Starting the stowaway pathway: the role of dispersal behavior in the invasion success of low-mobile marine species

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/s10530-020-02285-7 ” 1 Starting the stowaway pathway: the role of dispersal behavior in the invasion success of 1 low-mobile marine species 2 Ros M1,2*, Navarro-Barranco C2, González-Sánchez M2, Ostalé-Valriberas E2, Cervera-Currado 3 L1, Guerra-García JM2 4 1Departamento de Biología, Área de Ecología, Facultad de Ciencias del Mar y Ambientales, 5 Universidad de Cádiz, 11510, Puerto Real, Spain 6 2Laboratorio de Biología Marina, Departamento de Zoología, Facultad de Biología, Universidad 7 de Sevilla, Avda Reina Mercedes 6, 41012, Sevilla, Spain 8 *Corresponding author: [email protected] 9 10 Abstract 11 Despite the importance of dispersal behavior in the eco-evolutionary responses of species to 12 global environmental change, its role in the invasion dynamic of stowaway invaders has been 13 poorly studied, especially in low-mobile species (i.e. with direct development). Here we use 14 peracarid crustaceans as a target group to understand the role that local dispersal plays in the 15 initial stages of the stowaway pathway in species inhabiting transport hubs (specifically marinas). 16 Thus, we performed field experiments to quantitatively explore differences in species´ propensity 17 to disperse when comparing species with low and high invasion potential (considered here as the 18 ability for successful anthropogenic dispersal). At the community level, we found that widespread 19 introduced species (with high potential for invasion; HPI) exhibited higher propensity for local 20 dispersal than closely related species that fail to spread (with low potential for invasion; LPI). 21 From a functional perspective, high invasion potential, in synergy with omnivorous feeding and 22 a tube-dweller lifestyle, was instrumental in determining differences in trait composition between 23 assemblages that vary in their natural tendency to disperse. In addition to anthropogenic dispersal 24 on ships, we show that unaided dispersal may play a crucial role at the beginning of the stowaway 25 pathway, even in low-mobile species. Knowledge of dispersal behavior should be more fully 26 2 integrated into research on managing the risk of this growing invasion pathway. This may help to 27 predict rates of spread and provide new insights into the proximate causes of stowaway 28 organisms’ invasion success. 29 30 Keywords 31 Biological invasions; unintentional introductions; natural dispersal; hull fouling; recreational 32 boating; biological trait analysis (BTA) 33 34 INTRODUCTION 35 Human activities are profoundly changing the dispersal patterns of biological 36 communities, both directly by transporting species, and indirectly by altering habitat and/or 37 natural vectors (Bullock et al. 2018). For example, human-mediated dispersal has altered the 38 diversity of organisms involved in long-distance dispersal, explaining the widespread distribution 39 of some low-mobile species (Winston 2012). In aquatic ecosystems, many of these species are 40 fouling invertebrates which are unintentionally transported as stowaways on ships, either in the 41 hulls of boats and ships or in ballast water, beyond their native distribution ranges (Hulme et al. 42 2008). Indeed, stowaway transport is a growing invasion pathway that is expected to increase 43 with globalized trade (Tingley et al. 2018). The Strategic Plan for Biodiversity (2011–2020) 44 emphasizes the need to tackle the means by which invasive alien species are introduced through 45 human activities (Hulme 2015). Thus, identifying the pathway and understanding its role in 46 determining potential invaders may be of use in preventing new invasions (Carlton 1996; 47 Padayachee et al. 2017). A crucial step in starting the stowaway pathway is uptake by a human- 48 mediated vector. However, little information exists about potential traits that may increase 49 success at this initial stage. This poses a challenge because of the vast number of species that 50 could be unintentionally introduced as stowaways and the significant lack of knowledge of their 51 biology and life history traits (Leung et al. 2014; McGeoch et al. 2016). 52 3 In marine ecosystems, high relative abundance in a transport hub (such as marinas or 53 harbours) and vector characteristics (e.g. maintenance history of the boats) are considered 54 important factors in determining the probability of unintentional transport, since more abundant 55 species have a higher overall chance to colonize human vectors (Wonham et al. 2000; Floerl and 56 Inglis 2005). However, little is known about the factors driving movement of non-indigenous 57 species (NIS) to reach these vectors. For example, it remains unclear why some closely-related 58 species (for example with similar development modes) inhabiting a transport hub with relatively 59 similar abundances, fail to spread whereas others succeed. In these types of habitats, it is common 60 to find species with restricted distribution, lacking a history of human-mediated introductions 61 (species with low potential for invasion, hereinafter LPIs), coexisting with closely-related human- 62 mediated widely introduced species (species with high potential for invasion, hereinafter HPIs). 