Patterns of association of native and exotic boring polychaetes on the southeastern Pacific coast of Chile: the combined importance of negative, positive and random interactions

Introduction

Studies on biological invasions have mainly focused on the importance of negative interactions between non-indigenous species and native biota (e.g., Elton, 1958; Seabloom et al., 2003; Olyarnik et al., 2009). According to the biotic resistance hypothesis (e.g., Catford, Jansson & Nilsson, 2009), native biota can deter exotic species invasion, through competition and predation as local mechanisms (Elton, 1958; Shea & Chesson, 2002; Byers & Noonburg, 2003; Olyarnik et al., 2009). However, biotic resistance is frequently evaluated as the relationship between native and exotic species diversities (i.e., diversity resistance hypothesis), where a negative relationship is interpreted as biotic resistance of the native community to invasion success (Elton, 1958; Kennedy et al., 2002; Herben et al., 2004; Fargione & Tilman, 2005; Fridley et al., 2007). Nevertheless, apparent contradictions in empirical studies have led researchers to discuss the importance of other interactions, such as the influence of facilitation on population- and community-level variables (Bruno, Stachowicz & Bertness, 2003).

Negative relationships may be observed at the local scale, due to negative biotic interactions, while at a regional scale the overall relationship may be positive, since regional heterogeneity in abiotic conditions may present more niche opportunities for both native and exotic species richness (Shea & Chesson, 2002; Byers & Noonburg, 2003). Particularly in cases where negative interactions do not totally exclude exotics, biotic resistance may act to constrain the abundance of established exotics (Levine, Adler & Yelenik, 2004; Olyarnik et al., 2009). On the other hand, biotic resistance can also be explained by differentiation in the occupation of niches or fitness between native and invasive species (Byun, De Blois & Brisson, 2017). This is inferred from the classical theory of competitive interactions, where limits in niche overlap, or the use of resources between native species and invasive species, determine the results of the interspecific interaction (MacArthur & Levins, 1967; Weltzin et al., 2003). According to this theory, an invading species will not establish where a resident species occupies a similar niche or has similar traits. Therefore, when there is niche overlap, the fitness difference between the resident species and the invader determines which one will be competitively excluded (MacDougall, Gilbert & Levine, 2009). This idea implicitly assumes that all resident species have negative effects on invader success, and that only the negative interactions and abiotic factors can deplete populations and remove species, which largely ignores the effect of other ecological interactions, or the possibility of no (or incomplete) exclusion. In fact, it has been suggested that the coexistence of native and invasive species depends on the number and strength of simultaneous negative interactions (Mills, Rader & Belk, 2004). Moreover, the competitive exclusion assumption is unlikely to be true in natural communities given that species that repel invaders are as common as the species that facilitate their colonization (Henriksson et al., 2016), and exclusion could occur if only one limiting resource exists or if a single species is a superior competitor for several resources (e.g., Seabloom et al., 2003). Consequently, the coexistence of native and invasive species could be a frequent result in natural communities; changing the general question of how do native species exclude exotics (biotic resistance hypothesis), for the complementary question of how do native and exotic species coexist? The first intuitive answer to the latter question is: a predominance of positive species interactions (i.e., facilitation) can allow native and exotic species to coexist (Bruno, Stachowicz & Bertness, 2003). However, there is a second answer that is in agreement with both the competitive exclusion principle (Gause, 1934) and the biotic resistance hypothesis, which proposes that these species coexist due to weak competitive interactions (e.g., Liao, Bogaert & Nijs, 2015). Furthermore, a third hypothesis suggests that competitive intransitivity networks (i.e., non-linear hierarchical competition, in which there is no single best competitor) allow for species co-existence (Soliveres et al., 2015; Ulrich, Jabot & Gotelli, 2017), even under modest levels of intransitivity (Laird & Schamp, 2006).

Species are either positively, negatively, or randomly associated with one another, according to their niches (Sfenthourakis, Tzanatos & Giokas, 2006; Gotelli & Ulrich, 2010; Veech, 2013). Therefore, coexistence can result from three potential outcomes of native and exotic species’ interactions: (1) facilitation through positive species associations (Hypothesis 1); (2) weak competitive or neutral interactions through weak negative species associations with hierarchical competition (Hypothesis 2); and (3) if competitive intransitivity is promoting native and exotic species coexistence the ecological guild will show significantly strong negative species associations, with non-hierarchical competition (i.e., non transitivity or competitive networks, where species are ambiguous in their order of competitive abilities) (Hypothesis 3). We evaluated each of these hypotheses using as a study model the guild of native and exotic boring polychaetes present in calcareous hosts on the southeastern Pacific coast of Chile.

