resource availability will foster community invasibility. Some exotic and native species rely on The theory is based on the assumption that an invading species must have access to the presence of species-specific traits available resources (e.g., light, nutrients, water for plants) and that a species will have within the resident community [33– greater success in invading a community if it does not encounter intense competition 35]. For example, in the alpine zone of for these resources from resident species. An increase in resource availability can occur the Chilean Andes, the establishment either because the rate at which resources are supplied from external sources is faster of the exotic forb, , Taraxacum officinale than the rate at which the resident assemblage can use them, or because the resident depends on the presence of cushions assemblage’s use of resources declines. formed by the native plant, Azorella Functional traits are defined as the characteristics of an organism that determine [36]. monantha its performance in response to the environment and/or its effects on ecosystem Including positive effects of natives functioning. Variation between individuals or species in traits such as phenology, on exotic species drastically modifies architecture, resource acquisition, and allocation will influence the success of a predictions based on resource population or community. Community structure can be simplified by categorizing depletion (Figure 1C–1H; only F1 and species into functional groups based on suites of correlated traits. F2 curves are illustrated for the sake of brevity), since facilitation of exotics by natives can counterbalance the effects of competition. Facilitation can richness [34,37,38], and large-scale large-sized species. Alternatively, cause invasibility to deviate from the observational studies [34,39–41], which competition could inherently near-universal prediction of a decline have frequently found the opposite in operate at a smaller spatial scale than with increasing species richness nature. A positive correlation between facilitation, and thus be more likely to (Figure 1C and 1D). The distribution native and exotic diversity could arise drive the results of small-scale studies. of facilitating traits across the native at large spatial scales because the species pool determines the species response of both native and exotic From Theory to Practice: richness level at which invasibility species to heterogeneity in abiotic Implications for Management deviates from linearity. Even when factors at such scales overwhelms Taking into account positive native– resources are monopolized by a small the positive effects of diversity on exotic relationships has important number of species or functional invasion resistance that prevail at implications for intervention strategies groups, invasibility can be sustained by smaller scales [27,42,43]. Alternatively, targeting biological invasions. Such facilitation (Figure 1E) and, indeed, this positive correlation could be strategies are based on different rise at high species richness (Figure due to facilitation of exotics by approaches: direct eradication of 1F). Also, facilitation can boost natives [19,28]. Facilitation is a scale- the invader (by means of either invasibility at intermediate (Figure 1G) dependent process, because the larger biological or mechanical/chemical to high (Figure 1H) levels of species the area over which the observation/ tools) or deliberate modification of richness, when high native species manipulation is conducted, the larger physical and biological features of richness insignificantly reduces the the number of native species (and the receiving system [44–46]. The availability of resources. potential facilitating traits) that are feasibility of the physical elimination How does this conceptual model included. Scant experimental evidence of an invader is independent of the relate to our current understanding for a positive native–exotic species attributes of resident assemblages of the biodiversity–invasibility relationship could be, therefore, (although eradication techniques may relationship? Conflicting results due to the fact that the spatial scales not be). In contrast, actions commonly have emerged between small-scale at which biodiversity manipulations prescribed to control invaders by experimental studies, which have are generally carried out are too targeting physical and biological typically found a negative relationship small to sample most of the native features of natural systems, such as the between native and exotic species species/functional traits or to include manipulation of disturbance regimes