63 For example, the native tunicate Clavelina sabbadini (LPIs) and the introduced Clavelina oblonga 64 (HPIs), or the native caprellid crustacean Caprella acanthifera (LPIs) and the introduced Caprella 65 scaura (HPIs), frequently occur in sympatry in Mediterranean marinas (López-Legentil et al. 66 2015 and Ros et al. 2013 respectively). Within these species assemblages, it remains unclear why 67 LPIs fail to spread whereas HPIs succeed. Knowledge of different types of dispersal dynamics 68 (e.g. anthropogenic and natural dispersal) of successful and failed potential stowaways can be 69 crucial for understanding and managing the risk of the stowaway pathway. Thus, taking into 70 account that the risks posed by NIS depend largely on their ability to expand their non-native 71 ranges, better knowledge of their dispersal behavior may help predict their ecological or economic 72 impacts (Lockwood et al. 2009) and lead to the development of effective post-invasion 73 management actions (Myers et al. 2000). However, the importance of natural dispersal (i.e. active 74 and/or passive unaided dispersal) in the invasion process of stowaway organisms may go 75 unnoticed, especially when they have low capacity for autonomous dispersion. 76 Natural dispersal plays an important role in understanding community assembly, 77 metapopulation and metacommunity dynamics, and it is critical for better comprehension of 78 invasion ecology (Hanski 1999; Puth and Post 2005; Clobert et al. 2012). It is a complex 79 4 phenomenon which is likely to be driven by a combination of species traits and environmental 80 factors (Bowler and Benton 2005; Benard and McCauley 2008) and, therefore, it is a highly 81 context-dependent process. It can involve multiple dispersal components like propensity 82 (tendency or motivation to move about) and ability (to travel long distances) (Tung et al. 2018). 83 Here, we will focus on the natural local dispersal propensity (i.e. tendency to move about short 84 distances). We differentiate this behavior from the anthropogenic long-distance dispersion ability, 85 defined as a species’ ability to successfully move beyond its native distribution range owing to 86 human-mediated transportation (and used here as a proxy of its invasion potential). The 87 propensity of individuals to move about has important ecological and evolutionary consequences, 88 including the ability of a species to spread along its fundamental niche and colonize new areas 89 (Kokko and Lopez-Sepulcre 2006). It may also promote success in the early stages of the process 90 leading to unintentional introductions (Chapple et al. 2012), as well as later on, in the post- 91 establishment dispersal (i.e. secondary spread) (Theoharides and Duke 2007). However, despite 92 the fact that species’ natural dispersal propensity is commonly suggested as an important factor 93 influencing their geographical ranges (e.g. Brown et al. 1996; Gaston 2003), this expectation in 94 most cases has not been empirically evaluated (Lester et al. 2007; Fletcher et al. 2013), especially 95 in relation to unintentional biological invasions (see Chapple et al. 2011). For example, little 96 comparative information exists to demonstrate that widespread introduced species exhibit a high 97 dispersal propensity and/or better dispersal abilities than native or closely related species that fail 98 to spread (Rehage and Sih 2004; Bubb et al. 2006; Brousseau and McSweeney 2016). Taking into 99 account that dispersal is a context-dependent process and that anthropogenic dispersal is 100 intimately related to human-altered environments, marinas and harbours provide unique scenarios 101 for understanding if natural dispersal propensity influence the invasion success of inadvertent 102 marine species introductions. Comparisons of natural dispersal propensity to move about between 103 LPIs and HPIs associated with these habitats could shed light on the question. 104 Dispersal dynamics in marine ecosystems are generally complex and poorly understood 105 (Weersing and Toonen 2009). Two of the major issues related to natural dispersal propensity 106 5 include linking species traits to movement propensity patterns and studying the differences in 107 anthropogenic vs. natural movement (Chapple et al. 2012). A potential approach is to infer these 108 patterns from the colonization of new habitats by community assemblages in field experiments 109 (e.g. Jenkins and Buikema 1998; Cáceres and Soluk 2002; Havel et al. 2002), especially if they 110 focus on species that can colonize rapidly (Puth and Post 2005). Fouling communities associated 111 with submerged structures of marinas are frequently comprised of opportunistic species with 112 above average abilities to colonize artificial habitats. One of the dominant groups among these 113 communities, both in terms of abundance and number of species, are peracarid crustaceans, 114 especially the orders Amphipoda, Isopoda and Tanaidacea (e.g. Conradi et al. 2000; Martínez- 115 Laiz et al. 2019). Unlike other crustaceans, peracarids lack a pelagic larval stage; development is 116 direct with young emerging with the adult morphology. This generally limits their ability to 117 disperse, posing a challenge to their spread. They