Evaluating these hypotheses is especially important for the southeastern Pacific coast, where competitive exclusion (i.e., the biotic resistance hypothesis) has been proposed as one of the principal explanations for low invader success (Castilla & Neill, 2009). Preliminary analysis of invasion patterns on the southeastern Pacific coast suggests that the number and impact of exotic species is lower than on other shores in both the northern and southern hemispheres (Castilla & Neill, 2009). Nevertheless, deliberate introduction of species for aquaculture is on the rise (e.g., Camus, 2005; Gozlan, 2017), and currently there are land-based and water-based aquaculture facilities in both northern and southern Chile. This is worrisome for the southeastern Pacific coast considering that, globally, aquaculture activities are considered an important gateway and reservoir for exotic species (Naylor, Williams & Strong, 2001). Boring polychaetes are common epibionts on these mollusks, as well as on other invertebrate species with calcareous shells, both in culture facilities and natural environments (Moreno, Neill & Rozbaczylo, 2006; Diez, Radashevsky & Orensanz, 2011). A review by Moreno, Neill & Rozbaczylo (2006) reports nine species of boring polychaetes present in calcareous substrata along the coast of Chile, of which at least six species were non-indigenous. These researchers report that the relationships between boring polychaete species and their hosts are not highly specific, in accordance with the general pattern for shell-boring polychaetes (Simon & Sato-Okoshi, 2015), with several examples of native and exotic boring polychaete species coexisting and utilizing multiple hosts. Although boring polychaetes utilize the same type of habitat, potential ecological interactions between exotic and native polychaetes are unknown and represent an interesting study model to evaluate whether the apparently low success of invasive species on the southeastern Pacific coast is due to negative interactions. The aim of this study is to evaluate the importance of positive, negative, or random species associations to explain the coexistence of native and exotic boring polychaetes inhabiting invertebrate hosts on the southeastern Pacific coast of Chile.

Materials & Methods

Boring polychaete sampling, processing, and quantification

Boring polychaetes were obtained from mollusk hosts present in and near aquaculture centers at three sites in the northern region and three sites in the southern region along southeastern coast of Chile (Fig. 1). Native host shells were collected by hand while hookah diving (i.e., a diving system where air is supplied from a hose attached to an air compressor located on land or on a boat) or SCUBA diving, and transported to the laboratory where they were preserved in 10% formalin for subsequent boring polychaete extraction. Exotic host shells (Crassostrea gigas Thunberg, 1793) were obtained from aquaculture centers present at each of the sites, and a total of 784 shells of 19 species were collected and processed. To extract the boring polychaetes each individual shell was chemically treated, with initial submersion in 5% nitric acid (HNO₃) for a period of 12 h, and later neutralized with 5% anhydrous sodium sulfate (NH₂SO₄) for an additional 12 h. This process dissolves the calcite crystals of the shell without damage to the boring polychaetes, which were then separated by hand under a dissecting microscope for identification and quantification. Species identification was made by drawing diagnostic structures using a camera lucida installed on a stereoscopic microscope (Leica model M3Z) and using keys by Gravier (1908), Fauchald (1977), Rozbaczylo (1980), Rozbaczylo (1985) and Fitzhugh & Rouse (1999). Identification was facilitated by bibliographic information on native and exotic boring polychaetes from different sources, especially for Spionidae polychaetes (Rozbaczylo, Schmiede & Sánchez, 1980; Rozbaczylo, Méndez & Bravo, 1994; Basilio, Cañete & Rozbaczylo, 1995; Sato-Okoshi & Takatsuka, 2001; Cárdenas & Cañete, 2004; Radashevsky & Olivares, 2005; Bertrán, Vargas & Quijón, 2005; Moreno, Neill & Rozbaczylo, 2006) and Cirratulidae polychaetes of the genus Dodecaceria (Carrasco, 1977; Rozbaczylo & Carrasco, 1996; Moreno, Neill & Rozbaczylo, 2006). Boring polychaete species were classified as native or exotic according to the identification and records published by Orensanz et al. (2002), Castilla et al. (2005), Moreno, Neill & Rozbaczylo (2006) and Çinar (2012).