PLoS Biology | www.plosbiology.org 1138 June 2008 | Volume 6 | Issue 6 | e162 (e.g., fire, grazing, mowing) and species, enabling the assessment of the 19. Bruno JF, Stachowicz JJ, Bertness MD (2003) Inclusion of facilitation into ecological theory. nutrient availability or the restoration counterbalance between positive and Trends Ecol Evol 18: 119-125. of native species richness, are negative interactions. 20. Valiente-Banuet A, Rumebe AV, Verdú M, commonly grounded in resource-based Callaway RM (2006) Modern quaternary Acknowledgments lineages promotes diversity through facilitation invasion theory [2,46,47]. They do not of ancient Tertiary lineages. Proc Natl Acad Sci take into account the dual nature of U S A 103: 16812-16817. We wish to thank M. A. Davis and D. Tilman species interactions, and might yield 21. Brooker RW, Maestre FT, Callaway RM, Lortie for providing advice at different stages of CL, Cavieres LA, et al. (2008) Facilitation in unanticipated surprises. For example, this work, and I. Bertocci, J. A. Commito, plant communities: The past, the present, and the use of ecosystem engineers [48] E. Maggi, and three anonymous reviewers the future. J Ecol 96: 18-34. 22. Simberloff D, Von Holle B (1999). Positive (e.g., species that create or modify for insightful comments on previous drafts interactions of nonindigenous species: habitats) or facilitation in general of the paper. This is publication number Invasional meltdown? Biol Inv 1: 21-32. [49] in restoration practises, while MPS-08033 of the MarBEF Network of 23. Lenz TI, Facelli JM (2003) Shade facilitates Excellence. an invasive stem succulent in a chenopod enhancing the recovery of targeted shrubland in South Australia. Aust Ecol 28: native species and, likely, overall References 480-490. biodiversity, could unintentionally 24. Ruesink JL (2007) Biotic resistance and 1. Jenkins PT (1996) Free trade and exotic facilitation of a non-native oyster on rocky create new opportunities for invaders species introductions. Conserv Biol 10: 300-302. shores. Mar Ecol Prog Ser 331: 1-9. [50]. This could be the case with 2. Mack RN, Simberloff D, Lonsdale WM, Evans 25. Hooper DU, Vitousek PM (1998) Effects of H, Clout M, et al. (2000) Biotic invasions: plant composition and diversity on nutrient the native eelgrass, Zostera marina, Causes, epidemiology, global consequences, cycling. Ecol Monogr 61: 128-149. which enhances the establishment of and control. Ecol Appl 10: 689-710. 26. Davis MA, Grime JP, Thompson K (2000) Sargassum muticum on soft-sediments 3. Lonsdale WM (1999) Global patterns of plant Fluctuating resources in plant communities: invasions and the concept of invasibility. A general theory of invasibility. J Ecol 88: 524- by trapping drifting fragments of the Ecology 80: 1522-1536. 534. invasive macroalga [51]. Hence, the 4. McKinney JL, Lockwood ML (1999) Biotic 27. Levine JM, D’Antonio CM (1999) Elton restoration of seagrass meadows, while homogenization: A few winners replacing many revisited: A review of evidence linking diversity losers in the next mass extinction. Trends Ecol and invasibility. Oikos 87:14-26. benefiting a large number of native Evol 14: 450-453. 28. Fridley JD, Brown RL, Bruno JF (2004) Null species [52], could also foster invasion. 5. Byrnes JE, Reynolds PL, Stachowicz JJ (2007) models of exotic invasion and scale-dependent Invasions and extinctions reshape coastal patterns of native and exotic species richness. Considering positive interactions marine food webs. PLoS ONE 2(3): e295. Ecology 85: 3215-3222. among species within the same trophic doi:10.1371/journal.pone.0000295 29. Arenas F, Sánchez I, Hawkins SJ, Jenkins level can clearly alter our expectations 6. Bruno JF, Fridley JD, Bromberg K, Bertness SR (2006) The invasibility of marine algal MD (2005) Insights into ecology, evolution assemblages: Role of functional diversity and of the role of native species richness and biogeography. In: Sax DF, Gaines SD, identity. Ecology 87: 2851-2861. in determining the success of exotic Stachowicz JJ, editors. Species invasions. 30. Ceccherelli G, Piazzi L, Balata D (2002) Spread plants and animals. The next step will Sunderland (MA): Sinauer Associates. pp. 13- of introduced Caulerpa species in macroalgal 40. habitats. J Exp Mar Biol Ecol 280: 1-11. be to incorporate multiple trophic 7. Witte F, Goldschmidt T, Goudswaard PC, 31. Bulleri F, Benedetti-Cecchi L (2008) levels and consumptive interactions. Ligtvoet W, van Oijen MJP, et al. (1992) Facilitation of the introduced green alga, Species extinction and concomittant ecological Caulerpa racemosa, by resident algal turfs: Taking into account herbivory changes in Lake Victoria. Neth J Zool 42: 214- experimental evaluation of underlying and/or predation would generate a 232. mechanisms. Mar Ecol Prog Ser. In press. multidimensional model; this could 8. Goldschmidt T, Witte F, Wanink J (1993) 32. Maron JL, Connors PG (1996) A native Cascading effects of the introduced Nile Perch nitrogen-fixing shrub facilitates weed invasion. likely occur as our understanding on the detritivorous/phytoplanktivorous Oecologia 105: 302-312. of the mechanisms regulating species in the sublittoral areas of Lake Victoria. 33. Callaway RM (1998) Are positive interactions interactions among species at different Cons Biol 7: 686-700. species-specific? Oikos 82: 202-207. 