live as epibionts on a wide variety of fouling 118 organisms (basibionts), especially those with an arborescent structure, including macroalgae, 119 hydroids or bryozoans. Moreover, they can also reach high densities on structurally complex 120 artificial substrata, such as ropes or nets (e.g. Cabezas et al. 2013). Only a subset of peracarid 121 species inhabiting transport hubs are frequently transported around the globe as inadvertent 122 stowaways on ships and boats, potentially causing significant impacts on recipient communities 123 (e.g. Ros et al. 2015). However, little is known about how this transport occurs. It seems that the 124 presence of complex artificial and biogenic substrata on the submerged surface of transport 125 vectors (e.g. ships’ hull), where they have in fact been found (e.g. Martínez-Laiz et al. 2019), 126 could serve as a refuge during transport. Moreover, they have also been found in ballast water 127 tanks (Carlton 1985; Gollasch et al. 2002), although less frequently. The role that natural dispersal 128 could play in the invasion dynamics of these invertebrates is not well understood. 129 We aim to investigate the role of species’ natural dispersal propensity in the successful 130 introduction and subsequent spread of low-mobile (i.e. with direct development) marine invaders. 131 To address this aim we conducted a comparative experiment to quantitatively explore differences 132 in local dispersal propensities between species with low and high potential for invasion (LPIs vs. 133 6 HPIs) inhabiting marina structures. Additionally, we performed a biological trait analysis (BTA) 134 to further explore the potential interplay between short-distance natural and long-distance 135 anthropogenic dispersal propensity from a functional perspective. We hypothesized that a high 136 propensity to move short distances may increase the chances of vector uptake (or vector 137 colonization) and in turn the success of accidental anthropogenic transportation in species 138 inhabiting transport hubs. Therefore, we expect that HPIs will exhibit greater natural dispersal 139 propensity than LPIs. 140 141 METHODS 142 Study system 143 In this study, we use peracarid crustaceans as our target group. We focused on this group 144 for several reasons. First, peracarids are one of the dominant taxa of mobile epibiont communities 145 associated with submerged artificial structures. Second, they rapidly colonize artificial submerged 146 structures. Third, their dispersal stages (mostly juveniles but also adults) can be identified to the 147 species level with the appropriate taxonomic keys (contrary to what happens in the case of larval 148 stages). Finally, because of their reduced mobility, natural dispersal usually occurs at spatial 149 scales that are easy to cover. 150 Field experiments were conducted in a focal point for species introductions, the harbour 151 of Ceuta, located on the North African coast of the Strait of Gibraltar, a major maritime route. 152 This harbour is located between two bays connected by a water channel, which increases the water 153 renewal across its middle zone, where a marina is located (Fig. 1). This design enables the 154 colonization of the marina structures by a high diversity of fauna from nearby areas (see Guerra- 155 García et al. 2003; Guerra-García and García-Gómez 2004). Owing to these factors, experiments 156 were focused in the Ceuta marina (35º 53’ N, 5º 18’ W). 157 Experimental design 158 7 In this study we used the colonization rates of passive collectors as a quantitative proxy 159 for natural dispersal tendency. Collectors were built of commercially available plastic bath puff, 160 consisting of a dense spherical folded mesh approximately 10 cm in diameter and with a mean 161 volume of 12 ml. The three-dimensionality of these meshes is intended to reflect the structural 162 complexity of the local fouling basibionts on which epibiont organisms live. These arborescent 163 basibionts can grow on the surface of boat hulls, increasing their structural complexity and 164 promoting the spread of their associated fauna as stowaway epibionts. Collectors were deployed 165 in the water column at a depth of 1 meter (the same depth at which boat hulls tend to be more 166 heavily fouled; Dafforn et al. 2009) for two short-term periods of three and fifteen days 167 respectively. Periods were selected to cover the range of days that transient boats spent in a 168 visiting marina (from a weekend to two weeks of mean stage; Ceuta marina staff, personal 169 communication). 170 Considering that colonization rates can be affected by the hydrodynamic morphology of 171 the marina structure (Koehl 2007), as well as by the landscape’s particular attributes that may 172 impede movement or increase dispersal mortality (Fahrig 2001), we select two different positions 173 within the marina. One position was located close to the water entrance and the other was located 174 in a more internal part of the marina (Fig. 1). Sixteen collectors were deployed for each period in 175 each position. Adjacent collectors were placed at least 2 meters apart. The experiment was carried 176 out at the end of summer (September 2015) and was repeated at the beginning of winter (January 177 2016) to check pattern consistency. Overall, a total of 128 collectors were analyzed. At the end 178 of each deployment period, collectors were carefully retrieved and meshes were enclosed in a 179 plastic pot before breaking the rope-tie. Just after collection, samples were preserved in 70 % 180 ethanol. At the laboratory, meshes were unraveled and all macrofaunal specimens were sorted 181 and counted. 