Results

Abundance

We observed a significant positive relationship between exotic and native polychaete abundances at both local and regional scales (Fig. 3), indicating a coexistence trend. The transitive vs. intransitive competition analyses performed on Transitivity software support, at both local and regional scales, indicated the presence of a transitive network competition structure (i.e., All hosts and sites τ_p = 0.917, Mean = 0.884, limits = 0.583–1.000, Prop <1 = 0.410; All hosts from the Southern area τ_p = 0.667, Mean = 0.864, limits = 0.583–1.000, Prop < 1 = 0.452; and All hosts from Northern area τ_p = 0.750, Mean = 0.869, limits = 0.667–1.000, Prop < 1 = 0.308). In all of these cases the dominant eigenvector supports a transitive competition structure, where P. rickettsi is the dominant species at all sites (and in the northern region). At the regional scale D. huelma emerges as the dominant species in the southern region (Table 2).

Discussion

Our results suggest that native and invasive boring polychaetes species coexist through weak competitive interactions (Hypothesis 2), together with transitive hierarchical competition (Table 2). This means that the boring polychate species present a hierarchical order in their competitive abilities, without evidence of positive species interactions (i.e., facilitation), allowing native and invasive species to coexist, as has been proposed by Bruno, Stachowicz & Bertness (2003). Our evidence of the importance of weak transitive competitive interactions allowing coexistence suggests that in many native communities where invasive species arrive, the expectation of competitive exclusion (Gause, 1934) may not be found because environmental resource availability prevents strong competition. From this perspective, weak competition, is a plausible alternative to competitive intransitivity networks for explaining species co-existence (Laird & Schamp, 2006; Soliveres et al., 2015; Ulrich, Jabot & Gotelli, 2017).

The exotic polychaete P. rickettsi interacts negatively with native and exotic species, suggesting a superior ability of this species to compete for resources. In this sense, life history characteristics can also be important for the outcome of some invasions (Kupferberg, 1997; Stohlgren et al., 1999), where invasive species not only have a superior ability to acquire resources but also more efficient resource conversion capacity (Byers, 2000). We propose that the competitive dominance of individuals of the exotic P. rickettsi is due to this species’ ability to monopolize space in host shells through the construction of galleries within the shell, which may impede the establishment of other individuals of boring polychaete species (e.g., exploitation competition). Indeed, P. rickettsi is a strict boring specialist that exclusively inhabits calcareous substrates and constructs U-shaped galleries, which may become more complex over time (Sato-Okoshi & Takatsuka, 2001; Diez, Radashevsky & Orensanz, 2011). Being a specialist in calcareous substrates, this species can monopolize the resource, resulting in the high levels of infestation found in different hosts (e.g., 52% Bertrán, Vargas & Quijón, 2005; 54% Diez, Radashevsky & Orensanz, 2011). This evidence, together with the observed positive relationship between native and exotic polychaete abundance, provides strong support for the hypothesis that P. rickettsi is a successful invader by being competitively dominant, but variations in local conditions prevent the exclusion of other boring polychaetes, potentially because this species is specialized and highly effective at boring on calcareous substrates, while the other species are more generalist.

Boccardia tricuspa is the least specialized species in relation to the use of calcareous substrate (Moreno, Neill & Rozbaczylo, 2006), and it has been reported in different types of sandy, rocky and mud sediment substrates, as well as in coralline algae, between sponges and between tube-dwelling polychaetes (Rozbaczylo, Méndez & Bravo, 1994; Villamar, 2000; Villamar & Cruz, 2007). Therefore, by being more generalist in its use of resources, B. tricuspa is potentially less competitive on calcareous substrates, and it can survive in other substrates by niche differentiation, allowing coexistence by decreasing the competitive pression (i.e., weak competition hypothesis). Similarly, the interactions of D. opulens in this guild are not host specific, being associated with both multiple substrates, including shellfish valves and crustaceans (Rozbacyzlo & Carrasco, 1996; Hernández, Muñoz & Rozbaczylo, 2001; Moreno, Neill & Rozbaczylo, 2006), and also using calcareous substrates like a secondary borer (as reported for several species of the genus, (Martin & Britayev, 1998), or using cracks and crevices for refuge (Hernández, Muñoz & Rozbaczylo, 2001).