9. Pimentel D, Lach L, Zuniga R, Morrison D 34. Levine JM (2000) Species diversity and trophic levels advances. The effects of (2000) Environmental and economic costs of biological invasions: Relating local process to resource availability and facilitation nonindigenous species in the United States. community pattern. Science 288: 852-854. Bioscience 50: 53-65. 35. Levin PS, Coyer JA, Petrik R, Good TP (2002) on the ability of an exotic species to 10. Tilman D (1997) Community invasibility, Community-wide effects of nonidigenous become established within a recipient recruitment limitation, and grassland species on temperate rocky reefs. Ecology 83: community would be, in fact, modified biodiversity. Ecology 78: 81-92. 3182-3193. 11. Vitousek PM, Mooney HA, Lubchenco J, 36. Cavieres LA, Quiroz CL, Molina-Montenegro by the outcome of a complex web Melillo JM (1997) Human domination of MA (2008) Facilitation of the non-native of direct and indirect interactions, Earth’s ecosystems. Science 277: 494-499. Taraxacum officinale by native nurse cushion varying in direction and strength. 12. Elton CS (1958) The ecology of invasions by species in the high Andes of central Chile; Are animals and plants. London: Methuen. 181 p. there differences between nurses? Funct Ecol But given the prominent role of 13. Huston MA (1997) Hidden treatments in 22: 148-156. facilitation in mitigating consumer ecological experiments: Re-evaluating the 37. Kennedy TA, Naeem S, Howe KM, Knops JMH, ecosystem function of biodiversity. Oecologia Tilman D, et al. (2002) Biodiversity as a barrier effects on prey populations, a broad 108: 449-460. to ecological invasion. Nature 417: 636-638. view would likely bolster the realized 14. Loreau M (2000) Biodiversity and ecosystem 38. Stachowicz JJ, Fried H, Osman RW, Whitlatch net role of positive interactions and functioning: Recent theoretical advances. RB (2002) Biodiversity, invasion resistance, Oikos 91: 3-17. and marine ecosystem function: Reconciling further modify diversity–invasibility 15. Fargione JE, Tilman D (2005) Diversity pattern and process. Ecology 83: 2575-2590. relationships. decreases invasion via both sampling and 39. Smith MD, Knapp A (1999) Exotic plant Predictions of future invasion complementarity effects. Ecol Lett 8: 604- species in a C4-dominated grassland: 611. Invasibility, disturbance, and community scenarios and management strategies 16. Bertness MD, Callaway RM (1994) Positive structure. Oecologia 120: 605-612. based on a single side of the coin interactions in communities. Trends Ecol Evol 40. Stohlgren TJ, Binkley D, Chong GW, Klakhan 9: 91-193. MA, Schell LD, et al. (1999) Exotic plant (negative interactions) will yield 17. Callaway R (1995) Positive interactions among species invade hot spots of native plant limited predictive power and problem- plants. Bot Rev 61: 306-349. diversity. Ecol Monogr 69: 25-46. solving capability. A unified theory of 18. Richardson DM, Allsopp N, D’Antonio 41. Stohlgren TJ, Bernett TD, Kaertesz JT (2003) CM, Milton SJ, Rejmánek M (2000) Plant The rich get richer: Patterns of plant invasion invasibility must, therefore, include the invasions—The role of mutualisms. Biol Rev in the United States. Front Ecol Environ 1: 11- attributes of both native and invading 75: 65-93. 14.

PLoS Biology | www.plosbiology.org 1139 June 2008 | Volume 6 | Issue 6 | e162 42. Shea K, Chesson P (2002) Community ecology 46. Bakker JD, Wilson SD (2004) Using ecological 50. Badano EI, Villaroel E, Bustamante RO, theory as a framework for biological invasions. restoration to constrain biological invasion. J Marquet PA, Cavieres LA (2007) Ecosystem Trends Ecol Evol 17:170-176. Appl Ecol 41: 1058-1064. engineers facilitate invasions by exotic 43. Davies KF, Chesson P, Harrison S, Inouye 47. Richardson DM, Bonds WJ (1991) plants in high-Andean ecosystems. J Ecol 95: BD, Melbourne BA, et al. (2005) Spatial Determinants of plant distribution: Evidence 682-688. heterogeneity explains the scale dependence of from pine invasions. Am Nat 137: 639-669. 51. Tweedley JR, Jackson EL, Attrill MJ (2008) the native-exotic diversity relationship. Ecology 48. Byers JE, Cuddington K, Jones CG, Talley Zostera marina seagrass beds enhance the 86: 1602-1610. TS, Hastings A, et al. (2006) Using ecosystem attachment of the invasive alga Sargassum 44. Thomas MB, Willis AJ (1998) Biocontrol— engineers to restore ecological systems. Trends muticum in soft sediments. Mar Ecol Prog Ser Risky but necessary? Trends Ecol Evol 13: 325- Ecol Evol 21: 493-500. 354: 305-309. 329. 49. Halpern BS, Silliman BR, Olden JD, Bruno JP, 52. Orth RJ, Carruthers TJB, Dennison WC, Duarte 45. Myers JH, Simberloff D, Kuris AM, Carey JR Bertness MD (2007) Incorporating positive CM, Fourqurean JW, et al. (2006). A global (1998) Eradication revisited: Dealing with interactions in aquatic restoration and crisis for seagrass ecosystems. Bioscience 56: exotic species. Trends Ecol Evol 15: 316-320. conservation. Front Ecol Environ 5: 153-160. 987-996.

PLoS Biology | www.plosbiology.org 1140 June 2008 | Volume 6 | Issue 6 | e162