182 In order to characterize the local pool of epibiont species associated with the marina 183 structures, we carried out a complementary sampling focused on the nearest potential source of 184 ‘propagules’ (i.e. dispersing individuals): the fouling basibionts associated with the lateral 185 8 submerged surface of the floating pontoons (consisting mainly of bryozoans, hydrozoans, and 186 macroalgae). This also allows us to understand which of the epibionts inhabiting the pontoons are 187 the fastest to colonize empty three-dimensional niches (i.e. the passive collectors). In this case, 188 we scraped samples of 20x20 cm of encrusting sessile organisms (basibionts) and their associated 189 macrofauna (epibionts) and placed them in plastic bags. We collected six random replicates per 190 position before deploying the collectors for each of the two experiments. All samples were 191 preserved in 70 % ethanol. As above, all animals were sorted and counted. 192 Classification of individuals collected 193 For both collectors and scrapings, peracarid crustaceans were identified to the species level and 194 the other taxa were grouped into larger categories. Species were classified, based on the absence 195 or presence of a documented history of long-distance anthropogenic transport, as species with low 196 potential for invasion (LPIs) and species with high potential for invasion (HPIs), respectively. 197 LPIs included native species for the study area (i.e. the Mediterranean and the North East Atlantic 198 Ocean, since Ceuta is located in the middle of these two bodies of water) that have never been 199 transported outside their native range (N in Table S1). HPIs included three categories of human- 200 mediated widespread species: (i) native species to the study area, but introduced overseas (i.e. 201 exported species) (Nex in Table S1), (ii) cryptogenic species, which are not demonstrably native 202 or introduced in the study area, but are characterized by a long history of documented 203 anthropogenic transport (C in Table S1), and (iii) introduced species in the study area (NIS in 204 Table S1). All HPIs share wide but discontinuous distribution patterns closely associated with 205 anthropogenic transport vectors, as well as having been considered NIS somewhere around the 206 globe (references used can be found in Table S2 References). The status was not applied to 207 individuals with insufficient taxonomic certainty, which were considered undetermined (U). This 208 approach allows for assessing potential determinants of invasive potential on a global scale (van 209 Kleunen et al. 2010). Those individuals that could not be identified to the species level were 210 considered undetermined (U) and were excluded from subsequent analysis. 211 Statistical analyses 212 15 higher tendency to disperse than LPIs. At a community level, when we compared the abundance 373 and species richness of HPIs and LPIs between the collectors and the local basibionts, we found 374 that this may be true in outer marina positions, where HPIs dominated over LPIs, but not in the 375 inner parts. In the local basibionts associated with the internal positions of the marina we found a 376 higher abundance of LPIs than HPIs (regardless of the season), a pattern that was not reflected in 377 the collectors, where again HPIs dominated over LPIs. In this case, a high natural dispersal 378 propensity would be more related to the species’ abilities to move and detect new substrata rather 379 than density effects (Martin-Smith 1994). This could be the case of the HPIs Paracerceis sculpta, 380 with high relative abundance in the collectors but scarcely found in the local basibionts. However, 381 it is worth nothing that this pattern could be also an artifact due to the impossibility of sampling 382 all possible local habitats. Thus, we cannot rule out that the species exists in high abundance in 383 other local basibionts that were not sampled. Moreover, the supply of dispersing epibionts, and 384 therefore the ‘propagule pressure’ on experimental collectors, may also be affected by the 385 hydrodynamic patterns within a marina. Marinas’ spatial features, including floating structures, 386 usually attenuate water currents in the inner parts and concentrate the flow on access channels 387 (Huguet et al. 2020). This hydrodynamic pattern facilitates the transport to the inner zones of mud 388 sediments and pollutants (e.g. Kenworthy et al. 2018; Mali et al. 2017), and promotes larval 389 recruitment of sessile invertebrates (Floerl and Inglis 2003). Low-mobile epibionts with direct 390 development, such as peracarid crustaceans, can resist these hydrodynamic forces when they are 391 attached to their basibionts, which in turn possibly facilitates their transport as stowaways in the 392 ship and boat hulls. Hydrodynamic forces would act primarily on the pool of individuals that 393 exhibits a propensity to detach from their basibiont and move into the water column (i.e. 394 individuals with high dispersal propensity). Due to their reduced swimming capacity, some of 395 these individuals can be transported passively towards inner marina positions where 396 hydrodynamic energy is reduced according to Huguet et al. 2020 and our personal observations. 