In this study even though the deterministic weak transitive competition is important for explaining the coexistence of native and exotic polychaetes, approximately 50% of the interactions observed in the boring polychaete guild were random. This means that species patterns can be accounted for by stochastic processes (Hubbell, 2001; Hubbell, 2006). This idea emphasizes that neutral-stochastic processes are important, and thus competition does not play a fundamental role, allowing species to accumulate randomly over time because: (1) there is a the high availability of resources (i.e., calcareous substrates for boring polychaetes); and (2) individuals from some species are randomly chosen from an assemblage with some probability of death (extinction), birth, migration and/or mutation. One consequence of this low-competition scenario is species coexistence, as observed in this study. Consequently, this guild of boring polychaetes could be functioning neutrally. The consideration of a broader conceptual framework that accommodates both biological structuring processes (niche based community assemblage by strong competitive interactions) and neutral stochastic processes (neutral based community assemblage with weak competitive interactions) in organizing species assemblages (Stokes & Archer, 2010; Alonso, Etienne & McKane, 2016) could provide a better understanding of the mechanisms limiting or enabling exotic species introductions.

Most other studies that evaluate the biotic resistance hypothesis do so by describing the relationship between native and exotic species diversity. However, this generalized approach can, and has, caused confusion (e.g., scale dependent results; Shea & Chesson, 2002; Byers & Noonburg, 2003) and has two potential drawbacks: (1) detection of the predicted negative pattern when the underlying mechanism is not species exclusion; and (2) not detecting the pattern when exclusion mechanisms are occurring that do not affect diversity, but do affect other ecological characteristics. In the first case, if a negative relationship in diversity is observed, one cannot immediately conclude negative biotic interactions, since this pattern could be the result of mechanisms such as habitat heterogeneity (“habitat checkerboards”), where species are associated with different abiotic features of the sites, leading to less co-occurrence than expected by chance (Gotelli & Entsminger, 2010). In this case, additional co-occurrence analyses, such as those used in this study, can provide a useful test to the competition hypothesis, where lower co-occurrence than expected by chance can be evidence of competitively structured communities. In the second case, the lack of a negative relationship, or weak negative relationship, between native and exotic diversity does not necessarily indicate that exclusion is not occurring; rather the mechanism may affect other variables, such as abundance. Especially in cases where total exclusion does not occur, as in this study, species abundance may be affected, which may not be detected by comparing native and exotic diversities. This situation can occur because competition rarely limits immigration or leads to local extinction (Levine, Adler & Yelenik, 2004; Bruno et al., 2005), therefore the result of negative interactions would only be observed in variables such as abundance, and may even result in coexistence, as seen in this study. Furthermore, in cases where the assemblage of interest is inherently species-poor (such as this study), the pattern may not be statistically detectable when diversity or richness is used as the response variable. In cases where no relationship is observed, either due to the lack of an effect on diversity or a lack of statistical power to detect this effect, analyses to evaluate other hypotheses (such as the biotic resistance hypothesis, positive species interactions, weak competitive interactions, or competitive intransitivity), can be supplemented by assessing the relationship between native and exotic abundances, and looking for constraints on abundance using transitivity analysis or other techniques (e.g., quantile regression).

Conclusions

Our results support the idea that the response of a community to the arrival of exotic species is not necessarily given by changes in diversity (i.e., competitive exclusion; Levine, Adler & Yelenik, 2004), but rather by the relative importance of interactions (e.g., species interact negatively, such as P. rickettsi vs. all other species in this study, or randomly, such as all other species interactions in this study). This supports findings in the multispecies interactions with invaders in natural communities, where negative interactions are as common as positive ones (Henriksson et al., 2016). Coexistence may be a frequent result of interactions between native and exotic species, although perhaps less apparent than competitive exclusion. In this sense, the probabilistic point-of-view used in this study also provides a statistical tool for evaluating coexistence as a result of invasive and native species interactions, an idea which has been proposed in invasion ecology, but largely lacks empirical support and methodologies for detecting underlying mechanisms (Godoy, 2019).

Finally, we found support for the weak competition hypothesis, which is related to the weighted species richness hypothesis sensu Henriksson et al. (2016). This hypothesis describes biotic resistance as a weighted sum of the resident species, where the species-specific weights describe the relative contribution of each species to biotic resistance. In our study, we found evidence both in presence/absence and abundance data that P. rickettsi is a successful invader by being competitively dominant, but not excluding other species due to general weak competition. This evidence allows us to suggest that the low species richness of introduced species on the southeastern Pacific coast of Chile may be explained by the particular geographic isolation of this coast (i.e., low propagule pressure) and abiotic resistance (Castilla & Neill, 2009).

Supplemental Information