397 This may help explain the greater percentage of HPIs found in experimental collectors in internal 398 positions compared to those deployed close to the water entrance. Thus, while LPIs dominate in 399 16 the local pool of epibionts attached to basibionts in internal positions, HPIs dominate in the pool 400 of dispersal epibionts in both internal and external positions. This highlights the potential 401 relationship between the invasive potential and the dispersal propensity. Moreover, this may 402 imply that boats in transit would have a higher chance of being infested by HPIs than by LPIs, 403 but also that those moored in inner positions may harbor a higher percentage of HPIs than those 404 moored close to the water entrance. 405 At specific level, all the dominant species associated with collectors belonged to HPIs 406 (Jassa slatteryi, Paracerceis sculpta, Phtisica marina, Apocorophium acutum and Caprella 407 scaura), while in the local basibionts there were not a clear dominance of HPIs on LPIs or vice 408 versa. The dominance of HPIs over LPIs in the collectors supports that HPIs exhibited higher 409 local dispersal propensity and ability to colonize new habitats than LPIs. For example, despite the 410 LPI amphipod Apherusa mediterranea was one of the dominant species in the local basibionts, it 411 was virtually absent in experimental collectors. Therefore, under the same availability of vectors 412 and a high density of individuals in the local basibionts, a low propensity for dispersal and/or poor 413 colonization abilities may explain why some species that are well adapted to live in transport hubs 414 fail to spread. For these species, reaching the anthropogenic vector of transport may be a 415 determinant filter stopping stowaway transport events. However, it is also possible that these LPIs 416 present low plasticity in colonizing artificial structures (such as the experimental collectors or the 417 boat hulls) or present low affinity for these structures. For example, in marine sessile fouling 418 communities, Dafforn et al. (2009) found a greater recruitment of invasive species over natives 419 on shallow moving artificial structures (specifically Perspex plates). In the few cases with a 420 combination of restricted distribution (LPIs) and a strong dispersal tendency and/or good 421 colonization abilities, the filter will probably take place during transport and/or the establishment 422 stage, limiting their current distribution to the native range. Conversely, we also observed a few 423 cases of HPIs with low dispersal propensity (such as Elasmopus rapax), suggesting that these 424 species would have no affinity for the experimental collectors or that they are very adapted to deal 425 with conditions during transport, increasing their success in each stowaway event. 426 17 From a functional perspective, different biological traits can also influence dispersal 427 tendency patterns at both interspecific and intraspecific levels (see Wilson et al. 2009). At the 428 interspecific level, we found that assemblages colonizing experimental collectors in a very short- 429 term period (those colonizing C3) were characterized by animals with a high propensity for 430 anthropogenic LDD (i.e. high invasion potential), and also presenting a tube-dwelling life style 431 and an omnivorous feeding strategy. Franz and Mohamed (1989) suggested that tube-dweller 432 amphipods present a high propensity to disperse because of the limitation of suitable sites for 433 building their tubes. Moreover, the facility to form attached tubes in hard substrata allows the 434 colonization of otherwise unsuitable habitats, and protects against adverse environmental 435 conditions such as strong currents or desiccation (Moore and Eastman 2015). The tube provides 436 adaptive benefits for LDD by increasing the likelihood of successful attachment, retention to the 437 human vector, and survival during transport. Similar benefits can be influential for cavity-dweller 438 species, such as the isopod P. sculpta, which occupy sponge and barnacles’ cavities and likely the 439 artificial cavities of boat hulls. In a similar way, high trophic plasticity may favor feeding during 440 transport. In this sense, omnivory appears to be a common trait in successfully established 441 invasive alien crustaceans (Hänfling et al. 2009). This may also reflect the difficulty to define a 442 good invader simply on the basis of a life history trait. It is probably the synergistic combination 443 of several traits that allows HPIs to overcome the different filters at the initial stages of the 444 invasion pathway. 445 Further studies on the biology of marine stowaways may improve stowaway risk 446 assessments in order to prevent new introductions. However, our results may inform boat cleaning 447 programs to prevent infestation during transient boating activity. Considering that even low- 448 mobile invaders could colonize empty niches in short-time periods, transient boats (as well as 449 ropes, buoys and other submerged structures associated with boating activity) should be cleaned 450 before leaving each marina. Following a precautionary approach, cleaning must be done even if 451 the boat’s hull is apparently clean, because there may be small invaders that have managed to 452 colonize it. This must be considered in future and ongoing biofouling management strategies. 453 18 Conclusions 454 Overall we found that, for low-mobile species inhabiting transport hubs, widespread 455 introduced species seem to exhibit a higher local dispersal propensity than closely related species 456 that fail to spread. This may increase the frequency of stowaway transport events, emphasizing 457 the critical role of natural dispersal at the beginning of the stowaway pathway, and later in the 458 secondary propagation among artificial habitats. Increasing global connectivity through human 459 transportation networks makes it challenging to address the risk posed by the stowaway pathway 460 (Hulme et al. 2008; Hulme 2015; Tingley et al. 2018). Knowledge of species’ dispersal propensity 461 inhabiting human transport hubs may help to identify high-risk stowaway organisms, providing a 462 basis for optimizing stowaway pathway risk assessments. 463 464 Acknowledgments 465 This study was funded by the Instituto de Estudios Ceutíes, the Andalusian Government 466 (Project I+D+I, US-1265621, co-financed by the Fondo Europeo de Desarrollo Regional - 467 FEDER-), and a Juan de la Cierva postdoctoral grant (IJCI-2017-34656) awarded to MR by the 468 Spanish Government. Additional support was provided by the Ministerio de Ciencia, Innovación 469 y Universidades (Project CGL2017-82739-P co-financed by the Agencia Estatal de Investigación 470 -AEI- and FEDER). We are extremely grateful to José Carlos García-Gómez and José Manuel 471 Ávila for hosting us in the Estación Marina de Ceuta, and to the editor and the anonymous referees 472 for their constructive and very helpful comments and suggestions. We sincerely thank the Ceuta 473 marina staff and the Port Authority of Ceuta for allowing us to carry out the field experiments. 474 We also thank Clara Gavira O'Neil for carrying out the English revision of the manuscript. 475 476 References 477 Benard MF, McCauley SJ (2008) Integrating across life-history stages: Consequences of natal 478 habitat effects on dispersal. Amer Naturalist 171:553–67. 479 19 Blackburn TM, Cassey P, Lockwood J (2009) The role of species traits in the establishment 480 success of exotic birds. Global Change Biol 15: 2852–2860. 481 Bowler DE, Benton TG. 2005. Causes and consequences of animal dispersal strategies: Relating 482 individual behavior to spatial dynamics. Biol Rev 80:205–25. 483 Bremner J, Rogers SI, Frid CLJ (2006) Methods for describing ecological functioning of marine 484 benthic assemblages using biological traits analysis (BTA). Ecol Indic 6:609–622. 485 Brousseau DJ, McSweeney L (2016) A comparison of reproductive patterns and adult dispersal 486 in sympatric introduced and native marine crabs: implications for species characteristics 487 of invaders. Biol Invasions 18:1275–1286. 488 Brown JH, Stevens GC, Kaufman DM (1996) The geographic range: size, shape, boundaries, and 489 internal structure. Annu Rev Ecol Evol Syst 27:597–623. 490 Bubb DH, Thom TJ, Lucas MC (2006) Movement, dispersal and refuge use of co-occurring 491 introduced and native crayfish. Freshwater Biol 51:1359–1368. 492 Bullock J, Bonte D, Pufal G, Chapman D, Garcia C, Garcia D, Delgado M (2018) Human- 493 mediated dispersal and the rewiring of spatial networks. Trends Ecol Evol 33:958–970. 494 Cabezas MP, Navarro-Barranco C, Ros M, Guerra-García JM (2013b) Long-distance dispersal, 495 low connectivity and molecular evidence of a new cryptic species in the obligate rafter 496 Caprella andreae Mayer, 1890 (Crustacea: amphipoda: Caprellidae). Helgoland Marine 497 Research 67:483–497. 498 Cáceres CE, Soluk DA (2002) Blowing in the wind: a field test of overland dispersal and 499 colonization by aquatic invertebrates. Oecologia 131:402–408. 500 Carlton JT (1985) Transoceanic and interoceanic dispersal of coastal marine organisms: the 501 biology of ballast water. Oceanography and Marine Biology - An Annual Review 23: 502 313–371. 503 20 Carlton JT (1996) Pattern, process, and prediction in marine invasion ecology. Biol Conserv 504 78:97–106. 505 Chapple DG, Simmonds SM, Wong BB (2011) Know when to run, know when to hide: can 506 behavioral differences explain the divergent invasion success of two sympatric lizards? 507 Ecol Evol 1:278–289. 508 Chapple DG, Simmonds SM, Wong BB (2012) Can behavioral and personality traits influence 509 the success of unintentional species introductions? Trends Ecol Evol 27:57–64. 510 Chessel D, Dufour AB, Thioulouse J (2004) The ade4 Package – I: One-Table Methods. R News 511 4:5–10. 512 Clarke KR, Gorley RN. 2006. PRIMER v6: User Manual/Tutorial (Plymouth Routines in 513 Multivariate Ecological Research). PRIMER-E, Plymouth. 514 Clobert J, Le Galliard JF, Cote J, Meylan S, Massot M (2009) Informed dispersal, heterogeneity 515 in animal dispersal syndromes and the dynamics of spatially structured populations. Ecol 516 lett 12(3), 197-209. 517 Clobert J, Baguette M, Benton TG, Bullock JM (2012) Dispersal ecology and evolution. Oxford 518 University Press, Oxford. 519 Conradi M, López-González PJ, Cervera-Currado JL, García-Gómez JC (2000) Seasonality and 520 spatial distribution of peracarids associated with the bryozoan Bugula neritina in 521 Algeciras Bay. J Crustacean Biol 20:334–349. 522 Dafforn KA, Johnston EL, Glasby TM (2009) Shallow moving structures promote marine invader 523 dominance. Biofouling 25:277–287. 524 Fahrig L (2001) How much habitat is enough? Biol Conserv 100:65–74. 525 Fletcher LM, Forrest BM, Bell JJ (2013) Natural dispersal mechanisms and dispersal potential of 526 the invasive ascidian Didemnum vexillum. Biol Invasions 15:627–643. 527 21 Floerl O, Inglis GJ (2005) Starting the invasion pathway: the interaction between source 528 populations and human transport vectors. Biol Invasions 7:589–606. 529 Floerl O, Inglis GJ (2003) Boat harbour design can exacerbate hull fouling. Australian Ecology 530 28:116–127. 531 Franz DR, Mohamed Y (1989) Short-distance dispersal in a fouling community amphipod 532 crustacean, Jassa marmorata Holmes. J Exp Mar Bio Ecol 133:1–13. 533 Gaston KJ (2003) The Structure and Dynamics of Geographic Ranges. Oxford University Press, 534 Oxford. 535 Gollasch S, Macdonald E, Belson S, Botnen H, Christensen J et al. (2002) Life in Ballast Tanks. 536 In: Leppakoski E, Gollasch S, Olenin S (eds) Invasive Aquatic Species of Europe: 537 Distribution, Impacts and Management. Kluwer Academic Publishers, Dordrecht, The 538 Netherlands. 539 Guerra-García JM, Corzo J, García-Gómez JC (2003) Short-term benthic recolonization after 540 dredging in the harbour of Ceuta, North Africa. Mar Ecol 24:217–229. 541 Guerra-García JM, García-Gómez JC (2004) Crustacean assemblages and sediment pollution in 542 an exceptional case study: a harbour with two opposing entrances. Crustaceana 77(3): 543 353–370. 544 Guerra-García JM, García-Gómez JC (2009). Recolonization of macrofauna in unpolluted sands 545 placed in a polluted yachting harbour: A field approach using experimental trays. Estuar 546 Coast Shelf S 81:49–58. 547 Hamilton WD, May RM (1977) Dispersal in stable habitats. Nature 269, 578. 548 Hänfling B, Edwards F, Gherardi F (2011) Invasive alien Crustacea: dispersal, establishment, 549 impact and control. BioControl 56:573–595. 550 Hanski IA (1999) Metapopulation Ecology. Oxford University Press, Oxford. 551 22 Havel JE, Shurin JB, Jones JR (2002) Estimating dispersal from patterns of spread: spatial and 552 local control of lake invasions. Ecology 83:3306–3318. 553 Huguet JR, Brenon I, Coulombier T (2020) Influence of Floating Structures on Tide-and Wind- 554 Driven Hydrodynamics of a Highly Populated Marina. J Waterw Port Coast 555 146:05019004. 556 Hulme PE (2015) Invasion pathways at a crossroad: policy and research challenges for managing 557 alien species introductions. J Appl Ecol 52:1418–1424. 558 Hulme PE, Bacher S, Kenis M, Klotz S, Kühn I, Minchin D, ..., Pyšek P (2008). Grasping at the 559 routes of biological invasions: a framework for integrating pathways into policy. J Appl 560 Ecol 45:403–414. 561 Jenkins DG, Buikema AL (1998) Do similar communities develop in similar sites? A test with 562 zooplankton structure and function. Ecol Monogr 68:42–443. 563 Kenworthy JM, Rolland G, Samadi S, Lejeusne C (2018) Local variation within marinas: effects 564 of pollutants and implications for invasive species. Mar Pollut Bull 133:96–106. 565 Koehl MRA (2007) Mini review: hydrodynamics of larval settlement into fouling communities. 566 Biofouling 23:357–368. 567 Kokko H, López-Sepulcre A (2006) From individual dispersal to species ranges: perspectives for 568 a changing world. Science 313:789–791. 569 Lester SE, Ruttenberg BI, Gaines SD, Kinlan BP (2007) The relationship between dispersal 570 ability and geographic range size. Ecol Lett 10:745–758. 571 Levinton JS (1970) The paleoecological significance of opportunistic species. Lethaia 3:69–78. 572 Llewelyn J Phillips BL, Alford RA, Schwarzkopf L, Shine R (2010) Locomotor performance in 573 an invasive species: cane toads from the invasion front have greater endurance, but not 574 speed, compared to conspecifics from a long-colonised area. Oecologia 162:343–348. 575 23 Lockwood JL, Cassey P, Blackburn TM (2009) The more you introduce the more you get: the 576 role of colonization pressure and propagule pressure in invasion ecology. Divers Distrib 577 15:904–910. 578 López-Legentil S, Legentil ML, Erwin PM, Turon X (2015) Harbor networks as introduction 579 gateways: contrasting distribution patterns of native and introduced ascidians. Biol 580 Invasions 17:1623–1638. 581 MacArthur RH (1960) On the relative abundance of species. Am Nat. 94:25–36. 582 Mack R, Simberloff D, Lonsdale WM, Evans H, Clout M, Bazzaz FA (2000) Biotic invasions: 583 causes, epidemiology, global consequences, and control. Ecol Appl 10:689–710. 584 Mali M, Dell'Anna MM, Mastrorilli P, Damiani L, Piccinni AF (2017) Assessment and source 585 identification of pollution risk for touristic ports: Heavy metals and polycyclic aromatic 586 hydrocarbons in sediments of 4 marinas of the Apulia region (Italy). Mar Pollut Bull 587 114:768–777. 588 Martin-Smith KM (1994) Short-term dynamics of tropical macroalgal epifauna: patterns and 589 processes in recolonization of Sargassum fussifolium. Mar Ecol Prog Ser 110:177–185. 590 Martínez-Laiz G, Ulman A, Ros M, Marchini A (2019) Is recreational boating a potential vector 591 for non-indigenous peracarid crustaceans in the Mediterranean Sea? A combined 592 biological and social approach. Mar Pollut Bull 140: 403–415. 593 Moore PG, Eastman LB (2015) The tube-dwelling lifestyle in crustaceans and its relation to 594 feeding. In: Thiel M, Watling L. (Eds.), The Natural History of the Crustacea Volume 2: 595 Lifestyles and Feeding Biology (pp. 57–99). Oxford University Press. 596 Murray CC, Therriault TW, Martone PT (2012) Adapted for invasion? Comparing attachment, 597 drag and dislodgment of native and nonindigenous hull fouling species. Biol Invasions 598 14:1651–1663. 599 24 Myers JH, Simberloff D, Kuris AM, Carey JR (2000) Eradication revisited: dealing with exotic 600 species. Trends Ecol Evol 15:316–320. 601 Myles-Gonzalez E, Burness G, Yavno S, Rooke A, Fox MG (2015) To boldly go where no goby 602 has gone before: boldness, dispersal tendency, and metabolism at the invasion front. 603 Behav Ecol 26:1083–1090. 604 Padayachee AL, Irlich UM, Faulkner KT, Gaertner M, Procheş Ş, Wilson JR, Rouget M (2017) 605 How do invasive species travel to and through urban environments? Biol Invasions 606 19:3557–3570. 607 Palmer MA, Allan JD, Butman CA (1996) Dispersal as a regional process affecting the local 608 dynamics of marine and stream benthic invertebrates. Trends Ecol Evol 11:322–326. 609 Puth LM, Post DM (2005) Studying invasion: have we missed the boat? Ecol Lett 8:715–721. 610 Rehage JS, Sih A (2004) Dispersal behavior, boldness, and the relationship to invasiveness, a 611 comparison of four Gambusia species. Biol Invasions 6:379–391. 612 Ros M, Vázquez-Luis M, Guerra-García JM (2013) The role of marinas and recreational boating 613 in the occurrence and distribution of exotic caprellids (Crustacea: Amphipoda) in the 614 Western Mediterranean: Mallorca Island as a case study. J Sea Res 83: 94–103. 615 Ros M, Vázquez-Luis M, Guerra-García JM (2015) Environmental factors modulating the extent 616 of impact in coastal invasions: The case of a widespread invasive caprellid (Crustacea: 617 Amphipoda) in the Iberian Peninsula. Mar Pollut Bull 98:247–258. 618 Ruiz GM, Fofonoff PW, Carlton JT, Wonham MJ, Hines AH (2000) Invasion of coastal marine 619 communities in North America: apparent patterns, processes, and biases. Annu Rev Ecol 620 Evol Syst 31:481–531. 621 Theoharides KA, Dukes JS (2007) Plant invasion across space and time: factors affecting 622 nonindigenous species success during four stages of invasion. New Phytol 176:256–273. 623 31 Figure 3 -100-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100 Relative abundance (%) Local basibionts Experimental collectors Local basibionts Collectors (3 days) Collectors (15 days) HPIs LPIs Jassa slatteryi Paracerceis sculpta Phtisica marina Apocorophium acutum Caprella scaura Monocorophium acherusicum Dynamene edwardsi Caprella equilibra Elasmopus rapax Stenothoe valida Stenothoe georgiana . Pariambus typicus Gammaropsis maculata Aora spinicornis Microdeutopus chelifer Ischyrocerus inexpectatus Ericthonius argenteus Apherusa mediterranea Apolochus picadurus Ampithoe ramondi Paranthura costana 32 Figure 4 d = 0.5 Collectors (3 days) Collectors (15 days) Local basibionts FPC1 = 44.91% FPC2 = 26.62 % Burr Herb Nofed Det Carn Omni Free Tub Cavid Ship Raft Abs Lat Dors Cyl Slend Craw Swim FP LB C15 C3 33 Figure S1 34 Table 1 Table 2 Table 3 Source of deviation % HPI Abundance % HPI Richness df MS F P MS F P Days = Da 1 2008.2 4.21 0.04* 9.03 0.02 0.89 Season = Se 1 1018.13 2.13 0.15 1696.53 3.62 0.06 Position = Po 1 2493.95 5.23 0.02* 2869.03 6.13 0.01* Da x Se 1 962.51 2.02 0.16 544.5 1.16 0.28 Da x Po 1 815.07 1.71 0.19 1740.5 3.72 0.06 Se x Po 1 10.69 0.02 0.88 72 0.15 0.69 Da x Se x Po 1 197.51 0.41 0.52 2467.53 5.27 0.02* Residual 120 477.27 468.41 Cochran’s C-test C = 0.2503 n.s C = 0.2456 n.s. Transformation None None Average similarity Average dissimilariy C3 C15 LB C3 vs C15 C3 vs LB C15 vs LB (37.15 %) (36.60 %) (21.89 %) (72.85 %) (90.75 %) (86.03 %) SPECIES Contribution to similiarity (%) Contribution to dissimilarity (%) Apherusa mediterranea - - 16.33 - - 7.81 (>LB) Apocorophium acutum* - - - - - 7.87 (>LB) Chondrochelia cf. savignyi - - 13.15 - 8.51 (>LB) 9.11 (>LB) Elasmopus rapax* - - 28.39 - 8.92 (>LB) 9.36 (>LB) Jassa slatteryi* 87.94 47.87 - 26.54 (>C3) 26.74 (>C3) 16.15 (>C15) Paracerceis sculpta* - - - 14.58 (>C3) 8.51 (>C3) - Phtisica marina* - 14.93 - 9.80 (>C15) - 11.08 (>C15) Average similarity Average dissimilariy C3 C15 LB (68.42%) C3 vs C15 C3 vs LB C15 vs LB (64.72 %) (73.77%) (35.18%) (39.49%) (35.02%) FUNCTIONAL TRAITS Contribution to similiarity (%) Contribution to dissimilarity (%) LDD by ship fouling 18.90 17.81 - - - 8.11 (>C15) LDD absent - - - - 9.29 (>LB) 9.21 (>LB) Swimming 16.48 14.70 19.13 - - - Tube dwelling - 12.32 - 10.44 (>C3) 12.71 (>C3) 12.09 (>C15) Free living - - 15.49 8.18 (>C15) 13.52 (>LB) 13.09 (>LB) Omivorous 18.65 13.44 - 9.50 (>C3) 8.46 (>C3) - Laterally flatted - - 16.49 10.76 (>C3) 8.84 (>LB) 8.99 (>LB) Slender / elongated - - - 6.92 (>C15) - - Dorsally fltatted - - - 6.69 (>C15) - - 35 Table S1 36 Table S2 37